Back to bug 2010084

Who When What Removed Added
Red Hat Bugzilla 2021-10-03 15:28:52 UTC Pool ID sst_high_availability_rhel_8
Red Hat One Jira (issues.redhat.com) 2021-10-03 15:29:10 UTC Link ID Red Hat Issue Tracker RHELPLAN-98713
tbsky 2021-10-03 15:35:17 UTC Comment 0 updated
tbsky 2021-10-03 15:35:30 UTC Comment 0 updated
tbsky 2021-10-03 15:36:30 UTC Comment 0 updated
Ken Gaillot 2021-10-04 15:32:56 UTC Doc Type --- If docs needed, set a value
Keywords Triaged
Reid Wahl 2023-03-16 04:27:50 UTC Assignee kgaillot nwahl
CC nwahl
Reid Wahl 2023-03-23 23:18:44 UTC Status NEW POST
RHEL Program Management 2023-04-03 07:28:07 UTC Resolution --- WONTFIX
Status POST CLOSED
Last Closed 2023-04-03 07:28:07 UTC
Reid Wahl 2023-04-03 08:08:04 UTC Status CLOSED POST
Resolution WONTFIX ---
Keywords Reopened
Ken Gaillot 2023-04-24 14:56:39 UTC CC kgaillot
Doc Text Cause: When a cluster elects a new designated controller, all nodes send their current history to the new DC, which saves it to the CIB.

Consequence: If actions were already in progress when a new DC is elected, and finish after the nodes send their current history to the new DC, the actions' results could be lost.

Fix: DC elections are not finalized until all pending actions are complete.

Result: No action results are lost.
Target Release --- 8.9
Doc Type If docs needed, set a value Bug Fix
Markéta Smazová 2023-05-16 13:04:30 UTC CC msmazova
Dean Jansa 2023-05-16 13:11:02 UTC Link ID Red Hat Issue Tracker CLUSTERQE-6692
Chris Lumens 2023-05-26 13:32:29 UTC Fixed In Version pacemaker-2.1.6-1.el8
Status POST MODIFIED
errata-xmlrpc 2023-05-26 13:39:58 UTC Status MODIFIED ON_QA
Steven J. Levine 2023-06-30 14:46:03 UTC CC slevine
Flags needinfo?(kgaillot)
Ken Gaillot 2023-07-03 15:17:42 UTC Flags needinfo?(kgaillot)
Steven J. Levine 2023-07-07 19:19:04 UTC Doc Text Cause: When a cluster elects a new designated controller, all nodes send their current history to the new DC, which saves it to the CIB.

Consequence: If actions were already in progress when a new DC is elected, and finish after the nodes send their current history to the new DC, the actions' results could be lost.

Fix: DC elections are not finalized until all pending actions are complete.

Result: No action results are lost.
.Pacemaker Designated Controller elections no longer finalized until all pending actions are complete

When a cluster elects a new Designated Controller (DC), all nodes send their current history to the new DC, which saves it to the CIB. As a consequence, if actions were already in progress when a new DC is elected, and the actions finish after the nodes send their current history to the new DC, the actions' results could be lost. With this fix, DC elections are not finalized until all pending actions are complete and no action results are lost.
Markéta Smazová 2023-07-19 10:24:00 UTC Status ON_QA VERIFIED
Steven J. Levine 2023-07-27 14:47:22 UTC Docs Contact slevine
Red Hat Bugzilla 2023-08-10 15:41:28 UTC QA Contact cluster-qe cluster-qe

Back to bug 2010084