TLDR: Today's minor update workflow (overcloud stack update -i) is used to update packages and apply puppet in order to bring a node uptodate
The way it works today is basically that yum_update.sh is called on each node and once all nodes are through the update we will reapply puppet (amongst other things). This works well but is a bit limited in certain scenarios. Imagine the following need:
- We need to ship a fixed resource-agents package and run certain pcs commands to amend some pacemaker configuration
- The problem is that we cannot do this inside yum_update.sh because there is no guarantee that we have the newer resource-agents package on all nodes (we can only guarantee it is there on a single node)
Probably the only way to fix such a need/scenario is to make sure the updated resource-agent is pulled in via yum_update.sh and then abuse pacemaker_resource_restart.sh to do any changes that are necessary.
One thing we could probably do is to document exactly how/when to ship fixes in the minor/update workflow and how/when to ship them via an overcloud deploy running against an existing cloud.
This is an automated cleanup. This bug report has been closed because it
is older than 18 months and there is no open code change to fix this.
After this time it is unlikely that the circumstances which lead to
the observed issue can be reproduced.
If you can reproduce the bug, please:
* reopen the bug report (set to status "New")
* AND add the detailed steps to reproduce the issue (if applicable)
* AND leave a comment "CONFIRMED FOR: <RELEASE_NAME>"
Only still supported release names are valid (FUTURE, PIKE, QUEENS, ROCKY, STEIN).
Valid example: CONFIRMED FOR: FUTURE