<@phsmoura:fedora.im>
20:03:02
!startmeeting Fedora Infrastructure Ops Daily Standup Meeting
<@meetbot:fedora.im>
20:03:03
Meeting started at 2024-11-13 20:03:02 UTC
<@meetbot:fedora.im>
20:03:03
The Meeting name is 'Fedora Infrastructure Ops Daily Standup Meeting'
<@phsmoura:fedora.im>
20:03:06
!info agenda is at https://board.net/p/fedora-infra-daily
<@phsmoura:fedora.im>
20:03:06
<@phsmoura:fedora.im>
20:03:06
!chair nirik smooge phsmoura
<@phsmoura:fedora.im>
20:03:06
!meetingname fedora_infrastructure_ops_daily_standup_meeting
<@phsmoura:fedora.im>
20:03:06
!info meeting is 30 minutes MAX. At the end of 30, its stops
<@phsmoura:fedora.im>
20:03:06
!info reminder: speak up if you want to work on a ticket!
<@meetbot:fedora.im>
20:03:07
The Meeting Name is now fedora_infrastructure_ops_daily_standup_meeting
<@phsmoura:fedora.im>
20:03:10
hello
<@james:fedora.im>
20:03:16
!hi
<@zodbot:fedora.im>
20:03:16
James Antill (james)
<@phsmoura:fedora.im>
20:04:39
!info https://pagure.io/fedora-infrastructure/issues?status=Open&priority=1
<@phsmoura:fedora.im>
20:04:39
!topic Tickets needing review
<@nirik:matrix.scrye.com>
20:04:48
morning
<@phsmoura:fedora.im>
20:04:51
!ticket 12282
<@zodbot:fedora.im>
20:04:52
<@zodbot:fedora.im>
20:04:52
**fedora-infrastructure #12282** (https://pagure.io/fedora-infrastructure/issue/12282):**f41 - createRepo/mergeRepo failures related to zstd in mirrors**
<@zodbot:fedora.im>
20:04:52
● **Last Updated:** Never
<@zodbot:fedora.im>
20:04:52
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
20:04:52
● **Opened:** 23 hours ago by rcallicotte
<@nirik:matrix.scrye.com>
20:05:19
I've not had time to look too much at this. I guess low low ops for now since no one else is yelling yet?
<@nirik:matrix.scrye.com>
20:05:40
possibly some difference between f41 base repos and f41-updates repos.
<@james:fedora.im>
20:05:40
I wasn't sure if this was just an upstream thing?
<@nirik:matrix.scrye.com>
20:06:04
well, base 41 works, not f41 updates, so that sounds like our bodhi config somehow...
<@nirik:matrix.scrye.com>
20:06:35
not sure if there's any differences that could explain that
<@james:fedora.im>
20:08:23
Maybe? But I'm not sure how much of our infra. they are using ... if the repos. were bad we'd get a lot of issues
<@james:fedora.im>
20:09:10
It could be something weirdly specific to updates repos. ... but it's going to be a pretty hard for us to work out what it is that fixes their problems, IMO.
<@james:fedora.im>
20:09:58
At first I thought they were using their own instance of koji and it was failing ... but reading again, it looks like they have their own tool which does the tasks koji does?
<@nirik:matrix.scrye.com>
20:10:00
well, updates and base repos are made with completely different tools... so I wouldn't be surprised if there's a difference, but yeah, odd that no one else is seeing it.
<@nirik:matrix.scrye.com>
20:10:11
no, they have their own koji.
<@james:fedora.im>
20:10:16
ahh
<@rcallicotte:fedora.im>
20:10:38
!hi
<@zodbot:fedora.im>
20:10:39
Robby Callicotte (rcallicotte) - he / him / his
<@nirik:matrix.scrye.com>
20:10:40
it's the koji tasks that are failing (for f41 updates only)
<@nirik:matrix.scrye.com>
20:10:45
hey Robby Callicotte!
<@james:fedora.im>
20:10:49
Yeh, my first instinct then is to tell them to file it with upstream koji and or give us more idea what we are doing wrong/different.
<@james:fedora.im>
20:11:00
Probably not the most welcoming response though.
<@james:fedora.im>
20:11:43
Worst case I could look at it a bit more today/tomorrow and ask questions.
<@nirik:matrix.scrye.com>
20:11:51
we could also cc mikem there to see if he knows off hand what it could be?
<@james:fedora.im>
20:12:08
Can't hurt 👍️
<@phsmoura:fedora.im>
20:13:10
ok, do you want to add a comment there?
<@nirik:matrix.scrye.com>
20:13:38
I can sure.
<@phsmoura:fedora.im>
20:14:08
!ticket 12283
<@zodbot:fedora.im>
20:14:09
● **Last Updated:** Never
<@zodbot:fedora.im>
20:14:09
<@zodbot:fedora.im>
20:14:09
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
20:14:09
● **Opened:** 16 hours ago by ryanlerch
<@zodbot:fedora.im>
20:14:09
**fedora-infrastructure #12283** (https://pagure.io/fedora-infrastructure/issue/12283):**risky-file-permissions: File permissions unset or incorrect. [openshift playbooks linting]**
<@nirik:matrix.scrye.com>
20:14:10
Robby Callicotte: were your ears burning?
<@rcallicotte:fedora.im>
20:14:23
hehe yep! :)
<@james:fedora.im>
20:14:52
I think this is just add mode=0644 to the end of the line?
<@james:fedora.im>
20:15:21
But then I saw a big PR and wasn't sure if Ryan had worked it out already, so didn't comment.
<@james:fedora.im>
20:15:54
But, low/low seems good anyway?
<@phsmoura:fedora.im>
20:16:05
+1 low low ops
<@nirik:matrix.scrye.com>
20:16:34
yeah, it's not solved, but I think just adding a mode= there should fix it.
