[14:40:33] <jelto>	 dzahn, arnoldokoth: Puppet fails on all gitlab-runner (eqiad, codfw and wmcs) due to some error in the docker class. Most probably I wont habe time to fix this today. Can you take a look? Otherwise I'll fix this tomorrow
[14:40:47] <jelto>	 mutante ^ wrong ping ;)
[14:44:03] <jelto>	 most probably because of the docker version refactoring from cgoubert
[15:18:06] <_joe_>	 jelto: claime is not here, maybe post in #serviceops
[15:51:48] <hashar>	 jelto: for splitting the gitlab backups, I had the inspiration from https://docs.gitlab.com/ee/raketasks/backup_gitlab.html#excluding-specific-directories-from-the-backup  which list the few buckets of data being backed up (db, uploads, builds, artifacts, repositories).
[15:52:32] <hashar>	 so db and repositories probably need to be backed up often  while we can most probably afford to loose the CI job output logs (and maybe artifacts as well)
[15:52:47] <hashar>	 if the backup is smaller, it is probably easier to manage
[15:53:29] <hashar>	 there is also a `GZIP_RSYNCABLE=yes` which we apparently already use \o/
[15:53:44] <hashar>	 anyway, that was a great presentation
[15:53:46] <jelto>	 hashar: thanks for the input, I added this option to my notes and will most probably create a task for that
[15:53:48] <jelto>	 thanks!
[15:54:28] <jelto>	 mutante, arnoldokoth: runner are fixed, see also -serviceops :)
[18:15:35] <wikibugs>	 10GitLab (Project Migration), 10Fundraising Tech - Chaos Crew, 10Release-Engineering-Team: Create new GitLab project group: Fundraising-Tech - https://phabricator.wikimedia.org/T316695 (10jgleeson)