> I see. Basically this is what `PinningIsBlockingBundledPluginMonitor`
> was designed to help with, though it sounds like you are getting some
> error earlier and more intrusively than that was designed to handle.
> Can we treat this particular error as a robustness bug?

It's not a notable problem with Monitor External Job because the bundled 
version was increased in real Jenkins. In no-bundling Jenkins OTOH, the old 
version would stick around and exhibit this issue.

> Alternately, be more aggressive and force updates of “split-bundled”
> plugins newer than an installed version, disregarding historical
> pinned status (and deleting the existing monitor). I think that is a
> much more reasonable proposal when we are no longer bundling plugins
> just for fun, and we are only talking about updating to a version very
> soon after the split—from 1.0 to 1.0.1, say.

We also discussed this as "We're not doing compat-bundling upgrades just for 
fun" (so no version increases just to get some cool functionality in there) and 
it seems like a viable approach. Worst case is we'll need to backport the 
critical fix to an older x.y.1 so we don't include a bunch of unrelated stuff.

>> What are we going to do when we notice there's something broken with the 
>> first release of a detached (and therefore compat-bundled) plugin? Fix and 
>> release the plugin, and increase the compat-bundled version?
> 
> That is what we have done in the past.

Right, the question starts in the sentence after that. Bundling and pinning 
allowed for upgrading of these (or at least showing a stern warning), while new 
compat-bundling would not, at least not as currently proposed.

-- 
You received this message because you are subscribed to the Google Groups 
"Jenkins Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jenkinsci-dev/9FAD9A40-4E03-4663-8490-CB459959E09C%40beckweb.net.
For more options, visit https://groups.google.com/d/optout.

Reply via email to