Sign Up
Log In
Log In
or
Sign Up
Places
All Projects
Status Monitor
Collapse sidebar
SUSE:SLE-15-SP2:GA
pacemaker.16889
pacemaker.changes
Overview
Repositories
Revisions
Requests
Users
Attributes
Meta
File pacemaker.changes of Package pacemaker.16889
------------------------------------------------------------------- Tue Oct 20 09:50:28 UTC 2020 - Yan Gao <ygao@suse.com> - fencer: restrict certain IPC requests to privileged users (CVE-2020-25654, bsc#1177916) - pacemakerd: ignore shutdown requests from unprivileged users (CVE-2020-25654, bsc#1177916) - executor: restrict certain IPC requests to Pacemaker daemons (CVE-2020-25654, bsc#1177916) * bsc#1177916-CVE-2020-25654_pacemaker.patch ------------------------------------------------------------------- Tue Sep 08 11:10:39 UTC 2020 - Yan Gao <ygao@suse.com> - extra: quote shell variables in agent code where appropriate (bsc#1175557) * bsc#1175557-0006-Low-extra-quote-shell-variables-in-agent-code-where-.patch - extra: use ":=" where appropriate in agent code * bsc#1175557-0005-Low-extra-use-where-appropriate-in-agent-code.patch - extra: add vim modelines to agents * bsc#1175557-0004-Low-extra-add-vim-modelines-to-agents.patch - extra: remove trailing whitespace from agent code * bsc#1175557-0003-Refactor-extra-remove-trailing-whitespace-from-agent.patch - extra: use 4-space indents in resource agent code * bsc#1175557-0002-Refactor-extra-use-4-space-indents-in-resource-agent.patch - extra: update agent boilerplate (copyright/license notices) * bsc#1175557-0001-Doc-extra-update-agent-boilerplate-copyright-license.patch ------------------------------------------------------------------- Wed Aug 12 13:45:41 UTC 2020 - Yan Gao <ygao@suse.com> - rpm: use the user/group ID 90 for haclient/hacluster to be consistent with cluster-glue (bsc#1167171) ------------------------------------------------------------------- Tue Jul 7 09:47:46 UTC 2020 - Callum Farmer <callumjfarmer13@gmail.com> - Fixes for %_libexecdir changing to /usr/libexec ------------------------------------------------------------------- Thu Jun 18 10:54:09 UTC 2020 - Yan Gao <ygao@suse.com> - rpm: add spec option for enabling CIB secrets ------------------------------------------------------------------- Tue Jun 02 07:42:10 UTC 2020 - Yan Gao <ygao@suse.com> - rpm: put user-configurable items at top of spec ------------------------------------------------------------------- Wed May 20 12:38:55 UTC 2020 - Yan Gao <ygao@suse.com> - move bcond_with/without up front for e.g. pcmk_release ------------------------------------------------------------------ Mon May 11 15:07:22 UTC 2020 - Yan Gao <ygao@suse.com> - fencer: avoid possible use-of-NULL when parsing metadata (bsc#1171372) * bsc#1171372-0011-Fix-fencer-avoid-possible-use-of-NULL-when-parsing-m.patch - libstonithd: make the assert message from stonith__device_parameter_flags() more clear (bsc#1171372) * bsc#1171372-0010-Refactor-libstonithd-make-the-assert-message-from-st.patch ------------------------------------------------------------------- Fri May 08 07:32:19 UTC 2020 - Yan Gao <ygao@suse.com> - libstonithd: respect `pcmk_host_argument=none` on `validate` (bsc#1171372) * bsc#1171372-0009-Fix-libstonithd-respect-pcmk_host_argument-none-on-v.patch - libstonithd: add `port` or `plug` parameter according to metadata on `validate` if no `pcmk_host_argument` specified (bsc#1171372) * bsc#1171372-0008-Fix-libstonithd-add-port-or-plug-parameter-according.patch - Pacemaker Explained: update the default value and the description of `pcmk_host_argument` (bsc#1171372) * bsc#1171372-0007-Doc-Pacemaker-Explained-update-the-default-value-and.patch - fencer: add `port` or `plug` parameter according to metadata for RHCS-style fence-agents (bsc#1171372) * bsc#1171372-0006-Fix-fencer-add-port-or-plug-parameter-according-to-m.patch - libstonithd: add function to check supported parameters according to the metadata of a fence agent (bsc#1171372) * bsc#1171372-0005-Refactor-libstonithd-add-function-to-check-supported.patch ------------------------------------------------------------------- Mon May 04 17:11:07 UTC 2020 - Yan Gao <ygao@suse.com> - stonith_admin: --delay is an optional option for --fence/--unfence/--reboot commands (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0017-Doc-stonith_admin-delay-is-an-optional-option-for-fe.patch - controller: avoid possible use-of-NULL on logging fencing message (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0016-Fix-controller-avoid-possible-use-of-NULL-on-logging.patch ------------------------------------------------------------------- Mon May 04 17:03:40 UTC 2020 - Yan Gao <ygao@suse.com> - cts: add `plug` parameter for fence_dummy agent (bsc#1171372) * bsc#1171372-0004-Test-cts-add-plug-parameter-for-fence_dummy-agent.patch - libstonithd: `plug` parameter of RHCS-style fence-agents is shown as non-required in the metadata (bsc#1171372) * bsc#1171372-0003-Fix-libstonithd-plug-parameter-of-RHCS-style-fence-a.patch - libstonithd: `action` parameter of RHCS-style fence-agents is shown as non-required in the metadata (bsc#1171372) * bsc#1171372-0002-Fix-libstonithd-action-parameter-of-RHCS-style-fence.patch - libstonithd: functionize fudging metadata of RHCS-style agent to make specific parameter non-required (bsc#1171372) * bsc#1171372-0001-Refactor-libstonithd-functionize-fudging-metadata-of.patch ------------------------------------------------------------------- Thu Apr 16 11:46:28 UTC 2020 - Yan Gao <ygao@suse.com> - fencer: update cpg_topology_delay test to also verify pcmk_delay_base is added (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0015-Test-fencer-update-cpg_topology_delay-test-to-also-v.patch - controller: requested priority fencing delay defaults to 0 (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0014-Feature-controller-requested-priority-fencing-delay-.patch - stonith_admin: --delay option defaults to 0 (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0013-Feature-stonith_admin-delay-option-defaults-to-0.patch - fencer: any delays from pcmk_delay_base/max are added to requested fencing delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0012-Feature-fencer-any-delays-from-pcmk_delay_base-max-a.patch - scheduler: do not differentiate the case where all the nodes have equal priority (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0011-Feature-scheduler-do-not-differentiate-the-case-wher.patch - scheduler: priority-fencing-delay defaults to 0 meaning disabled (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0010-Feature-scheduler-priority-fencing-delay-defaults-to.patch ------------------------------------------------------------------- Thu Apr 16 11:19:36 UTC 2020 - Yan Gao <ygao@suse.com> - Pacemaker Explained: document priority-fencing-delay cluster option (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0009-Doc-Pacemaker-Explained-document-priority-fencing-de.patch - fencer: add cpg_topology_delay test to verify enforced fencing delay with fencing topology (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0008-Test-fencer-add-cpg_topology_delay-test-to-verify-en.patch - fencer: handle any enforced fencing delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0007-Feature-fencer-handle-any-enforced-fencing-delay.patch - stonith_admin: add --delay option to support enforced fencing delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0006-Feature-stonith_admin-add-delay-option-to-support-en.patch - controller: request fencing with any enforced priority fencing delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0005-Feature-controller-request-fencing-with-any-enforced.patch - libstonithd: introduce fence_with_delay() operation (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0004-Feature-libstonithd-introduce-fence_with_delay-opera.patch - scheduler: add regression test for priority-fencing-delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0003-Test-scheduler-add-regression-test-for-priority-fenc.patch - scheduler: implement priority-fencing-delay (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0002-Feature-scheduler-implement-priority-fencing-delay.patch - scheduler: add priority-fencing-delay cluster option (jsc#ECO-1611, jsc#SLE-12237) * jsc#ECO-1611-0001-Feature-scheduler-add-priority-fencing-delay-cluster.patch ------------------------------------------------------------------- Thu Apr 09 11:48:23 UTC 2020 - Yan Gao <ygao@suse.com> - libcrmcluster: use uint64_t type for corosync ringid (membership id) when updating node state (bsc#1168771) * bsc#1168771-0001-Fix-libcrmcluster-use-uint64_t-type-for-corosync-rin.patch ------------------------------------------------------------------- Tue Jan 28 16:09:11 UTC 2020 - Yan Gao <ygao@suse.com> - get ready for implicit -fno-common with upcoming GCC 10 (bsc#1160410) * bsc#1160410-0001-Build-get-ready-for-implicit-fno-common-with-upcomin.patch ------------------------------------------------------------------- Fri Jan 10 14:18:35 UTC 2020 - Yan Gao <ygao@suse.com> - attrd: properly declare global variables as extern in header (bsc#1160410) * bsc#1160410-0001-Refactor-attrd-properly-declare-global-variables-as-.patch ------------------------------------------------------------------- Thu Dec 05 12:49:59 UTC 2019 - Yan Gao <ygao@suse.com> - scheduler: make sure cluster-wide maintenance-mode=true overrides per-resource settings (bsc#1154881) * bsc#1154881-0001-Fix-scheduler-make-sure-cluster-wide-maintenance-mod.patch ------------------------------------------------------------------- Wed Oct 16 12:34:13 UTC 2019 - Yan Gao <ygao@suse.com> - cts: ComponentFail: killed corosync doesn't respawn immediatelly. * 0001-Test-cts-ComponentFail-killed-corosync-doesn-t-respa.patch - fencer: improved log messages regarding fencing actions (bsc#1151007) * bsc#1151007-0005-Log-fencer-improved-log-messages-regarding-fencing-a.patch - fencer: no need to check the length of a non-empty list for pending fencing actions (bsc#1151007) * bsc#1151007-0004-Refactor-fencer-no-need-to-check-the-length-of-a-non.patch - fencer: Indicate fencing target in the logs when scheduling and executing fencing command (bsc#1151007) * bsc#1151007-0003-Log-fencer-Indicate-fencing-target-in-the-logs-when-.patch - fencer: Make sure concurrent fencing commands get triggered to execute (bsc#1151007) * bsc#1151007-0002-Fix-fencer-Make-sure-concurrent-fencing-commands-get.patch - fencer: Don't let a fencing command with a long delay block any other commands pending on the device (bsc#1151007) * bsc#1151007-0001-Fix-fencer-Don-t-let-a-fencing-command-with-a-long-d.patch ------------------------------------------------------------------- Mon Jul 15 09:35:42 UTC 2019 - Yan Gao <ygao@suse.com> - scheduler: wait for probe actions to complete to prevent unnecessary restart/re-promote of dependent resources (bsc#1130122, bsc#1032511) * bsc#1130122-0001-Fix-scheduler-wait-for-probe-actions-to-complete-to-.patch * bsc#1130122-0002-Test-scheduler-wait-for-probe-actions-to-complete-to.patch ------------------------------------------------------------------- Fri Jun 07 10:39:45 UTC 2019 - Yan Gao <ygao@suse.com> - stonith_admin --help: specify the usage of --cleanup (bsc#1135317) * bsc#1135317-0001-Low-stonith_admin-help-specify-the-usage-of-cleanup.patch - controller: confirm cancel of failed monitors (bsc#1133866) * bsc#1133866-0001-Fix-controller-confirm-cancel-of-failed-monitors.patch - libcrmcommon: return error when applying XML diffs containing unknown operations (bsc#1127716) * bsc#1127716-0003-Fix-libcrmcommon-return-error-when-applying-XML-diff.patch - libcrmcommon: avoid possible use-of-NULL when applying XML diffs (bsc#1127716) * bsc#1127716-0002-Fix-libcrmcommon-avoid-possible-use-of-NULL-when-app.patch - libcrmcommon: correctly apply XML diffs with multiple move/create changes (bsc#1127716) * bsc#1127716-0001-Fix-libcrmcommon-correctly-apply-XML-diffs-with-mult.patch ------------------------------------------------------------------- Wed May 01 07:22:56 UTC 2019 - Yan Gao <ygao@suse.com> - controller: avoid memory leak when duplicate monitor is scheduled * 0001-Low-controller-avoid-memory-leak-when-duplicate-moni.patch - libcrmcommon: return proper code if testing pid is denied (bsc#1131353, bsc#1131356) * bsc#1131353-bsc#1131356-0001-Low-libcrmcommon-return-proper-code-if-testing-pid-i.patch ------------------------------------------------------------------- Tue Apr 30 13:17:22 UTC 2019 - Yan Gao <ygao@suse.com> - libcrmcommon: avoid use-of-NULL when checking whether process is active (bsc#1131353, bsc#1131356) * bsc#1131353-bsc#1131356-0001-Fix-libcrmcommon-avoid-use-of-NULL-when-checking-whe.patch ------------------------------------------------------------------- Thu Apr 18 12:07:35 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190417.13d370ca9: - controld: fix possible NULL pointer dereference (bsc#1131353, bsc#1131356) - pacemakerd vs. IPC/procfs confused deputy authenticity issue (CVE-2018-16877, bsc#1131356) (CVE-2018-16878, bsc#1131353) - libservices: fix use-after-free wrt. alert handling (CVE-2019-3885, bsc#1131357) - executor: systemd starts should return UNKNOWN_ERROR instead of NOT_RUNNING - pacemaker-remoted: improve proxy accept failure message - executor: consider stonith resource stopped only if stop succeeded - executor: improve stonith operation rc and status mapping - executor: reschedule recurring stonith monitor if not connected - executor: put recurring stonith failsafe stop in correct place ------------------------------------------------------------------- Thu Apr 11 15:32:23 UTC 2019 - Yan Gao <ygao@suse.com> - Rebase: bug-728579_pacemaker-stonith-dev-id.patch - Revert "use common service interface for fence-agents and RAs" (bsc#1132123) * 0002-Revert-use-common-service-interface-for-fence-agents.patch - Revert "service-lib: avoid call-pattern leading to use-after-free" * 0001-Revert-Fix-service-lib-avoid-call-pattern-leading-to.patch ------------------------------------------------------------------- Tue Apr 9 10:00:32 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190408.1b68da8e8: - scheduler: avoid error log in harmless situation - libcrmcommon: use INT_MIN/INT_MAX instead of -1 for out-of-range integers - service-lib: avoid call-pattern leading to use-after-free - libp-i: Renamed to libpacemaker. ------------------------------------------------------------------- Thu Apr 4 11:13:13 UTC 2019 - Jan Engelhardt <jengelh@inai.de> - Drop %if..%endif guards around %packages which does not affect the build result. ------------------------------------------------------------------- Thu Apr 4 09:09:49 UTC 2019 - Yan Gao <ygao@suse.com> - spec: no need to explicitly require the virtual dependency on libqb ------------------------------------------------------------------- Thu Apr 4 09:05:55 UTC 2019 - Yan Gao <ygao@suse.com> - Rebase: * bug-728579_pacemaker-stonith-dev-id.patch - Update to version 2.0.1+20190402.e091f4f0c: - scheduler: one group stop shouldn't make another required - libcrmcommon: compare_version: drop superfluous heap allocation - libfencing: enumerate installed RHCS-style agents alpha-sorted - xml diff'ing: prevent 'id' attribute reordering in "diff-removed" - libcrmcommon: downplay "Creating RNG parser context" as debug prio - controller: improve failed recurring action messages - controller: improve lost action logs - use common service interface for fence-agents and RAs - libcrmcommon: allow whitespace in name/value pairs - libcrmcommon: pcmk_nvpair_t should handle NULL values ------------------------------------------------------------------- Wed Mar 13 11:19:11 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190312.059e2e26b: - spec: move stonith_admin to -cli where it belongs - tools: Add crm_rule. ------------------------------------------------------------------- Tue Mar 12 10:41:19 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190311.e91ee92e5: - libfencing: Change return type on stonith_agent_exists. - scheduler: cl#5301 - respect order constraints when relevant resources are being probed (bsc#1117934, bsc#1128374) ------------------------------------------------------------------- Wed Mar 6 16:34:53 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190304.1ac166cd3: - build: Update pkgconfig files for libpe. - libp-i: Add a pkgconfig file for libp-i. - libs: Rename libpengine to libpacemaker-internal. - tools: Add new validate options to crm_resource. - Add HealthIOWait provider which checks hosts's IOWait - tools: crm_resource --clear should print out what it does ------------------------------------------------------------------- Tue Mar 5 13:49:11 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.1+20190304.9e909a5bd (Pacemaker-2.0.1): - libcrmcommon: complete interrupted live migrations correctly ------------------------------------------------------------------- Thu Feb 28 11:38:50 UTC 2019 - Yan Gao <ygao@suse.com> - Update to version 2.0.0+20190225.22ee9a769: - scheduler: skip all native constraints for unmanaged resources - scheduler: sort nodes when creating constraints ------------------------------------------------------------------- Thu Jan 31 08:07:34 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190130.57cc9c14b: - CTS: really don't require nodes to be specified if listing tests (bsc#1114840) - controller: really avoid closing attrd IPC for temporary failures - controller: clear election dampening when DC is lost ------------------------------------------------------------------- Mon Jan 28 10:24:10 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190125.788ee2c49: - attrd: clear election dampening when the writer leaves - libcrmcluster: *really* write only one election storm black box - attrd: log previous writer *before* clearing it ------------------------------------------------------------------- Fri Jan 25 14:36:54 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190124.c3c9c6e66: - scheduler: don't order non-DC shutdowns before DC fencing - scheduler: improve fence action ordering - scheduler: don't disable waiting for DC fencing - libcrmservice: cancel DBus call when cancelling systemd/upstart actions - libcrmservice: null-terminate string *before* printing it - libcrmservice: improve logs when cleaning up an operation - spec: add hint about sbd compatibility ------------------------------------------------------------------- Thu Jan 24 11:24:16 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190122.3f993c74f: - crm_mon: remove duplicity of fence-action-state in xml-output - cts: temporarily disable any enabled cluster serivces when running remote tests - cts: service counts as enabled only if it's explicitly enabled ------------------------------------------------------------------- Mon Jan 14 09:52:35 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190111.750c5bd01: - cts-exec: still run the tests for the other resource classes even without python systemd bindings (bsc#1121808) ------------------------------------------------------------------- Fri Jan 11 15:57:23 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190111.f07d766f3: - controller: directly acknowledge unrecordable operation results - resources: clean serialized file on SIGTERM in Dummy - libcrmcommon: downgrade empty output logging to trace level ------------------------------------------------------------------- Fri Jan 11 10:11:05 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190110.7bbbeb6f6: - tools: Clear all prefer constraints when performing a move (bsc#1155290) - tools: Fix moving a resource with a lifetime constraint (bsc#1155290) - attrd: check for alert changes after CIB is replaced - attrd: start new election if writer is lost - attrd: handle shutdown more cleanly (bsc#1173668) - attrd: don't start a new election when receiving a client update - attrd: clear lost updates - attrd: don't delay re-attempted writes unless original failed - attrd: connect to the CIB before connecting the cluster ------------------------------------------------------------------- Tue Jan 8 14:38:38 UTC 2019 - ygao@suse.com - Update to version 2.0.0+20190103.5b36d4e10: - scheduler: downgrade clone pre-allocation message - libpe_status: downgrade remote node fence-before-clear message - libpe_status: improve trace messages when finding actions - libpe_status: avoid use-after-free when logging at trace level - libpe_status: avoid double free of stop_needed list - tools: stonith_admin -I doesn't require an agent - pacemakerd,libcrmcommon: don't touch kernel.sysrq - fencing: use next device's name where intended - tools: Implement crm_resource --expired. - CTS: ignore monitor failures when testing remote node fencing - CTS: lower remote connection failure detection time - attrd: Ignoring delayed updating of attributes when integrating disjointed clusters without stonith. - pacemakerd: only one active process check needed - cts: use bash in the pacemaker-cts-dummyd - cts: clear constraints on cluster nodes (bsc#1121272) - cts: resume any possibly frozen pacemaker_remoted when cleaning up the test (bsc#1121272) - cts: simulate failure of pacemaker_remoted by freezing it with SIGSTOP (bsc#1121272) - scheduler: order resource moves after connection starts - scheduler: get rid of now-unused stonith_complete pseudo-action - scheduler: get rid of now-unused all_stopped pseudo-action - scheduler: order start after particular stonith op - controller: be more tolerant of malformed executor events - scheduler: don't send clone notifications to a stopped remote node - tools: avoid memory leak at failure exit in crm_resource --restart - controller: improve debug messages for remote connection events - CTS: don't require nodes to be specified if only listing tests (bsc#1114840) - tools: Remove extra newlines from crm_resource output. - tools: Change crm_resource output for attributes. - cts: set default number of iterations to 1 - crm_report: Add sanitize drbd to remove values of sensitive attributes. - crm_report: remove deprecated drbd-overview and add drbd configuration - scheduler: avoid container stop on wrong node - scheduler: check bundle remote connection actions properly - scheduler: downgrade addr replacement message - scheduler: don't trace implicit resource XML - spec: split -schemas standalone noarch package - pacemaker-fenced: fix easily plausible NULL pointer dereference - tools: crm_simulate -P should return consistent exit code - tools: end crm_ticket error messages with newline - libcrmcommon: make blackbox logger compatible with newer libqb - tools: make crm_verify return reliable exit codes - libpe_status: add constructor/destructor for pe_working_set_t ------------------------------------------------------------------- Fri Nov 9 12:13:25 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20181108.62ffcafbc: - fenced: trigger device-update on resource-defaults changed - fenced: adhere meta-attributes when triggering device-update ------------------------------------------------------------------- Thu Nov 8 10:34:21 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20181105.fb362c0c9: - scheduler: formally deprecate score in rsc_order - pacemaker-remote: skip remote_config_check for guest-nodes ------------------------------------------------------------------- Fri Nov 2 16:05:37 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20181101.a21fd0099: - rhbz#1644076 - Ensure the bundle stop/demote notifications are directed to the correct host - libcrmservice: override removal failure is not a problem - libcrmcommon,execd: pass remaining timeout to drain function - tools: log user-specified time correctly in crm_simulate - libcrmcluster: log microseconds sensibly - libcrmservice: order systemd resources relative to pacemaker_remote - scheduler: order guest pseudo-fencing properly after clean-up - scheduler: avoid unnecessary recovery of cleaned guest nodes - scheduler: improve bundle address fixing messages - attrd: record tentative writer only if election not already lost - attrd: skip election check if peer election op had error - attrd: disregard lost nodes' election votes - attrd: improve election messages - attrd: initialize election before broadcasting protocol - controller: remove node's vote as soon as it is inactive - controller: don't register DC input twice - controller: don't stop election timer for nonexistent election - controller: improve election messages - libcrmcluster: don't write more than one election storm black box - libcrmcluster: only count no-vote if an election is in progress - libcrmcluster: validate election cluster messages better - libcrmcluster: improve election log messages ------------------------------------------------------------------- Tue Oct 16 07:40:06 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20181012.dc40c8037: - schedulerd: Improve internal bundle ordering - execd: Allow the SBD check to be explicitly disabled for bundles - fenced: Handle fencing requested with nodeid by utilizing the membership cache of known nodes (bsc#1094208, bsc#1107270, bsc#974108) * Drop merged bsc-1094208-Refactor-fenced-Handle-fencing-requested-with-nodeid.patch - controld: able to manually confirm unseen nodes are down (bsc#1094208, bsc#1107270) * Drop merged bsc-1094208-Fix-controld-able-to-manually-confirm-unseen-nodes-a.patch ------------------------------------------------------------------- Fri Oct 5 12:57:16 UTC 2018 - ygao@suse.com - fenced: Handle fencing requested with nodeid by utilizing the membership cache of known nodes (bsc#1094208, bsc#1107270, bsc#974108) * bsc-1094208-Refactor-fenced-Handle-fencing-requested-with-nodeid.patch - controld: able to manually confirm unseen nodes are down (bsc#1094208, bsc#1107270) * bsc-1094208-Fix-controld-able-to-manually-confirm-unseen-nodes-a.patch ------------------------------------------------------------------- Thu Oct 4 09:35:38 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20180927.b67d8d0de: - logrotate: set a maximum size for logs - tools: ensure crm_resource --force-* commands get stderr messages - libcrmcommon: properly check whether resource supports parameters - tools: "return" from crm_mon after calling functions that don't - alerts: send all MIB OIDs with all SNMP alerts - resource-agents: add "s"-suffix where missing in metadata - libcommon: do not write to /proc/sys/kernel/sysrq when unneeded - pacemaker-based: drop declared, errant option never backed in tree - crm_mon: don't exit directly from cib_connect on error - scheduler: honor asymmetric orderings even when restarting ------------------------------------------------------------------- Thu Sep 13 09:59:39 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20180911.b7803d27c: - libcrmcommon: allow configurable bounds for DH prime length - CIB,executor: improve remote server-side messages - CIB: give more detail when TLS handshake fails - libcrmcommon: get DH prime bit length from GnuTLS API - libcrmcommon: improve TLS session initialization failure handling - CIB: raise exit escalation timer to 10s - cts,tools: make static analysis happy - crm_mon: no sense in having --as-xml in daemon-mode - attrd: Adding the ignore_delay parameter to reduce the number of scans of the attribute hash table. - crm_mon: rhbz#1625231 - disable fence-history in file-mode ------------------------------------------------------------------- Thu Sep 6 12:43:10 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20180831.e41c7d6d6: - XML: schemas.c: ensure proper upgrade-enter fallback & guard strdup arg - controller: trigger join after both peer and join notifications - controller: handle shutdown completion only when needed - libcrmcluster: reset "when lost" when a peer becomes a member - libcrmcluster: improve CPG membership messages - controller: improve membership messages - stonith-history: keep history recorded within boundaries - tools: crm_resource -C should wait for all replies (bsc#1140519) - libcrmcommon: explicitly include stdint.h in util.h (bsc#1108303) - attrd: improve log messages - attrd: delay re-attempted writes - crm_mon: in case of html-output avoid missleading output - tools: make crm_resource messages tool-agnostic - scheduler: show promotable clones as "(promotable)" rather than "Master/Slave" - tools: make crm_resource --lifetime specify rule with local time zone - crm_mon: rhbz1419548: show standby-node with active resources - crm_mon: rhbz#1486869 - common language on connection-errors - avoid possible printf vs. scanf format specifiers disrepancy - also utilize configured format specifiers serving uint32_t - configure: fix overkill format specifier serving uint64_t check - controller: shut down DC if unable to update node attributes - libcrmcommon: assume unprivileged user if can't get user info - libcrmcommon,libcib: improve messages when unable to get user info - Initial podman support - scheduler: sort node names with leading zeroes - scheduler: add regression test for anonymous instance numbering - fenced: g_hash_table_insert has no returnvalue till glib 2.40 - scheduler: sort node names numerically and case-insensitively - scheduler: Act on clone instances in numerical order - scheduler: assign anonymous clone instances properly - scheduler: print pending node if no active node - scheduler: print clone instances individually when action is pending - stonith-API: free history reply in client - controller: avoid error message if node state not known in peer callback - tools: document crm_master -r option - stonith_admin: no sense in calling without action option - fence-API: empty history-query-reply is OK without history-tag - scheduler: reword messages when finding anonymous clone instances - scheduler: make inc6 regression test input more sane - tools: simulation should use same history name as live cluster - scheduler: avoid unnecessary probe loop for anonymous clones - scheduler: improve efficiency of checking for clone instance on node - scheduler: never probe more than a single instance for anonymous clones - crm_mon: default to showing pending fence actions - tools: already latest schema is not failure for cibadmin --upgrade - pacemaker-based: inform originator of CIB upgrade failure - controller: set exit-reason when faking failure - crm_mon: default to showing fence-failures - fence-history: automatically sync history - crm_mon: use new fence-history notifications - fence-history: add cleanup & sync & history-notify - RPM: don't package servicelog-related binaries - agents: remove references to corosync plugin daemons from controld meta-data - controller: log variables *after* setting them - cts-scheduler: add test for forced anonymous clone - scheduler: clear globally-unique correctly when forcing anonymous - scheduler: respect lower clone-max for forced anonymous clones - scheduler: reword force-anonymous message - scheduler: only some agent standards support unique clones - controld: No update by already confirmed events. - Present exec_time in event environment ------------------------------------------------------------------- Fri Jul 27 14:19:11 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20180726.3d81c89b8: - Shutdown corosync after a fatal error - tools: ipmiservicelogd FTBFS on ppc64le - tools: notifyServicelogEvent FTBFS on ppc64le - tools: detect CIB directory correctly - tools: avoid memory leak when crm_resource checks resource - controller: avoid any chance of use-of-NULL when doing LRM op - resources: controld doesn't need to check for corosync 1 plugin - resources: use correct variable name in Stateful - cts-scheduler: restore ability to pass arguments to single test run - resources: use ocf_is_true in SysInfo - spec: drop superfluous dependency of -libs-devel on -cts - tools: update URL in crm_simulate help - crm_mon: rhbz#1461964 - add fence history features (bsc#968055) ------------------------------------------------------------------- Fri Jul 27 11:23:32 UTC 2018 - ygao@suse.com - Update to version 2.0.0+20180706.8cf3fe749 (Pacemaker-2.0.0): - tools: Multiple -V increase debug level (#1526) - scheduler: clarify multiple-anonymous-instances message - scheduler: avoid use-after-free when on-fail=block - controller: add node name to meta-data call environment - controller: filter certain synthesized failures from CIB - controller: filter certain action timeouts from CIB - controller: don't record pending notifies in CIB - executor: cancel recurring monitors if fence device registration is lost - controller: avoid memory leak if start-up fails - XML: schemas.c: fix possible memleak + less magic with numeric constants - tools: stonith_admin -h/-H options are queries - libstonithd: avoid memory leak when getting meta-data - XML: resources-3.1.rng: relaxed schema after tight upgrade enforcement - XML: schemas.c: take severities of XSLT messages into account properly - XML: assets/upgrade-2.10-htmldiff.xsl: express message severity visually - XML: upgrade-2.10{,-leave}.rng: distinguish message severity - XML: upgrade-2.10-roundtrip.xsl et al.: new tests incl. browser view - XML: upgrade-2.10.xsl: simplify thanks to diehard assumptions wrt. input - XML: upgrade-2.10-leave.xsl: post-upgrade output restoring stylesheet - XML: upgrade-2.10-enter.xsl: pre-upgrade input normalizing stylesheet - XML: schemas.c: implement & employ full round-trip apply_upgrade - XML: schemas.c: switch to using add_schema_by_version - XML: schemas.c: prepare the ground for unfold-transform-refold logic - XML: schemas.c: crm_schema_init: simplify for/break, call add_schema 1x - XML: schemas.c: shortcut upgrade transformation when not needed - st_client, stonith_admin: cleanup memory of stonith-history - XML: cib_ops.c: make use of cib_verbose flag for RNG/XSLT handling - XML: xml_internal.h: accept changing priority over particular chunks - CIB: no schema violating oversimplified patching with @id-ref - schemas.c: cleanup XSLT extension modules' data and whatnot - spec: move crm_node to -cli package - attrd: support removing node by ID - tools: query controller for crm_node -n - controller: add CRM_OP_NODE_INFO operation - xml.c: fix cleaning of own vs. XSLT's XML node's _private load - xml.c: properly "staticize" or drop eligible functions - scheduler: print to stdout when appropriate - tools: treat INFINITY correctly in crm_failcount - libcrmservice: support mount, path, and timer systemd unit types - libcrmcluster: avoid use-of-NULL when searching for remote node (bsc#1128772) - spec: add procps-ng & psmisc dependencies as suitable - libcrmcommon: lower maximum IPC flush delay to 1.5 seconds - CTS: fix python3 usage - libstonithd: show error if agent not found for meta-data call - fencing: add stonith_admin --validate option - libstonithd: add client API for device validation - based_callbacks.c: restore -Werror buildability, avoid truncation - tools: handle stonith_admin exit codes better - tools: stonith_admin should print error if stonithd connection fails - tools: improve stonith meta-data error messages ------------------------------------------------------------------- Mon Jun 4 12:36:13 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180531.63ff11d35: - scheduler: ensure orphaned recurring monitors have interval set - scheduler: improve action creation messages - XML: regression.sh + upgrade-2.10.xsl + test-2: browser-based test drive - XML: upgrade-2.10.xsl: use namespacing and/or modes for easier mash-ups - XML: regression.sh + upgrade-detail.xsl: fix minor cosmetic issues - scheduler: "requires" shouldn't apply to guest node fencing - libpe_status: handle "requires" properly for anonymous clones - scheduler: avoid unnecessary delay of start actions - scheduler: order implicit actions more strictly - all: prefer appropriate node when multiply active - libpe_status: find active instances properly according to requires - libpe_status: handle pending migrations correctly - tools: correctly check crm_resource --move for master role - scheduler: update wiki URL in log message - libcrmcommon: use proper IPC buffer size when reading - CIB manager: improve notification failure messages - liblrmd: improve TLS key reading - XML: upgrade-2.10.xsl: rsc op instance_attrs [2/2]: remaining meta attrs - XML: upgrade-2.10.xsl: rsc op instance_attrs [1/2]: meta-like requires - XML: upgrade-2.10.xsl: resource meta_attributes: in-situ exchange - XML: upgrade-2.10.xsl: simplify "should diagnostics be emitted" decision - XML: upgrade-2.10.xsl: related commits in "./regression.sh -X" output - XML: regression.sh + upgrade-2.10.xsl: remove commented out leftovers - XML: resources-3.0.rng: omit "isolation*" resource meta_attributes - XML: upgrade-2.10.xsl: normalize indentation of new rsc meta attributes - executor: give more detail when TLS handshake fails - fenced: be more paranoid about freeing memory - cts: use python2/3-compatible comparison ------------------------------------------------------------------- Fri May 18 18:44:31 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180517.adf4d77b8: - spec: unify quoting in {pre,post}* scriptlets' action soliciting - spec: add a simple retention of pre-existing log move targets - libcrmcommon,libcrmcluster,liblrmd: recognize daemons under new names - XML: upgrade-2.10.xsl: resource meta_attributes: tweak diagnostic msg - XML: upgrade-2.10.xsl: re-apply on scheduler CTS test cases - XML: upgrade-2.10.xsl: resource meta_attributes: validity conformance ------------------------------------------------------------------- Mon May 14 08:54:18 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180511.acd8ff5a7: - execd: avoid memory leak when testing remote key - cts-scheduler: avoid unsupported usage of requires - controld: avoid memory leak when synthesizing failure - XML: upgrade-2.10.xsl: resource meta_attributes: isolation* items - XML: upgrade-2.10.xsl: make diagnostics more direct wrt. context - XML: upgrade-2.10.xsl: remaining "marking" space with op meta_attributes - XML: upgrade-2.10.xsl: simplify 2/2: HelperIdentity, no INNER-RECURSION - XML: upgrade-2.10.xsl: simplify 1/2: resource instance_attributes - XML: regression.sh: avoid doubts when tests expected at stdin with '-' - XML: regression.sh: modularize "XSLT processor", allow substitutes - tools: improve error messages from crm_resource --move - tools: use output redirection correctly in crm_standby - tools: handle multiple values properly in crm_attribute - tools: bash portability issue in crm_failcount - libcrmservice: don't consider a cancelled recurring operation as failed - tools: cibsecret checks pacemaker-controld instead of crmd - tools: cibsecret --help/--version doesn't require cluster to be running (bsc#1102915) - spec: package cts-exec-helper with main package instead of -cli - spec: move old logs after upgrade - scheduler: improve failed op message - libcrmservice: find absolute paths when used with "service:" - libcrmservice: separate LSB-specific code into own source files - execd: handle systemd actions correctly when used with "service:" ------------------------------------------------------------------- Wed May 9 10:48:06 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180504.70fe3d6e9: - all: take advantage of new libqb features - contoller,scheduler: guard hash table deletes (bsc#1136712) - controller: don't abort after delay if we're no longer DC - libcrmcommon,pacemakerd: kernel task name is at most 15 characters - cts: install dummy systemd daemon via helper script - cts: externalize log watcher script - Maint: make systemctl react to pacemaker-remote as