22:10:08 <paladox> why is it so slow, what is it doing seeing as it's not copying the files
22:10:54 <RhinosF1> paladox: the prep steps to make sure everything is up to date that should be
22:13:29 <RhinosF1> It's because you always currently have to deploy to the deployment server first
This should be fixed in my refactor of the code. Filing a task for tracking.
| Status | Assigned | Task | ||
|---|---|---|---|---|
| Resolved | Unknown Object (User) | T9446 Upgrade to MediaWiki 1.39.1 | ||
| Resolved | Unknown Object (User) | T8982 Migrate revision_actor_temp usage back to the revision table | ||
| Resolved | Unknown Object (User) | T8743 Upgrade to MediaWiki 1.38.1 | ||
| Resolved | RhinosF1 | T8369 explicitly selected deploys should not imply canary |
Based on https://github.com/RhinosF1/puppet/blob/patch-111/modules/mediawiki/files/bin/deploy-mediawiki.py,
Move https://github.com/RhinosF1/puppet/blob/patch-111/modules/mediawiki/files/bin/deploy-mediawiki.py#L188 to before the prep steps,
Then only run any pre flight stage + canary deploy if canary is in the serverlist (add it to all).
We can simply skip canary when it comes up in https://github.com/RhinosF1/puppet/blob/patch-111/modules/mediawiki/files/bin/deploy-mediawiki.py#L57 or pop it out.
Moving to low priority as 1) this isn't really causing a significant problem 2) it doesn't seem that this will be fixed very soon since it will only come with 'version 2'.