-
Bug
-
Resolution: Done-Errata
-
Major
-
rhel-8.10
-
pacemaker-2.1.7-5.2.el8_10
-
Yes
-
None
-
ZStream, Regression
-
rhel-sst-high-availability
-
ssg_filesystems_storage_and_HA
-
21
-
26
-
2
-
QE ack, Dev ack
-
False
-
-
Yes
-
Red Hat Enterprise Linux
-
None
-
Approved Blocker
-
Pass
-
None
-
Bug Fix
-
-
Proposed
-
-
All
-
None
What were you trying to do that didn't work?
Since upstream version 2.1.7 (RHEL 8.10 / 9.4), crm_node dash i / dashdash cluster-id, which is supposed to show the local node's cluster ID, can show "Node is not known to cluster" instead.
Steps to reproduce
- Run crm_node -i repeatedly
Expected results
crm_node -i always gives the local node ID
Actual results
crm_node -i sometimes gives an error
Other notes
I (Reid) could reproduce this on Fedora 39 only after installing the pacemaker-2.1.7-5.fc39.x86_64 package from the fedora repo via dnf. When I built and installed from source, the uninitialized local variable that causes the problem was always set to 0 by gcc.
- clones
-
RHEL-47249 crm_node mishandles node IDs
- Closed
- links to
-
RHBA-2024:137362 pacemaker update