with underscored one - make some old daemon references traceable with "man -K OLDNAME" - io.c: restore -Werror buildability, put conversion specifier last ------------------------------------------------------------------- Tue May 8 14:28:49 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180430.c0d7ba798: - libcrmcommon: avoid memory leak on failed IPC send - daemons: add configure option to create links at old daemon names - controld: rename "crmd-*" cluster options - all: change log file message format - daemons: improve connection loss messages - libcrmcommon: improve connection loss message - pacemaker_remote: correct documentation URL in systemd unit file - libcrmcommon: check file/directory writability better - spec: no more buildroot XML to modify - cib: rename cib daemon to pacemaker-based - cts-stonithd: rename cts-stonithd to cts-fencing - fenced: rename stonith-test to cts-fence-helper - fencing: rename stonithd to pacemaker-fenced - rename cts-pengine to cts-scheduler - pengine: rename pengine daemon to pacemaker-schedulerd - crmd: rename crmd daemon to pacemaker-controld - pacemaker-execd: rename pacemaker_remoted to pacemaker-remoted - rename cts-lrmd to cts-exec - pacemaker-execd: rename lrmd_test to cts-exec-helper - lrmd: rename lrmd daemon to pacemaker-execd - attrd: rename attrd daemon to pacemaker-attrd ------------------------------------------------------------------- Wed May 2 12:48:53 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180430.b12c320f5: * XML: upgrade-2.10.xsl: fix "mark space" issue (bsc#1091661) ------------------------------------------------------------------- Sat Apr 28 21:07:31 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180427.a34ccd926: - crmd: avoid double free after ACL rejection of resource delete - lib: Changed to lowercase comparison. - all: correct misspellings - controld: remove deprecated configdir parameter - controld: deprecate and ignore configdir parameter - controld: consistently indent with spaces instead of tabs - libcrmcommon: don't record pending notify actions as completed - crmd: improve messages when a node is unexpectedly lost - cib: improve shutdown messages - cts-regression: allow specifying to run multiple tests again (bsc#1091360) ------------------------------------------------------------------- Wed Apr 18 15:36:43 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180417.24df451b5: - alerts: Delete redundant code. - liblrmd: cast time_t to long long when printing - cts: separate dummy systemd service into separate files * Drop obsolete fate-323526_pacemaker-tests-shebang-python3.patch - crm_diff: rhbz#1561617 - Ignore attribute placement when comparing in 'cib' mode - spec: remove reference to long-gone file - spec: mark -cts & -doc packages as noarch - spec: fix malformed conditional macros for systemd-less cases - alerts: Change timestamp of Alert of snmp. ------------------------------------------------------------------- Thu Apr 12 08:30:58 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180411.2f77f4922: - libcrmcommon: free signal triggers when cleaning up mainloop - libcrmcommon: fix memory leak in schema workaround - fencing: avoid memory leaks when freeing remote operation - fencing: free dynamic memory at stonithd shutdown - crmd: delete resource from lrmd when appropriate (bsc#1117381) - Test: rhbz#1565187 - Ensure failures that cause fencing are not removed until after fencing completes - rhbz#1565187 - Ensure failures that cause fencing are not removed until after fencing completes ------------------------------------------------------------------- Mon Apr 9 09:19:28 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180406.633845120: - libcrmcommon: add a byte to pid file holder - attrd,cib,libcrmcommon: avoid %z - crmd: avoid memory leak when synthesizing operation result - tools: avoid memory leak on invalid resource - crmd: log rc code as extended information - cib: avoid file descriptor leak on temporary file creation - pengine: avoid redundant test - libcrmcommon: handle compression failures properly - libcrmcommon: improve empty string check - various: improve NULL checks - libcrmcommon: improve patch xpath searches - libcrmcommon,tools: use format strings appropriate to variables - libcrmcommon: use correct argument type for duration - libcrmcluster: avoid potential use-of-NULL - libcrmcommon,libstonithd: avoid resource leaks - tools: get sensor lun in ipmiservicelogd before using it - schema upgrade: temporary fix to avoid crash on newly present text ------------------------------------------------------------------- Fri Apr 6 15:39:19 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180406.ce623681e: - fencing: convert fence_legacy to python - fencing: fix typos in messages ------------------------------------------------------------------- Fri Apr 6 14:49:31 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180406.18fe1cc2b: - XML: upgrade-2.10.xsl: add transform for automatic upgrades to latest schema - Log: establish correct character string literals also with qb_logt sink - pengine: fix swapped warning message arguments leading to segfault (bsc#1090538) - cts-pengine: add regression test for notifications for unrunnable actions - cts-pengine: update regression tests for notification changes - rhbz#1545449 - Do not perform notifications for events we know wont be executed - bundle: Added add-host option. - liblrmd: add API function for new lrmd op - lrmd: add new lrmd op to list recurring ops - spec: require corosync >= 2.0.0 - crmd,libcrmcommon,libpe_status: guint can't be negative - xml: Allow local "kind" in resource_set within rsc_order (bsc#1085515) - Pacemaker Explained: "symmetrical" defaults to "false" for serialize orders (bsc#1085515) - pengine: "symmetrical" defaults to "false" for serialize orders (bsc#1085515) - pengine: Avoid potential use-of-NULL in unpack_simple_rsc_order() (bsc#1085515) - cts-pengine: add new regression test for interval handling - libpe_status: handle ISO8601 intervals when checking for blocking failures - pengine: properly cancel recurring monitors - pengine: improve recurring operation messages - pengine: check for duplicate intervals properly - libcrmcommon: avoid misuse of crm_get_msec() - pengine: treat reconnect_interval more consistently with action intervals - libpe_status: use default timeout rather than 0 if timeout spec is bad - libpe_status: use correct default timeout for monitors ------------------------------------------------------------------- Fri Mar 23 13:19:39 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180322.d37067b2a: - spec: Make sure shadow package is installed before adding user and group ------------------------------------------------------------------- Tue Mar 13 18:53:03 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180313.2c83c6872: - cts-pengine: add regression test for changing bundle replica count - libpe_status: handle unique bundle children correctly ------------------------------------------------------------------- Tue Mar 13 00:46:45 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180312.31c732adc: - spec: make shadow dependency explicit - spec: drop requirements for bison and flex ------------------------------------------------------------------- Sat Mar 10 18:37:13 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180309.e51687162: - crmd,liblrmd,libcrmcommon: improve remote node disconnection logs - libcrmcommon: add function to get text equivalent of client type - crmd: always write faked failures to CIB whenever possible - liblrmd: add function to create resource info structure - crmd: functionize setting fake operation status - crmd: always assert when operation can't be created - crmd: avoid core dump if remote connection doesn't exist - crmd: improve lrmd failure handling - crmd: return rich error codes from get_lrm_resource() - crmd: match only executed down events - crmd: functionize pieces of do_lrm_invoke() - crmd: don't abort for v2 diff LRM refresh if actions pending - crmd: validate CIB diffs better ------------------------------------------------------------------- Wed Mar 7 13:06:04 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180305.9cd0f6cb8: - libcrmcommon: avoid memory leak when schema transform not found - tools: warn if crm_resource --wait called in mixed-version cluster - Install /etc/pacemaker directory for storing authkey file (bsc#1082883) ------------------------------------------------------------------- Wed Feb 28 13:57:19 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180226.f788652cf: - crmd: log transition ID when aborting - crmd: always log unmatched down events at least at debug level - pengine: deprecate require-all in rsc_order (fate#324508) - libpe_status: deprecate stonith-action=poweroff (fate#324508) - tools: already latest schema is not failure for cibadmin --upgrade - libcrmcluster: prefer corosync name over ring0_addr (fate#324508) - spec: add gcc to BuildRequires - attrd: Synchronize attributes held only on own node. ------------------------------------------------------------------- Fri Feb 23 04:28:37 UTC 2018 - ygao@suse.com - Build: Ship cibsecret by pacemaker instead of cluster-glue (bsc#1082456) * Drop obsolete pacemaker-cibsecret-tool-temp-disabled.patch ------------------------------------------------------------------- Tue Feb 20 00:09:50 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180216.ae50b7f5d: - various: improve string allocation - XML: upgrade-2.10.xsl: re(drop) rsc_colocation/@score-attribute{,-mangle} - XML: upgrade-*.xsl: bump the validate-with field from within stylesheet - XML: upgrade-*.xsl: omit XML declaration - upgrade-*.xsl: don't be picky about comments, PIs, etc. ------------------------------------------------------------------- Wed Feb 14 04:30:48 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180213.965befd59: - libcrmcommon: use /tmp when creating temporary XML file - tools: crm_master should always work on node attribute - tools: crm_resource --cleanup couldn't match clone instances - tools: crm_resource --refresh should ignore --operation and --interval - crmd: quorum gain should always cause new transition - rhbz#1527072 - Correctly observe colocation constraints with bundles in the Master role - Bug rhbz#1519812 - Prevent notify actions from causing --wait to hang - libpe_status: recover after failed demote when appropriate - lrmd: always use most recent remote proxy - libpe_status: limit resource type check to primitives - PE: correct mispelled constant - PE: Ensure stop operations occur after stopped remote connections have been brought up - PE: Remote connection resources are safe to to require only quorum - PE: Only allowed nodes need to be considered when ordering resource startup after _all_ recovery - PE: Ordering bundle child stops/demotes after container fencing causes graph loops - PE: Passing boolean instead of a pointer ------------------------------------------------------------------- Sat Feb 10 06:43:16 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180209.7acb22379: - use proper way to make output files executable - spec: move lrmd_test to cli package - spec: make log directories owned by cli package - spec: move all test data into cts package - pengine: move regression test to cts as cts-pengine - lrmd: move regression test to cts as cts-lrmd - fencing: move regression test to cts as cts-stonithd - tools: move CLI regression test to cts as cts-cli - rename top-level testing files and move to cts directory - tools: drop crmadmin -i and -d options (fate#324508) ------------------------------------------------------------------- Fri Feb 9 00:05:41 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180208.f6a261987: - Feature: crmd: modify feature set handling ------------------------------------------------------------------- Wed Feb 7 02:12:04 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180205.b44c57e35: - pengine,doc,tools: rename "master score" to "promotion score" (fate#324508) - libpe_status: rename bundle 'masters' to 'promoted-max' (fate#323415) - pengine,libpe_status: rename master-{max,node-max} to promoted-* (fate#324508) - various: merge clone resource type and master/slave resource type (fate#324508) - attrd,pacemakerd,libcrmcommon: make "Exiting" last message - don't use corosync logging settings (fate#324508) - crmd,libpe_rules: drop support for notification-agent/recipient - tools: improve crm_failcount exit code usage ------------------------------------------------------------------- Tue Jan 30 00:36:28 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180129.c801ab11e: - allow configuring specific python intepreter version (fate#323526) * Drop obsolete fate-323526_pacemaker-pacemaker-configure-AM_PATH_PYTHON-python3.patch - use auto-configured python executable in python shebangs (fate#323526) * Update fate-323526_pacemaker-tests-shebang-python3.patch - xml: drop score-attribute and score-attribute-mangle - pengine: avoid misuse of function parameter - libcrmservice,pacemakerd: improve privilege dropping ------------------------------------------------------------------- Sat Jan 20 02:40:06 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180119.9146c3c20: - crmd: Avoid double-removal of glib event source on stopping ------------------------------------------------------------------- Fri Jan 19 00:23:32 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180118.9c3f7a49c: - libcrmcommon: put exit/return handling in own source/header files - Build: bump build dependencies - Rebase: * pacemaker-Wno-format-signedness.patch ------------------------------------------------------------------- Thu Jan 18 13:28:17 UTC 2018 - ygao@suse.com - configure: Check for python3 with AM_PATH_PYTHON (fate#323526) * fate-323526_pacemaker-pacemaker-configure-AM_PATH_PYTHON-python3.patch ------------------------------------------------------------------- Sat Jan 13 11:38:33 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180112.bd7115b8c: - tools: improve help text - libpe_status: deprecate restart-type and role_after_failure ------------------------------------------------------------------- Wed Jan 10 05:18:59 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180109.57d48a1b1: - libcluster,libcrmcommon: improve BZ2 error messages - all: use consistent exit status codes - libcrmcommon,tools: improve XML write error handling - stonithd,libcrmcommon: auto-calculate stonith-watchdog-timeout when requested (fate#324508) - logrotate: switch user when rotating logs ------------------------------------------------------------------- Tue Jan 9 14:43:27 UTC 2018 - ygao@suse.com - cts: Operate pacemaker service on startup to prevent triggering StopWhenUnneeded of corosync service * pacemaker-cts-StartCmd.patch ------------------------------------------------------------------- Fri Jan 05 03:16:07 UTC 2018 - ygao@suse.com - Update to version 1.1.18+20180104.7ba28d854: - tools: improve crm_standby option handling - tools: improve crm_standby help and error messages - tools: improve crm_master option handling - tools: improve crm_master help and error messages - tools: remove deprecated stonith_admin option (fate#324508) - tools: remove deprecated crm_verify option (fate#324508) - tools: remove deprecated crm_attribute --node-uuid option (fate#324508) - tools: avoid using deprecated options - tools: remove deprecated crm_resource options (fate#324508) - libpe_status: use consistent monitor for implicit remotes - all: overhaul log locations (fate#324508) ------------------------------------------------------------------- Wed Jan 3 12:28:59 UTC 2018 - ygao@suse.com - spec: Switch dependencies from python to python3 (fate#323526) - tests: Switch shebang from /usr/bin/python to /usr/bin/python3 (fate#323526) * fate-323526_pacemaker-tests-shebang-python3.patch ------------------------------------------------------------------- Tue Jan 02 10:13:35 UTC 2018 - bliu@suse.com - Update to version 1.1.18+20180101.494eacba4: - migrate cts to python3 (fate#323526) ------------------------------------------------------------------- Sat Dec 23 12:03:19 UTC 2017 - ygao@suse.com - Update to version 1.1.18+20171221.c91a650ec: - libcrmcommon: drop special handling of pre-1.1.7 failed recurring op history - libcrmcommon: drop special handling of pre-1.1.4 migration history entries - crmd: drop support for DCs that don't support throttling - cib,libcrmcommon: drop support for PCMK_legacy variable ------------------------------------------------------------------- Fri Dec 22 05:15:12 UTC 2017 - ygao@suse.com - libcrmcluster: use crm_strdup_printf() instead of calloc() - attrd: avoid small memory leak at start-up - systemd unit files: restore DBus dependency - libcrmservice: avoid memory leak on DBus error - pengine: validate more function arguments - Upstream version cs: 098d557cc495944e1b6ebd9d05b819d6da74825e ------------------------------------------------------------------- Mon Dec 18 13:22:27 UTC 2017 - ygao@suse.com - libpe_status: drop support for "requires" operation meta-attribute (fate#324508) - tools: crm_resource --cleanup with no resource specified (bsc#1071503) - tools: crm_resource --cleanup (bsc#1071503) - crmd,libpe_status: drop support for legacy option aliases (fate#324508) - libstonithd,libpe_status: drop support for pcmk_poweroff_action (fate#324508) - libstonithd,libpe_status: drop support for pcmk_arg_map (fate#324508) - libstonithd,libpe_status: drop support for pcmk_*_cmd (fate#324508) - libpe_status,libpe_rules: drop support for default-action-timeout (fate#324508) - libpe_status: use correct default timeout for probes - libpe_status: drop support for operation meta-attributes in instance_attributes (fate#324508) - libpe_status: fix precedence of operation meta-attributes - libpe_status,libpe_rules: remove support for is-managed-default (fate#324508) - libpe_status,libpe_rules: drop support for default-resource-stickiness (fate#324508) - crmd: drop support for LRMD_MAX_CHILDREN environment variable (fate#324508) - tools: remove crm_mon SMTP support (fate#324508) - tools: remove crm_mon SNMP support (fate#324508) - lrmd: remove deprecated -r/--dummy command-line option (fate#324508) - tools: crm_resource --cleanup for non-primitive resources (bsc#1071503) - pengine: unfence before probing or starting fence devices - Upstream version cs: 4528eb3b726197ab077243ab923ed291473f0ed7 ------------------------------------------------------------------- Thu Nov 30 15:50:43 UTC 2017 - ygao@suse.com - attrd: ensure node name is broadcast at start-up (CLBZ#5330) - Upstream version cs: 407ded8de381543e71c5b94b3ade296fcfd2c828 ------------------------------------------------------------------- Thu Nov 30 15:43:02 UTC 2017 - ygao@suse.com - libpe_status: remove support for default-resource-failure-stickiness (fate#324508) - libpe_status: remove support for resource-failure-stickiness (fate#324508) - various: remove support for pre-Pacemaker-1.0 CIB XML (fate#324508) - libpe_status,tools: remove support for legacy ticket state entries (fate#324508) - various: remove isolation wrapper support (fate#324508) - systemd unit files: enable TasksMax=infinity (bsc#1028138, bsc#1066710) * Drop obsolete bug-1028138_pacemaker-pacemaker.service-TasksMax.patch - Upstream version cs: 53a3fe14b15b087bbde9ccb524d11d0a413cc6e9 ------------------------------------------------------------------- Wed Nov 29 12:31:43 UTC 2017 - ygao@suse.com - RA: NodeUtilization RA is now shipped by resource-agents package (bsc#1070347) * Drop bug-1015842_pacemaker-NodeUtilization-RA.patch ------------------------------------------------------------------- Wed Nov 29 12:05:00 UTC 2017 - ygao@suse.com - crmd: default record-pending to TRUE (fate#324508) - tools: enable new crm_resource --cleanup/--refresh behavior (fate#324508) - Upstream version cs: f36ad364748880815bf00151c516a1f2deae2fed ------------------------------------------------------------------- Thu Nov 23 13:48:52 UTC 2017 - rbrown@suse.com - Replace references to /var/adm/fillup-templates with new %_fillupdir macro (boo#1069468) ------------------------------------------------------------------- Mon Nov 20 16:20:53 UTC 2017 - ygao@suse.com - Update to version 1.1.18 - libpe_status: always use default key location with bundles - attrd,stonithd: more efficient regular expression parsing - libpe_status: avoid potential NULL dereference - tools: use proper variable in message from crm_resource -g - tools: use uber-parent for crm_resource -a/-A - tools: use g_main_loop_is_running() properly - libpe_status: monitors are rescheduled, not reloaded - crmd: determine restart/private digests correctly - systemd: add TasksMax comment to pacemaker_remote unit (bsc#1028138, bsc#1066710) - tools: make regression test insensitive to line number changes - pengine,tools,libpe_status: avoid unnecessary use of pe_find_current - pacemaker.service: Recommend not to limit tasks (bsc#1028138, bsc#1066710) * Update bug-1028138_pacemaker-pacemaker.service-TasksMax.patch - tools: set the correct OCF_RESOURCE_INSTANCE env when crm_resource --force-* executes RA - typo: overriden -> overridden - portability: the difference of time_t values is given by difftime() - pengine: avoid not-really-possible use-of-NULL - PE: Have bundles log to stderr so that 'journalctl -M' works (rkt) - tools: don't expect reply to failed send - pengine: use newer Pacemaker Remote terminology - pengine: memory leak when writing graph to file - tools: don't reinvent the glib wheel - tools: implement clean-up dry-run correctly - pengine: avoid potential use-of-NULL - Update regression test outputs - PE: Improved logging of reasons for stop/restart actions - PE: Allow all resources to stop prior to probes completing - PE: Correctly defer processing of resources inside containers - crm_resource: Ensure we wait for all messages before exiting - crm_resource: Have cleanup operate only on failures - PE: Have bundles log to stderr so that 'docker logs' works - PE: Resources are allowed to stop before their state is known everywhere - PE: Use the node we already have and know isnt NULL - PE: Flag resources that are acting as remote nodes - crmd: Scale all cib operation timeouts - PE: Bare metal remotes _can_ run resources now and must be probed - PE: Bundles only need to wait for other containers on the same node to be probed - crmd: Scale timeouts with the number of remotes too - PE: There is no need for port mapping directives when net=host is specified - PE: Do not always expire failed operations of nested remotes - PE: Consolidate REMOTE_CONTAINER_HACK logic - PE: Exclude resources and nodes from the symmetric_default constraint in some circumstances - crm_resource: Don't cleanup on nodes not meeting the discovery conditions - PE: Restore the ability to send the transition graph via the disk if it gets too big - crm_resource: Prevent disconnection from crmd during cleanup - PE: Do not probe connection resources until the container is active - PE: Detailed resource information should include connection resource state - Tools: Allow crm_resource to operate on anonymous clones in unknown states - tools: Clean up everywhere if we don't know anything about the resource - crm_resource: See what cleanup would have done for a saved configuration - PE: Only pass requests for promote/demote flags onto the bundle's child - Upstream version cs: 2b07d5c5a908998891c3317faa30328c108d3a91 (Pacemaker-1.1.18) ------------------------------------------------------------------- Thu Nov 2 15:15:06 UTC 2017 - ygao@suse.com - pengine: if ignoring failure, also ignore migration-threshold - libcrmcommon,liblrmd,lrmd: improve messages for failed remote sends - libcrmcommon,lrmd: meaningful error codes when sending remote messages - libcrmcommon: lower watchdog messages when default - tools: set meta_timeout env when crm_resource --force-* executes RA - fencing: don't print event twice with stonith_admin --verbose - fencing: memory leak in stonith_admin --env - fencing: improve stonith_admin help - crmd: hard error if remote start fails due to missing key - libcrmcommon,liblrmd: report meaningful async connection errors - libcrmcommon: return meaningful error codes to connection callbacks - libcrmcommon: async connection callback must get negative error codes - libcrmcommon,liblrmd: improve remote connection messages - pacemaker_remote: warn if TLS key can't be read at start-up - lrmd: tweak TLS listener messages - pengine: don't probe offline or unclean guest nodes (fate#324441) - agents: ifspeed - Try to detect interface name from ip address. - pengine: don't keep unique instances on same node - pengine: improve messages when assigning resources to nodes - Upstream version cs: 23be51236edc34477a99d6979bc0e2cc1e1a7684 ------------------------------------------------------------------- Tue Oct 24 10:29:38 UTC 2017 - ygao@suse.com - libpe_status: avoid use-after-free (again) - lrmd: don't reject protocol 1.0 clients (bsc#1009076) - tools: print descriptive messages for OCF exit code - tools: differentiate trace log level for RAs - tools: show RAs debug logs only when verbose mode for crm_resource --force-* - tools: suppress an invalid error message for crm_resource --force-* options - pengine: don't schedule reload and restart in same transition - libcrmservice: list systemd agents alphabetically - libcrmcommon: add function to sort list alphabetically - pengine: new regression test for unrecoverable remote connection - libcrmservice: avoid infinite loop on bad DBus reply - libcrmservice: list systemd unit files, not only active units - libpe_status: print container type in lowercase when outputting XML - tools: re-enable crm_resource --lifetime option (CLBZ#5251, bsc#950128) - libcrmcluster: improve error checking when updating node name - tools: fix crm_resource --why output when resource and node are specified - libcrmservice: parse LSB long description correctly - libcrmservice: follow LSB standard for header block more strictly - libcommon: Erase old exit_reason - Upstream version cs: 36d2962a8613322fc43d727d95720d61a47d0138 ------------------------------------------------------------------- Fri Oct 13 12:13:50 UTC 2017 - ygao@suse.com - crm_mon: error-exit child if execl should return - crmd: avoid memory leak when caching meta-data - tools: ensure crm_resource data set is initialized - libcib: ensure xpath result isn't empty - tools: missing break statement in attrd_updater - libcrmcommon: check sscanf() result when parsing schema version - tools: check sscanf return value when parsing failed op list - tools: treat bundles with replicas like clones when restarting - tools: allow crm_resource to be called without arguments - tools: update regression test for recent message changes - tools: improve crm_resource --show-metadata error messages (CLBZ#5287) - libcrmcommon: add functions for using agent specifications - tools: improve crm_resource help (CLBZ#5251, bsc#950128) - libcrmcommon: improve cluster option parsing - Upstream version cs: 27b631162104d90e944cbc107c2d46e983f94ca7 ------------------------------------------------------------------- Fri Oct 6 14:25:23 UTC 2017 - ygao@suse.com - spec: Disable deprecated crm_mon SNMP/ESMTP support (configure alerts instead) ------------------------------------------------------------------- Fri Oct 6 14:23:14 UTC 2017 - ygao@suse.com - pengine: add tests for no-quorum-policy=suicide - libpe_status: properly detect when nodes should suicide - libpe_status: always log startup-fencing value - libpe_status: generalize ability to log warning only once - pengine: reset loss-policy from fence to stop if no fencing - pengine: avoid use-of-NULL when unpacking tickets - tools: ensure crm_resource works if no command is specified - libpe_status: validate no-quorum-policy=suicide correctly - tools: use-after-free in crm_diff - RA: ClusterMon - Correctly handle "update" parameter - attrd: "peer peer loss" -> "peer loss" - Tools:Refactored Crm_resource to have proper assignments for require_* - Upstream version cs: c4a09cfec99570a2509831f4b8974457dcdb1af6 ------------------------------------------------------------------- Fri Sep 29 11:56:01 UTC 2017 - ygao@suse.com - Log: Reduce verbosity of developer logging - PE: We most definitely need to probe connection resources - PE: Ensure the bundle nodes get set with the correct discovery mode - PE: Implement probing of container remote nodes (fate#324441) - attrd: Tweak log messages for clarity - crm_resource: Do not send history erasure commands for nodes where this is none - PE: Resources in bundles should also respect failcounts - PE: Do not send notifications to unclean bundles - PE: Prevent graph loops when fencing the host out from underneath a bundle - Tools: Allow bundle resources to read/write master scores based on the same logic as attrd (cf34f4c) - PE: bundle: Send an additional host list when container-attribute-target=host - PE: Correctly interpret colocation with bundles - PE/attrd: Allow bundle resources to read/write attributes based on the physical host rather than the container name - Upstream version cs: 4328c6a65424d1caf9785107664ec1b7fc824664 ------------------------------------------------------------------- Thu Sep 28 10:22:07 UTC 2017 - ygao@suse.com - Rebase: * bug-728579_pacemaker-stonith-dev-id.patch - Fix: -Wformat-truncation=2: prevent format truncation - snprintf: use correct signedness in print format specifiers - snprintf: use sizeof(X) as size + prevent off-by-ones - crmd,libcrmcommon,libcluster,tools: handle PID as string properly - tools: use right size for string buffer in fake_transition.c - crm_resource: fix a memleak - pengine: sprintf -> snprintf - libcrmcommon: fix possible infloop in buffer_print - crm_snprintf_offset like parts: fix off-by-ones - attrd: improve update messages - stonithd,lrmd,pengine: spelling/formatting corrections - build: install ifspeed resource agent - libcrmservices: avoid assert for HB resource with no parameters - crmd: re-enable local collection of meta-data for remote nodes - crmd,pengine,libpe_status,libpe_rules: remove support for versioned attributes - crmd: avoid memory leak when node state is unknown - crmd: avoid assert if LRM query fails - libpe_status: more error checking when unpacking bundles - libpe_status: bundles now respect PCMK_authkey_location - libcrmcommon,liblrmd,lrmd: validate PCMK_remote_port - PE: Bundles: add support for rkt bundle type - libcrmcommon,libpe_rules: unpack new alert filter syntax - xml: allow specifying alertable events - Fix: a proper calculation of a speed in case of hfi1 for ifspeed RA. - pengine: detect proper clone name at startup - pengine: don't ignore permanent master scores at startup - pengine: handle resource migrating behind a migrating remote connection - stonith-ng: advertise pcmk_on_action via metadata - liblrmd: Make sure the operation of a remote resource returns if setup of the key fails (bsc#1053463) - libcrmcommon,pengine: improve log messages - PE: only print to stdout if run from command line - crmd: eliminate size restriction on node state xpath - pengine: add regression test for remote connect interval wait - pengine: re-enable unrecoverable remote fencing - pengine: Make sure calculated resource scores are consistent on different architectures (bsc#1054389) - PE: Allow resources inside bundles to receive notifications - PE: Update unfencing device change test output - attrd,crmd: erase attributes at attrd start-up, not first join - attrd: accept connections only after CIB connection is active - attrd: make CIB connection function self-contained - attrd: improve messages - stonith-ng: make fencing-device reappear properly after reenabling - crmd: don't need meta-data after resource is deleted - crmd: only get meta-data for OCF and stonith agents - libcrmcommon: generate agent key without :: if no provider - pengine: avoid fence loop for remote nodes - attrd: don't log error if CIB doesn't have an alerts section - PE: Only retrigger unfencing on nodes that ran operations with the old parameters - PE: Output resource changes in a friendlier format - PE: Attempt to indicate why resources are changing state - Mechanism for recording why an action is required - Support unfencing of remote nodes - PE: Ignore optional unfencing events and report the fencing type - Upstream version cs: 20cffae1e9c64217339173c197edac8612ad02b9 ------------------------------------------------------------------- Mon Jul 31 09:40:31 UTC 2017 - ygao@suse.com - logging: ensure blackbox gets generated on arithmetic error - crm_mon: overcome crm_system_name no longer influenced with argv - crm_mon: make CGI bail out on suspicious arguments - crm_mon: make CGI bail out properly and w/o help screen - crm_mon: canonical casing of Content-Type CGI header field - libcrmcommon: handle schema versions properly - attrd: don't reconnect to lrmd if already connected - PE: Ensure nested container connections run on the same host - crmd: send alerts via lrmd - lrmd: execute alerts as cluster daemon user - libservices: add ability to specify user to run an action - libcrmcommon: improve user lookup messages - extra: correct ClusterMon metadata - attrd: avoid memory leak when disconnecting from lrmd - attrd,tools: avoid memory leaks from use of crm_itoa() - attrd: be consistent about attr/host logging - attrd: delay lrmd connect until needed - attrd: add timestamp to alert just when needed - attrd: improve log messages - libcrmcommon,libpe_rules: avoid crash when alerts section has comments - libcrmcommon: add API call for next named XML child - liblrmd: only set environment variables with a value - attrd,liblrmd: handle adding env var at head of list - lrmd: drain alerts at shutdown - attrd,lrmd: generate alert sequence number in lrmd - liblrmd: better trace message when creating command - stonith-ng: avoid double-free of pending-ops in free_device - crmd: log correct feature set revision - PE: update start-failure-is-fatal help - cib: avoid unnecessary memory allocation - stonith-ng: add pcmk_delay_base as static base-delay (fate#324488) - XML: upgrade-1.3.xsl: be explicit about non 1:1 wrt. ACL @attributes - enable upgrade XSLTs to use <xsl:message> meaningfully - Tools: Added option of --why/Y to crm_resource to inform the status of resources if running or not - various: fix spelling in messages and comments - cib: reword trace message - PE: Prevent use-after-free when creating bundle constraints - add versioned operations support - TE: don't bump counters when action or synapse is invalid - obtain metadata in CRMd via async call - tools: add version options for cibsecret - alerts: Support of meta attributes(kind, attribute_name). - attrd,lrmd,lib: Addition of the transmission function of the attribute alert to attrd. - extra: Change alert_xxx.sh.sample for attribute alerts. - extra: Change PCMK-MIB.txt for attribute alerts. - XML: upgrade-1.3.xsl: simplify branching - XML: upgrade-1.3.xsl: drop superfluous branching - XML: upgrade-1.3.xsl: simplify "weak identity" transform rules - XML: upgrade-1.3.xsl: easy simplifications + redundant declarations drop - re-enable versioned attributes - Upstream version cs: d134f83b4f0188f6d7cd381c8cdda1a2a7c38bd5 ------------------------------------------------------------------- Wed Jul 19 14:06:09 UTC 2017 - ygao@suse.com - Update to version 1.1.17 - PE: Ensure unrecoverable remote nodes are fenced even if no resources can run on them - iso8601: strftime needs fully populated struct tm (bsc#1058844) - crmd: abort transition whenever we lose quorum - PE: Ensure bundle nodes and child resources are correctly cleaned up - PE: Generic ordering for anything with bundles - PE: Default to non-interleaved bundle ordering for safety - PE: Functional inter-bundle ordering when either side has no child - PE: Basic inter-bundle ordering when both sides have children - PE: Implement bundle ordering - libpe_status: properly disallow other resources from bundle nodes - libcib: get remoteness correctly from node status - libpe_status: avoid memory leaks when creating bundle remote resource - libpe_status: disallow resources on bundle nodes - pengine: make checks a little safer - PE: Clones within bundles may also have notifications enabled - PE: Correctly implement pe_order_implies_first_printed - PE: Do not re-add a node's default score for each location constraint - PE: Bundle location constraints should only apply to the IP and docker resources - PE: Bundles: Allow clones to be colocated with bundles - PE: Bundles: Implement colocation of primitives and groups with bundles - cib,libcib: support option for IPC eviction threshold - libcib: correctly search for v2 patchset changes - libcib: always use current values when unpacking config - libcrmcluster: fix spelling error in error message - crmd: DC should update stonith fail count before aborting transition - cib: Broadcasts of cib changes should always pass ACLs check (bsc#1042054) - Upstream version cs: b36b869ca8fe74caa077c286304f653f0be117ef (Pacemaker-1.1.17) ------------------------------------------------------------------- Fri Jun 2 12:58:32 UTC 2017 - ygao@suse.com - pengine,libpe_status: revisit fencing messages - PE: Improve logging of node fencing and shutdown - PE: Fence unrecoverable remote nodes with no resources - libpe_status: propagate multiple-active=block for bundles - libpe_status: set parent for bundle components - attrd: broadcast local overrides of sync'ed attributes - crmd: attribute write-out not needed with atomic attrd - PE: Prevent use-after-free when creating bundle constraints - PE: Ensure remote nodes are fenced when the connection cannot be recovered - PE: Improved fencing logging - PE: Assume resources on remote nodes do not need to be restarted until absolutely necessary - PE: Correctly compare a point with NULL instead of FALSE - libpe_status: correctly flag remote connection resources - pengine: improve internal bundle constraints - pengine: re-phrase trace log for internal constraints - libpe_status: set #kind to "container" for bundle nodes - pengine,libpengine: avoid potential use-of-NULL - pengine: avoid use-of-NULL in bundle without primitive - libpengine: don't double score when adding first allowed node - pengine: improve detection of invalid constraints - PE: Prevent use-of-NULL when expanding containers - PE: Partially restore 62ed004 to ensure remote connections are available before attempting resource recovery - stonith-ng: Trigger on constraints added to cib (bsc#1042374) - Upstream version cs: e47bd5bc77a3d1b38589b3716f5112ed39329e7b ------------------------------------------------------------------- Mon May 15 12:41:35 UTC 2017 - ygao@suse.com - RA: NodeUtilization - Use xl if available (bsc#1015842) * Drop obsolete: ** bug-1015842_pacemaker-NodeUtilization-RA-1.patch ** bug-1015842_pacemaker-NodeUtilization-RA-2.patch ** bug-1015842_pacemaker-NodeUtilization-RA-3.patch * Add bug-1015842_pacemaker-NodeUtilization-RA.patch - Drop obsolete pacemaker-lrmd.h-include-libxml.patch - lrmd,pengine: remove versioned parameters from documentation and tests - lrmd,pengine: don't support versioned attributes by default - Containers: Relax docker/remote colocation constraint when appropriate - Containers: Ensure remote resources are only allocated to full cluster nodes - lrmd: Ensure verbosity options are handled after crm_log_init() - pengine: unmanage guest-container puts guest in maintenance - pengine: unique ids for meta-attributes of guest-connection - pengine: