2024-07-23T07:45:52 acidsys: let me know when you allow odin to send mails, I will do a dry run 2024-07-23T08:19:22 *** teepee_ is now known as teepee 2024-07-23T10:02:33 Karthik: feel free to add a commit for it 2024-07-23T10:09:55 acidsys: i will try. should the .py extension -> to jinja ? 2024-07-23T10:11:29 for your script, it would be nice if you make it .py.jinja and add the {{ pillar['managed_by_salt'] }} on the top after the shebang (to have it render our banner) 2024-07-23T10:11:54 (not technically mandatory, just our convention) 2024-07-23T10:24:46 acidsys: done. 2024-07-23T10:24:51 acidsys: sure 2024-07-23T10:40:24 https://matrix.opensuse.org/_matrix/media/v3/download/poddery.com/GiEbEhbDWHykTklehdmneucT/image.png 2024-07-23T10:40:32 like this ^ 2024-07-23T10:40:34 * ^? 2024-07-23T10:42:50 yep or just add it to your description comment block if you don't want it to get interpreted as python in your IDE 2024-07-23T10:54:03 acidsys: done. 2024-07-23T10:55:25 and pushed 2024-07-23T10:57:52 looks good 2024-07-23T18:49:57 acidsys: how long does the syncing usually take? 2024-07-23T18:50:10 feels like a bumpy day today 2024-07-23T18:50:19 Karthik (కార్తీక్): as long as build.o-o stays up I'm happy enough, even if everything else is rocky. but yes 2024-07-23T18:52:41 I'm afraid you can't be too happy then - the OBS main page works, but looking at a repo ends up with a timeout 2024-07-23T18:53:39 Is there a new big outage? 2024-07-23T18:53:48 we're almost back 2024-07-23T18:56:02 bmwiedemann: you could have the honors to check download.o.o please? :) 2024-07-23T19:01:01 heroes infra is fully back now 2024-07-23T19:01:17 bmwiedemann: nevermind, they are still working on cirrus 2024-07-23T19:04:32 team is still working on the OBS side of things 2024-07-23T19:04:51 keeping status.o.o incident open til then 2024-07-23T19:04:52 Greetings. I was just about to ask what's the status with the outage, but channel logs, nice! So, godspeed then, hoping for a easy recovery! :-) 2024-07-23T19:05:21 hi chris_goe, we're mostly back besides download/OBS. thanks :) 2024-07-23T19:05:31 \o/ 2024-07-23T19:06:02 mailing lists are still listed as down, is this intentional? (lists.o.o works, but I guess we have a nice _external_ mail queue) 2024-07-23T19:06:56 I'm surprised this hasn't hit the forums yet, that's where I looked first. 2024-07-23T19:09:26 chris_goe: the forums were gone too :D 2024-07-23T19:10:08 cboltz: thanks, overlooked. yes, mail queue will take a bit to recover 2024-07-23T19:17:53 hah, OK. Sadly, IRC isn't linked on www.o.o, but it's good it was mentioned on status.o.o! 2024-07-23T19:18:32 starting to look good! GJ to everyone involved 2024-07-23T19:19:39 :) 2024-07-23T19:21:12 kinda curious about the original culprit... "storage backend software downgrade", is it something akin to HP Nimble arrays that is used? I would have guessed you'd run e.g. ZFS on white box hardware 2024-07-23T19:21:34 maybe a Ceph cluster 2024-07-23T19:26:32 is the storage backend imposed by the Prague DC (as in you rent capacity)? 2024-07-23T19:26:37 it's two NetApp storage clusters, and apparently a bug in their new software version causing it to crash 2024-07-23T19:26:44 nvm nerding out a bit, none of my business :D 2024-07-23T19:27:04 ah 2024-07-23T19:27:10 it's operated by SUSE. we as openSUSE can use it for our needs too 2024-07-23T19:27:43 oh boy... so even the corpo side got hit? 2024-07-23T19:30:33 let's say, it was an interesting day for everyone 2024-07-23T19:32:53 let's all sit back, grab some popcorn and watch the Finger Pointing Olympics? :D 2024-07-23T19:34:03 "Why not during the Crowdstrike mess?!?! That could have flown under the radar!" 2024-07-23T19:34:25 luckily we don't have a finger pointing culture 2024-07-23T19:34:37 but yes people said they want to fix the netapp thing to finally rest in peace and just blame crowdstrike :P 2024-07-23T19:36:23 no finger pointing, wow that's rare, cool! so you get some resources from SUSE's pool rather than have a dedicated testbed? 2024-07-23T19:37:06 ffs here I go again, don't mind me 2024-07-23T19:37:15 human or technical resources? 2024-07-23T19:37:27 technical 2024-07-23T19:38:45 from the technical side, SUSE provides fancy colocation including hardware for openSUSE 2024-07-23T19:39:10 at least that is how it's in Prague now. in Provo, we still have a legacy setup where SUSE operates the hardware, this will change after migration to a new location as well 2024-07-23T19:39:49 ok 2024-07-23T19:42:25 download.o.o/OBS are now back as well, but the team is still doing some post-work 2024-07-23T19:47:13 well, thanks all for your work and patience! cya around :) 2024-07-23T19:47:34 thanks for stopping by :) 2024-07-23T19:47:44 np! 2024-07-23T21:25:09 FYI: I just created a basic AppArmor profile for apache on status2 2024-07-23T21:25:33 thank you! 2024-07-23T21:25:44 it covers the obvious things (restarting apache, people visiting status.o.o), but probably still needs some additions 2024-07-23T21:25:52 which also means it's in complain mode for now 2024-07-23T21:27:15 I also noticed there's a profile for /usr/sbin/mariadbd which couldn't be loaded because it included two non-existing abstractions (rpm, zypp) 2024-07-23T21:27:37 I commented out these abstractions, and switched the profile to complain mode to avoid that it breaks mariadb 2024-07-23T21:28:09 interesting why mariadbd would need rpm or zypp abstractions 2024-07-23T21:28:54 maybe one of them (while they existed) had something that matched a log event, and whoever created the profile mis-choose the abstraction 2024-07-23T21:30:14 cool beans, thanks to everyone involved getting this all back online! 2024-07-23T21:30:30 I had to add a few rules to the mariadb profile after removing these abstractions, but nothing too bad 2024-07-23T21:32:11 for now, audit.log stays silent - but we should keep an eye on it in the next days 2024-07-23T22:00:08 *** teepee_ is now known as teepee