Closed kavishgr closed 1 year ago
Yes, this is extremely confusing. I generally agree with you, however there's a high degree of subtlety and complexity in the interaction between the two components today, and "user control" becomes a major third one.
This also relates to https://github.com/coreos/zincati/issues/928 I think possibly.
And a definite factor that's often bit me in the past is how zincati leaves around the deployment finalization lock even if it's explicitly stopped.
I think as a bottom line today though, if you want to take the wheel, you need to at least temporarily stop zincati...sorry.
I think this is basically https://github.com/coreos/zincati/issues/498
Yeah looks like it, closing as duplicate
I am experiencing an issue with rpm-ostree while trying to manually update the system. Currently, I have configured my zincati to perform updates automatically on weekends. However, when I check for updates with rpm-ostree, it notifies me that there is an update pending:
but attempting to upgrade results in a message stating that no upgrade is available:
This is my update strategy:
Ideally, I should be able to update the system manually regardless of the zincati update strategy. However, this is not happening. The only way I can trigger a manual update is by disabling the zincati configuration by renaming it:
and restarting the zincati service:
When I check with the
rpm-ostree status -v
command, the state shows as busy, indicating that it is about to update to a new deployment:Following this, I receive a broadcast message notifying me that an update is about to take place:
This behavior is unexpected. I guess I should be able to update the system at any time without having to alter the zincati configuration.