update regression tests for probes of guest-connections - pengine: have guest-node connection-resources probed - crmd: clean up throttle memory on exit - libcib: avoid use-after-free when deleting CIB connection - systemd unit files: add dependency on resource-agents-deps - pengine: Allow use of resource params and meta-params in location rules - libcrmcommon: avoid evicting IPC client if messages spike briefly - libcrmcommon: don't delay next flush by more than 5 seconds - libcib: downgrade ACL status message to trace - crmd: don't destroy election structure twice - crmd: avoid use-after-free when disconnecting from CIB - libcib: functionize destroying op callback table - crmd,libcrmcommon: update throttle when CPUs are hot-plugged - crmd: remove I/O load checks - libcrmcommon: add function to get number of CPU cores - crmd: functionize checking throttle thresholds - crmd: functionize setting throttle load target - CTS: update corosync fail patterns - crmd: check correct timer for being started - crmd: improve fencing result message - lrmd: Have pacemaker-remote reap zombies if it is running as pid 1 - pengine: add test for per-op fail count - libcrmcommon: filter attributes with '#' from XML fields - extra: Dummy stop shouldn't fail if monitor is in progress - extra: use whitespace consistently in Dummy - crmd: don't fence old DC if it's shutting down as soon-to-be DC joins - crmd: avoid DC sending offer to itself twice - crmd: don't restart transition if no fence devices - crmd: skip restart at (not above) stonith-max-attempts - crmd: allow clearing all stonith fail counts - crmd: track stonith fail counts on all nodes - crmd: forget stonith failures when forgetting node - crmd: consider target when checking stonith failures - crmd: check for too many stonith failures only when aborting for that reason - PE: Remote: Allow remote nodes that start containers with pacemaker remote inside - fencing: ignore empty 'action' parameter in fence devices - crmd: fix remove attr_id from update_attr_delegate() - Fix log showing the node status so that it is easily distinguishable from other logs - crmd: don't abandon fencing after one "no devices" failure - libcrmcommon: more useful message if can't open dump file - libcib: use new convenience function for sanitizing ID - libpe_status: implement active check for bundles - tools: update crm_mon RNG schema for bundles - crmd: add set_join_state() - crmd: fix line wrapping - crmd: update_attr_delegate() instead of set_standby() - PE: Containers: Allow custom docker run commands - logrotate: rotate pacemaker logs inside bundled containers - libpe_status: improve bundle docker defaults - tools: crm_resource should free result if re-running function returns one - libpe_status: avoid memory leak when counting failures - crmd: remove functions from crmd_utils.c - PE: Containers: Ensure replicas_per_host=1 is respected during rediscovery and startup - libcib: change the format of the state id in set_standby() - PE: Containers: Allow the internal and external ports to optionally differ - PE: Containers: Allow custom control ports when using the machine's local address - crmd: set_standby in first_join - PE: Bundles: Better support for net=host, allow the pacemaker-remote port to change - Remote: Allow the port to listen on to be specified on the commandline - libpe_status: allow more than 254 containers in a bundle - crmd,liblrmd: handle crmd-local proxy requests properly - common: Prevention of the double definition. - pacemaker_remoted: Allow compilation with glib older 2.36 - PE: Containers: Default 'replicas' to the same number of 'masters' - PE: Containers: Correctly find the right anonymous child resource running in a container - PE: Clone: Simplified allocation function - PE: Contaienrs: Better checks when assuming a container will start - PE: Preferred nodes are only accepted if their scores are equal to the otherwise best candidate - PE: Containers: Remove containers on termination by default - PE: Container: Add support for non-default docker networks and supplying external IPs to container resources - PE: Containers: Allow specifiying the maximum number of peers per host - PE: Containers: Force a sane hostname - PE: Containers: Add support for nic and netmask parameters - PE: container: Parse network and mount options - PE: Allow the graph to assume container nodes will come up and schedule actions for it - PE: Force node scores for containerized resources - PE: Basic allocation of containers - Core: Track the line number of callers of crm_(set|clear)_bit - xml: Correctly find the first child element - PE: Basic container parsing and expansion (fate#323415) - PE: Correctly test for clone functionality - Improve recovery when demote fails with OCF_NOT_RUNNING - Upstream version cs: ce86684f5d0672791e9fd3043a84f0931cdd97ea ------------------------------------------------------------------- Thu Mar 23 16:02:22 UTC 2017 - ygao@suse.com - libcrmcluster,crmd: log join phase as text rather than integer - pacemaker-remote: pacemaker_remoted shutdown while unmanaged - tools: disable agent stdout/stderr for crm_resource --validate - tools: Add "--validate" command to crm_resource - tools: crm_resource should set OCF_RESKEY_crm_feature_set - tools: make crm_failcount compatible with per-operation failcounts - tools: support clearing failure of a single operation with crm_resource -C - crmd: support clearing failure of a single operation - crmd: make clearing last failure more efficient - attrd,libcrmcommon: support clearing failure of a single operation - attrd,crmd,libraries,tools: track resource failures per-operation - tools: support crm_failcount -q as advertised - Upstream version cs: 77ea74d01bcb6b7100e1c7acab0e66e79aceaf8b ------------------------------------------------------------------- Fri Mar 17 16:10:49 UTC 2017 - ygao@suse.com - NodeUtilization: Never set hv_memory to a negative value (bsc#1015842) * bug-1015842_pacemaker-NodeUtilization-RA-3.patch - NodeUtilization: Fix mixed indentation, set executable bit (bsc#1015842) * bug-1015842_pacemaker-NodeUtilization-RA-2.patch - RA: NodeUtilization - Use xl if available (bsc#1015842) * Drop obsolete bug-1015842_pacemaker-NodeUtilization-RA.patch * Add bug-1015842_pacemaker-NodeUtilization-RA-1.patch ------------------------------------------------------------------- Fri Mar 17 11:07:49 UTC 2017 - ygao@suse.com - spec: Merge the SLE and Factory versions * enable_fatal_warnings=0 ------------------------------------------------------------------- Thu Mar 16 14:25:19 UTC 2017 - ygao@suse.com - crmd: new cluster option stonith-max-attempts with default value 10 (fate#323417) - crmd: Change of the log level and addition of uuid. - crmd: use new attrd op to clear fail counts - tools: crm_resource -C should clear last-failure - libcrmcommon: add API for new attrd clear-failure operation - attrd: add more detail to legacy attrd messages - attrd: add API for clearing failures - Upstream version cs: e55d506cc390f0a0335b4ec6e365ea89da30b6b1 ------------------------------------------------------------------- Mon Mar 13 13:46:36 UTC 2017 - ygao@suse.com - pacemaker.service: Add presetting TasksMax=4915 (bsc#1028138) * bug-1028138_pacemaker-pacemaker.service-TasksMax.patch ------------------------------------------------------------------- Tue Mar 7 00:01:15 UTC 2017 - ygao@suse.com - Build: lrmd: Include libxml/tree.h in lrmd.h * pacemaker-lrmd.h-include-libxml.patch ------------------------------------------------------------------- Mon Mar 6 22:58:46 UTC 2017 - ygao@suse.com - libfencing,fencing: properly remap "action" in configuration - libservices: ensure recurring actions table is created before using - libservices: improve error messages when creating operation - libservices: properly detect in-flight systemd/upstart ops when kicking - libservices: properly cancel in-flight systemd/upstart op - libservices: handle in-flight case first when cancelling an operation - libservices: prevent use-after-free when freeing an operation - libservices: ensure completed ops aren't on blocked ops list - libcrmcommon: assert if can't generate operation key - libcrmcommon: improve remote connection polling messages - crmd: clear failure only for requested node - crmd: improve message when clearing failures - pengine,libpe_status: make failcount clearing messages more helpful - pengine,libpe_status: don't clear same fail-count twice - cib: properly mark variable as volatile - libcib: avoid memory leak in query_node_uuid() - tools: avoid NULL dereference in crm_resource debug message - cib,libcrmcommon,lrmd: handle IP addresses better in messages - tools: crm_attribute should prefer node name from environment - libcrmcommon: Correctly delete XML comments according to their positions (bsc#1024037) - libcrmcommon: Correctly compare XML comments to prevent crmd from getting into infinite election loop (bsc#1024037) - crmd,libcrmcluster: minor log message improvements - tools: avoid deprecated function in notifyServicelogEvent - libcrmcommon,pengine,tools: pass local node name to resource agents - use callback instead of sync call - libcib: make querying node uuid more efficient - tools: allow regular expression with crm_attribute - attrd: support regular expressions in legacy attrd - attrd: support Pacemaker Remote node attributes in legacy attrd - use status scope to start node in standby mode - attrd: ignore unsupported requests in legacy attrd - attrd,crmd: implement peer remove requests in legacy attrd - crmd: don't clear remote node transient attributes unnecessarily - crm_mon: protect against non-standard or failing asctime - libservices(sync): partially prevent killing foreign process - libservices (sync): ensure no zombie is left behind - cib: improve re-sync handling - tools: implement crm_failcount command-line options correctly - tools: use correct regular expression for fail counts in crm_resource - cib: never disable legacy mode with corosync 1 stacks - cib: improve warning when legacy diff fails - libcrmcommon: ensure filename is not NULL before opening - crmd: be more resilient when checking an LRM command's "from" - crmd: improve crmd's LRM-related log messages - tools: properly ignore version with crm_diff --no-version (bsc#888726) - Ability to start node in standby (fate#323416) - Upstream version cs: a7825f4f0a8db272d1ce0b1f93e8044cbcaa5144 ------------------------------------------------------------------- Mon Mar 6 13:04:16 UTC 2017 - ygao@suse.com - libservices: properly watch writable DBus handles - tools: don't reference hidden parameter in crm_resource help text - Alert: Set SNMP_PERSISTENT_DIR directory for the snmp-trap tool. - dbus: Prevent lrmd from hanging on dbus calls (bsc#1015264) - spec: add %check to run a subset of tests on rpmbuild - crmd: update cache status for guest node whose host is fenced - pengine: consider guest node unclean if its host is unclean - pengine: create a pseudo-fence for guest node recovery - pengine: guest node fencing doesn't require stonith enabled - pengine: remove unnecessary assert - mcp: Correction of the difference in access permission setting. - stonith_admin: Addition of the list-targets option. - stonith: Check for missing params in new device and dup - libservices: treat systemd service reloading as OK (bsc#1059187) - crmd: increase severity when fencing didn't happen (bsc#1011240) - rng: Create resources-2.7.rng to update template class validation - libcrmcommon: Don't report error if the output file isn't syncable - crmd: bump feature set - pengine: disable migration for versioned resources - add versioned parameters support - Upstream version cs: 2817a2081e9e70e7412f0d0a253495611afab55e ------------------------------------------------------------------- Mon Mar 6 12:58:37 UTC 2017 - ygao@suse.com - wd: Fix to not compare NULL when environment variable is not set - mcp,pacemaker_remote: order after time-sync - configure: fix fouled parameter expansion * Drop obsolete pacemaker-configure-systemd-detection.patch - xml: rng: Keep score-attribute{,-mangle} in the existing RelaxNG schema files for graceful upgrades - Upstream version cs: 94ff4df51a55cc30d01843ea11b3292bac755432 (Pacemaker-1.1.16) ------------------------------------------------------------------- Mon Mar 6 12:57:23 UTC 2017 - ygao@suse.com - RA: NodeUtilization - Use xl if available (bsc#1015842) * bug-1015842_pacemaker-NodeUtilization-RA.patch ------------------------------------------------------------------- Mon Mar 6 12:53:50 UTC 2017 - ygao@suse.com - spec: Prevent overwriting existing sysconfig files by conditionally running %fillup_only (bsc#1022807, bsc#980341) ------------------------------------------------------------------- Mon Mar 6 12:52:25 UTC 2017 - ygao@suse.com - spec: cts brings an RA that needs python-systemd ------------------------------------------------------------------- Wed Feb 22 15:39:36 UTC 2017 - dimstar@opensuse.org - Explicitly own %{_defaultdocdir}/%{name}. ------------------------------------------------------------------- Tue Nov 8 10:21:00 UTC 2016 - ygao@suse.com - Revert "Build: configure: make systemd detection akin to + before Upstart one" * pacemaker-configure-systemd-detection.patch - libcrmcommon: fix CVE-2016-7035 (improper IPC guarding) (bsc#1007433) - sysconfig: minor tweaks (typo, wording) - various: issues discovered via valgrind and coverity - attrd_updater: fix usage of HAVE_ATOMIC_ATTRD - Upstream version cs: b6f251a13fadca0cb74702b0c007a55275c8ea09 ------------------------------------------------------------------- Fri Oct 21 09:35:01 UTC 2016 - ygao@suse.com - Rebase: * pacemaker-NodeUtilization-RA.patch - extra: add resource agent for managing a node attribute - crmd: cl#5185 - Record pending operations in the CIB before they are performed (bsc#1003565) - ClusterMon: fix to avoid matching other process with the same PID - mcp: improve comments for sysconfig options - spec: require libqb >= 0.13.0 - tools: crm_standby --version/--help should work without cluster - spec: reorganize so %pre etc. blocks are self-contained - spec: require python >= 2.6 - spec: comment better - libpengine: ensure child is not null before looking up value - libpengine: only log startup-fencing warning once - Upstream version cs: 722276c442bf023931842313bfe0dfb7d90a7c15 ------------------------------------------------------------------- Mon Sep 26 10:51:34 UTC 2016 - ygao@suse.com - pacemaker.service: do not mistakenly suggest killing fenced - pacemaker.spec: consume "systemdunitdir" via %_unitdir - libcrmcommon: report errors consistently when waiting for data on connection (bsc#986644) - libservices: use DBusError API properly - remote: Correctly calculate the remaining timeouts when receiving messages (bsc#986644) - libfencing: report added node ID correctly - crm_mon: Do not call setenv with null value - pacemaker_remote: allow remote clients to timeout/reconnect - pengine: Do not fence a maintenance node if it shuts down cleanly (bsc#1000743) - install README in doc instead of now-removed AUTHORS file - ping: Avoid temporary files for fping check (bsc#987348) - all: clarify licensing and copyrights - schemas.c: treat CIB with unrecognized validate-with better - schemas.c: avoid downgrade validate-with in a corner case - xml.c: better info in cli_config_update on no validation success - fix infloop when no schema validates on update_validation - xml.c: internal self-or-less reference inflicted infloop guard - crmd: Resend the shutdown request if the DC forgets - ping: Avoid temp files in fping_check (bsc#987348) - crmd: Ensure the R_SHUTDOWN is set whenever we ask the DC to shut us down - crmd: clear remote node operation history only when it comes up - libcib,libfencing,libtransition: handle memory allocation errors without CRM_CHECK() - tools: make crm_mon XML schema handle resources with multiple active - tools: make crm_mon resources section more consistent - crmd: bump feature set - pengine: set OCF_RESKEY_CRM_meta_notify_active_* for multistate resources - LSB init: try to be a little more compliant - pengine: avoid null dereference in new same-node ordering option - lrmd,libcluster: ensure g_hash_table_foreach() is never passed a null table - crmd: don't log warning if abort_unless_down() can't find down event - lib: Correction of the deletion of the notice registration. - stonithd: Correction of the wrong connection process name. - crmd: Keep a state of LRMD in the DC node latest. - pengine: avoid transition loop for start-then-stop + unfencing - libpengine: allow pe_order_same_node option for constraints - pengine: avoid sisyphean score-attribute-mangled parsing - xml: score-attribute{,-mangle} has no meaning for rsc_colocation - Upstream version cs: 4cce2b26f6f468c342adb24bf74234a12a0070c2 ------------------------------------------------------------------- Fri Aug 26 16:20:25 UTC 2016 - ygao@suse.com - cts: Restart systemd-journald with "systemctl restart systemd-journald.socket" (bsc#995365) * bug-995365_pacemaker-cts-restart-systemd-journald.patch ------------------------------------------------------------------- Fri Jul 8 09:58:46 UTC 2016 - ygao@suse.com - libcrmcommon: properly handle XML comments when comparing v2 patchset diffs - crmd: don't abort transitions for CIB comment changes - libcrmcommon: log XML comments correctly - libcrmcommon: remove extraneous format specifier from log message - crm_mon: Avoid logging errors for any CIB changes that we don't care about (bsc#986931) - crmd: When node load was reduced, crmd carries out a feasible action. - Upstream version cs: 58fef06198fef17873d0d2e2d0cc5c7db73c5724 ------------------------------------------------------------------- Fri Jul 1 23:10:54 UTC 2016 - ygao@suse.com - ping: Log sensible error when /tmp is full (bsc#987348) - crm_mon: Do not log errors for the known CIB changes that should be ignored (bsc#986931) - logging: Enable synchronous logging when blackboxes are enabled - libpengine: print target role and blocked status in XML output - tools: add maintenance mode and disabled/blocked resources to crm_mon XML/HTML output - tools: break up long lines in crm_mon console output - libservices: include socket units when listing all systemd agents - fencing: fence_legacy - Search capable devices by querying them through "list" action for cluster-glue stonith agents (bsc#986265) - extra: correct quoting of timestamp-format in alert-snmp-example - pengine,tools: Display pending resource state by default when it's available (bsc#986201) - rng: Allow rsc-pattern in locations for version 2.6 - pengine: Expand regexp references in location constraints - pengine: Allow to set base score for progressive node health strategy - systemd: Allow unit names ending in .socket. - watchdog: Correction to do to change action of the system reboot by fail_fast - Upstream version cs: 630d0d112092042b6f79c2a02bb442e82fee93ec ------------------------------------------------------------------- Thu Jun 30 13:16:25 UTC 2016 - tchvatal@suse.com - Remove conditionals that are not really needed as they always kept the values - Remove the cman conditional as we do not even have the dependencies to build with it in OBS - Convert deps to pkgconfig style ones and remove deps that are not mentioned by the testsuite or the configure.ac ------------------------------------------------------------------- Wed Jun 22 12:25:47 UTC 2016 - ygao@suse.com - Update to version 1.1.15 - crmd: mention that graceful remote shutdowns may cause connection failures (bsc#981489) - alerts - drop 5min timeout to 30s since it is configurable now - Bug bz#5292 - crmd: wait for alerts to finish - libais,libcluster,libcrmcommon,liblrmd: don't use %z specifier - alerts: ensure sample alert agents handle undefined recipient - tools: crm_resource -T option should not be hidden anymore - PE: Correctly update the dependant actions of un-runnable clones - alerts: make timestamp-format-string work properly if strftime-escape is the last character - extra: make alert-file-example cope with spaces in timeformat-string - attrd,ipc: Prevent possible segfault on exit (bsc#986056) - alerts: fix SMTP recipient defaulting - tools: properly handle crm_resource --restart with a resource in a group - tools: don't assume all resources restart on same node with crm_resource --restart - tools: avoid memory leaks in crm_resource --restart - tools: remember any existing target-role when doing crm_resource --restart - tools: correctly count starting resources when doing crm_resource --restart - Upstream version cs: e174ec84857e087210b9dacee3318f8203176129 (Pacemaker-1.1.15) ------------------------------------------------------------------- Tue Jun 14 11:35:04 UTC 2016 - tchvatal@suse.com - Clean-up with spec-cleaner - Drop heartbeat conditions, no heartbeat around - Sort CMAN conditionals to be actually logical - Run fdupes to not waste so much space - Add logrotate to reqs of pacemaker-cli - Cleanup rpmlintrc to contain only bare needed minimum - Run basic sanity check in %check phase ------------------------------------------------------------------- Tue Jun 14 11:09:45 UTC 2016 - tchvatal@suse.com - Drop sysvinit support we now run on sle12+ only - Cleanup the define conditionals that are never used - Drop profiling and coverage options it is not enabled by default and better to run separately ------------------------------------------------------------------- Mon Jun 13 10:31:52 UTC 2016 - ygao@suse.com - libcrmcommon: really ensure crm_time_t structure is fully initialized by API calls - extra: correct syntax error in sample alert script - crmd: avoid timeout on older peers when cancelling a resource operation - crmd: use proper resource agent name when caching metadata - crmd: avoid memory leak when sending fencing alert - liblrmd: avoid memory leak when closing or deleting lrmd connections - extra: improve determination of successful monitors in SNMP alert script - tools: avoid memory leak when crm_mon unpacks constraints - pengine: avoid use-after-free with location constraint + sets + templates - pengine: better error handling when unpacking sets in location constraints - pengine: avoid memory leak when invalid constraint involves set - Build: rename sample alert scripts and install them in datadir - pengine : Correction of the record judgment of the failed information. - Upstream version cs: dd46368eead0d756735216a4bc765466b99274fc ------------------------------------------------------------------- Tue Jun 7 08:40:25 UTC 2016 - ygao@suse.com - alert-helpers: some fixes in pcmk_smtp_helper.sh - libpengine: log message when stonith disabled, not enabled - pengine: only set unfencing constraints once - tools: don't require node to be known to crm_resource when deleting attribute - libcrmcommon: ensure crm_time_t structure is fully initialized by API calls - attrd: don't leak memory when returning due to malformed update request - crmd: always free result of crm_itoa() - xml: adapt crm_mon.rng per recent reorder of the summary elements - Upstream version cs: 1a05449500f015414b6df68b6d4feb64b713f533 ------------------------------------------------------------------- Fri Jun 3 12:07:19 UTC 2016 - ygao@suse.com - Rebase: * bug-977201_pacemaker-controld-self-fencing.patch - pengine: properly order stop actions relative to stonith - crm_resource: don't pretend that target-role:started would prevent promotion - crm_mon: consistently print ms resource state - controld: use "stonith_admin -H '*' -V" (bsc#977201) - stonith_admin: allow -H '*' to show history for all nodes (bsc#977201) - controld: fix "[: -eq: unary operator expected" issue from PR 1022 (bsc#977201) - extra: Support of the task designation of a resource and the fencing. - extra: Change pcmk_snmp_helper.sh (2). - Upstream version cs: 6a8c1e9d35c65cbfd4019cfe797cb7bb9dfd4322 ------------------------------------------------------------------- Sun May 29 20:06:12 UTC 2016 - ygao@suse.com - controld: Prevent unwanted self-fencing if "stateful_merge_wait" state of dlm is not available (bsc#977201) * bug-977201_pacemaker-controld-self-fencing.patch ------------------------------------------------------------------- Sun May 29 14:44:18 UTC 2016 - ygao@suse.com - libservices: clean up non-signalfd code paths - spec: fence_pcmk only eligible for Pacemaker+CMAN - crmd: Acknowledge cancellation operations for remote connection resources (bsc#976865) - controld: improve DLM check with stateful_merge_wait (bsc#977201) - attrd, libcrmcommon: validate attrd requests better - crmd: clear remote node transient attributes on disconnect (bsc#981489) - Revert "Fix: attrd: Correctly implement mass removal of a node's attributes" (bsc#981489) - crm_mon: Fix time formatting on x32 - resources: use OCF version tagging correctly - libcommon: crm_procfs_pid_of() would always return last /proc entry checked - crmd: correctly dig into cib-query-answer if alerts section is present timestamp-format defaults to "%H:%M:%S.%06N" - crmd: Take start-delay into account for the timeout of the action timer (bsc#977258) - libservices: make systemd override file world-readable to avoid log warning - libcrmcommon: correct directory name in log message - Log: make crmd messages more user-friendly - Log: make messages in common quorum/fencing situations more user-friendly - Log: make shutdown messages more user-friendly - pengine: lf#2560/lf#2588 - Correctly set the environment variable "OCF_RESKEY_CRM_meta_timeout" when "start-delay" is configured (bsc#977258) - Log: make operation messages more user-friendly - Log: make startup messages more user-friendly - services: Correctly clean up service actions for non-dbus case - uninitialized use of nano_digits - attrd: Update also dampens when synchronizing attributes (bsc#977846) - Upstream version cs: 389294fd1ebefddad67fff51306f585b758ca860 ------------------------------------------------------------------- Wed May 4 10:38:16 UTC 2016 - ygao@suse.com - Rebase: * pacemaker-Wno-format-signedness.patch - RA: SysInfo - Reset the node attribute "#health_disk" to "green" when there's sufficient free disk (bsc#975079) - pengine: Organize order of actions for slave resources in anti-colocations (bsc#977800) - attrd: Fix the synchronization of attributes (bsc#977846) - pengine: Organize order of actions for master resources in anti-colocations (bsc#977800) - pengine: Respect asymmetrical ordering when trying to move resources (bsc#977675) - fencing: Record the last known names of nodes to make sure fencing requested with nodeid works (bsc#974108) - crmd: first implementation of integrated multi-agent, multi-target alerting mechanism (fate#320855) - ping resource: Use fping6 for IPv6 hosts (bsc#976271) - pengine: Consider resource failed if any of the configured monitor operations failed (bsc#972187) - ping resource: avoid undefined func w/ OCF_FUNCTIONS=/dev/null - spec: make Publican docs not depend on pre-existing packages - cib/fencing: Set status callback before connecting to cluster (bsc#974108) - xml: use spaces instead of tabs in schema files - extra: Support of the dampen change by RA. - Upstream version cs: 9e382568ac3cd54c222496501088ab3fc158b2e2 ------------------------------------------------------------------- Thu Mar 24 17:26:44 UTC 2016 - ygao@suse.com - crmd: update cluster state in CIB after fencing an unseen node - crmd: don't abort transition for expected remote node shutdown - libtransition: potential memory leak if unpacking action fails - Upstream version cs: 5a6cdd11b4f93b4d55e1b2654078018c339931d4 ------------------------------------------------------------------- Thu Mar 17 10:04:23 UTC 2016 - ygao@suse.com - attrd: crash on exit if initialization fails - build: fix scalability regression - pengine: Suppress assert if a stateful clone is not being demoted/stopped (bsc#971129) - attrd: Support of the dampen change by attrd. - PE: Correctly handle the ordering of demote actions for failed and moving containers - Upstream version cs: d076e9dbf9827f7801c89c6d635df370aeca5e90 ------------------------------------------------------------------- Mon Mar 7 15:07:27 UTC 2016 - ygao@suse.com - crmd,tools: Set attributes for remote nodes directly into cib if it's legacy attrd (bsc#967775) - libcib: Correctly determine the node type (bsc#967775) - resources: match agents' default for globally_unique to pacemaker's - crm_mon: Help admins identify why resources are not being started or responding to commands - PE: Memory leak - crm_resource: Prevent segfault when --resource is not correctly supplied for --restart command - Build: Create CRM_CORE_DIR during installation - Upstream version cs: eb89393a82eb41ff50d041b91765ccac52e873e0 ------------------------------------------------------------------- Mon Feb 29 11:53:36 UTC 2016 - ygao@suse.com - cpg: Ensure group name is never unset - remote: Simplify calls to accept() and inet_ntop() by using "struct sockaddr_storage" (bsc#964183) - remote: cl#5269 - Notify other clients of a new connection only if the handshake has completed (bsc#967388, bsc#1002767, CVE-2016-7797) - remote: Correctly display the IP address of the remote client (bsc#964183) - spec: declare also BSD license as suitable - libservices: Add error handling for pipe() failed. - libservices: Check resource agent is executable or not before pipe open. - crmd: Prevent use-after-free when an unexpected remote client takes over (bsc#964183) - pacemakerd: Correction of the stonithd process judgment of pacemakerd. - crmd: Do not turn on stonith-watchdog-timeout automatically if SBD is detected - remote: Correctly check for SBD configuration consistency - resources: Fix state file metadata in Dummy, Stateful and HealthCPU resource agents - systemd: Modified so that the cluster to stop before the rsyslog - common: Correctly locate sbd via its pidfile - common: Assume processes are active if we get EACCESS - Fencing: Simplify watchdog integration - crmd: Only poopulate stonith-watchdog-timeout if SBD is active - remote: Ensure SBD is consistently configured on remote nodes - remote: Allow cluster and remote LRM API versions to diverge (bsc#1009076) - cib: Do not send notifications when the dryrun flag is present - Upstream version cs: cf67796823c4429714857411e614cfacfbc59d89 ------------------------------------------------------------------- Mon Feb 15 09:31:25 UTC 2016 - ygao@suse.com - crm_shadow: fix broken --display command - ipc: Respect the parameter of pick_ipc_buffer() but don't let it affect the global value - libcluster: crm_peer_uname() should return NULL if uuid is invalid - remote: recover from remote node fencing more quickly - crmd: update node state for migrated pacemaker_remote nodes - crmd: update node state for stopped pacemaker_remote nodes - crmd: update node state for started pacemaker_remote nodes - crmd: add new message type for updating pacemaker_remote node state - libcluster: set remote peer cache state based on CIB node state - libcluster: send notifications when pacemaker_remote nodes are started - crmd: report pacemaker_remote node state in CIB status section - Upstream version cs: c851a117bc77bdefe15181a7d753f1aed867cde6 ------------------------------------------------------------------- Thu Feb 11 13:28:13 UTC 2016 - ygao@suse.com - spec: Move the common directories under /var/lib/pacemaker into pacemaker-cli package - liblrmd: Prevent potential use-after-free (bsc#964183) - crmd: Disconnect the relevant remote proxies as well when disconnecting a remote node (bsc#964183) - crmd: Prevent potential use-after-free (bsc#964183) - liblrmd: Prevent potential use-after-free issues (bsc#964183) - PE: Fix conditions for internal sanity check (bsc#971129) - attrd: ensure remote nodes are in correct peer cache (bsc#981489) - crmd: set remote flag when gracefully shutting down remote nodes (bsc#981489) - fencing: Correctly track active stonith actions (bsc#938545) - fencing: Functionize adding and removing active pids of device (bsc#938545) - crm_shadow: Do not invoke shells with --noprofile option other than bash (bsc#961392) - lib/common: check return value of dlsym() and not dlerror() - lib/ipc: allow PCMK_ipc_buffer to be smaller than default max sizes - lib/services: use self-pipe sigchld handling when signalfd isn't available - spec: associate subpackage with correct licenses - spec: rely on system-wide RPM macros if available - spec: C++ compiler is not something currently needed - RHBZ#1286316: Do an ordered shutdown of systemd resources - header == NULL when parsing compressed message - attrd: Correctly implement mass removal of a node's attributes (bsc#981489) - attrd: Hook up the client name so we can track requests - crmd: Set the shutdown transient attribute in response to LRMD_IPC_OP_SHUTDOWN_REQ from remote nodes (bsc#981489) - PE: Honor the shutdown transient attributes for remote nodes (bsc#981489) - pacemaker_remote: support graceful stops (bsc#981489) - crmd: support graceful pacemaker_remote stops (bsc#981489) - lrmd,liblrmd: add lrmd IPC operations for requesting and acknowledging shutdown (bsc#981489) - fencing: Support concurrent fencing actions on each device (bsc#938545) - pengine: Support concurrent fencing (bsc#938545) - Upstream version cs: c54f10440a7bba5a1491ea8506925c7d6d9ca2a3 ------------------------------------------------------------------- Thu Jan 21 15:11:39 UTC 2016 - ygao@suse.com - Update to version 1.1.14 - fencing: free only after last usage makes coverity happy - crmd: 0 is a valid fd makes coverity happy - tools: crm_resource compile issue - crm_resource --list-agents: don't print uninitialized memory - extra: update SNMP MIB with new pacemakerTrap OID containing pacemakerNotificationTrap - fencing, libfencing: remap fence agent error codes before async callback (bsc#962309) - libfencing: topology is not used - fencing: topology is not used - libcrmcommon: when caching attrd connection, cache connection flags as well - cts: Plugin-based cluster has its own PacemakerUp pattern - Upstream version cs: 70404b0e5ecbe5e51d07c6c2233eea640cbd720a (Pacemaker-1.1.14) ------------------------------------------------------------------- Fri Dec 18 17:48:52 UTC 2015 - ygao@suse.com - Rebase: * bug-806256_pacemaker-log-level-notice.patch * bug-728579_pacemaker-stonith-dev-id.patch - core: Mistake of the parameter order. - RHBZ#1287315: stonithd: Trigger cib_devices_update in case of deletion of just an attribute - CLBZ#5253: stonithd: Do not intermingle stdout & stderr coming from stonith-RAs (bsc#962309) - resources: allow for top output with or without percent sign in HealthCPU - controld: Return error when DLM status is wait fencing (RHBZ#1268313) - crmd: Don't update cluster/peer CIB fields when processing stonith results - Upstream version cs: e16447de89d25db3b3578c7e080c1a4c2f85806f ------------------------------------------------------------------- Wed Dec 2 11:55:11 UTC 2015 - ygao@suse.com - ping: Clarify the description of host_list parameter in ping metadata (bsc#956459) - systemd: Prevent segfaul when logging failed operations - systemd: The underlying systemd calls default to 90s for their timeouts - cib: Do not terminate due to badly behaving clients - Correctly check for recycled PIDs in crm_pid_active - crmd: Consider only fence and shutdown transition events to be "down" events - pengine: Support of multiple-active=block for resource groups (bsc#942491) - crmd: use 300s timeout with notification scripts, not 300ms - stonith_admin: add ability to target topology by pattern or attribute - fencing,libfencing: add API support for targeting topology by pattern or attribute - Upstream version cs: a7d6e6b56fd2779a07970df7de4e0fb0b22742d3 ------------------------------------------------------------------- Mon Nov 16 17:49:49 UTC 2015 - ygao@suse.com - crm_report: Call "crm report" if available (bsc#950483) ------------------------------------------------------------------- Thu Nov 12 11:18:49 UTC 2015 - ygao@suse.com - fencing: crm_resource --show-metadata drops documentation strings for fencing agents (bsc#950375) - Upstream version cs: 6f22ad7312b703289da5d24e918a78117a0825e1 ------------------------------------------------------------------- Thu Nov 5 22:51:27 UTC 2015 - ygao@suse.com - CTS: add "try except" to deal with errors that raw_input gets EOFError and add "--yes" to skip interaction (bsc#953192) - Upstream version cs: 577898dc4ac1efdb5989a996cb5f9ac60ed2eb7a ------------------------------------------------------------------- Wed Nov 4 11:42:56 UTC 2015 - ygao@suse.com - crm_resource: Correctly clean up all children for anonymous cloned groups (bsc#950128) - services: Directly add operations into in-flight list in the execution functions (bsc#952874) - systemd: Return PCMK_OCF_UNKNOWN_ERROR instead of PCMK_OCF_NOT_INSTALLED for uncertain errors on LoadUnit (bsc#952884) - systemd: Return FALSE for systemd_mask_error() if it only maps an error instead of masking one (bsc#952884) - dbus: Correctly return the result for pcmk_dbus_find_error() no matter if there's a DBusError passed in (bsc#952884) - services: Prevent potential use-after-free when trying to add an operation into in-flight list (bsc#952874) - systemd: Directly return an error if the connection to System DBus is closed (bsc#952874) - services: Correctly determine if operations are in-flight (bsc#952874) - systemd: Reconnect to System DBus if the connection is closed (bsc#931271) - lrmd: Finalize all pending and recurring operations when cleaning up a resource (bsc#950450) - fencing: don't poke peers when remote node status changes - attrd: don't try to communicate with attrd on a remote node - libcluster,crmd: never update join info or processes for remote nodes - cib: Increased paranoia when peer updates fail to apply in compatability mode - crmd: do not accept remote peer data as enough for startup - crmd: remote nodes can't be DC so never add them to stonith cleanup list - libcluster: never autoreap from remote peer cache - crmd: improve generation of node state - CTS: get Reattach test working again and up-to-date (bsc#953192) - libcommon: Ignore CDATA of metadata of the resource. - cib: Downgrade the log message on forwarding CRM_OP_NOOP requests from INFO to DEBUG (bsc#949267) - fencing: Return a provider for the internal fencing agent "#watchdog" instead of logging an error (bsc#949441) * Drop bug-949441_pacemaker-fencing-provider-internal.patch which has been merged upstream - Upstream version cs: 7bd00f3f3010bc10d13b87e2ff501b597ea6ca03 ------------------------------------------------------------------- Thu Oct 8 10:31:45 UTC 2015 - ygao@suse.com - fencing: Return a provider for the internal fencing agent "#watchdog" instead of logging an error (bsc#949441) * bug-949441_pacemaker-fencing-provider-internal.patch - spec: Move the normal resource agents into pacemaker-cli package (bsc#947197) - spec: Move logrotate configuration file into pacemaker-cli package (bsc#947197) - spec: Move attrd_updater, crm_attribute and crm_master into pacemaker-cli package (bsc#947197) - spec: Move xml schema files and PCMK-MIB.txt into pacemaker-cli package (bsc#947197) ------------------------------------------------------------------- Thu Oct 8 10:29:54 UTC 2015 - ygao@suse.com - crm_resource: Correctly observe --force when deleting and updating attributes (bsc#948197) - crm_resource: Correctly delete existing meta attributes regardless of their position in the heirarchy (bsc#948197) - fencing,libfencing: use level tag consistently in internal XML - crm_resource: Correctly update existing meta attributes regardless of their position in the heirarchy (bsc#948197) - crmd: properly detect CIB update failures for remote nodes - cibadmin: Prevent potential use-of-NULL in print_xml_output() (bsc#947180) - Fencing: The device list is expected to be comma separated - Tools: Do not send command lines to syslog (bsc#986676) - Tools: Repair the logging of 'interesting' command-lines - PE: resolve memory leak - libfencing: make topology level removal compatible with latest stonithd API - fencing: parse from correct XML location when registering/unregistering topology - tools: improve error handling when modifying configuration - crmd,libcrmcommon,libservices,tools: potential memory leaks (bsc#948197) - crmd,fencing: avoid potential null dereference in string searches - typo: OPSTIONS -> OPTIONS in crm_mon.sysconfig - cib: Check if the configuration changes with cib_config_changed() only for v1 diffs (bsc#946224) - liblrmd: don't print error if remote key environment variable unset - libcib: properly handle temporary file - libcrmcommon: better validation of environment variable value - crmd: avoid potential null dereference - libcib: potential user input overflow - controld: replace bashism - PE: drop probe_complete actions - PE: Allow resources to start and stop as soon as their state is known on all nodes - Fencing: Correctly construct simple topologies with stonith_admin - Fencing: Correctly remove topologies and detect invalid entries - remote: Revise a misleading message in the ocf:pacemaker:remote resource agent (bsc#946332, bsc#967383) * Drop bug-946332_pacemaker-resource-agent-remote-message.patch which has been merged upstream - remote: Correctly display the usage of the ocf:pacemaker:remote resource agent (bsc#946332, bsc#967383) * Drop bug-946332_pacemaker-resource-agent-remote-usage.patch which has been merged upstream - libcib: find_nvpair_attr_delegate: check buffer overflow - libcib: find_nvpair_attr_delegate: check alloc failure - Upstream version cs: 7906df92a713052db1edae851579ea229d8a5393 ------------------------------------------------------------------- Fri Sep 18 08:22:19 UTC 2015 - ygao@suse.com - remote: Revise a misleading message in the ocf:pacemaker:remote resource agent (bsc#946332, bsc#967383) * bug-946332_pacemaker-resource-agent-remote-message.patch - remote: Correctly display the usage of the ocf:pacemaker:remote resource agent (bsc#946332, bsc#967383) * bug-946332_pacemaker-resource-agent-remote-usage.patch ------------------------------------------------------------------- Thu Sep 17 09:26:52 UTC 2015 - ygao@suse.com - crm_resource: Allow the resource configuration to be modified for --force-{check,start,..