[03:24:09] <diskdance[m]>	 [[meta:Community_Wishlist_Survey_2023/Larger_suggestions/Explore_evasion_methods_of_state-level_censorship_across_Wikimedia_movement]]
[03:24:09] <diskdance[m]>	 Does this have something to do with the SRE team?
[03:26:43] <diskdance[m]>	 https://meta.wikimedia.org/wiki/Community_Wishlist_Survey_2023/Larger_suggestions/Explore_evasion_methods_of_state-level_censorship_across_Wikimedia_movement
[10:45:04] <claime>	 godog: I presume https://phabricator.wikimedia.org/T287539#7339799 is no longer relevant and swift/swift-ro can now be treated as any other service for the switchover?
[10:47:01] <godog>	 claime: yes AFAIK you are correct, cc Emperor though
[10:48:20] <claime>	 There's also this in switchd.services : 'docker-registry',  # swift replica goes codfw => eqiad and needs manual switching (and blubberoid has to switch with docker-registry too)
[10:48:51] <claime>	 So if we move swift, then blubberoid and docker-registry need to move too, if i understand correctly
[10:51:00] <godog>	 mmhh that I'm less sure about tbh
[10:53:34] <_joe_>	 claime: hold your horses if you're looking at the cookbook for switching services
[10:54:11] <claime>	 _joe_: I'm not sure I have enough wifi outside to both hold my horses and work at the same time
[10:54:14] <claime>	 But yes, holding
[10:54:26] <claime>	 :p
[10:54:41] <_joe_>	 context: the cookbook I'm writing for depooling/pooling all services in a DC should cover most bases and I'll re-write the switch cookbook to use it
[10:55:02] <_joe_>	 that's what I get for saying "hold your horses" to someone who actually has horses :P
[10:57:10] <godog>	 lol
[11:05:05] <vgutierrez>	 hmm puppet-merge is failing in puppetmaster1001
[11:05:16] <vgutierrez>	 https://www.irccloud.com/pastebin/kYmipFzQ/
[11:05:37] <vgutierrez>	 probably related to trying to decode as UTF-8 the output of openssl rand 64
[11:06:58] <vgutierrez>	 trying to bypass the issue by using -q / --quiet
[11:07:04] <vgutierrez>	 I get another error
[11:07:09] <vgutierrez>	 $ sudo -i puppet-merge -q
[11:07:09] <vgutierrez>	 Internal error!
[11:07:20] <vgutierrez>	 any ideas on how to proceed? :)
[11:07:50] * jbond looking
[11:08:30] <vgutierrez>	 jbond: offending change is https://gerrit.wikimedia.org/r/c/labs/private/+/886000
[11:11:02] <Emperor>	 claime: Swift replication (swiftrepl as-was, now rclone) runs weekly (when I'm not trying to debug it) on Monday UTC-morning; the script that's run compares confctl --object-type mwconfig select 'name=WMFMasterDatacenter' get to /etc/wikimedia-cluster when deciding to proceed or not.
[11:12:18] <Emperor>	 NB also that MW always tries to write to both swift ms clusters
[11:16:03] <jbond>	 vgutierrez: i have manualy pushed it through for now, ill send a permenent fix shortly
[11:16:16] <vgutierrez>	 jbond: cheers and sorry /o\
[11:16:45] <vgutierrez>	 I was expecting to break puppet-merge with an innocent change in labs/private
[11:16:47] <vgutierrez>	 *wasn't
[11:17:06] <jbond>	 :D no problem and no it shuldn't have
[11:50:16] <claime>	 still about the switchover, this will be the first switchover with multi-DC right? (CommRel asking if anything major changed since the last time)
[11:53:00] <marostegui>	 claime: yep
[11:58:30] <claime>	 Do y'all see anything else major that changed?
[13:55:28] <Emperor>	 https://www.softwareheritage.org/2023/02/02/software-heritage-is-now-using-gitlab/ <-- Software Heritage on their Phab -> gitlab migration
[15:02:36] <diskdance[m]>	 Will we finally move away from Phab?
[16:06:48] <dancy>	 There's no plan to move away from Phab.
[16:13:10] <zabe>	 I like phab
[16:42:33] <inflatador>	 It's no longer being updated
[16:43:21] <inflatador>	 https://github.com/phacility/phabricator Not sure if that's a huge deal but just FYI
[16:47:17] <mutante>	 https://phabricator.wikimedia.org/T328595
[16:47:28] <mutante>	 https://phabricator.wikimedia.org/T302528
[16:47:54] <inflatador>	 oh cool, had not heard about Phorge
[16:48:30] <mutante>	 https://we.phorge.it/ :p
[16:49:09] <mutante>	 the idea would be to replace Maniphest, the task part of Phab
[16:49:21] <mutante>	 but not necessarily the other apps and for sure not Diffusion (repos)
[16:49:30] <mutante>	 because we already have gitlab and gerrit
[16:49:40] <mutante>	 and github..
[16:49:48] <mutante>	 and 4 different places for repos is just a bit much 
[16:50:13] <mutante>	 so Diffusion is out.. but Maniphest is not.. and I also want to keep the pastebin
[16:50:31] <cdanis>	 we use the pastebin in production heh :)
[16:50:35] <cdanis>	 dbctl diffs, for once
[16:52:30] <mutante>	 yea, pastebin should stay and we should definitely not lose the existing pastebins. that would be very bad
[16:54:08] <RhinosF1>	 mutante: upgrading phab to phorge is fairly easy
[16:54:17] <claime>	 16:50:31           cdanis │ we use the pastebin in production <  I don't know how to feel about this
[16:54:19] <RhinosF1>	 It should be as simple as a phab upgrade
[16:55:58] <cdanis>	 claime: https://gerrit.wikimedia.org/g/operations/puppet/+/production/modules/base/files/phaste.py installed on every machine as `phaste` :D
[16:56:08] <inflatador>	 That reminds me: Italians (or anyone else with a .it domain), what is a good registrar for .it?
[17:35:38] <jynus>	 does someone have handy a practical example to add coverage to CI for a python project?
[18:09:06] <cdanis>	 jynus: https://gerrit.wikimedia.org/r/c/integration/config/+/701597
[18:09:48] <jynus>	 that part I have it, it is the tox and coverage run that I am getting a bit lost
[18:10:05] <jynus>	 let me check that source code
[18:15:09] <jynus>	 I think I make it work, but I may have something weird in my code because some test is identified as code and not as test
[18:52:53] <jynus>	 I found the issue- on the newer nose version, a parameter has changed, but it behaves slightly differently
[19:35:17] <jynus>	 I am at 48% coverage- more than mediawiki, mayhaps? XD https://integration.wikimedia.org/ci/job/tox-docker/30158/console