<@nirik:matrix.scrye.com>
20:16:46
they just want you to be explicit about what mode it should have.
<@phsmoura:fedora.im>
20:17:29
!ticket 12284
<@zodbot:fedora.im>
20:17:30
**fedora-infrastructure #12284** (https://pagure.io/fedora-infrastructure/issue/12284):**no-changed-when: Commands should not change things if nothing needs doing. [openshift playbooks linting]**
<@zodbot:fedora.im>
20:17:30
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
20:17:30
● **Last Updated:** 15 hours ago
<@zodbot:fedora.im>
20:17:30
● **Opened:** 15 hours ago by ryanlerch
<@zodbot:fedora.im>
20:17:30
<@james:fedora.im>
20:18:06
Yeh, this sucks ... and we have similar problems in zabbix
<@nirik:matrix.scrye.com>
20:19:28
so yeah, command isn't idempotent...
<@james:fedora.im>
20:19:37
It's basically that a bunch of the playbook isn't idempotent
<@nirik:matrix.scrye.com>
20:20:57
some of the things could be moved to manual playbooks... like the bodhi pod scaling stuff.
<@nirik:matrix.scrye.com>
20:21:07
but yeah... need more thinking on it I guess.
<@james:fedora.im>
20:21:40
med/med? Kind of weird as all the work is upfront, and almost all the gain happens over time.
<@nirik:matrix.scrye.com>
20:21:58
sure.
<@phsmoura:fedora.im>
20:22:06
ok
<@phsmoura:fedora.im>
20:22:44
!info https://pagure.io/releng/issues?status=Open
<@phsmoura:fedora.im>
20:22:56
!releng 12457
<@zodbot:fedora.im>
20:22:57
● **Opened:** 5 hours ago by markkemel
<@zodbot:fedora.im>
20:22:57
**releng #12457** (https://pagure.io/releng/issue/12457):**Fix incorrectly created epel10 branch for bluechi package**
<@zodbot:fedora.im>
20:22:57
<@zodbot:fedora.im>
20:22:57
● **Last Updated:** a minute ago
<@zodbot:fedora.im>
20:22:57
● **Assignee:** Not Assigned
<@nirik:matrix.scrye.com>
20:23:53
This is a weird on. pagure should prevent people from making that kind of branch. ;(
<@phsmoura:fedora.im>
20:23:56
is it done?
<@nirik:matrix.scrye.com>
20:24:01
This is a weird one. pagure should prevent people from making that kind of branch. ;(
<@nirik:matrix.scrye.com>
20:24:30
no. we need to figure out how they were able to do it and fix that. ;)
<@nirik:matrix.scrye.com>
20:24:43
I guess it could be a different ticket, but seems ok to just keep this one for now.
<@phsmoura:fedora.im>
20:25:18
ah I see..
<@phsmoura:fedora.im>
20:25:49
ok so med med?
<@nirik:matrix.scrye.com>
20:25:54
sure
<@phsmoura:fedora.im>
20:26:30
!releng 12458
<@zodbot:fedora.im>
20:26:32
● **Opened:** 3 hours ago by decathorpe
<@zodbot:fedora.im>
20:26:32
● **Assignee:** Not Assigned
<@zodbot:fedora.im>
20:26:32
**releng #12458** (https://pagure.io/releng/issue/12458):**Unretirement request: rust-libhandy-sys and rust-libhandy**
<@zodbot:fedora.im>
20:26:32
<@zodbot:fedora.im>
20:26:32
● **Last Updated:** Never
<@nirik:matrix.scrye.com>
20:27:21
low low?
<@phsmoura:fedora.im>
20:27:33
+1
<@phsmoura:fedora.im>
20:28:13
!topic Planning, Upcoming work and Open floor
<@nirik:matrix.scrye.com>
20:28:45
been fires for me today... pagure.io load/scapers, random broken things. ;(
<@james:fedora.im>
20:30:09
Doing ipset-country port ... should hopefully have something by next week ... zabbix 7.0 in staging has the idempotent problem, so I can't just run a playbook. Not sure what to do, get nirik to manually do things on the machine or get more access so I can? Fix the playbook?
<@nirik:matrix.scrye.com>
20:30:40
I'm happy to get you more access if there's manual steps needed...
<@nirik:matrix.scrye.com>
20:30:59
if it's just lint failures, they shouldn't block anything, we can merge them
<@james:fedora.im>
20:32:21
No, I think it's more than lint. Playbook is setup to do an install only, so we'd need to reinstall the machine, or manually run the steps on the machine to upgrade it in place.
<@phsmoura:fedora.im>
20:33:07
nirik: I was reading the sop to add f40/41 test instances and changed 2 files to add them, so should upload the patch to batcave and then run ./do-domains?
<@nirik:matrix.scrye.com>
20:34:10
James Antill: yeah, either way... if restarting is ok thats easy, but if we want to run the upgrade steps manually thats fine too.
<@nirik:matrix.scrye.com>
20:34:31
phsmoura: put the patch somewhere? perhaps just add it to the ticket? and I can merge it and push it.
<@james:fedora.im>
20:35:25
nirik: Okay, cool, I'll probably ping you tomorrow about it.
<@nirik:matrix.scrye.com>
20:35:44
sure, sounds good.
<@phsmoura:fedora.im>
20:37:09
ok, we passed the time
<@phsmoura:fedora.im>
20:37:20
thanks everyone :)
<@nirik:matrix.scrye.com>
20:37:33
oh yeah, oops.
<@nirik:matrix.scrye.com>
20:37:53
thanks phsmoura and James Antill
<@phsmoura:fedora.im>
20:37:56
!endmeeting