} calls - Fencing: Correctly parse 'by-attribute' fencing levels - Fencing: Use dedicated fields for fencing level patterns and attribute matching - pacemaker_remote: memory leak in ipc_proxy_dispatch() - crmd: don't add node ID to proxied remote node requests for attrd - membership: Safely autoreap nodes without code duplication - PE: Resolve memory leak - pacemakerd: Do not forget about nodes that leave the cluster - corosync: Display node state and quorum data if available - crmd: Implement reliable event notifications - extra: Added snmp trap of type "NOTIFICATION-TYPE" to MIB and snmp helper script - Date: Correctly set time from seconds-since-epoch - PE: Bug cl#5247 - Imply resources running on a container are stopped when the container is stopped - Upstream version cs: 20c2178f076ff32fdf9ba9a467c193b8dac2f9e5 ------------------------------------------------------------------- Wed Sep 9 13:24:17 UTC 2015 - ygao@suse.com - lrmd: Upgrade start/finish log tags from INFO to NOTICE (bsc#943295) * bug-943295_pacemaker-lrmd-log-notice.patch ------------------------------------------------------------------- Fri Aug 28 11:49:26 UTC 2015 - ygao@suse.com - services: Reduce severity of noisy log messages (bsc#943319) - xml: Mark xml nodes as dirty if any children move (bsc#942382) - pengine: The failed action of the resource that occurred in shutdown is not displayed. - tools: memory leak in crm_resource - crm_resource: Correctly check if a resource is unmanaged or has a target-role - fencing: add ability to target by attribute - crmd: Resolve memory leak in remote_proxy_cb() - ipc: Do not constantly increase suggested size for PCMK_ipc_buffer every time we find it's insufficient (bsc#940992) - Upstream version cs: e8888b98cd7dbdfb986c1a46bdb4b113b96e914f ------------------------------------------------------------------- Thu Aug 13 12:56:09 UTC 2015 - ygao@suse.com - crm_resource: Indicate common reasons why a resource may not start after a cleanup - pengine: Ensure fencing of the DC precedes the STONITH_DONE operation - crm_resource: Compensate for -C and -p being called with the child resource for clones - ipc: Fix output formats (bsc#940992) - fencing: ensure "required"/"automatic" only apply to "on" actions - fencing: Remove unnecessary casts (bsc#940711) - crm_simulate: Prevent segfault on arches with 64bit time_t (bsc#940711) - ipc: Correctly compare values for the size of ipc buffer and prevent suggesting a negative value when it's insufficient (bsc#940992) - xml: Reduce severity of noisy log message (bsc#950551) - crm_resource: Correctly clean up failcounts for inactive anonymous clones - crmd: Avoid cancelling transitions when we tell nodes to shutdown - systemd: drop legacy SysVStartPriority from service units - crm: Set the attribute from remote node. - Upstream version cs: 4ad07f5f464c84b80c739922bd9fd708e8d36e5a ------------------------------------------------------------------- Thu Jul 30 09:46:31 UTC 2015 - ygao@suse.com - dbus: Remove redundant ref/unref of pending call records - stonithd: potential device list corruption - cib: don't destroy peer cache in peer status callback - systemd: Ensure pending structs are correctly unreferenced - systemd: Track pending operations so they can be safely cancelled - crmd: memory leak in CIB callbacks - attrd: memory leak in legacy attrd's CIB callback - attrd: memory leak in CIB callback - libcib: allow free function for user data supplied to CIB callbacks - xml: Prevent use-of-NULL in crm_xml_dump() - crm_mon: Memory leaks - Upstream version cs: 3fb7f71bf9ca2a3d6e1187fa3b74c7841d6af5a7 ------------------------------------------------------------------- Wed Jul 22 10:22:39 UTC 2015 - ygao@suse.com - pengine: properly handle blocked clone actions - pengine: Correctly bypass fencing for resources that do not require it - libservices: potential memory leaks when executing systemd actions - libservices: memory leak in DBus-related log messages - crmd: memory leaks in recurring operation history - pengine: Addtion of the clone resource clone-min metadata option - Upstream version cs: 3d781d3fe61504c841e3c89b2e8ccff8195d4af6 ------------------------------------------------------------------- Mon Jul 20 08:46:38 UTC 2015 - ygao@suse.com - Rebase bug-806256_pacemaker-log-level-notice.patch - Build: Update version numbers to 1.1.13 in version.m4 and sample spec file - fencing: Broadcast uname as soon as a peer's state has changed (bsc#938410) - fencing: remap sequential topology reboots to all-off-then-all-on - fencing: remap sequential topology reboots to off - fencing: support remapping reboots when processing fencing query results - fencing: treat remapped reboots as reboots when appropriate - fencing: support remapping reboots in fencing query results - crmd: check for errors when searching for CIB daemon's load file - libcib,libfencing,tools: memory leaks from xmlGetNodePath() - attrd: use-after-free in attrd_client_update() - fencing: memory leak when searching for a target's fencing topology - lrmd: memory leak when freeing command structure - libcluster: work around peer cache destruction in client status callbacks - fencing: add support for operation phases to device properties - fencing: add support for operation phases - crmd: Do not enforce restart for resources that support reload and don't support any unique parameters - crmd: Do not enforce restart for changes of non-unique parameters - cts: change the stack from openais (white-tank) to corosync (plugin v0) in set_stack of environment.py (bsc#936149) - Upstream version cs: bd8c4b5a30c9cacd0257fe50ab50d3a6436db42a ------------------------------------------------------------------- Fri Jul 10 14:03:00 UTC 2015 - ygao@suse.com - crmd: Make resource_supports_action() really general - crmd: Remove a duplicate invocation of snprintf() - pengine: clear failures after reconnect_interval expires - RA: Prevent an error message on stopping "Dummy" resource - PE: Ignore comment blocks when unpacking the cib - fencing: use "off" timeout when remapping "reboot" to "off" - fencing: do not set call options twice in query - crmd: don't synthesize op failures during shutdown - lrmd: prevent double free after unregistering stonith device for monitoring (bsc#1035822) - libcommon: crm_ipc_get_fd() must set errno on error - PE: Record the execution date in PE files so we can re-use it in crm_simulate - pengine: allow guest remote nodes using containers/vms to be nested in a group resource - pengine: Resolve memory leak in expand_notification_data() - cib: Prevent use-after-free and return -EINVAL when attempting to delete the whole "/cib" (bsc#934609) - cib: Prevent use-after-free when invoking "cibadmin --delete-all --xpath" (bsc#934609) - crmd: Prevent segfault when erasing resources from disconnected remote nodes - pengine: do not timeout remote node recurring monitor op failure until after fencing - crmd: No need to re-process actions we've already marked as failed - crmd: All pseduo ops must be allowed to complete - crm_simulate: Optionally include action IDs in the dot graph - fencing: properly decide whether a topology fencing device has been found - fencing: properly sort peers by number of fencing devices found - crmd: Record actions in the CIB as failed if we cannot execute them - pengine: do not stop notify a fenced node that the rscs on the fenced node stopped - pengine: fixes segfault in pengine when fencing remote node - remote: do not fail operations because of a migration - remote: handle merging duplicate recurring monitors for connection rsc - pengine: cl#5235 - Prevent graph loops that can be introduced by "load_stopped -> migrate_to" ordering - PE: Exclude nodes which don't match any exclusive discovery rules - Log: improve IPC connection error messages - Log: make syslog less noisy - lrmd: cancel currently pending STONITH op if stonithd connection is lost - fencing: Correct the all_topology_devices_found() implementation - crmd: Pre-emptively fail in-flight actions when lrmd connections fail - PE: Include a list of all and available nodes with notifications - lrmd: set recv timeout upper bound for tls connections - services: prevent recurring monitors from colliding with other operations - crmd: handle resources named the same as cluster nodes - crmd: make actions incompletable after remote-node fails - crm_simulate: Correctly invoke the policy engine library - libcluster: overhaul peer cache management - crmd: Prevent use-of-NULL - crm_report: Enable password sanitizing by default - PE: Keep the size of the secure list small in the cib - PE: Reduce noise when sanitized policy engine files are replayed - attrd: handle removals of cluster peers and remote peers consistently - libcluster: avoid possibility of invalidating a hash table iterator - membership: Ensure peer cache management supports all stacks - libcluster: do not re-add peer entries for exiting nodes - PE: Skip unrunnable actions when one-or-more is in effect - PE: Ensure recurring monitor operations are cancelled when clone instances are de-allocated - systemd: No need to trigger assert if cannot obtain an unit (bsc#931271) - dbus: Use operation timeout for dbus reply timeout (bsc#931271) - fencing: correct choice of fencing topology level - crm_shadow: Fix "crm_shadow --diff" (bsc#931166) - cib: Calculate and add digest for a patchset after accepting changes for the target xml (bsc#931114) - remote: enable reloads for select remote connection options - crmd: fence remote node after connection failure only once - remote: reconnect_delay option for remote nodes to delay reconnect after fence - tools: ignore stderr from crm_mon when running via systemd - xml: update RelaxNG schema for crm_mon XML output - Fencing: Correctly handle changes to the fencing topology - fencing: Allow complex configurations to apply to multiple hosts - tools: Update regression tests for ACL fixes - tools: improve failed actions display in crm_mon - tools: implement printing tickets in crm_mon HTML/XML output - tools: implement printing negative location constraints in crm_mon HTML/XML output - tools: implement printing operations and failcounts in crm_mon HTML/XML output - tools: implement printing node attributes in crm_mon HTML/XML output - tools: improve cluster summary display in crm_mon - tools: improve guest node output in crm_mon - tools: prettify plaintext output - tools: prettify crm_mon HTML output - tools: prettify crm_mon --simple-status output - crm_report: Allow log file discovery to be disabled - crm_resource: Support debugging promote and demote operations - fencing: Allow semi-colon delimiter for pcmk_host_list - Upstream version cs: ef86b494a411ce920eb18b9b6541a3a0425a41e0 ------------------------------------------------------------------- Fri Jul 10 11:06:45 UTC 2015 - ygao@suse.com - Update to version 1.1.13 - Build: Update library versions - Build: Move new struct members to the end for backwards compatibility - Fencing: Gracefully handle invalid metadata from agents - Upstream version cs: 6052cd16c2f455809f8088af76ce86483bf98353 (Pacemaker-1.1.13) ------------------------------------------------------------------- Wed Jun 3 14:40:55 UTC 2015 - ygao@suse.com - pengine: cl#5130 - Only check the capacities of the nodes that are allowed to run the resource (FATE#313105) - cib: Correctly add "update-origin", "update-client" and "update-user" attributes for cib (bsc#931177) - fencing : Correction of the log output of the practice node of stonith. - Upstream version cs: a0449d443fa5c834207def5fe36a117a53c334d6 ------------------------------------------------------------------- Mon May 18 08:23:52 UTC 2015 - ygao@suse.com - xml: Correctly create v1 patchsets when deleting attributes (bsc#931170) - xml: Prevent potential use-of-NULL (bsc#929960) - Upstream version cs: 677f53216662d121a47b4c7683ec3749c614437e ------------------------------------------------------------------- Fri May 15 13:07:29 UTC 2015 - ygao@suse.com - cib: No need to enforce cib digests for v2 diffs in legacy mode (bsc#929960) - cib: Ability to upgrade cib validation schema in legacy mode (bsc#929960) - xml: Do not dump deleted attributes (bsc#929960) - cib: Revert d153b86 to instantly get cib synchronized in legacy mode (bsc#929960) - cib: Enable cib legacy mode for plugin-based clusters (bsc#929960) - Build: CRM_STATE_DIR is created dynamically when needed - Tools: Repair expected output for ACLs - Build: Prevent rpm packaging conflicts - pengine: cl#5130 - Choose nodes capable of running all the colocated utilization resources (FATE#313105) * Drop obsolete pacemaker-colocated-utilization.patch - Upstream version cs: 0b79da3e73fe2ced8b9602aa59d55873d857b3e7 ------------------------------------------------------------------- Mon May 11 10:51:23 UTC 2015 - ygao@suse.com - crmd/pengine: implement on-fail=ignore without allow-fail (bsc#981731) - pengine: Prevent use-after-free in sort_rsc_process_order() (bsc#930368) - crmd: Prevent the old version of DC from being fenced when it shuts down for rolling-upgrade (bsc#929955) - xml: Fix upgrade-1.3.xsl to correctly transform ACL rules with "attribute" (bsc#929681) - crmd/pengine: handle on-fail=ignore properly (bsc#981731) - crmd: respect start-failure-is-fatal even for artifically injected events (bsc#981731) - crmd: don't update fail count twice for same failure (bsc#950450, bsc#981731) - crmd: report operation rc as advertised instead of status (bsc#981731) - xml: add RelaxNG schema for crm_mon XML output - Upstream version cs: c9ad9cd503f071391efe17d3c45d578d6f152acd ------------------------------------------------------------------- Wed May 6 09:57:05 UTC 2015 - ygao@suse.com - pengine: process rsc parameters during unpack (bsc#929842) - pengine: handle setting container attribute in the correct location (bsc#929842) - lrmd: move isolation wrappers to hidden directory (bsc#929842) - mainloop: resolves race condition mainloop poll involving modification of ipc connections (bsc#929835) - Upstream version cs: f47ea56b4ec23c7d8e0c129535a9b49cdbd80482 ------------------------------------------------------------------- Thu Apr 30 11:31:46 UTC 2015 - ygao@suse.com - fencing: Correctly make args for fencing agents (bsc#925751) - fence_legacy: Avoid passing 'port' to cluster-glue agents (bsc#925751) - corosync: Bug cl#5232 - Somewhat gracefully handle nodes with invalid UUIDs - HealthSMART: OCF_RESKEY_drives could contain multiple values - pengine: restart master instances on instance attribute changes - spec: move report.common and report.collector to cli package - pengine: only kill process group on timeout when on-fail does not equal block. - lrmd: Hint to child processes that using sd_notify is not required (bsc#956626) - Agents: Fix the HealthSMART metadata - make targetted reprobe for remote node work, crm_resource -C -N <remote node> - crmd: include remote nodes when forcing cluster wide resource reprobe - crm_mon: Support of the start from systemd of crm_mon - Upstream version cs: 195559d47b0e9186b5d131981f6f7fee96ecad76 ------------------------------------------------------------------- Mon Apr 13 08:57:35 UTC 2015 - ygao@suse.com - Rebase: * pacemaker-NodeUtilization-RA.patch - docker-wrapper: per resource logging of lrmd and remote client tool - pengine: Avoid trailing whitespaces when printing resource state - pengine: cl#5207 - Display "UNCLEAN" for resources running on unclean offline nodes - docker-wrapper: add reuse option for presistent containers - docker-wrapper: only override exit reason if docker daemon is inactive - pengine: handle isolated resource scenario, cloned group of isolated resources - pengine: handle cloned isolated resources in a sane way - crm_report: More patterns for finding logs on non-DC nodes - lrmd: Use original timestamps when reporting multi-step systemd operations - lrmd: report original timeout when systemd operation completes - systemd: improve unit files - PE: Turn off legacy unpacking of resource options into the meta hashtable - acl: Do not delay evaluation of added nodes in some situations (bsc#927828: CVE-2015-1867) - docker-wrapper: add demote, promote, and notify actions for multistate rsc - docker-wrapper: properly forward key value pairs during isolation - attrd: fixes double free in attrd legacy - lrmd: do not let isolation wrappers inherit the wrapper attribute, or it makes a loop - pengine: pass node name of pcmk host into isolated resources - lrmd: isolation wrappers are moved from containers to isolation OCF namepace - pengine: Fix colocation with unmanaged resources - stonith-ng: Correctly track node state - stonith-ng: No reply is needed for CRM_OP_RM_NODE_CACHE - crm_resource: Fix output format of the warning from --force-start - crm_resource: Display the locations of all anonymous clone children when supplying the children's common ID (bsc#870339) - cib: Correctly track node state - Upstream version cs: 7a2e3ae6b736621933b24f4cdc9ea260021d878d ------------------------------------------------------------------- Fri Mar 27 15:09:25 UTC 2015 - ygao@suse.com - Rebase: * bug-806256_pacemaker-log-level-notice.patch * pacemaker-NodeUtilization-RA.patch * pacemaker-colocated-utilization.patch - crmd: All peers need to track node shutdown requests (bsc#917625) - crmd: Cached copies of transient attributes cease to be valid once a node leaves the membership (bsc#917625) - crmd: Wait for all pending operations to complete before poking the policy engine - lrmd: preserve exit reason string when isolation wrappers are in use - docker-wrapper: properly separate docker and resource specific attributes - docker-wrapper: set authkey file permissions and properly set container 'node name' during start - systemd: Trick systemd into not stopping our services before us during shutdown (bsc#913251) - mcp: Allow a configurable delay when debugging shutdown issues - systemd: Kindly ask dbus NOT to kill the process if the dbus connection fails (bsc#931271) - systemd: Tell systemd not to take DBus down from underneath us (bsc#913251) - cib: Correctly set up signal handlers - PE: Do not record duplicate copies of the failed actions - lrmd: enable ipc proxy for docker-wrapper privileged mode - lrmd: properly handle poke requests in lrmd client when using ipc - spec: add docker-wrapper directory to spec file - pengine: disable migrations for resources with isolation containers - pengine: disable reloading of resources within isolated container wrappers - pengine: ability to launch resources in isolated containers - extra: docker container tech wrapper script for pcmk remote - tools: crm_mon prints Stopped clones only if --inactive was specified - tools: display node names more consistently in crm_mon output - tools: Improve crm_mon output with certain option combinations - tools: make crm_mon last updated header consistent across formats - remote: pcmk remote client tool for use with container wrapper script - crmd: Reset stonith failcount to recover transitioner when the node rejoins (bsc#921102) - systemd: fix crash caused when canceling in-flight operation - Upstream version cs: 3e93bc1dd6c1c3f2ee101f754567aa007b24bb5e ------------------------------------------------------------------- Fri Mar 13 09:38:52 UTC 2015 - ygao@suse.com - PE: Support non-actionable degraded mode for OCF - lrmd, services: interpret LSB status codes properly - tools: add crm_resource --wait option - tools: attrd_updater supports --query and --all options - attrd: attrd supports queries - pengine: remove unnecessary whitespace from notify environment variables - tools: free allocated memory correctly in crm_resource - attrd: properly write attributes for peers once uuid is discovered - cib: assignment where comparison intended - attrd: only send ack for recognized operations - attrd: make each client operation handler its own function - tools: attrd_updater supports private attributes - attrd supports private attributes (not written to CIB) - pengine: Reschedule specific monitors when resource definitions have changed (bsc#914761) - pengine: Resolve memory leaks (bsc#914761) - Upstream version cs: e32080b460f81486b85d08ec958582b3e72d858c ------------------------------------------------------------------- Fri Feb 27 09:44:50 UTC 2015 - ygao@suse.com - properly record stop args after reload - pengine: Do not reschedule monitors that are no longer needed while resource definitions have changed (bsc#914761) - Build: Fix building when heartbeat support is disabled - fencing: cl#5134 - Support random fencing delay to avoid double fencing (FATE#317136, FATE#318381, bsc#970733) - extra: Use crm_simulate for the showscores script - attrd: refresh should force an immediate write-out of all attributes - service: Leave synchronous calls to be free'd by the caller - service: Additional pointer safety - crmd: Add "on_node" attribute for *_last_failure_0 lrm resource operations (bsc#918839) - crm_mon: Fill CRM_notify_node in traps with node's uname rather than node's id if possible (bsc#918839) - pengine: ensure if B is colocated with A, B can never run without A - cib: automatically update .sig files in file backends when clients update live CIB - cib: file-based backends now detect whether file is live CIB - lrmd: add back support for class heartbeat agents - lrmd: provide fake meta data for ra class heartbeat - crm_resource: Use-after-free when restarting a resource - cli: Ensure subsequent unknown long options are correctly detected - crmd: Bug rhbz#1181824 - Ensure the DC can be reliably fenced - attrd_updater: Allow attributes to be set for other nodes - native_print: report target-role as well - membership: fix crm_update_peer_proc to NOT ignore flags if partially set - crmd: don't trigger a stonith_reconnect if no longer required - Upstream version cs: fa1311a0159588385f24cd8c3530233f4b90f3a9 ------------------------------------------------------------------- Tue Feb 24 12:26:27 UTC 2015 - rguenther@suse.com - Add pacemaker-Wno-format-signedness.patch to disable -Wformat-signedness added by GCC 5 and enabled by -Wformat=2 ------------------------------------------------------------------- Fri Jan 30 10:37:57 UTC 2015 - ygao@suse.com - pengine: Handle ordering between stateful and migratable resources (bsc#881160) - crmd: Check "watchdog" daemon option before setting "have-watchdog" property - crmd: Correctly add the local option that validates against schema for pengine to calculate - Upstream version cs: 3021d731fdeba29e1136c117a62f3201785be6ea ------------------------------------------------------------------- Fri Jan 16 12:44:34 UTC 2015 - ygao@suse.com - crm_standby: update accepted options to match documentation - cts: Fix bashisms in LSBDummy script - pengine: fixes pengine crash with orphaned remote node connection resource - PE: Do not aggregate children in a pending state into the started/stopped/etc lists - acl: Correctly implement the 'reference' acl directive - Upstream version cs: 03ec61210c82471161f6cf3366e8afc30d737d67 ------------------------------------------------------------------- Mon Jan 12 11:05:14 UTC 2015 - ygao@suse.com - pengine: imply stop in container node resources when host node is fenced - Build: Fix building if ncurses doesn't provide pkg-config files - xml: Do not change the ordering of properties when applying v1 cib diffs - xml: Do not prune leaves from v1 cib diffs that are being created with digests - Upstream version cs: 4109f1539a625bfa3e884d19ab2fc42d7a4daec4 ------------------------------------------------------------------- Fri Jan 9 12:10:09 UTC 2015 - ygao@suse.com - xml: add 'require-all' to xml schema for constraints - pengine: require-all feature for ordered clones - xml: Ensure ACLs are reapplied before calculating what a replace operation changed - cib: Ensure file-based backends treat '-o section' consistently with the native backend - crmd: resolves memory leak in crmd. - pacemakerd: resolves memory leak of xml structure in pacemakerd - ipc: use server enforced buffer during ipc client send - pengine: do not recording pending migrate_to ops for pacemaker remote - crmd: never stop recurring monitor ops for pcmk remote during incomplete migration - pengine: prefer migration target for remote node connections - pengine: properly handle ordering during remote connection partial migration - crmd: Ensure remote connection resources timeout properly during 'migrate_from' action - crmd: gracefully handle remote node disconnects during op execution - pengine: Fence baremetal remote when recurring monitor op fails - crmd: Handle remote connection failures while executing ops on remote connection - remote: Treat recv msg timeout as a disconnect - pengine: properly recover remote-nodes when cluster-node proxy goes offline - systemd: Correctly handle long running stop actions (bsc#948989) - crm_resource: Include group members when calculating the next timeout - crm_resource: Ensure --restart sets/clears meta attributes - crm_resource: Clean up memory in --restart error paths - Build: Use a wrapper for realloc() which calls abort when we run out of memory - services: Prevent potential use-of-NULL - crm_resource: Memory leak in error paths - crmd: Mistake of the set of the variable. - Upstream version cs: d0139e9a00c32b9bc1d7d206d6210126045b49be ------------------------------------------------------------------- Fri Nov 21 15:38:57 UTC 2014 - ygao@suse.com - crm_node: Correctly remove nodes from the CIB by nodeid - fencing: Allow nodes to be purged from the member cache - xml: cl#5231 - Unset the deleted attributes in the resulting diffs (bsc#905641, bsc#967254) - fencing: stonith -l option for custom log - cib: Avoid nodeid conflicts we don't care about - Membership: Detect and resolve nodes that change their ID - attrd: Simplify how node deletions happen - attrd: Clean out the node cache when requested by the admin - crm_resource: Allow resource restart operations to be node specific - crm_resource: Implement --timeout for resource restart operations - crm_resource: Calculate how long to wait for a restart based on the resource timeouts - crm_resource: Implement an intelligent restart capability - crm_failcount: Better error reporting when no resource is specified - cib: tls sock cleanup for remote cib connections - Upstream version cs: 7dd90226e10035340d6707d9fd2b04f2e0141194 ------------------------------------------------------------------- Fri Nov 14 09:44:03 UTC 2014 - ygao@suse.com - pengine: make resource-discovery work with location rules - services: Conditionally clean up services actions - Upstream version cs: 5093c095c5f517ab58375b2b69ecfc5516c86b5b ------------------------------------------------------------------- Fri Nov 14 08:58:40 UTC 2014 - dimstar@opensuse.org - Drop conditional BuildRequires of system-devel: it was required one line lower unconditionally. - Replace remaining systemd-devel BuildRequires with pkgconfig(systemd). ------------------------------------------------------------------- Mon Nov 10 11:36:27 UTC 2014 - ygao@suse.com - pengine: exclusive discovery implies rsc is only allowed on exclusive subset of nodes - dbus: handle dispatch queue when multiple replies need to be processed - dbus ref count leaks - dbus: Ensure both the read and write queues get dispatched - systemd: Ensure failed monitor operations always return - Perform systemd reloads asynchronously - Upstream version cs: cd7c9abbcfc620aa84194ee07aade142240db15a ------------------------------------------------------------------- Tue Oct 28 14:21:58 UTC 2014 - ygao@suse.com - watchdog: Allow startup without sbd - service: Prevent potential use-of-NULL in metadata lookups - dbus: Fail gracefully if malloc fails - crmd: Prevent use-of-NULL during reprobe - crmd: Correctly obtain SBD_WATCHDOG_TIMEOUT from the environment - pengine: add #kind=remote for baremetal remote-nodes - Officially support the resource-discovery attribute for location constraints - xml: fix xml.c memleak in "missing upgrade XSLT" case - Prefer to inherit the watchdog timeout from SBD - stonith-ng: Correctly observe the watchdog completion timeout - sbd: Attempt to locate sbd via its pid file - Fencing: If configured, wait stonith-watchdog-timer seconds for self-fencing to complete - Resolve some potential issues spotted by clang and coverity - PE: Watchdog integration is sufficient for fencing - Fencing: Correctly wait for self-fencing to occur when the watchdog is in use - Fencing: Ensure the hostlist parameter is set for watchdog agents - crm_resource: Allow -M location to be reinforced with -f even if rsc is already on node - mcp: Turn on sbd integration if pacemakerd finds it running - pengine: Fix the behaviors of multi-state resources with asymmetrical ordering (bsc#899403) - lrmd: Prevent glib assert triggered by timers being removed from mainloop more than once - cib: Improved tracing of callbacks - Upstream version cs: a9c81774b89f21f990be255f9862446d1a38afee ------------------------------------------------------------------- Fri Oct 17 12:55:03 UTC 2014 - ygao@suse.com - remote: advertise default port correctly, further adjust default timeout periods - remote: report timestamps for remote connection resource operations - pengine: allow remote-nodes to be placed in maintenance mode - xml: Prevent assert errors in crm_element_value() on applying a patch without version information (bsc#900296) - pengine: prevent disabling rsc discovery per node in certain situations - crm_resource: avoid deletion of lrm cache on node with resource discovery disabled. - pengine: per-node control over resource discovery - mainloop: fix potential segfault for fd = 0 - ping: Correct metadata for attempts parameter (bsc#899324) - ping: Pass extra options to fping (bsc#899324) - ping: Remove incorrectly advertised migrate_to|migrate_from (bsc#899324) - ping: Add use_fping parameter (bsc#899324) - ping: Correctly advertise multiplier default (bsc#899324) - pengine: resource discovery mode for location constraints - cib: Ensure upgrade operations from a non-DC get an acknowledgement - lrmd: cancel pending async connection during disconnect - systemd: Ensure we don't call dbus_message_unref() with NULL - systemd: Perform monitor operations without blocking - dbus: Notice when dbus connections get disabled - pengine: allow baremetal remote-nodes to recover without requiring fencing when cluster-node fails - crmd: do not remove connection resources during re-probe - pengine: only fence baremetal remote when connection can fails or can not be recovered - ipc: properly allocate server enforced buffer size on client - remote: advertise more reasonable timeout periods for remote-nodes connections - crmd: properly update job limit for baremetal remote-nodes - crmd: Remote-node throttle jobs count towards cluster-node hosting conneciton rsc - crmd: Ensure throttle_mode() does something on Linux - systemd: Perform actions asynchronously (bsc#956500) - crm_attribute: Correctly update config options when -o crm_config is specified - crmd: Fixes crash when remote connection migration fails - pengine: fixes segfault caused by malformed log warning - Allow fail-counts to be removed en-mass when the new attrd is in operation - fencing: Use the correct define for origin - fencing: Reject actions for devices that haven't been explicitly registered yet - cts : change default cluster stack from openais to corosync (bsc#894871) - crm_resource: Ensure fail-counts are purged when we redetect the state of all resources - corosync: Avoid unnecessary repeated CMAP API calls (bsc#870771) - cib: Do not update on disk backups if we're just querying them - mcp: Allow orphan children to initiate node panic via SIGQUIT - Support machine panics for some kinds of errors (via sbd if available) - Fencing: Allow the cluster to recover resources if the watchdog is in use - crmd: If configured, trigger the watchdog immediately if we loose quorum and no-quorum-policy=suicide - crmd: Ensure a watchdog device is in use if stonith-watchdog-timeout is configured - Fencing: Advertise the watchdog device for fencing operations - PE: Allow natural ordering of colocation sets - stonith-ng: Reset mainloop source IDs after removing them - Fencing: Force 'stonith-ng' as the system name - crm_mon: Repair notification delivery when the v2 patch format is in use (bsc#909286) - mcp: Teach pacemakerd to ignore SIGHUP - Rebase bug-806256_pacemaker-log-level-notice.patch - Upstream version cs: 28aee948cebd51305a2cdc3ec4f94252ce975655 ------------------------------------------------------------------- Fri Sep 5 07:33:50 UTC 2014 - ygao@suse.com - cluster: Invoke crm_remove_conflicting_peer() only when the new node's uname is being assigned in the node cache (bnc#870771) - mcp: set environment variable which the cluster-glue module needs - services: Refector services os execute function - Upstream version cs: 266d5c2484716111616bcd0f32d9b94bc49baa77 ------------------------------------------------------------------- Wed Aug 27 11:27:38 UTC 2014 - ygao@suse.com - logging: correctly initialized default log file (bnc#883567) - Upstream version cs: 362f181a387e0a0a9f04e0da84d6c5906d71dd20 ------------------------------------------------------------------- Mon Aug 18 08:02:31 UTC 2014 - ygao@suse.com - crm_resource: Print a clear message if the attribute is not found for the resource (bnc#890400) - lrmd: fix rescheduling of systemd monitor op during start - services: restore SIGPIPE default behavior for child processes - lrmd: prevent \n from being printed in exit reason output - lrmd: pass exit reason prefix to ocf scripts as env variable - Upstream version cs: dad4cebccdb5a4bcdb8fb94e676f6df006ac0846 ------------------------------------------------------------------- Thu Aug 7 06:47:29 UTC 2014 - ygao@suse.com - crm_diff: Support generating a difference without versions details if --no-version/-u is supplied (bnc#888726) - crmd: When container's host is fenced, cancel in-flight operations - crm_mon: add exit reason to resource failure output - lrmd: store failed operation exit reason in cib - Upstream version cs: 1040002ecc1a0dfec9f0561f79936cbe64d716db ------------------------------------------------------------------- Mon Jul 28 05:26:26 UTC 2014 - ygao@suse.com - Bug rhbz#1067544 - Tools: Correctly handle --ban, --move and --locate for master/slave groups - lrmd: Log with the correct personality - mainloop: implement gio_poll_dispatch_mod - Upstream version cs: 483f48a1921241f1f5976a100a171785dc327192 ------------------------------------------------------------------- Tue Jul 22 05:10:09 UTC 2014 - ygao@suse.com - Update to version 1.1.12 - Log: Turn off the default log file if a real one is specified - Log: Do not unconditionally send crm_perror output to stderr - Log: allows the user to specify a log level that is output to syslog - Upstream version cs: 561c4cfda1b09b31b60fd5f73644ee05348f41fb (1.1.12) ------------------------------------------------------------------- Thu Jul 17 06:09:01 UTC 2014 - ygao@suse.com - logrotate: maxsize is not supported by older versions (bnc#883567) - Tools: Non-daemons shouldn't abort just because xml parsing failed - Log: Default to emitting CRIT messages only until logging is fully configured - Logging: Bootstrap daemon logging before processing arguments but configure it afterwards (bnc#883567) - dbus: Return 'null' if we cannot connect via dbus - Upstream version cs: 026944ebfad7c4b2da5a6864a9d763bf37e0e088 ------------------------------------------------------------------- Thu Jul 10 10:02:09 UTC 2014 - ygao@suse.com - systemd: Ensure the retry logic is also exercised for the 'service' class - crm_simulate: Remove resource states and transient attributes when a node is marked as down - crm_verify: Perform a full set of calculations whenever the status section is present (bnc#886479) - PE: Correctly warn when resources require fencing but fencing is disabled (bnc#886479) - PE: Always use pe_fence_node to mark nodes as unclean - lrmd: Handle systemd reporting 'done' before a resource is actually stopped - Upstream version cs: c468fa06483152c29529f81aa0b5ccf37da8b2af ------------------------------------------------------------------- Thu Jul 3 05:41:16 UTC 2014 - ygao@suse.com - attrd: Remove offline nodes from node cache for "peer-remove" requests (bnc#870771) - membership: Prevent use-after-free in reap_crm_member() (bnc#870771) - cluster: Remove unknown offline nodes with conflicting unames from node cache (bnc#870771) - crmd: Remove state of unknown nodes with conflicting unames from CIB (bnc#870771) - Upstream version cs: 440da7f59ac4bed16a0289e33f04d2635b335c70 ------------------------------------------------------------------- Tue Jul 1 08:03:50 UTC 2014 - ygao@suse.com - extra: Add logrotate configuration file for /var/log/pacemaker.log (bnc#883567) - PE: Correctly detect empty tags - xml: Ensure realloc didn't fail - crmd: Only write fence_averride if open() returns a positive file descriptor - tools: Non-recurring memory leak - PE: Potential use-of-NULL - ipc: Buffer size cannot be negative - xml: Prevent possible array underflow - mcp: Let daemons know when they have been respawned by the mcp - cts: fix RemoteBaremetal step2 move resource action failed issue - PE: Correctly observe requires=nothing - crmd: Remove unknown nodes with conflicting unames from CIB (bnc#883240) - controld: Return OCF_ERR_INSTALLED instead of OCF_NOT_INSTALLED (bnc#883729) - lrmd: Prevent OCF agents from logging to random files due to "value" of setenv() being NULL (bnc#883524) - cts: fix shutdown pattern not found issue - Fix: ignore SIGPIPE with gnutls is in use - crmd: avoid double free caused by nested hash table removal - crm_report: Correctly detect large files - cib: cl#5222 - Prevent assert in xml_calculate_changes() on performing cib operations (bnc#881575) - Upstream version cs: c3f1a7f8912f3d6d17497dd8af0b69fe0708b2b1 ------------------------------------------------------------------- Fri Jun 13 03:16:29 UTC 2014 - ygao@suse.com - Bug cl#5219 - PE: Allow unrelated resources with a common colocation target to remain promoted (bnc#899175) - Bug cl#5222 - Enable legacy mode whenever a broadcast update is detected (bnc#881575) - xml: Prevent incorrect xml diffs from corrupting the cib (bnc#878362) - mainloop: implement gio_poll_dispatch_mod - cluster: Prevent search of unames from attempting to create node entries for unknown nodes (bnc#881884) - Upstream version cs: 97629ded958f76652350569a54303b3cd25d1d7e ------------------------------------------------------------------- Wed Jun 11 05:44:21 UTC 2014 - ygao@suse.com - crm_report: Gracefully handle rediculously large logfiles - Bug cl#5222 - cib: Repair rolling update capability (bnc#881575) - cib: Use a better option name for enabling the legacy cib diff code (cl#5222, bnc#881575) - remote: clear remote-node status correctly - pengine: properly handle fencing of container remote-nodes when the container is orphaned - remote: Handle request/response ipc proxy correctly - cib: Prevent assert on sending cib sync requests (cl#5222, bnc#881575) - crm_simulate: Clean up temporary shadow files on exit (bnc#879934) - pengine: cl#5200 - Before migrating utilization-using resources to a node, take off the load that will no longer run there if it's not introducing transition loop (Rework cl#5044, lf#2599, bnc#695440) - Fencing: Cache metadata lookups to avoid repeated blocking during device registration - pengine: properly place resource within a container when container is a remote-node. - mainloop: Better handle the killing of processes in the act of exiting - services: Remove cancelled recurring ops from internal lists as early as possible (bnc#888586) - services: Remove file descriptors from mainloop as soon as we have drained them (bnc#888586) - pengine: Don't report a resource as restarting when it is a moving multistate resource - pengine: avoid assert by searching for stop action on correct node during LogActions - Upstream version cs: 734e6b538b0a037f9e7d0709f450bd664f26c332 ------------------------------------------------------------------- Mon Jun 2 03:17:54 UTC 2014 - ygao@suse.com - cib: Drop the message serverity now that the cause is well understood - crmd: Avoid spamming the cib by triggering a transition only once per non-status change - Upstream version cs: 559f733e8818a96085134cd2a34efc280c512a2a ------------------------------------------------------------------- Thu May 29 10:38:11 UTC 2014 - ygao@suse.com - pengine: Prevent possible segfault (bnc#880258) - attrd: Ensure attribute dampening is always observed - Bug cl#5213 - Limit the scope of the previous commit to the masters role - membership: Avoid duplicate peer entries in the peer cache - Bug cl#5184 - Ensure pending probes that ultimately fail are correctly updated (bnc#804763, bnc#879034) - Upstream version cs: 9776707a6232f25812009145fafa8a64c5b53910 ------------------------------------------------------------------- Tue May 27 09:39:48 UTC 2014 - ygao@suse.com - Rebase: * pacemaker-colocated-utilization.patch - Bug cl#5213 - Ensure role colocation with -INFINITY is enforced - pengine: Mark unrunnable stop actions as "blocked" and show the correct current locations (bnc#878662) - pengine: Block restart of resources if any dependent resource in a group is unmanaged (bnc#878662) - systemd: Try to handle dbus actions that complete prior to configuring a callback - mainloop: Canceling in-flight operations should not fail if child process has already exited. - Bug rhbz#1097457 - Limit the scope of the previous fix and include a helpful comment - Bug cl#5212 - Do not promote instances when quorum is lots and no-quorum-policy=freeze - Bug rhbz#1097457 - Prevent invalid transition when resource are ordered to start after the container they're started in - Fix: fix small memory leaks - Upstream version cs: 25ebca2b83df9263aedb944ea4eeb53468e90e83 ------------------------------------------------------------------- Fri May 23 07:33:51 UTC 2014 - jsegitz@novell.com - added necessary macros for systemd files ------------------------------------------------------------------- Mon May 12 09:20:16 UTC 2014 - ygao@suse.com - pengine: Use "#site-name" in rules for setting site-specific instance attributes - pengine: Use "#cluster-name" in rules for setting cluster-specific instance attributes (FATE#316118) - crmd: Set "cluster-name" property to corosync's "cluster_name" by default for corosync-2 (FATE#316118) - pengine: Fixed so that it could change while a setup of "startup-fencing" starts. - crmd: Use correctly signed inputs to sscanf() - PE: Correctly handle origin offsets in the future - iso8601: Different logic is needed when logging and calculating durations - xml: Do not prune meaningful leaves from v1 patchsets - Upstream version cs: da5aaab97ca5b7858f2aac98107a1a7ffc3e4252 ------------------------------------------------------------------- Thu May 8 11:30:57 CEST 2014 - ro@suse.de - use _rundir macro ------------------------------------------------------------------- Mon May 5 07:42:25 UTC 2014 - ygao@suse.com - Rebase: * pacemaker-NodeUtilization-RA.patch * pacemaker-colocated-utilization.patch - xml: Clean up schema list at exit - pengine: Support site-specific instance attributes based on rules for stretch cluster - pengine: Support cluster-specific instance attributes based on rules (FATE#316118) - crmd: Ensure crm_config options are re-read on updates - Resolve all clang analyser defects - xml: Check all available schemas when doing upgrades - Changes to the ACL schema to support nodes and unix groups - tools: Allow crm_shadow to create older configuration versions - xml: Prevent use-after-free when comparing two xml documents - cib: Resolve memory leaks in query paths - xml: Correctly subtract xml with v1 patchsets - fencing: Advertise support for reboot/on/off in the metadata for legacy agents - fencing: default to 'off' when agent does not advertise 'reboot' in metadata - fencing: Execute all required fencing devices regardless of what topology level they are at - fencing: Fence using all required devices - fencing: fixed so that change of <resources> could be handled correctly. - Upstream version cs: 42ae0e3cb50bb9e19dba7d1ad1ea69255bef5370 ------------------------------------------------------------------- Wed Apr 16 06:38:45 UTC 2014 - ygao@suse.com - pengine: Fix memory leak in expand_tags_in_sets() - PE: Memory leak - PE: Do not initiate fencing for unclean nodes when fencing is disabled - crm_node: Only remove the named resource from the cib - pengine: Support site-specific instance attributes based on rules (FATE#316118) - pengine: Ignore failure-timeout only if the failed operation has on-fail="block" (bnc#871175) - Upstream version cs: f7859e2e06c8cf320482c3e4b447ddbe0a96fe83 ------------------------------------------------------------------- Mon Apr 14 06:10:25 UTC 2014 - ygao@suse.com - pengine: Support cib object tags (FATE#315101) - crmd: Correctly detect changes of ticket state (bnc#872794) - Upstream version cs: 3f7ec44b51825898e3e30784ca4df09e3f45bf40 ------------------------------------------------------------------- Fri Apr 11 15:07:26 UTC 2014 - ygao@suse.com - Rebase: * bug-812269_pacemaker-fencing-device-register-messages.patch - Build: Update versions in configure.ac and sample spec file - fencing: Do not unregister/register all stonith devices on every resource agent change - PE: Unfencing is based on device probes, there is no need to unfence when normal resources are found active - pengine: handles edge case where container order constraints are not honored during migration - pengine: Allow container nodes to migrate with connection resource - xml: Resolve memory leak when logging xml blobs - PE: Do not unfence nodes that are offline, unclean or shutting down - PE: Ensure unfencing only happens once, even if the transition is interrupted - crmd: Actions can only be executed if their pre-requisits completed successfully - crmd: Do not erase the status section for unfenced nodes - Fencing: Correctly record which peer performed the fencing operation - lrmd: Expose logging variables expected by OCF agents - logging: Simplify processing of logging directives - logging: daemons always get a log file, unless explicitly set to configured 'none' - PE: Ensure unfencing occurs before fencing devices are (re-)probed - fencing: Filter self-fencing at the peers to allow unfencing to work correctly - fencing: Automatically switch from 'list' to 'status' to 'static-list' if those actions are not advertised in the metadata - lrmd: Provide stderr output from agents if available, otherwise fall back to stdout - fencing: properly detect new fencing devices for v2 cib updates - acl: Prevent use-of-NULL - cts: automatically determine whether to run RemoteBaremetal (bnc#856353) - pengine: Correctly search failcount (bnc#880520) - Upstream version cs: ff74a591e670dbdb86f7cddeec989167de521216 ------------------------------------------------------------------- Wed Apr 9 04:41:38 UTC 2014 - ygao@suse.com - pengine: Support id-ref in nvpair with optional "name" (FATE#316118) Drop the patches that have been merged upstream: * pacemaker-pengine-nvpair-id-ref.patch * pacemaker-pengine-nvpair-id-ref-test.patch - Upstream version cs: 6442c5a67cdd4cbaeedb2c532e0dc0667a79c202 ------------------------------------------------------------------- Tue Apr 8 13:34:59 UTC 2014 - ygao@suse.com - xml: INFINITY could overflow for 16-bit ints - explicitly cast it - xml: Correctly determine the lowest allowed schema version - xml: Construct the table of supported schemas - cib: All schema upgrade requests must go via the DC - cib: Determine the best value of validate-with if one is not supplied - acl: Ensure known users are not able to read the cib by default - xml: Add missing schema file - Upstream version cs: d9fbba56804a7a7bcceaa5d4b7e29e3f15fc0e59 ------------------------------------------------------------------- Fri Apr 4 10:46:58 UTC 2014 - ygao@suse.com - crmd: Correctly react to successful unfencing operations - PE: Delay unfencing until after we know the state of all resources that require unfencing - PE: Fencing devices default to only requiring quorum in order to start - PE: Automatically re-unfence a node if the fencing device definition changes - pengine: Only report 'migrate' in transition summary if migration is actually taking place - crm_node: Set correct subsystem in node rm cache msg - crm_node: Prevent use-after-free in tools_remove_node_cache() - crmd: make resource ID case sensitive - fencing: Correction of the registration from a difference. - crm_report: Add support for ISO8601 time format in logs (bnc#870886) - PE: Potential memory leak - xml: Support resources that require unfencing - crmd: make node_state erase correctly - xml: Add the ability to have lightweight schema revisions - xml: Split up the RelaxNG schema files so they're easier to version individually - crm_get_msec: prevent integer overflow - corosync: Reduce log severity for a redundant message (bnc#870624) - attrd: Reduce log severity for normal messages (bnc#870624) - cib: When data was old, it fixed so that the newest cib might not be acquired. - crm_attribute: Default --lifetime to forever when using --node (bnc#870696) - Upstream version cs: 76579e813b5aaa1972d400055233519f58bbc0c6 ------------------------------------------------------------------- Mon Mar 31 07:19:00 UTC 2014 - ygao@suse.com - pengine: Do not fence remote-node entries left over in the node section that have no connection rsc - pengine: properly handle remote-node without connection resource. - xml: Update xml validation to allow '<node type=remote />' - cib: allow setting permanent remote-node attributes - Upstream version cs: d63eaf41fa8173de3e29873d0461bfb888b83b37 ------------------------------------------------------------------- Fri Mar 28 07:52:07 UTC 2014 - ygao@suse.com - xml: Allow first-acl-match-wins semantics to be supported if desired - crm_report: extract additional DLM lockspace data - xml: If validate-with is missing, we find the most recent schema that accepts it and go from there - tools: fix "crm_shadow --reset <name>" - controld: Remove '-q 0' from default dlm_controld arguments - acl: Authenticate pacemaker-remote requests with the node name as the client - controld: Verify killall is available on resource start - Prevent use-of-NULL spotted by coverity - crmd: Correctly detect changes to the configuration root - crmd: Fenced nodes that return prior to an election do not need to have their status section reset - xml: Do not log deleted attributes - xml: Correctly mark the parent's of deleted attributes as dirty - crm_diff: Allow the generation of xml patchsets without digests - crmd: Prevent memory leak when accepting a new DC - xml: Find the minimal set of changes when part of a document is replaced - xml: Create the correct diff when only part of a document is changed - Upstream version cs: 61e95e4864be0301be528874eed65266f4e62cf0 ------------------------------------------------------------------- Fri Mar 21 05:51:05 UTC 2014 - ygao@suse.com - pengine: fixes invalid transition caused by clones with more than 10 instances - services_action_cancel: Interpret return code from mainloop_child_kill() correctly - fencing: make hashtable mapping hostname->alias case insensitive for hostname - attrd: Make hostname hashtable case insensitive - membership: make caching peers by uname case insensitive - crmd: make lrm_state hash table not case sensitive - lrmd: clean up the agent's entire process group - xml: Enable resource sets in location constraints for 1.2 schema (bnc#867880) - Upstream version cs: 8bee850bb95ee01dcba03038ddfd65d22e195cf6 ------------------------------------------------------------------- Thu Mar 13 10:44:49 UTC 2014 - ygao@suse.com - Drop obsolete bug-792124_pacemaker-stonith-monitor-log.patch ------------------------------------------------------------------- Thu Mar 13 09:38:46 UTC 2014 - ygao@suse.com - mainloop: Fixes use after free in process monitor code - lrmd: Cancel recurring operations before stop action is executed - lrmd: Merge duplicate recurring monitor operations - services: Do not allow duplicate recurring op entries - crmd: Ack pending operations that were cancelled due to rsc deletion (bsc#1035822) - crm_mon: Add failures to --as-xml output - cib: Check if on-disk cib is corrupt in the separate on_disk_cib_corrupt() rather than in validate_on_disk_cib() (bnc#866618) - crmd: Do not start timers for already completed operations - acl: Read access must be explicitly granted - xml: Better move detection logic for xml nodes - cib: Do not disable cib disk writes if on-disk cib is corrupt (bnc#866618) - xml: Correctly calculate the adjusted offset of xml elements - crm_diff: Mirror how the cib operates for better testing - cib: Ingore patch failures if we already have their contents - cib: Prevent use-after-free when cleaning up after query operations - xml: Mark the parents of a partial replace as dirty - xml: Correctly enforce ACLs after a replace operation - cib: Do not try to use an empty value for CIB_shadow - crm_resource: Gracefully handle -EACCESS when querying the cib - xml: Correctly observe ACLs when creating filtered copies of xml documents - xml: Only test write permissions for attributes that are being changed - xml: Always allow new scaffolding - node with no attributes or only an id field - cib: Memory leaks caused by ACLs returning filtered copies - xml: Set flags to indicate when updates were rejected by an ACL - cib: Respect the CIB_user environment variable for ACLs - xml: Unpack and apply ACLs when tracking changes - cib: Implement CIB_file-mode ACLs for testing - remote: Enable migration support for baremetal connection resources by default - cibadmin: Allow priviliged clients to perform tasks as unpriviliged users - cibadmin: Remove dangerous commands that exposed unnecessary implementation internal details - cibadmin: Purge legacy command options deprecated since 2008 to make the short-forms available for new options - cib: Include a request with file-mode cib actions for the acl code to interrogate - cib: Switch over completely to the new ACL code - xml: Allow new objects to be created provided they subseqently match a configured ACL - cib: Use the new acl filtering functionality - xml: Support the creation of filtered copies based on ACLs - xml: Convert all ACLs to xpaths during unpack - xml: Add ACL flags for xpath selectors when tracking is enabled - cib: ACL improvements - cib: Simplify ACL handling for read-only cib queries - cib: Check ACLs prior to making the update instead of parsing the diff afterwards - cib: Default ACL support to on - Upstream version cs: 38c5972cd284ea6627b3309b619dad4d27442ab0 ------------------------------------------------------------------- Mon Mar 3 11:42:04 UTC 2014 - ygao@suse.com - Add symlinks rcpacemaker and rcpacemaker_remote to /usr/bin/service (bnc#866057) ------------------------------------------------------------------- Mon Mar 3 09:32:30 UTC 2014 - ygao@suse.com - Rebase bug-812269_pacemaker-fencing-device-register-messages.patch ------------------------------------------------------------------- Mon Mar 3 07:40:22 UTC 2014 - ygao@suse.com - xml: Memory leak when free'ing xml documents - systemd: Prevent use-of-NULL when determining if an agent exists - xml: Fix segfault in TRACKING_CHANGES() (bnc#865345) - PE: Memory leak when updating cancel operations - xml: Prevent memory leak in xml _private field - pengine: Performance increase, stack allocated score2char - crm_report: Only gather dlm data if dlm_controld is running - Fix: memory leak found by valgrind was corrected. - Fencing: Add the ability to call stonith_api_time() from stonith_admin - Fencing: Pass the correct options when looking up the history by node name - cib: Ensure 'cibadmin -R/--replace' commands get replies - xml: Correctly initialize and find v1-style patch version numbers - crm_mon: Add -R switch to show clone details - Build: Install compatibility.h - fencing: Teach stonith-ng about finding device and topology changes in the v2 patch format - crm_report: Additional dlm detail - crm_report: Allow ssh alternatives to be used - crmd: Correctly recognise operation cancellations we initiated - crmd: Improved logging of transition cancellations - crmd: Correctly detect lrm refreshes - crmd: Ignore 'move' operations to the cib, we only care about content changes - PE: Ensure instance numbers are preserved for cloned templates - Bug cl#5196 - PE: Check values after expanding templates - crm_mon: Ensure resource attributes have been unpacked before displaying connectivity data - Upstream version cs: ca8234daf2bbb2d0fd2fb47cdbb6667f892081bb ------------------------------------------------------------------- Mon Mar 3 06:14:23 UTC 2014 - ygao@suse.com - pengine: Support id-ref in nvpair with optional "name" (FATE#316118) * pacemaker-pengine-nvpair-id-ref.patch * pacemaker-pengine-nvpair-id-ref-test.patch ------------------------------------------------------------------- Fri Feb 21 07:05:40 UTC 2014 - ygao@suse.com - controld: Do not consider the dlm up until the address list is present - pengine: Support per-resource maintenance mode (bnc#863071) - pengine: cl#5187 - Prevent resources in an anti-colocation from even temporarily running on a same node (bnc#856135, bnc#862319) - controld: handling startup fencing within the controld agent, not the dlm - Upstream version cs: 0b7d85a159cc92e00ebded65c9a33edce7ba1c22 ------------------------------------------------------------------- Fri Feb 21 06:44:59 UTC 2014 - ygao@suse.com - cib: Erasing the cib is an admin action, bump the admin_epoch instead - mcp: Tell systemd not to respawn us if we exit with rc=100 - crm_mon: Allow the file created by --as-html to be world readable - crmd: Do not overwrite existing node state when fencing completes - PE: Default sequential to TRUE for resource sets for consistency with colocation sets - Bug rhbz#1036631 - Stop members of cloned groups when dependancies are stopped - PE: Ensure stop actions are viable when reporting what will happen to a resource - xml: Consolidate patchset logging and apply functionality - cib: Validate that everyone still sees the same configuration once all updates have completed - cib: Validate clients are correctly applying v2-style diffs at most once a minute - xml: Allow both patch formats to include and validate an xml digest - election: If an error occurrs in crm_uptime(), compare as {0,0} - xml: Correctly compare the current feature set when determining which patch format to use - crmd: Support failcount processing with the new patch format - crmd: Handle patchset directives that change the configuration top-level - cib: Make patchset validation a run-time switch - cib: Implement zero-copy status update - cib: Enable the more efficient xml patchset format - crmd: Add support for parsing resource operations in the v2 patch format - xml: New function for calculating patchset version details - cib: Auto-detect which patchset format to use - xml: Allow the source of xml library errors to be traced - cib: Send all r/w operations via the cluster connection and have all nodes process them - xml: Use an optimized xpath search function for applying patch sets - xml: Use the new patch application code path - xml: Detect changes in global replace operations - xml: allow either patch format to be created and applied automatically - xml: Create a more robust diff format - xml: Track changes to xml trees as they happen - Upstream version cs: 0d3f74a5caa7e0f7d1f5ccc97e0d2d018ed8bdfc ------------------------------------------------------------------- Thu Feb 20 12:56:39 UTC 2014 - ygao@suse.com - Update to version 1.1.11 - Drop the patches that have been merged upstream: * pacemaker-display-pending-ops.patch * pacemaker-crm_mon-brief.patch - pengine: Prevent memory leak on handling pending state - crmd: Prevent message relay from attempting to create node entries for unknown nodes (bnc#862383) - attrd: Support "--refresh" command of attrd_updater - crmd: Prevent manual fencing confirmations from attempting to create node entries for unknown nodes (bnc#860169) - Upstream version cs: 3ca8c3bf87057339c9f2a55cd137db8c9c9922f0 ------------------------------------------------------------------- Sat Feb 8 14:00:23 UTC 2014 - ygao@suse.com - ipc: Convince clang that our casting is intended (bnc#845525) * Drop pacemaker-Wno-cast-align.patch which has been merged upstream - cib: Use the correctly typed enum value during initialization - Build: Add dbus as an rpm dependancy - Fix: Bug rhbz#1057697 - Use native DBus library for systemd async support to avoid problematic use of threads - remote: Fail remote-node connection rsc monitor ops early on unexpected disconnect - tools: Return error when attempting to delete an unspecified attribute - remote: Allow baremetal remote-node connection resources to migrate - lrmd: Send event alerting estabilished clients that a new client connection is created. - Fix: Bug rhbz#1057697 - Use native DBus library for systemd support to avoid problematic use of threads - xml: Fix segfault in find_entity() - Upstream version cs: 6290953f2b035a92a8f093973b94c5cf38b79cd9 ------------------------------------------------------------------- Tue Jan 28 04:00:43 UTC 2014 - ygao@suse.com - pengine: Free memory allocated by find_actions() - cib: Fix remote cib based on TLS - Upstream version cs: f3eeaf49d1fb7bca713ecbbc819aef51650d26ef ------------------------------------------------------------------- Tue Jan 28 03:47:02 UTC 2014 - ygao@suse.com - stonith: Drop the severity of log for recoverable condition (bnc#857779) * Update bug-812269_pacemaker-fencing-device-register-messages.patch ------------------------------------------------------------------- Sat Jan 25 06:22:21 UTC 2014 - ygao@suse.com - Build: Suppress -Wcast-align warnings/errors on arm and ppc64le (bnc#845525) * Update pacemaker-Wno-cast-align.patch ------------------------------------------------------------------- Sat Jan 25 01:22:06 UTC 2014 - ygao@suse.com - services: Do not block synced service executions - Upstream version cs: 96cd1941308be48fdf1e827b88ebd9410cf2010a ------------------------------------------------------------------- Fri Jan 24 08:51:31 UTC 2014 - ygao@suse.com - crm_mon: Display brief output if "-b/--brief" is supplied or 'b' is toggled (FATE#314757) * Add pacemaker-crm_mon-brief.patch - tools: Display pending state in crm_mon/crm_resource/crm_simulate if --pending/-j is supplied (cl#5178) (FATE#315159) * Add pacemaker-display-pending-ops.patch ------------------------------------------------------------------- Mon Jan 20 09:55:26 UTC 2014 - ygao@suse.com - Drop the obsolete patch (bnc#857779): * bug-812269_pacemaker-fencing-update-cib.patch ------------------------------------------------------------------- Mon Jan 20 09:47:13 UTC 2014 - ygao@suse.com - crm_report: Force grep to interpret logs as text - pengine: Force record pending for migrate_to actions - pengine: cl#5186 - Avoid running rsc on two nodes when node is fenced during migration - cluster: Fix segfault on removing a node (bnc#858745) - services: Reset the scheduling policy and priority for lrmd's children without replying on SCHED_RESET_ON_FORK (bnc#858857) - fencing: Update stonith device list only if stonith is enabled (bnc#857779) - services: Correctly reset the nice value for lrmd's children (bnc#858857) - pengine: Correctly perform partial migrations when node's uname is not equal to the id. - fencing: Immediately fail remote fencing operation on peer timeout - cl#5055: Improved migration support. - stonith_admin: Ensure pointers passed to sscanf() are properly initialized - Fix: Prevent potential use-of-NULL - crmd: Prevent memory leak in error paths - services: Fixes segfault associated with cancelling in-flight recurring operations. (bnc#859923) - pengine: cl#5174 - Allow resource sets and templates for location constraints (FATE#315158) - Upstream version cs: a3cda7619e71399d54f209296aebf3ba713a0bf4 ------------------------------------------------------------------- Sat Jan 11 11:35:16 UTC 2014 - ygao@suse.com - crmd: Change the default value of node-action-limit - crm_ticket: Support multiple modifications for a ticket in an atomic operation (bnc#855099) - Upstream version cs: 3e8930190b8b53565ef4ca3fa00006c41511be9f ------------------------------------------------------------------- Thu Jan 9 16:30:35 UTC 2014 - ygao@suse.com - ipc: fix memory leak for failed ipc client connections. (bnc#857779) - pacemakerd: Fixed a problem that only pacemakerd did not inherit the logfacility of corosync.conf. - pengine: Fixes memory leak in regex pattern matching code for constraints. - crm_attribute: Do not swallow hostname lookup failures - crmd: Avoid deleting the 'shutdown' attribute - Upstream version cs: ad10acb1c4d668befd7ea1d39312356223d86615 ------------------------------------------------------------------- Wed Dec 11 18:28:52 UTC 2013 - ygao@suse.com - cts: Install cib_xml.py (bnc#854926) - report: Add support for xz compressed logs (bnc#854060) - attrd: Memory leak - crm_mon: Fix overflow on SMTP subject line - Upstream version cs: 65bb87eadcf6dfd9b5cd6030ec318579fd4fe7c8 ------------------------------------------------------------------- Thu Dec 5 11:11:40 UTC 2013 - ygao@suse.com - Fix: Removes unnecessary newlines in crm_resource -O output (bnc#853651) - Upstream version cs: 93a37bf6ea4dcaea8fe7462442ee01019ec167c5 ------------------------------------------------------------------- Thu Nov 28 13:06:59 UTC 2013 - ygao@suse.com - crmd: Memory leak - Upstream version cs: ce5d77c6038a0a0294d4cae238169c458ba03b72 ------------------------------------------------------------------- Mon Nov 25 15:21:14 UTC 2013 - ygao@suse.com - pengine: Recover unexpectedly running container nodes. - pengine: Disable container node probes due to constraint conflicts - ipc: Have the ipc server enforce a minimum buffer size all clients must use. - Bug cl#5181 - corosync: Ensure node IDs are written to the CIB as unsigned integers - Upstream version cs: f57ad6c17624c0cbfc58c46f0bad7e28453506bb ------------------------------------------------------------------- Tue Nov 19 14:12:00 UTC 2013 - ygao@suse.com - Update pacemaker-colocated-utilization.patch * pengine: Do not check the allocation of resources during a colocation preview (bnc#849595) ------------------------------------------------------------------- Tue Nov 19 13:45:56 UTC 2013 - ygao@suse.com - election: Correctly hook up the timeout callback - crmd: Improve throttling response on clusters of single core machines (bnc#838358) - fencing: Wait until all possible replies are recieved before continuing with unverified devices - Bug cl#5179 - Corosync: Attempt to retrieve a peer's node name if it is not already known - Core: Allow blackbox logging to be disabled with SIGUSR2 - PE: Correctly handle demotion of grouped masters that are partially demoted - crmd: When the DC gracefully shuts down, record the new expected state into the cib - crmd: When a peer expectedly shuts down, record the new join and expected states into the cib - attrd: Memory leak - attrd: Do something sane when older non-atomic versions send us messages - crmd: Correctly observe the configured batch-limit (bnc#838358) - attrd: Improve compatibility with older attrd versions - controld: Use the correct variant of dlm_controld for corosync-2 clusters - lrmd: Correctly calculate metadata for the 'service' class - crmd: Prevent over-polling of throttle inputs (bnc#838358) - crmd: Enable cluster-wide throttling when the cib heavily exceeds its target load (bnc#838358) - Upstream version cs: 5a69be5ff56ec7fa66bcf9c2069af6211e4ebc98 ------------------------------------------------------------------- Thu Nov 14 07:19:28 UTC 2013 - ygao@suse.com - crmd: Send the first throttle update (bnc#838358) - Upstream version cs: 2383f6c469bfb31da9efa8e4dd626de049e808c6 ------------------------------------------------------------------- Sun Nov 10 00:43:46 UTC 2013 - ygao@suse.com - Rebase: * bug-806256_pacemaker-log-level-notice.patch - Drop the obsolete patch: * bug-800323_pacemaker-crm_mon-save-cib.patch - Probe for baremetal remote-node connection resources. - remote: Add support for ipv6 into pacemaker_remote daemon - election/fencing: Fix the casts that would result in incorrect values for big-endian systems (bnc#842915) - Fencing: Apply correct score to the resource of group - crm_resource: Wait for the correct number of replies when cleaning up resources (bnc#842204) - cpg: Only process one message at a time to allow other priority jobs to be performed - remote: Handle endian changes between client and server and improve forward compatibility - crmd: A more reliable method of looking up migration options (bnc#838358) - crmd: Allow non-CIB based throttling to be disabled (bnc#838358) - crmd: Honor load thresholds lower than the cib maximum (bnc#838358) - crmd: Tell the DC our job limit instead of the number of cores (bnc#838358) - crmd: Track the CIB's CPU usage (bnc#838358) - crmd: Fold the enforcement of migration-limit into the throttling code (bnc#838358) - crmd: Don't start throttling until we're over the target threshold (bnc#838358) - crmd: Do not send throttle updates to versions that don't support them (bnc#838358) - ipc: Use the higher of the configured buffer size or the default - crmd: Tweak the throttling threshold multipliers (bnc#838358) - crmd: Simplify option processing for throttling options (bnc#838358) - crmd: Simplify the logic for calculating the number of available cores (bnc#838358) - crmd: Make the per-node action limit directly configurable in the CIB (bnc#838358) - crmd: Tweak the throttling code based on migration-limit and/or LRMD_MAX_CHILDREN (bnc#838358) - crmd: Slow down recovery on nodes with IO load (bnc#838358) - crmd: Make the utilization limit configurable (bnc#838358) - crmd: Use the load on our peers to know how many jobs to send them (bnc#838358) - crmd: Notify peers of our relative load (bnc#838358) - remote: Properly version the remote connection protocol - crmd: Track CPU usage on cluster nodes (bnc#838358) - ipc: Consolidate client/server flags and add version field to header - ipc: Suppress asserts for proxied connections - crmd: Do not block on proxied calls from pacemaker_remoted - lrmd: Do not interrogate NULL replies from the server - crm_mon: Reduce flickering in interactive mode - fencing: Ignore changes to non-fencing resources - cli: Do not enabled blackbox for cli tools - crm_resource: Wait for all replies when cleaning up resources (bnc#842204) - crm_mon: Correctly show the queued and exec times for an action - crmd: Correctly update expected state when the previous DC shuts down - crm_resource: Observe --master modifier for --move - PE: Location constraints with role=Started should prevent masters from running at all - Upstream version cs: 907f52608e97b4f280690618342e1377f184b554 ------------------------------------------------------------------- Tue Oct 22 15:03:31 UTC 2013 - schwab@suse.de - Disable -Wcast-align on problematic files (bnc#845525) ------------------------------------------------------------------- Tue Oct 22 10:50:29 UTC 2013 - ygao@suse.com - iso8601: Prevent dates from jumping backwards a day in some timezones - crm_master: Avoid sending redundant "--node" options to crm_attribute (bnc#846360) - ipc: Client buffer size must never fall below our max read size - ipc: Ensure buffer environment variable is read before setting max buffer len - fencing: Deep copy current topology level list on remote op - Upstream version cs: afdce1faecf82c778ed81eb5eb9b7518d36a1160 ------------------------------------------------------------------- Thu Oct 17 06:43:26 UTC 2013 - ygao@suse.com - crm_resource: Wait for a reply for each request to crmd (bnc#842204) - Upstream version cs: 7afb87c8caa96d3c1ed9a353ae6061023a729fd4 ------------------------------------------------------------------- Wed Oct 16 02:46:19 UTC 2013 - ygao@suse.com - Cluster: On cluster signoff, make sure that the g_source gets disconnected. - crmd: Using PCMK_ipc_buffer variable for remote-node proxy - plugin: Reset the scheduling policy of daemons to SCHED_OTHER with nice value -10 (bnc#843647) - ipc: Compress msgs based on client's actual max send size - pengine: Do not probe for connection resources - Upstream version cs: 490e695539630cfc158c5967a99237659fac5f30 ------------------------------------------------------------------- Thu Oct 3 01:58:54 UTC 2013 - ygao@suse.com - fencing: Allow fencing for node after topology entries are deleted - fencing: Do not broadcast suicide if the on action is being executed - remote: Option to enable/disable probes on container resource remote-nodes - lrmd: Correctly cancel monitor actions for lsb/systemd/service resources on cleaning up (bnc#842471) - services: Fix the executing of synchronous actions - Upstream version cs: 728ec77851540f9ce64ee1b1269d967841571b25 ------------------------------------------------------------------- Sun Sep 29 05:12:24 UTC 2013 - ygao@suse.com - PE: Add a special attribute for distinguishing between real nodes and containers in constraint rules - PE: Allow location constraints to take a regex pattern to match against resource IDs - Bug rhbz#1011618 - Consistently use 'Slave' as the role for unpromoted master/slave resources - cibadmin: Report errors during sign-off - Fencing: Observe pcmk_host_list during automatic unfencing - services: Prevent use-of-NULL when executing service actions - crm_resource: Provide a meaningful error if --master is used for primitives and groups - ipc: Raise the default buffer size to 128k (bnc#838358) - Upstream version cs: ecd44e642db3b65ee83aceed0e93376894a58c47 ------------------------------------------------------------------- Wed Sep 25 06:37:41 UTC 2013 - lmb@suse.com - plugin: stonith-ng needs membership updates (bnc#841010) - controld: allow the stonith-enabled check to be overridden (bnc#842186, fate#315195) - attrd: Avoid infinite write loop for unknown peers - tools: Detect orphaned remote-nodes when setting attributes - controld: don't start dlm if stonith-enabled is false (bnc#842186, fate#315195) - pengine: Probe container nodes - crmd: Tell attrd when remote-node is disconnected - attrd: Give remote-nodes ability to set attributes with attrd - Upstream version cs: c68919f46bbaa96ec4ff2c770f82588280277fe6 ------------------------------------------------------------------- Wed Sep 25 06:19:58 UTC 2013 - lmb@suse.com - crmd: Expose "expected-quorum-votes" option in the metadata only for plugin-based clusters - decode_transition_key: Older systems do not support '%ms' in sscanf - cib: Guard against overly long user specified strings - crm_shadow: Simplify prompt generation and impose a sane limit on length - fencing: Skip empty notifications from the cib - election: Do not update state when ACKs arrive - cib: Additional logging for invalid diff updates - cib: Correctly check for archived configuration files - crmd: Reliably detect and act on reprobe operations from the policy engine - crm_mon: Do not display duplicates of failed actions - crm_node: Display partition output in sorted order - cib: Allow values to be added/updated and removed in a single update - attrd: Use the truely atomic version where CPG is used for cluster communication - attrd: Untested reimplementation of a truely atomic attrd - election: Track the election state and allow it to be queried - crm_resource: Handle --ban for master/slave resources as advertised (rhbz#902407) - xml: Location constraints are allowed to specify a role - Review and refresh all patches to apply without fuzz. - Upstream version cs: 144c2fc4b2f50623cbc87bf20788717e0b058a38 ------------------------------------------------------------------- Tue Sep 24 13:27:53 UTC 2013 - ygao@suse.com - Rebase: * bug-728579_pacemaker-stonith-dev-id.patch * bug-806256_pacemaker-log-level-notice.patch * bug-812269_pacemaker-fencing-device-register-messages.patch - Drop the patches that have been merged upstream: * pacemaker-node-maintenance.patch * pacemaker-node-maintenance-tests.patch * pacemaker-crmd-pending-delete.patch * pacemaker-resource.patch - Drop the obsolete patches: * pacemaker-cts-startcmd.patch * pacemaker-disable-listing-fence-agents.patch * pacemaker-daemon-symlinks.patch * pacemaker-ptest-build.patch * pacemaker-lrmd-remote-disabled.patch * bug-815447_pacemaker-cib-strip-text-nodes.patch * bug-809635_pacemaker-xml-digest.patch - Drop the obsolete source: * heartbeat-doc-1.0.tar.gz ------------------------------------------------------------------- Tue Sep 17 08:48:47 UTC 2013 - ygao@suse.com - spec: Recommends fence-agents and sbd ------------------------------------------------------------------- Tue Sep 10 02:21:32 UTC 2013 - ygao@suse.com - pengine: Properly handle orphaned multistate resources living on remote-nodes - pengine: Map orphaned container fillers to container resource during shutdown to preserve ordering - pengine: Do not attempt to probe resources on remote-nodes that are shutting down - cib: Prevent potential double free in get_remote_uuid - pengine: Properly shutdown orphaned remote connection resources - crmd: Populate remote peer cache when building new cib - crmd: Clear remote-node status AFTER start completes, not before. - pengine: Probe resources on baremetal remote-nodes - crmd: Remote-node fencing glue - pengine: Allow fencing of baremetal remote nodes - crmd: Cache baremetal remote-nodes - tools: Fixes formatting of remote-nodes in crm_mon and crm_simulate - Upstream version cs: d9bb7632492b1f91a300fb43e9ad8a262a8f2928 ------------------------------------------------------------------- Wed Sep 4 07:10:20 UTC 2013 - ygao@suse.com - cib: Support XML comments in diffs (bnc#836965) - Upstream version cs: 5d0a2239543bd5003444af4a16780ac47fd37079 ------------------------------------------------------------------- Mon Sep 2 11:24:55 UTC 2013 - ygao@suse.com - crm_mon: add --hide-headers option to hide all headers - pengine: Order demote/promote of resources on remote nodes to happen only once the connection is up - tools: clear ncurses screen when crm_mon cannot connect - crmd: Invalid formatting of log message causes crash - remote: Properly handle expected late replies when TLS transport is in use. - cluster: Allow remote-nodes to be cached - Upstream version cs: a4eb44fd075e8cb8608f5e45522cade45658ff98 ------------------------------------------------------------------- Wed Aug 21 05:08:44 UTC 2013 - ygao@suse.com - Fencing: Support agents that need the host to be unfenced at startup - cpg: Correctly set the group name length - pengine: Distinguish between the agent being missing and something the agent needs being missing (bnc#833374) - crmd: Prevent PCMK_LRM_OP_NOT_INSTALLED for being mapped to PCMK_LRM_OP_ERROR - Bug cl#5171 - PE: Don't prevent clones from running due to dependant resources - services: Correctly invoke callbacks for aborted operations - services: Preserve errno when handling exec failres least it be reset as a side-effect of other calls - services: Prevent use-of-NULL in g_error_free - services: Detect missing agents and permission errors before forking - lrmd: Do not overwrite any existing operation status error - crmd: Correcty update the history cache when recurring ops change their return code - PE: Correctly account for the location preferences of things colocated with a group - crmd: Gracefully handle actions that cannot be initiated - crm_simulate: Do not complain about CRM_OP_REPROBE actions - PE: Gracefully handle bad values for XML_ATTR_TRANSITION_MAGIC - crm_mon: Correctly extract alternate name for pingd attributes - Fix: Prevent use-of-NULL when ping resources do not define a host list - Fix: Gracefully handle -ECHILD from waitpid() in crm_abort() - Fix: Command-line tools should stop after an assertion failure - systemd: Prevent glib assertion - only call g_error_free() with non-NULL arguments - Upstream version cs: 4f672bc85eefd33e2fb09b601bb8ec1510645468 ------------------------------------------------------------------- Mon Aug 5 06:04:07 UTC 2013 - ygao@suse.com - crmd: Don't add node_state to cib, if we have not seen or fenced this node yet - PE: Do not allow colocation with blocked clone instances - PE: Do not re-allocate clone instances that are blocked in the Stopped state - fence_legacy: stonithd now handles killing the process group for us - Fencing: On timeout, clean up the agent's entire process group - fencing: Put all fencing agent processes into their own process group - crmd: Prevent crash by passing log arguments in the correct order - PE: Do not restart resources that depend on unmanaged resources - crmd: Prevent recurring monitors being cancelled due to notify operations - mcp: Fix so that only pacemakerd restarts, at the time of pacemakerd failure - mcp: Support PCMK_fail_fast to reboot a machine at the time of process (subsystem) failure - cpg: Ensure the CPG group is always null-terminated - Upstream version cs: ac7aa1c94f317fc1d4024fbc09b012bc9f1b229e ------------------------------------------------------------------- Fri Jul 26 02:49:48 UTC 2013 - ygao@suse.com - Update to version 1.1.10 - crm_resource: Allow options to be set recursively - cib: Correctly read back archived configurations if the primary is corrupted - mainloop: Hand the channel off to g_io_watch and forget it. - mainloop: Force removal of ipc fd in dispatch_del callback - crmd: Clean up handling of pending operations at shutdown and improve logging - Bug cl#5161 - crmd: Prevent memory leak in operation cache - cib: Ensure we set up hacluster's groups in stand-alone mode - Upstream version cs: 368c72641139785ec5981813f311fae8cd1410f3 (1.1.10) ------------------------------------------------------------------- Mon Jul 22 05:28:01 UTC 2013 - ygao@suse.com - cpg: Correctly free sent messages (bnc#853675, bnc#856018) - crm_resource: Present an error if an admin tries to --force-start an already active resource - cluster: Correctly construct the header for compressed messages (bnc#851951, bnc#856018) - PE: Re-allow ordering constraints with fencing devices now that it is safe to do so - cl#5170 - Correctly support on-fail=block for clones - PE: Do the right thing when admins specify the internal resource instead of the clone - fencing: Correctly detect existing device entries when registering a new one - crm_node: Return 0 if --remove passed - remove the mainloop_trigger that are no longer needed. - corosync: Ensure removed peers are erased from all caches - cluster: Detect and warn about node names with capitals - Upstream version cs: 9abe6876d50ff2879f17b08dfa1500817a7fbeba ------------------------------------------------------------------- Thu Jul 11 00:30:44 UTC 2013 - ygao@suse.com - crmd: Prevent messages for remote crmd clients from being relayed to wrong daemons ------------------------------------------------------------------- Wed Jul 10 12:43:46 UTC 2013 - ygao@suse.com - lrmd: Prevent use-of-NULL in client library - Upstream version cs: 9d656180294770e39deced79a134d4cf531b8159 ------------------------------------------------------------------- Wed Jul 10 08:33:31 UTC 2013 - ygao@suse.com - Bug cl#5168 - Prevent clones from being bounced around the cluster due to location constraints - Upstream version cs: 90b8afe77ab8a3ff2fd30826187fd7726aab4f04 ------------------------------------------------------------------- Tue Jul 9 14:32:31 UTC 2013 - ygao@suse.com - crm_mon: Bug cl#5167 - Only print "stopped" node list for incomplete clone sets - Upstream version cs: d16f4c56b120c5ae761680cfb4d98922a185c6c2 ------------------------------------------------------------------- Thu Jul 4 14:09:36 UTC 2013 - ygao@suse.com - Fix: memory leak when used score2char() - Fencing: Avoid hashtable lookups in NULL - Fencing: Force use of the calculated command for support ops - Bug cl#5157 - Allow migration in the absence of some colocation constraints - PE: Detect constraints with inappropriate actions (ie. promote for a clone) - PE: Do the right thing when admins specify the internal resource instead of the clone - Fencing: User-defined arguments always trump calculated ones - cluster: Correctly remove duplicate peer entries - Upstream version cs: 0dc3203917438057ab476a0f0aac4a1c8e27f97a ------------------------------------------------------------------- Wed Jul 3 02:32:22 UTC 2013 - ygao@suse.com - plugin: Fix memory leaks - Upstream version cs: 5053ba656a854bdf4eacaf7d43b0dce1b7e58b9d ------------------------------------------------------------------- Tue Jul 2 04:14:55 UTC 2013 - ygao@suse.com - plugin: Handle messages in the common way in the plugin dispatch function by default - Upstream version cs: 704f45e36594018358b0ea9d578291af1d52196e ------------------------------------------------------------------- Sat Jun 29 18:02:34 UTC 2013 - lmb@suse.com - plugin: Fix small memory leak ------------------------------------------------------------------- Wed Jun 26 10:20:03 UTC 2013 - ygao@suse.com - plugin: Change the function declaration in the plugin header - crm_mon --neg-location drbd-fence-by-handler - crm_mon -A: print_attr_msg needs to lookup paramters, not meta - crm_mon -A: fix memleak, add g_free_list(attr_list) - crmd: CID#1036761 Dereference null return value - Corosync: CIB#1036763 Dereference before null check - Tools: CID#1036764 Unused pointer value, CID#1036762 Resource leak - Tools: CID#1036760 Truncated stdio return value - PE: Delete the old resource state on every node whenever the resource type is changed (bnc#822233) - Upstream version cs: 61e4b8f62b109ca5b910efd38c7e79ef403030fd ------------------------------------------------------------------- Fri Jun 21 04:44:12 UTC 2013 - ygao@suse.com - crmd: Ensure operations for cleaned up resources don't block recovery (bnc#825536) - logging: If SIGTRAP is sent before tracing is turned on, turn it on - membership: Correctly merge partial entries in the peer cache - attrd: Fixes deleted attributes during dc election - crmd: Stop trying to fence when no devices exist capable of fencing a node - PE: Allow active nodes in our current membership to be fenced without quorum - PE: If fencing is unavailable or disabled, block further recovery for resources that fail to stop - crm_resource: Avoid whitespace in transition keys when cleaning up and failing resources (bnc#851308) - crmd: Everyone who gets a fencing notification should mark the node as down (bnc#842915) - crmd: Initiate node shutdown if another node claims to have successfully fenced us (bnc#842915) - crm_resource: Correctly implement --master - crm_resource: Support OCF tracing when using --force-(check|start|stop) - lrmd: Make lrmd and pacemaker_remoted in charge of setting up rsc state folder - Allow uuid retrieval for nodes with no known name - crmd: Update the status section with details of nodes for which we only know the nodeid - crm_error: Add the ability to list and print error symbols - crmd: Further improve clean shutdowns - Core: Ensure the blackbox is saved on abnormal program termination - crm_mon: Implement the ability to change the display options during operation. - Upstream version cs: 6ea4b7ef34ce6e17f53f69d87a404d334db5d837 ------------------------------------------------------------------- Wed Jun 19 08:45:46 UTC 2013 - ygao@suse.com - xml: Purge diff markers even if there's no digest (bnc#824097) ------------------------------------------------------------------- Wed Jun 19 08:32:15 UTC 2013 - ygao@suse.com - crmd: Prevent use-of-NULL during emergency shutdown - crmd: Ensure all membership operations can complete while trying to cancel a transition - Repair handling of attrd_update_delegate() return codes - attrd: Correctly detect successful invocation of attrd_update() - Convert all exit codes to positive errno values - crm_attribute: Send details on duplicate values to stdout - crm_attribute: Return the actual error code rather than the negative of it - crm_resource: Implement the --master option for the --move,--ban,--clear commands - crm_resource: Make --move without --host a true alias for --ban without --host - crm_resource: Implement --ban for moving resources away from nodes and --clear (replaces --unmove) - cib: Allow multiple children to be removed from the object root (eg. constraints) - crmd: Prevent use-after-free now that mainloop terminates naturally - mainloop: Saner ordering of reference counting events in mainloop_destroy_trigger() - Correctly calculate crm_system_name - Core: Tolerate NULL in mainloop_destroy_trigger() - crmd: Allow mainloop to terminate naturally on graceful shutdown - crmd: Use standard error codes on exit - Ensure anything we free at shutdown is set to NULL too - Ensure crm_system_name is cleaned up at exit for valgrind - crmd: Ensure timers are stopped at exit - Fencing: Ensure API object is consistently free'd - Clean up internal mainloop structures at exit - Core: Correctly unreference GSource inputs - crmd: Clean up more memory before exit - crmd: Do not get stuck in S_POLICY_ENGINE if a node we couldn't fence returns - Upstream version cs: 47ed85ac6f4a9ac353945ae8bc4d7cea98b7cb7a ------------------------------------------------------------------- Tue May 28 08:50:43 UTC 2013 - ygao@suse.com - PE: Mark unrunnable stop actions as "blocked" (bnc#816511) - crm_resource: Do not crash when calling --force-start on an invalid resource - ping: Ensure OCF_RESKEY_failure_score has a value (bnc#831327) - PE: Unlink pengine output files before writing. - Fencing: Restore the ability to manually confirm that fencing completed (bnc#843699) - Fix: Resolve several memory leaks and improper dereferencing of pointers - PE: cl#5155 - Block the stop of resources if any depending resource is unmanaged (bnc#816511) - crmd: Insert async failures so that they appear in the correct order - cl#5151 - Ensure node names are consistently compared without case - crmd: Block after 10 failed fencing regardless of crmd-transition-delay - PE: Re-initiate _active_ recurring monitors that previously failed but have timed out - cl#5133 - PE: Correctly observe on-fail=block for failed demote operation - PE: Suppress meaningless IDs when displaying anonymous clone status - PE: Display a list of nodes on which stopped anonymous clones are not active instead of meaningless clone IDs - PE: Correctly handle resources that recover before we operate on them - Core: Detect child processes that terminate before our SIGCHLD handler is installed - Upstream version cs: 674e69854a0ce4636ee54e123e45b1612c8f1fd5 ------------------------------------------------------------------- Thu May 9 14:58:19 UTC 2013 - ygao@suse.com - crmd: Do not record pending delete operations in the CIB ------------------------------------------------------------------- Sat May 4 01:31:38 UTC 2013 - ygao@suse.com - cib: Strip text nodes on writing CIB to disk (bnc#815447, bnc#809635) - xml: Use xmlNodeDump() from libxml2 when calculating digest (bnc#809635) ------------------------------------------------------------------- Fri May 3 08:35:25 UTC 2013 - ygao@suse.com - PE: Ensure per-node resource parameters are used during probes - cl#5154 - Do not expire failures when on-fail=block is present - Upstream version cs: 2db99f16fe9b19a6320db62755f0f28ced8efe00 ------------------------------------------------------------------- Thu May 2 03:59:20 UTC 2013 - ygao@suse.com - cl#5153 - Correctly display clone failcounts in crm_mon - crmd: Simplify the logic for resetting node status in the CIB after fencing events (bnc#812913) - crmd: cl#5152 - Correctly clean up fenced nodes during membership changes (bnc#812913) - ipc: Reference the correct type when inspecting IPC messages - Core: Ensure the last field in transition keys is 36 characters - crm_mon: Check if a process can be daemonized before forking so the parent can report an error - crm_mon: Ensure stale pid files are updated when a new process is started - extra: add failure_score parameter into ping RA - crm_resource: Allow --cleanup without a resource name - Upstream version cs: 3297085490c2fe0af25fc805d2b53123da877f7a ------------------------------------------------------------------- Thu Apr 18 08:08:51 UTC 2013 - ygao@suse.com - fencing: Drop the severity of the messages on registering a stonith device (bnc#812269) ------------------------------------------------------------------- Thu Apr 18 07:09:27 UTC 2013 - ygao@suse.com - crmd: Ensure we return to a stable state if there have been too many fencing failures - fencing: Prevent empty call-id during notification processing - crm_resource: Make --refresh and --reprobe an alias for --cleanup [--resource name] - crmd: Indicate completion of refresh to callers - crmd: Indicate completion of re-probe to callers - crm_resource: Allow individual resources to be reprobed - ipc: Allow unpriviliged clients to clean up after server failures - crmd: Only performa dry run for deletions if built with ACL support - Upstream version cs: a92f3a0e327da2af171856c979fb8eaa00122534 ------------------------------------------------------------------- Wed Apr 17 05:34:18 UTC 2013 - ygao@suse.com - crmd: Fixes hash table assert when resetting stonith failures - crmd: Fixes memleak in failure cleanup - crmd: Do not update fail-count and last-failure for old failures (bnc#804003) - tools: cibadmin - Do not log for query commands by default (bnc#813045) - services: Fix signal FD leak - Upstream version cs: d63c13cbcde7663cfae9bb412583ad11914b6acc ------------------------------------------------------------------- Thu Apr 4 03:42:40 UTC 2013 - ygao@suse.com - tools: crm_mon - Prevent segfault on reconnecting cib (bnc#813131) ------------------------------------------------------------------- Mon Apr 1 10:27:14 UTC 2013 - ygao@suse.com - Fix: tools: crm_mon - Print a timing field only if its value is non-zero (bnc#809313) ------------------------------------------------------------------- Mon Apr 1 07:44:56 UTC 2013 - ygao@suse.com - tools: crm_mon - Save relevant cib XML into /tmp/cmon.* files if refreshing fails (bnc#800323) ------------------------------------------------------------------- Fri Mar 29 08:06:16 UTC 2013 - ygao@suse.com - fencing: Directly query the full cib on updating the cib cache (bnc#812269) ------------------------------------------------------------------- Fri Mar 29 03:33:25 UTC 2013 - ygao@suse.com - cib: Use unique error codes for write_cib_contents() - Fix: Ensure custom error codes are less than 256 - corosync: cl#5148 - Correctly remove a node that used to have a different nodeid (bnc#805278) - Upstream version cs: e8caee88c9e078fccf98a9da05543b73d4696f04 ------------------------------------------------------------------- Thu Mar 28 16:55:11 UTC 2013 - ygao@suse.com - services: Improve the performance of synchronous actions (bnc#803969) ------------------------------------------------------------------- Wed Mar 27 14:11:11 UTC 2013 - ygao@suse.com - lrmd: Change the default directory for nagios plugins (bnc#812005) ------------------------------------------------------------------- Wed Mar 27 08:00:17 UTC 2013 - ygao@suse.com - xml: Prevent lockups by setting a more reliable buffer allocation strategy - fencing: Fix memory leaks and invalid read - fencing: Correctly unpack device parameters before using them - lrmd: Avoid memory leak in resources_action_create() - Skip WNOHANG when waiting after sending SIGKILL to child processes - fencing: Fail the operation once all peers have been exhausted - Fix several use-after-free and use-of-NULL - fencing: Correctly filter devices when no nodes are configured yet - ipc: Restore the ability for members of the haclient group to connect to the cluster - Fencing: Re-use the policy engine libraries for filtering devices - Upstream version cs: c7910371a5ada346fe7d1e1872ec9a40570018e6 ------------------------------------------------------------------- Mon Mar 25 03:05:54 UTC 2013 - ygao@suse.com - corosync: Support "crm_node --remove" with a node name for corosync plugin (bnc#805278) - Fix memory leaks in stonithd, crmd and lrmd - services: Close the reading pipes when a synchronous action is finished - Doc: Pacemaker Explained: Add nagios resource class - Upstream version cs: cb7b3f48dea9eb1b6825242da2ef507887b409cc ------------------------------------------------------------------- Tue Mar 12 11:39:23 UTC 2013 - ygao@suse.com - Log: Change some messages to notice level (bnc#806256) ------------------------------------------------------------------- Mon Mar 11 10:49:35 UTC 2013 - ygao@suse.com - xml: Support nagios resource class in the 1.2 schema - xml: Support resource container in the 1.2 schema ------------------------------------------------------------------- Fri Mar 8 11:37:17 UTC 2013 - ygao@suse.com - Update to version 1.1.9 - fencing: Only initiate topology based fencing if all topology queries have arrived AND we are still in the query phase - fencing: Don't delay execution if all topology queries have arrived - crmd: Bug cl#5144 - Do not updated the expected status of failed nodes (bnc#807824) - Core: Support PCMK_logpriority to set syslog priority - Upstream version cs: 2a917ddb9d1a01f47db84b17468413522d63ddfa ------------------------------------------------------------------- Fri Mar 1 10:31:54 UTC 2013 - coolo@suse.com - update license to new format ------------------------------------------------------------------- Sat Jul 28 14:21:29 UTC 2012 - aj@suse.de - Include sys/resource.h for getrlimit etc. ------------------------------------------------------------------- Wed Jun 6 08:15:34 UTC 2012 - ygao@suse.com - Core: Revert to the original process flags for the support of rolling-upgrade - PE: Bug cl#5059 - Use the correct action's status when calculating required actions for interleaved clones - Upstream version cs: 61a079313275f3e9d0e85671f62c721d32ce3563 ------------------------------------------------------------------- Mon May 21 11:46:15 UTC 2012 - badshah400@gmail.com - Use BuildRequires: libgnutls-devel instead of gnutls-devel for Factory in keeping with gnutls package naming changes. ------------------------------------------------------------------- Fri Apr 20 07:14:26 UTC 2012 - ygao@suse.com - corosync: Determine node state for plugin based cluster types - tools: crm_node - Test stack membership only when displaying partition members - Overhaul the way peers are determined to be sufficiently active - PE: Observe the promotion score of probed resources - crmd: Detect when peers have an incorrectly set process list - corosync: Detect when the first corosync has not yet formed and handle it gracefully - Corosync: Move deprecated corosync variants to their own file - Upstream version cs: 4c8b599e6d79499f28f8178988d65fd33e0d9b24 ------------------------------------------------------------------- Thu Apr 19 12:22:14 UTC 2012 - ygao@suse.com - Update to version 1.1.7 - crmd: cl#5057 - Restart sub-systems correctly (bnc#755671) - corosync: Fix compilation when libqb is not available - pengine: cl#5058 - Fixes issue with on-fail option on operation being ignored. - tools: rhbz#754216 - Make crm_mon --as-xml option default to --one-shot mode - cts: cl#5047 - Suppress cts badnews output related to fencing when fencing is expected. - crmd: cl#5053 - Fixes fail-count not being updated properly. - crmd: cl#5051 - Fixes file leak in pe ipc connection initialization. - Tools: crm_simulate - Correctly detect status sections with duplicate node_state entries - crmd: Detect when the quorum API isn't configured for corosync 2.0 - RA: controld - use the correct dlm_controld when membership comes from corosync directly - Core: Fix the logging of XML and conditional corefile creation by asserts - pengine: cl#5045 - Better handling of interrupted migrations. - Build: rhbz#805147 - Use default value for HB_DAEMON_DIR define when clusterglue does not provide one. - cib: Even with no peers, we still need to wait for connected clients to disconnect - corosync: Fix complation warning when corosync doesn't use libqb - cib: Determine when there are no peers to respond to our shutdown request and exit - cib: Correctly shutdown in the presence of peers without relying on timers - Corosync: Correctly disconnect from corosync variants - cib: Ensure no additional messages are processed after we begin terminating - Corosync: Reset connection variables after disconnect - Core: Correctly track the length of a string without reading from uninitialzied memory (valgrind) - PE: Fix memory leak in order_rsc_sets (valgrind) - Tools: Fix memory leak in crm_simulate (valgrind) - PE: Correctly locate crm_simulate for automated regression tests - Core: Use the default schema location when PCMK_schema_directory is defined but empty - PE: Correctly migrate a resource that has just migrated - pengine: cl#5025 - Automatically clear failures when resource configuration changes. - Fix a number of potential problems reported by coverity - crmd: Also purge fencing results when we enter S_NOT_DC - crmd: Correctly determine when a node can suicide with fencing - PE: Bug rhbz#799070 - Report resources as active in crm_mon if they are located on an unclean node - Upstream version cs: 023d3b183e77cfdb3699080b26d98ca4bc534c57 ------------------------------------------------------------------- Wed Apr 18 06:30:14 UTC 2012 - ygao@suse.com - PE: Support of referencing resource templates in resource sets (bnc#729634) - stonith: A "start" operation of a stonith resource does a "monitor" on the device beyond registering it - PE: More validations on configurations related to resource templates - Core: Bug cl#5027 - Fix compilation when libqb is not available - PE: cl#5002 - increment_clone() does not terminate on invalid input - PE: Indicate the difference between resources that are restarting and reloading - crmd: Leave it up to the PE to decide which ops can/cannot be reload - PE: Detect reloads of previously migrated resources - Core: Bug cl#5016 - Prevent failures in recurring ops from being lost - crmd: Bug cl#5015 - Remove the failed operation as well as the resulting fail-count and last-failure attributes - Fencing: Supress errors when listing all registered devices - PE: Reload of a resource no longer causes a restart of dependant resources - Fencing: Implement support for advanced fencing topologies: eg. kdump || (network && disk) || power - Fencing: Add support for fencing levels to the admin tool - Fencing: Prevent duplicate free and uninitialized pointers - Fencing: Correctly remove the fencing levels when requested - Fencing: Not all devices might be running locally when creating a fencing level - cib: Don't halt disk writes if the previous digest is missing - Cluster: Hook up the callbacks to the corosync quorum notifications - Corosync: Correctly extract the node id from membership udpates - Core: Ensure signals are handled eventually in the absense of timer sources or IPC messages - Corosync: Default to using the nodeid as the node's uuid (instead of uname) - corosync: Consistently set the correct uuid with get_node_uuid() - Fencing: Standardize on /one/ method for clients to request members be fenced - Core: Strip text nodes from on disk xml files (bnc#769910) - Tools: Implement crm_node functionality for native corosync - corosync: Hook up CPG membership - fencing: More gracefully handle failed 'list' operations for devices that only support a single connection - corosync: Unset the process list when nodes dissappear on us - corosync: Correctly infer lost members from the quorum API - crmd: Election - perform the age comparison only once - crmd: Remove rusage differences of less than a second from election calculations - PE: Ensure post-migration stop actions occur before node shutdown - fencing: Correctly remove a single fencing level for a target - fencing: Add the fencing topology section to the 1.1 configuration schema - corosync: Ensure we catch nodes that leave the membership, even if the ringid doesn't change - corosync: Treat CS_ERR_QUEUE_FULL the same as CS_ERR_TRY_AGAIN - crmd: Fast-track shutdown if we couldn't request it via attrd - tools: Bug rhbz#754216 - Add XML as an output option to crm_mon - Core: Correctly determine when an XML file should be decompressed - corosync: Remove deprecated member of the CFG API - fencing: Provide enough data to construct an entry in the node's fencing history - pengine: cl#5014, Prevent asymmetrical order constraints from causing resource stops - pengine: cl#5007, Fixes use of colocation constraints with multi-state resources - PE: Bug cl#5038 - Prevent restart of anonymous clones when clone-max decreases - PE: Bug rhbz#789397 - Failcount and related info should be reset or removed when the resource is deleted - fencing: fence_legacy - Drop spurilous host argument from status query (bnc#750512) - pengine: Bug cl#5000 - Implements ability to create rsc_order constraint sets such that A can start after (B || C) has started. - Fencing: fence_legacy - Ensure port is available as an environment variable when calling monitor (bnc#750512) - PE: Bug cl#5028 - Unmanaged services should block shutdown unless in maintainence mode (bnc#738882) - Fix coverity defects in pengine, cib, stonith-ng and crm_resource - Fix use-after-free in crmd, stonith-ng and crm_simulate - Fix use-of-NULL in libcrmcluster and stonith-ng - Fix segfaults in stonith-ng and xml.c - Upstream version cs: 1e8a96ae55d6022f7036434b537938a1fa75a700 ------------------------------------------------------------------- Thu Mar 29 07:39:30 UTC 2012 - lmb@novell.com - Add provides to identify support level for tickets (bnc#752241) ------------------------------------------------------------------- Mon Mar 26 04:43:35 UTC 2012 - ygao@suse.com - Resolve coverity RESOURCE_LEAK, REVERSE_INULL and UNREACHABLE defects ------------------------------------------------------------------- Sun Mar 25 14:46:13 UTC 2012 - ygao@suse.com - Tools: crm_simulate - Support to grant/revoke/standby/activate tickets from the new ticket state section (bnc#752241) ------------------------------------------------------------------- Sun Mar 25 09:03:42 UTC 2012 - ygao@suse.com - Tools: Add crm_ticket to be profiled by valgrind - Tools: crm_ticket - Display the tickets that are referenced in rsc_ticket constraints but have never been granted yet (bnc#752241) - Tools: crm_mon - Support to display tickets (bnc#752242) ------------------------------------------------------------------- Sat Mar 24 10:51:24 UTC 2012 - ygao@suse.com - PE: Move the ticket state section to "/cib/status/tickets/ticket_state" (bnc#752241) - PE: Support to make tickets standby for relinquishing tickets gracefully (bnc#752241) - Tools: Implement a new crm_ticket (bnc#752241) - Tools: Update CLI regression tests for the new crm_ticket - Tools: Add CLI regression tests for ticket standby state - PE: Update regression tests for the new ticket state section - PE: Support to make tickets standby for relinquishing tickets gracefully (regression tests) ------------------------------------------------------------------- Tue Mar 20 10:11:52 UTC 2012 - ygao@suse.com - Tools: crm_ticket - Granting/revoking tickets need user to confirm unless "--force" option is used (bnc#752269) ------------------------------------------------------------------- Tue Mar 13 09:28:38 UTC 2012 - ygao@suse.com - Tools: Implement the support of granting/revoking tickets for crm_simulate (bnc#752241) ------------------------------------------------------------------- Tue Jan 17 05:33:09 UTC 2012 - ygao@suse.com - Core: Move several new fields since v1.1.6 to the end of the data types for ABI compliance - stonith: Default to 20s timeout for start and monitor operations of stonith resources which is same as "default-action-timeout" (bnc#733337) - Tools: ptest - Fix "-X, --xml-text" option to handle a XML string as its argument (bnc#736212) - cib: Fix segfault caused by destroying an already freed hash table (cl#5023) - stonith: Expose IDs of stonith resources to stonith agents through "$CRM_meta_st_device_id" environment variable (bnc#728579) ------------------------------------------------------------------- Mon Jan 9 08:26:22 UTC 2012 - tserong@suse.com - Only include glib.h in plugin.h (don't directly use glib/ghash.h) ------------------------------------------------------------------- Fri Dec 16 04:42:38 UTC 2011 - tserong@suse.com - RA: add NodeUtilization - stonith-ng: export stonith resource name to stonith agents - crmd: Send out all of the meta parameters to lrmd for stop actions (bnc#715501) - RA: Fix the metadata of HealthSMART and SysInfo RAs - stonith: Longer default timeout/interval of operations for stonith resources - Doc: many proof-reading improvements - stonith: Fix the stonith timeout issue (cl#5009, bnc#727498) - update the license information in pacemaker.spec (bnc#726795) - xml: Add rsc_ticket support into "pacemaker-1.2" schema - xml: Add resource template support into "pacemaker-1.2" schema - tools: Avoid problems caused by differences between node name and uname -n - PE: Resolve memory leak when retrieving failure counts - Stonith: Implement a manual override for when nodes are known to be safely off - PE: Implement limit number of concurrent live migrations (FATE#310118) - PE: Implement referencing templates in constraints - PE: Move master based on failure of colocated group - Fencing: Support admin configurable timeouts for API operations - Fencing: Support checking in-progress operations for all nodes - don't package crm shell code - add Requires: crmsh - Build man pages and html versions (bnc#723680) - Shell: template support - Upstream version cs: b988976485d15cb702c9307df55512d323831a5e ------------------------------------------------------------------- Sun Nov 20 06:47:28 UTC 2011 - coolo@suse.com - add libtool as buildrequire to avoid implicit dependency ------------------------------------------------------------------- Tue Sep 20 14:27:47 UTC 2011 - tserong@suse.com - Upgrade to 1.1.6. - PE: Demote from Master does not clear previous errors - crmd: Prevent secondary DC fencing resulting from CIB updates that are lost due to elections - crmd: Log duplicate DC detection as a WARNING not ERROR - crmd: Bug lf#2632 - Correctly handle nodes that return faster than stonith (bnc#710314) - Core: Treat GNUTLS_E_UNEXPECTED_PACKET_LENGTH as normal termination of a TLS session - cib: Call gnutls_bye() and shutdown() when disconnecting from remote TLS connections - cib: Remove disconnected remote connections from mainloop - cib: Attempt a graceful sign-off for remote TLS connections - Core: Ensure there is sufficient space for EOS when building short-form option strings (prevents segfault) - Core: Fix variable expansion in pkg-config files - PE: Resolve memory leak reported by valgrind - PE: Fix memory leak for re-allocated resources reported by valgrind - PE: Improve the merging with template's operations - crmd: Allow nodes to fence themselves if they're the last one standing (lf#2584) - stonith: Add an API call for listing installed agents - stonith: Allow the fencing history to be queried - stonith: Ensure completed operations are recorded as such in the history - stonith: Support --quiet to display just the seconds since epoch at which a node was last shot - stonith: Serialize actions for a given device - stonith: Add missing entries to stonith_error2string() (missing error messages) - Shell: geo-cluster support commands - Shell: support for rsc_ticket - Shell: let the pager decide how to handle output smaller than terminal - Shell: look for log segments with more care and don't throw exception on seek (bnc#713939) - Shell: several history improvements - tools: crm_attribute - Update tickets directly to cib instead of through attrd - Shell: support for LRM secrets in resource level - Shell: enable removal of unmanaged resources (bnc#696506) - Shell: add crm execute directory to the PATH if not already present (bnc#712605) - Shell: update log patterns for history - Shell: relax transition acceptance in history - tools: crm_simulate - Send logging to stdout when additional detail is requested so that grep works as intended - PE: Prevent services being active if dependancies on clones are not satisfied (bnc#707150) - PE: Prevent resource shuffling when the 'utilization' placement strategy is used - PE: Implement resource template - fencing: Add support for --reboot to stonith_admin - Fencing: Downgrade an error message for a valid scenario - crmd: Ensure we do not attempt to perform action on failed nodes (bnc#703236, lf#2612) - Fencing: admin - Disable stderr logging by default - Fencing: Fix use-of-NULL by g_hash_table_lookup - Tools: Correctly log corosync node IDs (lf#2621) - pacemaker:ping: Incorporate fping functionality from Dan Urist - crmd: Recurring actions shouldn't cause the last non-recurring action to be forgotten - crmd: Do not wait for actions that were pending on dead nodes - crmd: Cancel timers for actions that were pending on dead nodes - PE: Correctly recognise which recurring operations are currently active - PE: Ensure role is preserved for unmanaged resources - PE: Ensure unmanaged resources have the correct role set so the correct monitor operation is chosen - PE: Implement cluster ticket and deadman - Build: Add crm_ticket into spec file - Tools: crm_attribute - No need to determine the node for tickets management - Tools: crm_mon crm_resource - Display the "operation_key" instead of the operation "id" if available (bnc#707201) - CTS improvements. - Shell: reimplement the history latest command (bnc#710958) - Shell: Don't limit to last hour when examining hb_reports - Shell: improve capture log slices for transitions (bnc#710907) - Shell: allow specifying PE files as relative paths in order to disambiguate between PE inputs with the same number (bnc#710655) - Shell: remove peinputs "get" and "list" subcommands, just use 'v' for the long listing (bnc#711060) - Shell: remove transition "show" subcommand, if there is no subcommand it is assumed that the user wants to do "show" (bnc#711060) - Shell: detect (and ignore) empty transitions - Shell: review transition interface (bnc#710655) - Fencing: Improved pcmk_host_map parsing - Fencing: Consolidate pcmk_host_map into run_stonith_agent so that it is applied consistently - Shell: Update regression tests for new stonith metadata - pengine: Correctly determine the state of multi-state resources with a partial operation history - Core: Cancelled and pending operations do not count as failed - PE: Ensure restarts due to definition changes cause the start action to be re-issued not probes - Fencing: Prevent use-of-NULL by g_hash_table_lookup - Shell: Install support for pssh - PE: Support of monitor op with role="Stopped" - Shell: improve logic for when to use the pager - Shell: don't print error on Broken pipe - cib: Record and display the originator details of CIB changes for schema "pacemaker-1.2" - crmd: Display the name and vlaue of the transient attribute(s) that caused a new transition - stonith: Correctly handle synchronous calls - PE: A demote operation is implied by the stonith operation (lf#2606) - PE: Group migration after failures and non-default utilization policies (lf#2613, lf#2619) - Shell: Include log file querying tools and "history" extension. - Significant number of fixes for issues reported by Coverity. - ais: Handle IPC error before checking for NULL data (bnc#702907) - PE: save PE inputs again (bnc#703723) - PE: Before migrating an utilization-using resource to a node, take off the load which will no longer run there (lf#2599, bnc#695440) - cib: Record and display the originator details of CIB changes - cib: Update "cib-last-written" for live CIB on a config change - cib: Filter out unnecessary "update-*" fields on digest calculating - doc: man page for crm - cib: Check the validation version before adding the originator details of a CIB change - PE: Don't log resource states unless debug is on - Shell: implement -w,--wait option to wait for the transition to finish - crmd: Store only the last and last failed operation in the CIB - Core: Use pre-glib2-2.28 version of g_str_hash to ensure consistent score calculation for clone instances - Shell: repair template list command - PE: Optimize the placement strategy and handle stickiness for utilization-based placement correctly (bnc#669686) - Shell: print error instead of warning and exit with 1 if a resource cannot be deleted (bnc#680401) - Tools: attrd - Do not initiate updates for deleted attributes - Tools: Make progress when attrd_updater is called repeatedly within the dampen interval but with the same value (lf#2528) - tools: Do not pass bash specific switches to other SHELLs (lf#2556) - crmd: Watch for config option changes from the CIB even if we're not the DC (lf#2509) - PE: Prevent shuffling by choosing the correct clone instance to stop (lf#2574) - PE: Avoid group restart when clone (re)starts on an unrelated node (lf#2581, bnc#681595) - PE: Set reasonable default limits for the number of input files saved - tools: Do not recursively look for resource meta attributes (lf#2563) - PE: Do not write out duplicated input files (lf#2582) - cib: Use SIGPIPE for re-enabling disk writes after an error (lf#2229) - crmd: Fail actions that were scheduled for a failed/fenced node (lf#2559) - Fencing: Clean up the metadata creation - Stonith: Allow the fencing agent operations to be overridden on a per-device basis (lf#2557) - Tools: crm_resource - Indicate master/slave state when querying the parent resource (lf#2477) - Tools: crm_mon - Display the true number of configured resources (lf#2549) - RA: Fix handling of 'devices' in the HealthSMART agent - Shell: print warning if start or stop interval is not zero - Shell: Check for violations of uniqueness for instance parameters during commit - PE: Indicate resource failures even if on-fail="ignore" - ais: Ensure we drain the corosync queue of messages when glib tells us there is input - crmd: Catch fence operations that claim to succeed but did not really - Core: Use pre-glib2-2.28 version of g_str_hash to ensure consistent score calculation for clone instances - Shell: repair template list command - Shell: node clearstate must flush LRM state too (bnc#653795) - Upstream version cs: 9971ebba4494 (release 1.1.6) ------------------------------------------------------------------- Fri Aug 5 06:09:30 UTC 2011 - tserong@novell.com - Fix rpmlint non-ghost-in-var-run check (bnc#710272) ------------------------------------------------------------------- Wed Feb 9 12:03:57 UTC 2011 - tserong@novell.com - PE: Consolidate logging for resources active on multiple nodes - PE: More accurate logging of restarts and recovery for master/slave resources - PE: target-role alone is not sufficient to promote resources (lf#2544) - ais: Ensure that process death is communicated to peer nodes (lf#2550) - PE: Certain stop operations require STONITH, could overload the batch-limit otherwise (lf#2551, bnc#669229) - PE: Suppress unnecssary assertion failure message (lf#2558, bnc#670983) - Shell: Do not print XML for permission denied errors - ACL: Support access control on "cleanup" - crmd: Cancel recurring operations while we're still connected to the lrmd - crmd: Suppress an ERROR by teaching fsa_action2string() about complex actions - crmd: ACL: Prohibit non-privileged users to "fail" resources - Core: Do not store last-run for recurring operations, since it is misleading - Shell: fix logic for not printing the default monitor depth - Doc: Add ACL documentation - Upstream version cs: ecb6baaf7fc0 ------------------------------------------------------------------- Tue Feb 8 13:24:24 UTC 2011 - tserong@novell.com - Upgrade from 1.1.2.1 to 1.1.5. Note that the following changelog entries cover changes made in versions 1.1.3 and 1.1.4 as well. - crmd: Cancel recurring operations while we're still connected to the lrmd - crmd: Suppress an ERROR by teaching fsa_action2string() about complex actions - PE: Consolidate logging for resources active on multiple nodes - PE: More accurate logging of restarts and recovery for master/slave resources - PE: target-role alone is not sufficient to promote resources (lf#2544) - ais: Ensure that process death is communicated to peer nodes (lf#2550) - PE: Certain stop operations require STONITH, could overload the batch-limit otherwise (lf#2551, bnc#669229) - Shell: Do not print XML for permission denied errors - Doc: Add ACL documentation - Core: Do not store last-run for recurring operations, since it is misleading - crmd: Purge status for resourcs that no longer live in the LRMd (lf#2536) - Stonith: Teach stonith_admin to automagically obtain the current node attributes for the target from the CIB - tools: Prevent use-of-NULL in crm_simulate (lf#2527) - ACL: Convert uid directly with strtol() for avoiding unnecessary error logs - Tools: Make unmanaged resources stand out in crm_simulate graphs - PE: Correctly reconstruct the status of anonymous cloned groups (lf#2508) - Core: Correctly parse the resource name from notificiation action IDs (lf#2511) - crmd: Indicate the interval for the timer that just expired - crmd: Reschedule the PE_START action if its not already running when we try to use it - crmd: Ensure notify variables are accurate for stop operations (lf#2545) - crmd: Implement reprobe natively in the crmd to avoid transition loops (lf#2533) - crmd: Cancel recurring monitor ops at shutdown (lf#2309) - PE: Don't imply colocation requirements when applying ordering constraints with clones (lf#2493) - PE: Protect against constraints that require stonith devices to stop before 'normal' ones (lf#2529) - PE: Prevent unstable clone placement by factoring in the current node's score before all others (lf#2544) - Tools: Suport cibadmin --empty for creating a minimal CIB - PE: Do not override configured clone stickiness values (lf#2445) - PE: Avoid relying on stickness for stable clone placement (lf#2445) - Shell: work around lrmd/lrmadmin blocking when not run as root or hacluster - Shell: more lazy help system initialization - Access control list support for pacemaker's shell and CIB. - PE: Prevent segfault by validating the contents of ordering sets (lf#2495) - PE: Prevent use-after-free resulting from unintended recursion when chosing a node to promote master/slave resources - PE: Avoid flooding the logs with errors for orphan resources (lf#2518) - cib: Add "enable-acl" property - Shell: don't create empty optional sections (bnc#665131) - xml: Add ACL support into pacemaker-1.2 and 1.1 schemata. - RA: HealthSMART: Several fixes. - RA: SysInfo: Allow a list of filesystems/devices to be queried for free space - RA: ClusterMon: htmlfile is not mandatory (lf#2512) - Shell: merge properties and rsc/op_defaults on load update (lf#2454) - stonith: use ST_DEVICEID for short description - Shell: combine crmd and PE meta-data (lf#2373,2419) - PE: Only update the node's promotion score if the resource is active there - PE: Only use the promotion score from the current clone instance - Shell: ask for confirmation for node fence and clearstate (bnc#642268) - Shell: insert sleep between stop and start in resource restart - stonith: remove unnecessary logging - Shell: fix exception when HOME is not set - Shell: Allow creation of an empty shadow CIB ("cib new ... empty") - Shell: Allow cibstatus to create new node status entries. - Shell: fix rsc/op_defaults edit - Shell: repair management of cloned groups - Shell: wait for the stop transition to finish before invoking start in the resource restart command (lf#2516) - PE: Correct target_rc for probes of inactive resources - Tools: Prevent crm_resource commands from being lost due to the use of cib_scope_local - PE: Prevent use-of-NULL resulting from variable shadowing spotted by Coverity - Shell: allow adding node attributes in configure when the attributes set doesn't already exist - Core: Repair reference to child-of-child after removal of xml_child_iter_filter from get_message_xml() - Core: Remove extra calls to xmlCleanupParser resulting in use-after-free - Core: Reinstate the logic that skips past non-XML_ELEMENT_NODE children - Core: Update the default validation type to pacemaker-1.0 - Core: Prevent recently introduced use-after-free in replace_xml_child() - crmd: Update failcount for failed promote and demote operations - PE: Prevent use-of-NULL when there is status for an undefined node - Shell: warn on non-existing nodes in location constraints (lf#2486) - Shell: put completion bits into a separate module - Shell: repair ability to display multiple property sets - Build: Script for determining test coverage - Remove usage of the xml_child_iter macro to obtain better information from tracing tools. - Shell: Let crm_resource handle node & resource iteration during resource cleanup (skips down nodes: bnc#647378) - PE: Avoid needless string compares - cib: Fix a log message - Core: Provide arguments to do_crm_log_alias() in the correct order when logging xml - Add gcov support to spec file (disabled by default) - crmd: Determin the target rc for an action once - crmd: Prevent use of NULL by making copies of the operation's hash table - Core: Turn off unneeded checks when applying a diff - crmd: Allow operations in the PE state - crmd: Ensure the CIB is always writable on the DC by removing a timing hole - cib: Deprecate the set_slave_all() API call - PE: Ensure fencing of the DC preceeds the STONITH_DONE operation - PE: Complete all fencing before attempting atomic migrations - stonith-ng: fix two memory leaks. - Core: Preserve ordering when removing subtrees with a diff - Tools: Calculate the digest after applying a diff - Core: Correctly extract the version details from a diff - cib: Drop the severity of a log message - crmd: Don't abort transitions when probes are completed on a node - PE: Try to ensure atomic migration ops occur within a single transition - Core: Correctly compare against crm_log_level in the logging macros - cib: Revert part of a recent commit that purged still valid connections - PE: reduce logging severity for single rsc sets - Core: Prevent use-after-free when creating a CIB update for a timed out action - Shell: update shadow file directory processing - Tools: crm_resource - Improved output when listing constraints relevant to a given resource - Tools: crm_resource - Fix -x, the short form of --xml-file - Tools: crm_resource - Include location constraints in -A|-a output - Tools: crm_shadow - Avoid replacing the entire configuration (including status) - PE: Prevent clones from being stopped because resources colocated with them cannot be active - crmd: Include the correct transition details for timed out operations - Core: Drop the compression limit (lf#2506) - cib: Don't remove client's when notifications fail, they might just be too big (lf#2506) - cib: Do not calculate changes for status updates that wont be broadcast - cib: Speed up the establishment of remote connections - crmd: All pending operations should be recorded, even recurring ones with high start delays - stonith-ng: reduce logging. - Core: Provide significant performance improvements by implementing versioned diffs and digests - cib: Reject updates with a feature set greater than our own - Core: Some performance enhancements when comparing xml objects - crmd: Use the supplied timeout for stop actions - Core: Ensure ordering is preserved when applying diffs - crmd: There's no need to check the cib version from the 'added' part of diff updates - crmd: Additional logging for the cause of new transitions - cib: Drop invalid/failed connections from the client hashtable - PE: Use the original digest algorithm for parameter lists - cib: Ensure all diffs sent to peers have sufficient ordering information - cib: Reduce unnecessary copying when comparing xml objects - Core: Purge diff markers before calculating and comparing digests - Core: Minor speedup when comparing xml objects - cib: The feature set version check is only valid if the current value is non-NULL - Tools: crm_resource - Ensure xml-related memory is cleaned up at exit - Tools: crm_simulate - Ensure xml-related memory is cleaned up at exit - PE: lf#2478 - Implement improved handling and recovery of atomic resource migrations - PE: Fix memory leak when processing completed migration actions - Large number of performance improvements to the Policy Engine and CIB. - Core: Cache Relax-NG schemas - profiling indicates many cycles are wasted needlessly re-parsing them - Core: Don't compress large payload elements unless we're approaching message limits - Tools: crm_node - Fix the return code for operations on corosync clusters - Stonith: Fix two memory leaks - crmd: Record completed migrate ops in CIB status section as migrate_to/migrate_from, not as stop/start (lf#2481) - PE: Fix typo leading to use-of-NULL in the new ordering code (bnc#653635) - crmd: on version print everything to stdout - stonith: cleanup children on timeout in fence_legacy - Shell: use uname instead of id in nodes when building other ids - Shell: add "clearstate" command for nodes - crmd: Don't hide stop events that time out - allowing faster recovery in the presence of overloaded hosts - PE: Use hashtables instead of linked lists for performance sensitive datastructures - PE: Skip probe_resources() completely if probing is disabled - PE: Obtain massive speedup by prepending to the list of ordering constraints (which can grow quite large) - PE: Detect start/stop actions with intervals - PE: Optimize the logic for deciding which non-grouped anonymous clone instances to probe for - stonith: Use the timeout specified by the user - stonith: Add the timeout at the top level where the daemon is looking for it (lf#2473) - stonith: Ensure timeouts are included for fencing operations (lf#2473) - Logging tuning - crmd: Use an operation's interval when an invalid timeout was specified - crmd: Use the correct define/size for lrm resource IDs - Tools: Ensure the line we're grabbling the time from contains a time - Tools: crm_simulate - Fix use-after-free in when terminating - Stonith: Ensure stonith operations complete within the timeout and are terminated if they run too long (lf#2473) - crmd: Prevent assertion failure for stop actions - PE: Prevent segfault by ensuring the arguments to do_calculations() are initialized - PE: Correct use-after-free - Tools: crm_simulate - Resolve memory leak reported by valgrind - PE: Ignore failed actions on nodes that have been shutdown - crmd: Better logging of failed stonith operations (lf#2473) - stonith: Gracefully handle remote operations that arrive late (after we've done notifications) (lf#2473) - Tools: crm_simulate - Resolve memory leak reported by valgrind - PE: Repair on-fail=block for groups and primitive resources (lf$2476) - crmd: Ensure stop actions always have the relevant resource attributes (lf#2458) - Tools: Fix assertion failure in crm_resource (lf#2456) - PE: Do not demote resources because something that requires it can't run - PE: Display managed/unmanaged state for resources when grouped by node in crm_mon (lf#2479) - cib: Clean up disconnected remote connections (lf#2469) - Tools: crm_mon - Only display attributes of online nodes (and sort them) - RA: ping - Indicate that host_list is a required parameter in the metadata - stonith: Prevent segfault by not looking up operations if the hashtable hasn't been initialized yet (lf#2461) - Tools: Have crm_resource --locate indicate which instances are promoted (lf#2477) - PE: Ensure resource op timeout takes precedence over op_defaults (lf#2474) - Upstream version cs: bdd89e69ba54 ------------------------------------------------------------------- Thu Nov 18 00:56:14 CET 2010 - meissner@suse.de - SUSE_ASNEEDED also during %install, if by chance we rebuild things in there. ------------------------------------------------------------------- Mon Aug 16 20:08:41 UTC 2010 - lmb@novell.com - cib: Allow backup files to proceed after wrap-around (bnc#630322) - Upstream version cs: dfc6b2561373 ------------------------------------------------------------------- Mon Aug 9 20:17:36 UTC 2010 - lmb@novell.com - stonith-ng: record exit code and log output for failed port list query (bnc#625594) - Upstream version cs: 0db65fda4b15 ------------------------------------------------------------------- Fri Aug 6 13:24:07 UTC 2010 - lmb@novell.com - Shell: print error if a resource is not found in stop/start/manage (bnc#627757) - PE: Fix memory leak when unpacking recurring operations - PE: Fix potential use-of-NULL when calculating the set of required actions - Shell: set shadow directory properly (bnc#626638) - Core: Enable coredumps properly for MCP and AIS plugin (lf#2463) - Upstream version cs: 61360150595d ------------------------------------------------------------------- Thu Aug 5 14:55:22 UTC 2010 - tserong@novell.com - Explicitly require libglue2 to avoid potential conflict with libheartbeat2 (bnc#628243) ------------------------------------------------------------------- Wed Jul 28 13:55:08 UTC 2010 - lmb@novell.com - crmd: Prevent everyone from losing DC elections by correctly initializing all relevant variables - crmd: Ensure we activate the DC timer if we detect an alternate DC - PE: Wait until stonith is available, don't fall back to shutdown for nodes requesting termination - Upstream version cs: e0d731c2b1be ------------------------------------------------------------------- Wed Jul 21 10:15:37 UTC 2010 - lmb@novell.com - PE: Correctly order clone startup after group/clone start - stonith-ng: pass the configuration to the stonith program via environment variables; requires cluster-glue >= 1.0.6 (bnc#620781, CVE-2010-2496) - Logging improvements. - Upstream version cs: f9fbf97b0ce8 ------------------------------------------------------------------- Wed Jul 14 14:58:10 UTC 2010 - lmb@novell.com - Shell: allow adding force parameter to resource migrate - Shell: add the actions option to configure ptest (lf#2360) - Shell: test early for duplicate elements in edit (bnc#620460) - Shell: Use crm_attribute directly instead of the deprecated crm_standby and crm_failcount wrappers - PE: Correctly detect when there is a real failcount that expired and needs to be cleared - PE: Avoid creating invalid ordering constraints for probes that aren't needed - PE: Include the prefix for clone children when printing the clone - PE: Bug lf#2453 - Enforce clone ordering in the absense of colocation constraints (lf#2453, bnc#619581) - crmd: Only log the failcount op if something is wrong - crmd: Base DC election on process relative uptime - crmd: Make sure we always poke the FSA after a transition to clear any TE_HALT actions - crmd: Make sure the membership cache is accurate after a sucessful fencing operation - crmd: All nodes should see status updates, not just the DC - stonith-ng: Advertise stonith-ng options in the metadata - Core: Correctly unpack HA_Messages containing multiple entries with the same name - ais: Sign everyone up for peer process updates, not just the crmd - ais: Move the code for finding uid before the fork so that the child does no logging - Upstream version cs: e9b2f255a20e ------------------------------------------------------------------- Tue Jul 6 14:53:24 UTC 2010 - lmb@novell.com - PE: Fail unmanaged resources should not prevent other services from shutting down (lf#1959) - crmd: Handle asynchronous notification of resource deletion events (lf#2439) - Shell: add support for classes to the providers command - crmd: Pass in the correct result when processing synchronous resource deletion operations - ais: Fix unterminated recursion in init_ais_connection() - Tools: Make sure crm_resource waits for the completion of crmd operations it sent - stonith: Support the poweron/poweroff actions (bnc#618798) - stonith: Map poweron/poweroff back to on/off expected by the stonith tool from cluster-glue (bnc#618798) - PE: Correctly handle pseudo action creation - Tools: crm_node - Enable the -A parameter, the short form of --openais - CLI: report bad ISO8601 duration specifiers (bnc#618808) - crmd: Fix assertion failure when performing async resource failures - crmd: Include the action for crm graph operations - crmd: Allow non-DC nodes to clear failcounts too and tidy up logging - Fix some valgrind reported leaks. - Upstream version cs: ea6cdd9d9613 ------------------------------------------------------------------- Mon Jun 14 14:25:25 UTC 2010 - lmb@novell.com - Shell: templates: clvm,ocfs2: add the fs/vg resource to the base group, add timeouts (bnc#605580) - Shell: skip resource checks for property-like elements (lf#2420) - Shell: on no id match the first of property-like elements (lf#2420) - Shell: set proper ids for rsc/op_defaults (lf#2420) - Shell: add exit codes for non-interactive use - PE: No services should be stopped until probes finish (lf#2433) - PE: Support clones in resource sets (bnc#586613, lf#2404, lf#2409) - Tools: Allow ptest and crm_simulate to work with configs that lack a status section - Core: Reduce logging for failed xpath searches - PE: Support colocation sets with negative scores (lf#2435) - PE: Ensure deallocated resources are stopped - PE: Ordering dependencies on partially active groups not observed properly (lf#2422, bnc#607125) - PE: If there are no compatible peers when interleaving clones, ensure the instance is stopped - PE: Fix colocation for interleaved clones - PE: Fix colocation with partially active groups - PE: Don't overwrite existing meta attributes (like timeout) for notify operations (lf#2424) - PE: stop-all-resources shouldn't apply to stonith resources (lf#2426) - Fencing: Reduce stonith-ng logging (lf#2431) - ais: Improved processing when the peer crmd processes join/leave (lf#2401) - Tools: crm_resource - Report crmd connection errors - Tools: Depreciate the 'pingd' daemon and resource agent in favor of ocf:pacemaker:ping - cib: Fix the application of unversioned diffs - Several valgrind + coverity issues fixed. - Upstream version cs: 8b9ec9ccc506 ------------------------------------------------------------------- Tue Jun 8 02:43:35 UTC 2010 - tserong@novell.com - Install crm shell and cts in $(pyexecdir) instead of $(pythondir) to ensure installation in /usr/lib64 on x86_64 ------------------------------------------------------------------- Tue May 18 06:30:42 UTC 2010 - tserong@novell.com - Fix crm membership check (bnc#603685, reverts changesets 6e6980376f01 and 2e096a41a5f9) - Upstream version cs: e8717a29601a (1.1.2.1) ------------------------------------------------------------------- Fri May 14 13:08:29 UTC 2010 - lmb@novell.com - ais: fix list of active processes sent to clients (bnc#603685) - Upstream version cs: 2e096a41a5f9 ------------------------------------------------------------------- Wed May 12 13:57:23 UTC 2010 - lmb@novell.com - crmd: Prevent use-after-free of the stonith-ng connection (lf#2414) - Core: Prevent use-after-free reported by valgrind when doing xpath based deletions (lf#2414) - PE: Fix memory leaks reported by valgrind - Core: Fix memory leak in replace_xml_child() reported by valgrind - PE: Remove developemnt logging which mistakenly logged at "ERROR" level. - Upstream version cs: b59c27dc114a (1.1.2) ------------------------------------------------------------------- Mon May 10 13:19:09 UTC 2010 - lmb@novell.com - crmd: Prevent use-after-free of the PE connection after it dies (lf#2414) - Shell: support for clone instances in constraints (lf#2169) - Upstream version cs: 42a5aa3413f4 ------------------------------------------------------------------- Mon May 10 09:52:28 UTC 2010 - lmb@novell.com - Shell: new command default-timeouts to set default operation timeouts (bnc#479052) - crmd: ais: Use the threadsafe version of getpwnam - PE: Enforce mandatory promotion (colocation) constraints (lf#2403) - PE: Don't complain about missing agents during probes of a-symetric clusters (lf#2410) - Shell: don't error on missing resource agent with asymmetrical clusters (lf#2410) - ais: Ensure the list of active processes sent to clients is always up-to-date - PE: Allow constraints to apply to specific instances of clones (lf#2169) - PE: Allow ordering constraints to apply to specific instances of clones - PE: Clean up failcount after it expires (lf#2066) - PE: Before starting utilization-using resources on a node, take off the load which will no longer run there (bnc#594782) - PE: Don't be so quick to pull the trigger on nodes that are coming up - crmd: Use global fencing notifications to prevent secondary fencing operations of the DC - cib: Changes to group order are not detected or broadcast to peers (lf#2352) - Core: Only use the named schema if the validation passes - Shell: don't verify empty configurations (bnc#602711) - crmd: Send erase_status_tag() calls to the local CIB when the DC is fenced, since there is no DC to accept them - crmd: Include operation interval in logged action - crmd: Arrange for only one node to clean up after a fenced DC node - crmd: create the final directory too for resource agents (bnc#603190) - Tools: crm_mon: refresh screen on terminal resize (bnc#589811) - PE: Correctly locate clone instances by their prefix (lf#2412) - ais: Make a better attempt at determining an accurate born-on value from corosync - tools: crm_mon - Enable 'connectivity' mode for 'ping' resources too - Upstream version cs: e8624c181a3f ------------------------------------------------------------------- Tue Apr 27 08:54:11 UTC 2010 - lmb@novell.com - Fix small error so that regression test suite passes. ------------------------------------------------------------------- Tue Apr 27 03:07:40 UTC 2010 - ygao@novell.com - PE: Before starting utilization-using resources on a node, take off the load which will no longer run there (bnc#594782) ------------------------------------------------------------------- Mon Apr 26 10:03:44 UTC 2010 - lmb@novell.com - PE: Don't create scaffolding for ordering sets with a single resources (bnc#586613) - PE: Ignore colocation weights from clone instances - PE: Print allocation scores as +/-INFINITY when appropriate - Upstream version cs: fa1c810f1d1d ------------------------------------------------------------------- Fri Apr 23 10:09:51 UTC 2010 - lmb@novell.com - Fix regression test for build environment. - Upstream version cs: c11e18619efd ------------------------------------------------------------------- Wed Apr 21 13:25:26 UTC 2010 - lmb@novell.com - Fix several memory leaks exposed by valgrind. - Improve regression tests. - Low: Fencing: Delay system initialization until after we handle possible metadata queries - Tools: Do not print help on executing "crm_node -i" - crm_simulate: Allow bringing new node online (bnc#591162) - crmd: Improved detection of partially active peers (lf#2401) - PE: Fix intra-set colocation and ordering (lf#2384) - PE: Combine failcounts for all instances of an anonymous clone on a host (lf#2383) - crmd: Add missing fields to the metadata for use by the shell (lf#2387) - PE: Check for use-of-NULL in dump_node_scores() - PE: Rewrite native_merge_weights() to avoid Fix use-after-free - Upstream version cs: 2a90abe96048 ------------------------------------------------------------------- Mon Apr 12 11:09:09 UTC 2010 - lmb@novell.com - Automated test case improvements. - Upstream version cs: 536bf0b9d3ba ------------------------------------------------------------------- Fri Apr 9 08:38:55 UTC 2010 - lmb@novell.com - Shell: always reload status if working with the cluster (bnc#590035) - Shell: add new regression tests - Shell: don't allow empty attributes for properties/*_defaults (lf#2395) - Shell: add another attribute not present in the crmd metadata - Tools: crm_mon - Enable specification of community for SNMP traps - PE: Ensure important options (notify, unique, etc) are always exposed during resource operations (lf#2391) - Tools: Support short form of crm_simulate --quorum (bnc#591168) - ais: Convert endian for all of the integer fields in messages (bnc#472014) - Upstream version cs: 80e0bb600e6c ------------------------------------------------------------------- Wed Mar 31 19:18:35 UTC 2010 - lmb@novell.com - Shell: Default to using the status section from the live CIB (bnc#592762) - Upstream version cs: 739282ba9e2e ------------------------------------------------------------------- Wed Mar 31 13:03:43 UTC 2010 - lmb@novell.com - crmd: fix non-fatal assert in lrmd.c when resources are manually failed (bnc#590924) - Upstream version cs: f77530201221 ------------------------------------------------------------------- Tue Mar 30 13:12:24 UTC 2010 - lmb@novell.com - tools: crm_simulate - fix small xpath memory leak in inject_node_state - Shell: templates: new sbd template - Shell: templates: update apache documentation - crmd: Do not ignore action timeouts based on FSA state (could cause transitions that never terminate) - Upstream version cs: 322ff3419bc8 ------------------------------------------------------------------- Mon Mar 29 14:55:27 UTC 2010 - lmb@novell.com - Shell: add two attributes not present in the crmd metadata, which broke "configure verify" - Upstream version cs: 24ee74882b42 ------------------------------------------------------------------- Mon Mar 29 13:58:05 UTC 2010 - lmb@novell.com - Shell: fix bad cib passed to crm_verify - Upstream version cs: c02327f742a0 ------------------------------------------------------------------- Mon Mar 29 13:05:26 UTC 2010 - lmb@novell.com - Shell: verify only changed elements on commit (bnc#590033) - Shell: observe op_defaults when verifying primitives (bnc#590033) - Shell: update meta attributes list - Shell: verify meta attributes and properties (bnc#589867) - Shell: fix wrong exit code setting logic in resource cleanup - Shell/PE: Show utilization/capacity information (bnc#589873) - Shell: implement new cibstatus interface and commands (bnc#580492) - crmd: Do not allow the target_rc to be misused by resource agents - Shell: check python version at start (lf#2369) - extra/showscores.sh: Correctly display stickiness if unset - RA: SystemHealth: exit properly when the required software is not installed (bnc#587940) - ais: Look for the correct conf variable for turning on file logging - crmd: Ensure the cluster terminates when the PE isn't available (lf#2379) - crmd: Ensure we dont get stuck in S_PENDING if we loose an election to someone that never talks to us again - Upstream version cs: 3fea355803af ------------------------------------------------------------------- Fri Mar 19 19:59:58 UTC 2010 - lmb@novell.com - crmd: Remove race condition that could lead to multiple instances of a clone being active on a machine - PE: Consolidate and unify logging of fencing causes - hb2openais: add support for corosync as target - Shell: templates: add cmirror to ocfs2 and clvm (bnc#587493) - Shell: support for resource/node utilization - Shell: switch to single cibadmin for commit - Shell: enable comments (lf#2221) - Upstream version cs: 83f654529b5c ------------------------------------------------------------------- Mon Mar 15 11:23:49 UTC 2010 - lmb@novell.com - Shell: don't touch resources within a group when managing groups/clones - Shell: repair node show (bnc#587883) - Upstream version cs: 3002fabf85d4 ------------------------------------------------------------------- Wed Mar 10 15:16:33 UTC 2010 - lmb@novell.com - PE: Make sure migration ops use the correct meta options (bnc#586710) - Upstream version cs: aaf4d36b8494 ------------------------------------------------------------------- Tue Mar 9 21:39:20 UTC 2010 - dmuhamedagic@novell.com - Shell: fix commit for new clones of new groups (bnc#585471) - Shell: edit multiple meta_attributes sets in resource management (lf#2315) - Shell: don't cache node and resource ids (lf#2368) - Shell: new cibstatus import command (bnc#585471) - Shell: check timeouts also against the default-action-timeout property ------------------------------------------------------------------- Tue Mar 9 14:29:40 UTC 2010 - lmb@novell.com - PE: Avoid needless restart of primitive depending on a clone (lf#2317, bnc#586613) - PE: Only log orphans when they are found to be active (lf#2364) - PE: Ensure clones observe mandatory ordering constraints if the LHS is unrunnable (lf#2361) - Tools: Add utilization support for crm_attribute and crm_resource - Upstream version vs: c23bbc5262b2 ------------------------------------------------------------------- Fri Mar 5 15:45:15 UTC 2010 - lmb@novell.com - ais: Do not count votes from offline nodes and calculate current votes before sending quorum data. - PE: Repair handling of unordered groups in RHS ordering constraints. - Shell: update regression tests. - CTS: Make it work for openais. - Upstream version cs: a9f5a06a8f9e ------------------------------------------------------------------- Thu Mar 4 09:57:38 UTC 2010 - lmb@novell.com - Conflict with ocfs2-tools-o2cb prevented said package from being installed at all due to dependencies (bnc#582697). ------------------------------------------------------------------- Wed Mar 3 19:46:15 UTC 2010 - lmb@novell.com - Shell: implement lifetime for rsc migrate and node standby (lf#2353) - Shell: node attributes update in configure (bnc#582767) - Tools/PE: Do not warn if the configuration is fully forward-compatible. - Upstream version cs: 745c756006da ------------------------------------------------------------------- Mon Mar 1 12:04:11 UTC 2010 - lmb@novell.com - Shell: update previous node lookup procedure to include the id where necessary - Shell: improve batch mode. - Shell: add support for xml in cli. - Shell: repare clone/ms cleanup (bnc#583288) - Shell: load update should update existing elements. - crmd: expected-quorum-votes not correctly updated after membership change (lf#2359) - PE: Fix master-master anti-colocation (lf#2358) - Upstream version cs: 059e4095d2ca ------------------------------------------------------------------- Wed Feb 24 10:15:57 UTC 2010 - lmb@novell.com - Work around non-thread-safe setenv() calls that could cause a crash (bnc#582598) - Upstream version cs: ae0170640c3e ------------------------------------------------------------------- Tue Feb 23 15:35:05 UTC 2010 - lmb@novell.com - Shell: update supported CIB RE to include pacemaker 1.1 and 1.2 - ais: Fix the child shutdown escalation logic - Core: Give signal handlers higher priority - PE: Correctly implement optional colocation between primitives and clone resources - crmd: Don't send operation updates for deleted resources (bnc#578644) - Upstream version cs: 530add2a3721 ------------------------------------------------------------------- Mon Feb 22 14:00:00 UTC 2010 - lmb@novell.com - Shell: add tab completion for nodes (bnc#577003) - Shell: fix wrong var reference - Shell: improve configure commit (lf#2336) - Shell: add tab completion for op/rsc_defaults (bnc#552541) - Shell: add option to control sorting of cib elements (lf#2290) - Shell: add tab completion for nodes (bnc#577003) - RA: ping(d) - Fix metadata. - xml schema: support differentiation between stable and experimental feature sets. - Upstream version cs: 0591630c8bd3 ------------------------------------------------------------------- Thu Feb 18 22:04:16 UTC 2010 - lmb@novell.com - Conflict with ocfs2-tools-o2cb; a system needs to be configured either for Pacemaker or O2CB/Oracle RAC, but never both. ------------------------------------------------------------------- Thu Feb 18 14:27:00 UTC 2010 - lmb@novell.com - stonith-ng: Fix meta-data not always being retrieved properly. - Tools: crm: fix CIB upgrade command (bnc#578637) - PE: Improved graphs when probing partially up nodes - crmd: Improve handling of cancelled operations caused by resource cleanup (bnc#578644) - PE: Significant performance bugs for larger clusters with many nodes/resources/constraints removed. - Upstream version cs: 4bc7a38e2b10 ------------------------------------------------------------------- Tue Feb 9 10:27:47 UTC 2010 - lmb@novell.com - Add dependencies on python and python-xml so the shell works immediately (bnc#578006). ------------------------------------------------------------------- Mon Feb 8 15:07:35 UTC 2010 - lmb@novell.com - Incorporate code for hard resource limits (FATE#303384) in shell and pacemaker policy engine. - External stonith agents were not properly handled (bnc#577007). - Upstream version cs: 46679a8feec7 ------------------------------------------------------------------- Fri Jan 22 10:36:45 UTC 2010 - lmb@novell.com - crmd: Ensure we have the latest status after a transition abort - crm: optimize shell start-up time (lf#2303). - crm: simplify template for ocfs2 - crm: don't remove sets which contain id-ref attribute (lf#2304) - crm: add update method to template apply (lf#2289) - crm: make template new more robust, add documentation - crm: fix check for invalid constraints - crm: rename status to cibstatus (lf#2236) - crm: save options if run from the command line (lf#2310) - crm: add notify to the list of valid operations - crm: fix node delete (lf#2305) - crm_node: accept -f as the short form of --force - crm_node: rovide a way to determin the nodeid of past cluster members (lf#2307) - Upstream version cs: 6a6aa02deea9 ------------------------------------------------------------------- Mon Jan 11 15:49:30 UTC 2010 - lmb@novell.com - PE: Don't log uninstalled resource agents as errors (lf#2251) - fix refresh time for html interface of crm_mon - Tools: crm: don't consider warnings as errors (lf#2274) - Tools: crm: direct RA interface for ocf class resource agents (lf#2270) - cib: Check also whether an user's primary group is matched - PE: Bug lf#2153 - non-clones shouldn't restart when clones stop/start on other nodes - PE: Bug lf#2153 - Update regression tests - PE: Correctly anti-colocate with a group - Tools: crm: define missing index in validate_template (lf#2287) - Tools: crm: stay in crm if user specified level only (lf#2286) - Tools: crm: fix exit code on single shot commands - Tools: Bug lf#2286 - Allow the shell to accept template parameters on the command line - Tools: crm: allow entering level and input file (-f) - Tools: crm: relax check for RA existence (lf#2270) - Tools: crm: fix OCF_ROOT processing - Tools: crm: catch EOFError exception in raw_input (lf#2289) - Tools: crm: implement -F (--force) option - Tools: crm: rename some misnamed functions - PE: Prevent non-multistate resources from being promoted through target-role - Tools: crm: direct RA interface for stonith class resource agents (lf#2270) - fencing: stonithd: ignore case when comparing host names (lf#2292) - Tools: crm: direct RA interface for lsb class resource agents (lf#2270) - PE: Bug lf#2153 - non-clones shouldn't restart when clones stop/start on other nodes - improved - TE: Fix batch-limit processing. - Tools: crm: exclude locations when testing for pathological constraints (lf#2300) - PE: Only complain about target-role=master for non m/s resources - Tools: crm: load help from crm_cli.txt - pengine/crmd: move crm_log_init after version and metadata calls (lf#2272) - crmd: Ensure batch-limit is correctly enforced - PE: Don't overwrite the value of batch-limit - cts: Add the 'cmd' to the output of external commands - pengine: crmd: Ensure help text includes correct binary name - Tools: crm: don't fall apart if there's no help text file - Tools: crm: cache help tables (lf#2303) - Upstream version cs: 443ff5d69319 ------------------------------------------------------------------- Thu Dec 10 18:46:03 UTC 2009 - "lmb@novell.com" - PE: Correctly unpack ordering constraints for resource sets to avoid graph loops - Two new system health RAs. - PE: Don't log uninstalled resource agents as errors (lf#2251) - Upstream version cs: cd2aaf7e35cf ------------------------------------------------------------------- Tue Dec 8 08:44:07 UTC 2009 - "lmb@novell.com" - Tools: crm: fix parentheses parsing in resource sets - Tools: crm: support for description attribute in resources/operations (bnc#548690) - Tools: hb2openais: add EVMS2 CSM processing (and other changes) (bnc#548093) - Tools: crm: don't add score which doesn't exist - Tools: crm: copy score from/to resource_set - Tools: crm: update tab completion for constraints (support resource sets) - Tools: crm: handle resource_set resource references in delete and rename - PE: Provide a default action for resource-set ordering - Tools: crm: drop empty attributes elements - xml: Add stopped as a supported role for operations - Tools: hb2openais: don't allow empty rules, clones, or groups (lf#2215) - Tools: hb2openais: refuse to convert pure EVMS volumes - Tools: hb2openais: add inf score to order constraints - crm: Enhance "colocation" help to explain how it works. - crm_mon: Mail subject needs space for \0 byte. - fencing: stonithd: add metadata for stonithd instance attributes (and support in the shell) - Tools: crm: drop short description from the metadata if it equals long description - Tools: crm: relax checks for advertised monitor operations (lf#2222) - Tools: crm: support for comments (lf#2221) - Tools: crm: test operation name against a list of possible ops - tools: Dont waste vertical space in crm_mon output - xml: Treat node unames as text fields not IDs (bnc#552713) - PE: rsc_order constraints always use score at the top level (lf#2206) - xml: Create an always-true expression for empty rules when upgrading from 0.6 (lf#2215) - PE: Clone ordering should be able to prevent startup of dependant clones (lf#2209) - PE: Add regression test (lf#2209) - PE: Correctly identify the state of anonymous clones when deciding when to probe (lf#2216) - PE: Operations that require fencing should wait for 'stonith_complete' not 'all_stopped'. (lf#2225) - PE: Prevent clone peers from stopping while another is instance is (potentially) being fenced (lf#2225) - crmd: C_TIMER_POPPED is now quite normal thanks to the recheck timer, downgrade log message - PE: Silently fix requires=fencing for stonith resources so that it can be set in op_defaults - xml: Preserve the old stonith-enabled=false symantics from 0.6 during an upgrade (lf#2230) - xml: Allow the auto-id base to be specified, because generate-id() isn't always unique - PE: Ensure groups process location constraints so that clone-node-max works for cloned groups (lf#2213) - Tools: crm: relax check for RA actions - Upstream version cs: 9f61f272598a ------------------------------------------------------------------- Tue Nov 3 23:00:00 UTC 2009 - "lmb@novell.com" - PE: Provide better defaults for lt(e) and gt(e) comparisons (bnc#515172) - Upstream version cs: 405fe9a92d82 ------------------------------------------------------------------- Tue Sep 29 06:35:15 UTC 2009 - "lmb@novell.com" - Revert conflicts with ocfs2-tools-o2cb to ease automatic upgrades. The two should still not be installed in parallel; please see release notes! - High: ais: Gracefully handle changes to the AIS nodeid - Low: PE: Fix the metadata to advertise the correct type for stonith-timeout - High: cib: Repair the ability to connect to the cluster from non-cluster machines - Medium: Tools: Use -q as the short form for --quiet (for consistency) - Medium (bnc#520707): Tools: crm: new templates ocfs2 and clvm - Medium (LF 2185): Tools: crm: add resource promote/demote commands - Low (LF 2175): Tools: crm: add help for node status - High: xml: Mask the "symmetrical" attribute on rsc_colocation constraints (bnc#540672) - Upstream version cs: 05c8b63cbca7 ------------------------------------------------------------------- Thu Aug 27 15:58:01 UTC 2009 - "lmb@novell.com" - Upstream version cs: 595cca870aff - Dev: Tools: crm: do not allow empty attribute names - Low: Tools: crm: make check for operations id attribute stricter - Dev: Tools: crm: implement binary op types properly (thanks to Alain St-Denis) - Dev: Tools: crm: allow id-refs to stand by themselves (update for fix 37077ed56af5) - High: crmd: Terminate if we are ever evicted from the membership - High: crmd: Unset any existing DC value before querying for a new one - High: PE: Bug bnc#515172 - Fix the boolean-op attribute of rules - High: PE: Bug bnc#515172 - Correctly process location constraint rules which contain multiple expressions - High (bnc#507255): Tools: crm: implement date expressions - Low: Tools: crm: let the user know that curses is needed for colors - High: Tools: crm: new display type (uppercase keywords) - Medium: Extra: Add tools, an RA and tests for the System Health feature - Medium: Tools: crm_resource - Prevent use-of-NULL by requiring a resource name for the -A and -a options - Medium: PE: Prevent use-of-NULL in find_first_action() - High: PE: Only prevent migration if the clone dependancy is stopping/starting on the target node - Low: crmd: Include cib version of the updates which caused a transition abort - High: PE: Bug lf#2160 - Dont shuffle clones due to colocation - Low: PE: Fix find_first_action() when no node filter is specified - Low: PE: ptest - Suppress deleted ordering constraints from the .dot graph - High: PE: New implementation of the resource migration (not stop/start) logic - Medium: crmd: Note that dc-deadtime can be used to mask the brokeness of some switches - High: Tools: crm_resource - Advertise --move instead of --migrate - Medium: Extra: New node connectivity RA that uses system ping and attrd_updater - High: crmd: Bug bnc#527530 - Wait for the transition to complete before leaving S_TRANSITION_ENGINE - High: ais: Avoid excessive load by checking for dead children every 1s (instead of 100ms) - Medium: PE: Stop ptest from logging to syslog - Medium: PE: Correctly log resources that would like to start but can't - High: PE: Bug lf#2171 - Prevent groups from starting if they depend on a complex resource which can't - Low: PE: Add regression test for lf#2171 - High: PE: Bug lf#2170 - stop-all-resources option had no effect - High: PE: Disable resource management if stonith-enabled=true and no stonith resources are defined - High: Core: Bug lf#2169 - Allow dtd/schema validation to be disabled - Medium: PE: Bug lf#2180 - Include node information for all failed ops - Medium: PE: Bug lf#2178 - Indicate unmanaged clones - Medium: ais: Include version details in plugin name - High: PE: Bug lf#2106 - Not all anonymous clone children are restarted after configuration change - Medium: ais: Attempt to enable core file generation if it was disabled - Medium: crmd: Requery the resource metadata after every start operation - High: Tools: crm: add semantic checks depending on the meta-data from resource agents ------------------------------------------------------------------- Thu Aug 13 14:48:06 CEST 2009 - abeekhof@suse.de - Update source tarball to revision: b05d5a2b9c52 + High (bnc#507255): Tools: crm: implement date expressions + High: Build: Fix compilation when snmp and esmtp are not available + High: PE: Bug 2160 - Dont shuffle clones due to colocation + High: PE: Bug bnc#515172 - Correctly process location constraint rules which contain multiple expressions + High: PE: Bug bnc#515172 - Fix the boolean-op attribute of rules + High: PE: New implementation of the resource migration (not stop/start) logic + High: PE: Only prevent migration if the clone dependancy is stopping/starting on the target node + High: Tools: crm: new display type (uppercase keywords) + High: Tools: crm_resource - Advertise --move instead of --migrate + High: ais: Fix cluster connection when using corosync 1.0 + High: ais: Ubuntu needs a leading zero for directory modes + High: crmd: Terminate if we are ever evicted from the membership + High: crmd: Unset any existing DC value before querying for a new one + Medium: Extra: Add tools, an RA and tests for the System Health feature written by Mark Hamzy + Medium: Extra: New node connectivity RA that uses system ping and attrd_updater + Medium: PE: Prevent use-of-NULL in find_first_action() + Medium: Tools: crm_resource - Prevent use-of-NULL by requiring a resource name for the -A and -a options + Medium: crmd: Note that dc-deadtime can be used to mask the brokeness of some switches ------------------------------------------------------------------- Thu Aug 13 12:29:37 UTC 2009 - lmb@novell.com - Remove references to heartbeat packages. ------------------------------------------------------------------- Mon Aug 3 14:00:07 UTC 2009 - lmb@novell.com - Require python-curses for CLI color support. ------------------------------------------------------------------- Tue Jul 21 14:27:31 CEST 2009 - lmb@novell.com - Update to 0123af21a705. ------------------------------------------------------------------- Mon Jul 13 12:51:34 CEST 2009 - lmb@novell.com - Prevent pacemaker from being installed along with ocfs2-tools-o2cb to reduce likelihood of misconfiguration (bnc#520721). ------------------------------------------------------------------- Wed Jul 8 17:00:00 CEST 2009 - lmb@suse.de - Build: properly install crm_standby, crm_failcount, crm_master. ------------------------------------------------------------------- Wed Jul 8 10:45:00 CEST 2009 - lmb@suse.de - Update source tarball to revision: b05d5a2b9c52 + High: Core: Show help text and exit with rc 1 if option processing failed + High: PE: Fix reload for master/slave resources + High: Tools: Differentiate between --help and an unknown option + High: Tools: crm: support for color output + High: lrm: Look in the correct location for stonith agents + Low: Core: Reduce logging noise + Low: Tools: Ensure crm_resource -q prints a trailing newline + Low: Tools: ptest no longer supports --xml-pipe + Low: Tools: remove spurios newline output to stderr + Low: ais: Better handling of to_file: and to_syslog: options + Low: ais: set HA_LOGFACILITY too + Low: extra: Add a dummy parameter to 'Dummy' so we can test reload + Medium: Tools: crm: fix the verify exit code properly + Medium: cib: Supply an empty status section for replace operations + Dev: Tools: crm: recreate constraints on grouping resources ------------------------------------------------------------------- Wed Jun 10 16:05:25 CEST 2009 - abeekhof@suse.de - Update source tarball to revision: Pacemaker-1.0.4 + High: ais: bnc#488291 - don't rely on byte endianness on ptr cast + High: Tools: bnc#507255 - crm: import properly rsc/op_defaults + High: Tools: lf#2114 - crm: add support for operation instance attributes + High: ais: Bug lf#2126 - Messages replies cannot be routed to transient clients + High: attrd: Support the value++ and value+=... syntax required for failcounts + High: cib: Fix huge memory leak affecting heartbeat-based clusters + High: Core: Generate the help text directly from a tool options struct + High: crmd: Bug lf#2120 - All transient node attribute updates need to go via attrd + High: crmd: Fix another large memory leak affecting Heartbeat based clusters + High: PE: Bug bnc#495687 - Filesystem is not notified of successful STONITH under some conditions + High: PE: Make running a cluster with STONITH enabled but no STONITH resources an error and provide details on resolutions + High: PE: Prevent use-of-NULL when using resource ordering sets + High: Tools: attrd - Prevent race condition resulting in the cluster forgetting node's wish to shut down + High: Tools: crm_mon - Fix smtp notifications + High: Tools: crm_resource - Repair the ability to query meta attributes + Medium: Core: Include supported stacks in version information + Medium: Tools: Include current stack in crm_mon output + Medium: PE: Correctly log the actions for resources that are being recovered + Medium: PE: Correctly log the occurance of promotion events ------------------------------------------------------------------- Fri Apr 3 12:16:27 CEST 2009 - abeekhof@suse.de - Add a man page for crm_shadow - Update source tarball to revision: 9196be9830c2 + High: crmd: Bug bnc#480977 - Prevent extra, partial, shutdown when a node restarts too quickly (verified) + Medium: crmd: Bug bnc#490426 - Escalated shutdowns stall when there are pending resource operations + Medium: PE: Bug bnc#491441 - crm_mon does not display operations returning 'uninstalled' correctly ------------------------------------------------------------------- Tue Mar 31 14:53:08 CEST 2009 - abeekhof@suse.de - Update for SLE11 HAE RC9 - Update source tarball to revision: 61da7b5260df + High: PE: Bug bnc#490418 - Correctly determine node state for nodes wishing to be terminated + High: RA: Bug bnc#490641 - Shut down dlm_controld with -TERM instead of -KILL ------------------------------------------------------------------- Mon Mar 30 21:21:47 CEST 2009 - abeekhof@suse.de - Update source tarball to revision: f090699a99c5 + High: PE: Bug bnc#488721 - Fix id-ref expansion for clones, the doc-root for clone children is not the cib root + High: PE: Bug lf#2094 - Fix probe ordering for cloned groups + High: crmd: Bug bnc#489063 - Ensure the DC is always unset after we 'loose' an election ------------------------------------------------------------------- Fri Mar 20 21:16:26 CET 2009 - abeekhof@suse.de - Update for SLE11 HAE RC8 - Update to upstream version number - Update source tarball to revision: be123e8615a5 + High: PE: Bug lf#2089 - Meta attributes are not inherited by clone children + High: PE: Bug lf#2091 - Correctly restart modified resources that were found active by a probe + High: PE: Correctly handle terminate node attributes that are set to the output from time() + High: ais plugin: Fix quorum calculation (bnc#487003) + High: ais: Bug bnc#486858 - Fix update_member() to prevent spamming clients with membership events containing no changes + High: fencing: stonithd: is_heartbeat_cluster is a no-no if there is no heartbeat support ------------------------------------------------------------------- Mon Mar 16 16:15:00 CET 2009 - lmb@suse.de - Update for SLE11 HAE RC7 - Update source tarball to revision: 7e44933b0e46 - Tools: crm: print short help if there is no long help; replace "desc" with "help" - Tools: crm: issue warnings on all unrecognized elements of the cib configuration - Tools: crm: the new configure monitor command - Tools: crm: return exit code from configure commands - Tools: crm: check shadow name for sanity in the cib use command - Tools: crm: more shadow name sanity checks ------------------------------------------------------------------- Mon Mar 16 07:16:29 CET 2009 - abeekhof@suse.de - Update source tarball to revision: ec6b0bbee1f3 + High: PE: The ordered option for clones no longer causes extra start/stop operations + High: RA: pingd: Set default ping interval to 1 instead of 0 seconds + High: Tools: crm: the new configure node command + High: Tools: crm_mon - Prevent use-of-NULL when summarizing an orphan + High: ais: Bug bnc#482847, bnc#482905 - Force a clean exit of OpenAIS once Pacemaker has finished unloading ------------------------------------------------------------------- Tue Mar 10 13:54:38 CET 2009 - abeekhof@suse.de - Update for SLE11 HAE RC6.1 - Update source tarball to revision: 37e5ae18279e + High: Fix compilation on x64 - We dont need longclock_t intervals/precision for debug timers + High: cib: Bug bnc#482885 - Simplify CIB disk-writes to prevent data loss + High: Tools: The default re-ping interval for pingd should be 1s not 1ms ------------------------------------------------------------------- Mon Mar 9 15:19:28 CET 2009 - abeekhof@suse.de - Update for SLE11 HAE RC6 - Remove heartbeat-specific binary from the distribution - Update source tarball to revision: 3028846cc3f5 + High: PE: Bug bnc#478687 - Don't crash when clone-max is 0 + High: PE: Bug LF#2087 - Correctly parse the state of anonymous clones that have multiple instances on a given node + High: PE: Bug LF:2075 - Fix large pingd memory leaks + High: PE: Bug bnc#466788 - Make colocation based on node attributes work + High: PE: Correctly attach orphaned clone children to their parent + High: PE: Fix memory leak in LogActions + High: PE: Fix the determination of whether a group is active + High: PE: Look up the correct promotion preference for anonymous masters + High: PE: Simplify handling of start failures by changing the default migration-threshold to INFINITY + High: Tools: Bug bnc#483365 - Ensure the command from cluster_test includes a value for --log-facility + High: Tools: crm: add support for command aliases and some common commands (i.e. cd,exit) + High: Tools: crm: create top configuration nodes if they are missing + High: Tools: crm: fix parsing attributes for rules (broken by the previous changeset) + High: Tools: crm: new ra set of commands + High: Tools: crm: resource agents information management + High: Tools: crm: support for no value attribute in nvpairs + High: Tools: hb2openais: fix ocfs2 processing + High: Tools: pingd - prevent double free of getaddrinfo() output in error path + High: ais: Correctly handle a return value of zero from openais_dispatch_recv() + High: ais: Fix memory leak in error path + High: ais: IPC messages are only in scope until a response is sent + High: cib: Bug bnc#482885 - Simplify CIB disk-writes to prevent data loss. Required a change to the backup filename format + High: crm: Avoid infinite loop during crm configure edit (bnc#480327) + High: crmd: Avoid a race condition by waiting for the attrd update to trigger a transition automatically + High: crmd: Bug bnc#479543 - Correctly find the migration source for timed out migrate_from actions + High: crmd: Fix memory leaks found by valgrind ------------------------------------------------------------------- Thu Mar 5 16:42:23 CET 2009 - abeekhof@suse.de - heartbeat-common _must_ be a PreReq: instead of a Requires: so that the script which creates hacluster/haclient has completed before Pacemaker is installed. Otherwise the file permissions are all wrong. ------------------------------------------------------------------- Tue Mar 3 16:40:20 CET 2009 - abeekhof@suse.de - Add man page for the crm shell ------------------------------------------------------------------- Fri Feb 27 13:00:00 CET 2009 - lmb@suse.de - Update source tarball to revision: e0f361e9bd17 - High: Tools: cli: fix and improve delete command - High: Tools: crm: support rsc/op_defaults ------------------------------------------------------------------- Wed Feb 25 16:30:59 CET 2009 - abeekhof@suse.de - Add extra soft dependancies: graphviz, libdlm, heartbeat-resources ------------------------------------------------------------------- Mon Feb 23 14:37:32 CET 2009 - abeekhof@suse.de - Update for SLE11 HAE RC5 - Update source tarball to revision: 33e92390b2ed + High: Core: Prevent unterminated strings after decompression + High: PE: Clones now default to globally-unique=false + High: PE: Prevent use-of-NULL for failed migrate actions in process_rsc_state() + High: PE: Start the correct monitor operation for unmanaged masters + High: Tools: Cause the correct clone instance to be failed with crm_resource -F + High: ais: Centralize all quorum calculations in the ais plugin and allow expected votes to be configured int he cib + High: crmd: Ensure updates for lost migrate operations indicate which node it tried to migrating to + High: crmd: Fix memory leak in cib notication processing code + High: crmd: Fix memory leak in transition graph processing + Medium: Core: Bug BNC#474727 - Set a maximum time to wait for IPC messages + Medium: PE: Ensure clones always have a value for globally-unique + Medium: crmd: Add new expected-quorum-votes option + Medium: crmd: Automatically detect and use new values for crm_config options + Medium: crmd: Bug BNC#472473 - Don't wait excessive amounts of time for lost actions + Medium: crmd: Bug BNC#472473 - Reduce the transition timeout to action_timeout+network_delay ------------------------------------------------------------------- Fri Feb 6 18:12:28 CET 2009 - abeekhof@suse.de - Update source tarball to revision: 09be44da9b32 + High: PE: Remove an optimization for the terminate node attribute that caused the cluster to block indefinitly + High: Tools: Bug BNC#468066 - Don't use the result of uname() when its no longer in scope + High: Tools: Bug LF:2036 - crm_resource cannot set/get parameters for cloned resources + High: ais: Add optional support for the new AIS IPC layer + High: ais: Always set header.error to the correct default: SA_AIS_OK + High: ais: Ensure crm_peer_seq is updated after a membership update + High: ais: Make sure all IPC header fields are set to sane defaults + High: cib: Allow clients without mainloop to connect to the cib + High: cib: Fix four moderately sized memory leaks detected by Valgrind + High: Core: Fix more memory leaks foudn by valgrind + High: Tools: Bug BNC#473265 - crm_resource -L dumps core + High: Tools: Bug LF:2046 - Node attribute updates are lost because attrd can take too long to start + High: Tools: hb2openais: add constraints for clvmd to controld + High: Tools: hb2openais: improve media processing and set rrp_mode for multiple rings configuration + High: Tools: hb2openais: openais supports up to two rings ------------------------------------------------------------------- Mon Jan 26 11:33:42 CET 2009 - abeekhof@suse.de - Update source tarball to revision: 7eef42ed409d + High: Tools: Bug BNC#468066 - crm_standby - Don't use the result of uname() when its no longer in scope + High: PE: Bug BNC#465484 - Fix the no-quorum-policy=suicide option + High: PE: Only shoot online nodes with no-quorum-policy=suicide + High: ais: Bug BNC#456208 - Prevent deadlocks by not logging in the child process before exec() + High: ais: Detect and handle situations where ais and the crm disagree on the node name + High: crmd: Bug BNC#467995 - Delay marking STONITH operations complete until STONITH tells us so + High: crmd: Prevent node attribute loss by detecting attrd disconnections immediately - Enabled support for SNMP and SMTP notifications on resource state changes ------------------------------------------------------------------- Wed Jan 21 13:38:15 CET 2009 - abeekhof@suse.de - Update source tarball to revision: a2f46902d821 + High: Core: CID:10 - Prevent indexing into an array of schemas with a negative value + High: Core: CID:13 - Fix memory leak in log_data_element + High: Core: CID:15 - Fix memory leak in crm_get_peer + High: Core: CID:6 - Fix use-of-NULL in copy_ha_msg_input + High: Core: Fix crash in the membership code preventing node shutdown + High: PE: CID:19 - Fix use-of-NULL in determine_online_status + High: Tools: Bug 2003 - pingd doesn't correctly handle node-down events on multi-NIC systems + High: Tools: Bug 2021 - pingd does not detect sequence wrapping correctly, incorrectly reports nodes offline + High: Tools: attrd - Correctly determine when an attribute has stopped changing and should be committed to the CIB + High: Tools: crm_master shouldn't define a set id now that attrd is used. Defining one can break lookups + High: Tools: hb2openais: adjust path to the supporting script and set CONF variable + High: Tools: hb2openais: set no-quorum-policy to ignore for 2-node clusters + High: ais: Bug BNC#456243 - Ensure the membership cache always contains an entry for the local node + High: cib: CID:18 - Fix use-of-NULL in cib_perform_op + High: cib: CID:18 - Repair errors introduced in b5a18704477b - Fix use-of-NULL in cib_perform_op + High: crmd: Bug LF:2010 - Ensure fencing cib updates create the node_state entry if needed to preent re-fencing during cluster startup + High: crmd: Correctly handle reconnections to attrd + High: stonith: CID:3 - Fix another case of exceptionally poor error handling by the original stonith developers + High: stonith: CID:5 - Checking for NULL and then dereferencing it anyway is an interesting approach to error handling + High: stonithd: wrong checks for shmid (0 is a valid id) ------------------------------------------------------------------- Wed Dec 3 13:32:20 CET 2008 - abeekhof@suse.de - Reduce bloat - remove regession tests from source tarball - Update source tarball to revision: 6b8d46c7ab9c + High: crmd: Don't fast-track the processing of LRM refreshes when there are pending actions. + High: PE: Bug LF:2005 - Fix startup ordering of cloned stonith groups + High: PE: Bug LF:2006 - Correctly reprobe cloned groups + High: PE: Prevent on-fail settings being ignored after a resource is successfully stopped + High: Tools: Bug LF:2003 - pingd doesn't correctly detect failures when the interface is down + High: Tools: Bug LF:2001 - Transient node attributes should be set via attrd + High: Tools: crm cli: Bug bnc:450815 - don't generate id for the operations tag + High: Tools: crm cli: drop '-rsc' from attributes for order constraint + High: Tools: crm cli: fix wrong order of action and resource in order constraint + High: Tools: crm cli: return on syntax error when parsing score ------------------------------------------------------------------- Fri Nov 21 19:55:34 CET 2008 - abeekhof@suse.de - Update source tarball to revision: 27be064e0907 + High: ais: Repair and streamline service load now that OpenAIS startup functions correctly - Requires and is related to sr3936: Fix Bug bnc#444376 - openais segafults in lcr_ifact_release during shutdown ------------------------------------------------------------------- Fri Nov 21 14:00:11 CET 2008 - abeekhof@suse.de - Update source tarball to revision: 61d0af5c31c0 + High: cib: Ensure configuration updates contain the correct values of admin_epoch Otherwise the configuration can not be kept in sync across the cluster + High: crmd: Bug LF:1962 - Don't NACK peers because they're not (yet) in our membership. Just ignore them. + High: crmd: Prevent node re-probe loops by ensuring manditory actions always complete + High: PE: Bug LF:1996 - Correctly process disabled monitor operations + High: PE: Correctly calculate the number of available nodes for the clone to use + High: Tools: crm cli: catch system errors on file operations + High: Tools: crm cli: exit with an appropriate exit code + High: Tools: crm cli: new command: configure upgrade + High: Tools: crm cli: prevent key errors on missing cib attributes + High: Tools: crm cli: sometimes, on errors, level would change on single shot use + Medium: crmd: do_dc_join_filter_offer - Check the 'join' message is for the current instance before deciding to NACK peers + Medium: Tools: crm cli: replace inf with INFINITY + Medium: Tools: crm cli: supply provider only for ocf class primitives ------------------------------------------------------------------- Tue Nov 18 14:57:43 CET 2008 - abeekhof@suse.de - Add man pages for Pacemaker admin tools - Update source tarball to revision: eda05459ce58 + High: Core: Repair the ability to have comments in the configuration + High: Tools: hb2openais: update logging and set logfile appropriately + High: ais: Hook up the openais.conf config logging options + High: ais: Zero out the PID of disconnecting clients + High: cib: Ensure global updates cause a disk write when appropriate + High: crmd: Ensure PE processing does not occur for all error cases in do_pe_invoke_callback + High: crmd: Requests to the CIB should cause any prior PE calculations to be ignored + Medium: PE: Prevent lsb scripts from being used in globally unique clones + Medium: Tools: crm_mon - display the meaning of the operation's rc code instead of the status + Medium: ais: Hook up the options for specifying the expected number of nodes and total quorum votes + Medium: crmd: Fix the handling of unexpected PE exits to ensure the current CIB is stored ------------------------------------------------------------------- Sat Nov 8 10:24:36 CET 2008 - abeekhof@suse.de - Update source tarball to revision: e21021b77657 + High: crmd: Erase the correct node's status section when it returns ------------------------------------------------------------------- Fri Nov 7 14:09:15 CET 2008 - abeekhof@suse.de - Update source tarball to revision: 7508f211e536 + High: PE: Bug LF:1988 - Ensure recurring operations always have the correct target-rc set + High: PE: Fix optional colocation + High: PE: Improve up the detection of 'new' nodes joining the cluster + High: PE: Prevent assert failures in master_color() by ensuring unmanaged masters are always reallocated to their current location + High: Tools: crm_shadow - Show more line number information after validation failures + High: ais: Ensure HA_RSCTMP exists for use by resource agents + High: crmd: Bug:1975 - crmd should wait indefinitely for stonith operations to complete + Medium: PE: Bug N:437719 - Ensure scores from colocated resources count when allocating groups + Medium: Tools: Bug LF:1985 - crm_mon - Correctly process failed cib queries to allow reconnection after cluster restarts + Medium: Tools: Bug LF:1987 - crm_verify incorrectly warns of configuration upgrades for the most recent version + Medium: Tools: attrd - Automatically refresh all attributes after a CIB replace operation + Medium: Tools: crm (bnc#441028): check for key error in attributes management + Medium: crmd: Detect deleted resource operations automatically + Medium: crmd: Erase a node's resource operations and transient attributes after a successful STONITH + Medium: crmd: Fix the recording of pending operations in the CIB + Medium: xml: Relax the restriction on the contents of rsc_locaiton.node ------------------------------------------------------------------- Mon Oct 27 14:15:30 CET 2008 - abeekhof@suse.de - Update source tarball to revision: 9082ded546ec + High: Tools: LF:1969 - Attrd never sets any properties in the cib + Medium: Core: Provide better feedback if Pacemaker is started by a stack it doesn't support + Medium: Ensure meta attributes are used consistantly + Medium: PE: Allow group and clone level resource attributes + Medium: PE: Make a best-effort guess at a migration threshold for people with 0.6 configs + Medium: Tools: crm_mon - Fix the display of timing data + Medium: Tools: crm_verify - check that we're being asked to validate a complete config ------------------------------------------------------------------- Tue Oct 14 08:10:28 CEST 2008 - abeekhof@suse.de - Add heartbeat-common as a runtime dependancy This ensures the clustering user/group is installed and directories iare created with the right permissions ------------------------------------------------------------------- Mon Oct 13 11:58:59 CEST 2008 - abeekhof@suse.de - Add openais as a hard dependancy ------------------------------------------------------------------- Thu Oct 9 12:20:08 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 9a6c6d1dd871 (1.0.0) + High: Core: Allow the expected number of quorum votes to be configrable + High: Tools: Make pingd functional on Linux + Medium: crmd: Avoid calling GHashTable functions with NULL + Medium: tools: crm_resource - return with non-zero when a resource migration command is invalid + Medium: tools: crm_shadow - Allow the admin to start with an empty CIB (and no cluster connection) + Medium: PE: Use the correct attribute for action 'requires' (was prereq) ------------------------------------------------------------------- Sun Sep 28 18:57:45 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 8415675fe428 + High: Core: Only call xmlCleanupParser() if we parsed anything. Doing so unconditionally seems to cause a segfault + High: hb2openais.sh: improve pingd handling; several bugs fixed + High: hb2openais: fix clone creation; replace EVMS strings + High: stonithd: fix handling of timeouts + High: stonithd: fix logic for stonith resource priorities + High: stonithd: initialize value before reading fence-timeout + Medium: PE: Remove new options that are better set in rsc_defaults + Medium: PE: Detect clients that disconnect before receiving their reply + Medium: PE: ptest - check the input is valid before proceeding + Medium: crmd: Hook up the stonith-timeout option to stonithd + Medium: crmd: Prevent potential use-of-NULL in global_timer_callback + Medium: crmd: Rationalize the logging of graph aborts + Medium: xml: pacemaker-0.7 should be an alias for the 1.0 schema ------------------------------------------------------------------- Mon Sep 15 11:08:19 CEST 2008 - abeekhof@suse.de - Create a libpacemaker subpackage ------------------------------------------------------------------- Thu Sep 11 09:33:12 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 2087f5b1a5a0 + High: PE: Bug LF:1950 - Ensure the current values for all notification variables are always set (even if empty) + High: PE: Bug LF:1955 - Ensure unmanaged masters are unconditionally repromoted to ensure they are monitored correctly. + High: PE: Bug LF:1955 - Fix another case of filtering causing unmanaged master failures + High: PE: Bug LF:1955 - Umanaged mode prevents master resources from being allocated correctly + High: PE: Bug N:420538 - Anit-colocation caused a positive node preference + High: PE: Correctly handle unmanaged resources to prevent them from being started elsewhere + High: PE: crm_resource - Fix the --migrate command + High: ais: Move the service id definition to a common location and make sure it is always used + High: stonithd (LF 1951): fix remote stonith operations + Medium: Build: Make sure crm_config.h is only ever included once + Medium: PE: Bug LF:1950 - Notifications do not contain all documented resource state fields + Medium: PE: Bug N:417585 - Don't restart group children who's individual score drops below zero + Medium: PE: Implement a true maintenance mode + Medium: ais: Use the agreed service number + Medium: Tools: Implement a unified CLI shell ------------------------------------------------------------------- Mon Sep 1 10:25:58 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 1df197982079 - Try again to fix compilation in STABLE (this time also tested with mbuild) - STONITH is now enabled by default and the cluster will warn if no STONITH resource is found ------------------------------------------------------------------- Sat Aug 30 12:38:56 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 902885c90d96 - Fixes build ordering ssue that prevented compilation of lib/common ------------------------------------------------------------------- Wed Aug 20 17:05:00 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: f567855328b7 (0.7.1) + Added features needed to support libdlm and ocfs2-tools + High: admin: Fix use-after-free in crm_mon + High: cib: Use single quotes so that the xpath queries for nvpairs will succeed + High: crmd: Always connect to stonithd when the TE starts and ensure we notice if it dies + High: crmd: Make sure async-failures cause the failcount to be incrimented + High: PE: Bug LF:1941 - Handle failed clone instance probes when clone-max < #nodes + High: PE: Prevent use-of-NULL - order->rsc_rh will not always be non-NULL + High: PE: Unpack colocation sets correctly + High: Tools: crm_mon - Prevent use-of-NULL for orphaned resources + Medium: Core: Get more details regarding parser errors + Medium: Core: Switch to the libxml2 parser - its significantly faster + Medium: PE: Allow groups to be involved in optional ordering constraints + Medium: PE: Allow sets of operations to be reused by multiple resources + Medium: PE: Bug LF:1941 - Mark extra clone instances as orphans and don't show inactive ones + Medium: PE: Determin the correct migration-threshold during resource expansion + Medium: PE: Implement no-quorum-policy=suicide (FATE #303619) + Lots of bug fixes ------------------------------------------------------------------- Mon Jul 21 16:26:00 CEST 2008 - abeekhof@suse.de - Update source tarball to revision: 7a31d0c33e22 (0.7.0) + High: Performance enhancements + High: Fix memory leaks and use-after-free + High: Admin: All tools now use --node (-N) for specifying node unames + High: Admin: All tools now use --xml-file (-x) and --xml-text (-X) for specifying where to find XML blobs + High: cib: Implement CIB_shadow - a facility for making and testing changes before uploading them to the cluster + High: cib: Make registering per-op callbacks an API call and renamed (for clarity) the API call for requesting notifications + High: Core: Add a facility for automatically upgrading old configurations + High: Core: Adopt libxml2 as the XML processing library - all external clients need to be recompiled + High: PE: Bug LF:1459 - Allow failures to expire + High: PE: Replace failure-stickiness with a simple 'migration-threshold' + High: PE: Bug LF:1941 - Handle failed clone instance probes when clone-max < #nodes + High: PE: Make OCF_ERR_ARGS a node-level error rather than a cluster-level one + High: PE: Properly handle clones that are not installed on all nodes + Medium: Admin: Bug LF:1438 - Allow the list of all/active resource operations to be queried by crm_resource + Medium: Admin: Bug LF:1708 - crm_resource should print a warning if an attribute is already set as a meta attribute + Medium: Admin: Bug LF:1883 - crm_mon should display fail-count and operation history + Medium: Admin: Bug N:371785 - crm_resource -C doesn't also clean up fail-count attributes + Medium: Admin: crm_mon - include timing data for failed actions + Medium: cib: Bug LF:1348 - Allow rules and attribute sets to be referenced for use in other objects + Medium: cib: BUG LF:1918 - By default, all cib calls now timeout after 30s + Medium: cib: Remove some useless parts of the API + Medium: cib: Implement the ability for clients to connect to remote servers + Medium: Core: Add support for validating xml against RelaxNG schemas + Medium: PE: Bug LF:1328 - Don't fencing nodes in clusters without managed resources + Medium: PE: Bug LF:1461 - Give transient node attributes (in <status/>) preference over persistent ones (in <nodes/>) + Medium: PE: Bug LF:1884, Bug LF:1885 - Implement N:M ordering and colocation constraints + Medium: PE: Bug LF:1886 - Create a resource and operation 'defaults' config section + Medium: PE: Bug LF:1892 - Allow recurring actions to be triggered at known times + Medium: PE: Bug LF:1926 - Probes should complete before stop actions are invoked + Medium: PE: Prefer the current location if it is one of a group of nodes with the same (highest) score ------------------------------------------------------------------- Sat May 3 07:35:35 CEST 2008 - abeekhof@suse.de - Use a libexecdir consistent with openais ------------------------------------------------------------------- Tue Apr 29 08:01:10 CEST 2008 - abeekhof@suse.de - Updated to upstream version 0.6.3-1 - Update source tarball to revision: fd8904c9bc67 - SNMP has been moved to the management/pygui project - Many bugfixes, primarily related to master/slave resources and failcount handling - Fix libtool version check and supply a better default for the LCRSO directory ------------------------------------------------------------------- Sat Apr 12 19:14:17 CEST 2008 - abeekhof@suse.de - version 0.6.2-1 - Initial package check-in
Locations
Projects
Search
Status Monitor
Help
OpenBuildService.org
Documentation
API Documentation
Code of Conduct
Contact
Support
@OBShq
Terms
openSUSE Build Service is sponsored by
The Open Build Service is an
openSUSE project
.
Sign Up
Log In
Places
Places
All Projects
Status Monitor
