<@jnsamyak:matrix.org>
15:00:31
!startmeeting RELENG (2025-09-01)
<@meetbot:fedora.im>
15:00:34
Meeting started at 2025-09-01 15:00:31 UTC
<@meetbot:fedora.im>
15:00:34
The Meeting name is 'RELENG (2025-09-01)'
<@jnsamyak:matrix.org>
15:00:34
!meetingname releng
<@meetbot:fedora.im>
15:00:36
The Meeting Name is now releng
<@jnsamyak:matrix.org>
15:00:39
!chair nirik jnsamyak patrikp amedvede
<@jnsamyak:matrix.org>
15:00:44
!info Meeting is 60 minutes MAX. At the end of 60, it STOPS.
<@jnsamyak:matrix.org>
15:00:48
<@jnsamyak:matrix.org>
15:00:48
!info Agenda is at https://hackmd.io/vm6biLBcTYKtkQUH5kQkmw.
<@patrikp:matrix.org>
15:01:00
Hello. 👋
<@jnsamyak:matrix.org>
15:01:05
!topic Init process.
<@jnsamyak:matrix.org>
15:01:10
Do we have anything for the init? Any blockers/tasks/issues/requests/features that need releng intervention?
<@jnsamyak:matrix.org>
15:01:10
<@jnsamyak:matrix.org>
15:01:14
Hey folks!
<@amedvede:fedora.im>
15:01:18
Hey!
<@jnsamyak:matrix.org>
15:01:21
How's everyone upto?
<@patrikp:matrix.org>
15:01:58
I'm good, thanks. Weather has been nice.
<@patrikp:matrix.org>
15:02:22
I don't have anything for init, I have one thing for tickets.
<@amedvede:fedora.im>
15:02:44
Really want to sleep, woke up at 5 am its evening for me xd, how you doing?
<@amedvede:fedora.im>
15:03:12
also nothing for init
<@jnsamyak:matrix.org>
15:03:33
For the starters today is day 5 of fasting
<@jnsamyak:matrix.org>
15:03:42
so pretty mch handling with waht i can hehe
<@jnsamyak:matrix.org>
15:03:54
and the city is under the water
<@jnsamyak:matrix.org>
15:04:04
it's been crazy raining here since 2 days
<@amedvede:fedora.im>
15:04:58
why are you fasting, is it connected to religion or something?
<@jnsamyak:matrix.org>
15:05:58
yeah
<@jnsamyak:matrix.org>
15:06:14
In my religion there are 10 days each day has it's own meaning
<@jnsamyak:matrix.org>
15:06:35
in which we try to fast and look into how we can work on our soul bad deed etc etc
<@jnsamyak:matrix.org>
15:06:52
in case you are interested: https://cdn.ymaws.com/www.jaina.org/resource/resmgr/Religious_Article/Das_Lakshan_Parva_by_Dr._Bha.pdf
<@amedvede:fedora.im>
15:07:40
ooh bad to head, for me at least it can be an execuse to not go to gym xd
<@jnsamyak:matrix.org>
15:07:52
But for work I have some stuff (one pr and issue) I worked over the weekend, and I want your opinions:
<@jnsamyak:matrix.org>
15:07:52
- https://pagure.io/releng/issue/12925
<@jnsamyak:matrix.org>
15:07:52
- https://pagure.io/releng/pull-request/12924
<@jnsamyak:matrix.org>
15:08:27
each has explanation there why these are things I want to do, I think we can retire ODCS from our scripts and docs as well but wanted to confirm first
<@amedvede:fedora.im>
15:09:33
Oh, I though we get rid of PDC related scripts during PDC decomission
<@jnsamyak:matrix.org>
15:10:15
I think it stayed there and no one cleaned up but before moving to move to new forge i want to get these done
<@amedvede:fedora.im>
15:11:53
As I mentioned in ticket, I think restructuring of scripts its a good idea, i know from my own exerience that it might be challenging to find script you need and sometimes it easier to write new one instead
<@patrikp:matrix.org>
15:13:24
The scripts reorg, massive +1 for the idea, long overdue. For the structure itself I need a bit more time to ponder on it, however I took a look at the proposed structure and there was nothing I strongly disagreed with that I could see off hand. The deadline for this is before Forgejo migration?
<@patrikp:matrix.org>
15:13:24
PDC and ODCS retirement is +1 for me, I will double check the PR and give the +1 there.
<@patrikp:matrix.org>
15:13:24
<@jnsamyak:matrix.org>
15:14:04
yeah there might be scripts which might be duplicates but with this we can clean those up gradually the challenge is that we need to carefully do these migration because some of the scripts are touched by external repos and we need to change those there as well
<@jnsamyak:matrix.org>
15:14:31
<@jnsamyak:matrix.org>
15:14:31
> The deadline for this is before Forgejo migration?
<@jnsamyak:matrix.org>
15:14:31
I personally would like to get it done before it
<@amedvede:fedora.im>
15:14:40
1 thing comes to my mind: do we want to have QA scripts and infrastructure scripts in releng repo or it would be better to pass it to different repos?
<@jnsamyak:matrix.org>
15:14:46
because we will be changing urls/repos links that time
<@jnsamyak:matrix.org>
15:15:10
so it would be good to have these done once and for all
<@jnsamyak:matrix.org>
15:15:31
What scripts are you talking about?
<@patrikp:matrix.org>
15:15:42
This will be tricky.
<@patrikp:matrix.org>
15:15:42
That part is concerning, I've been thinking about it. How can we check for if some script clones the repo and runs a script using a path? I.e. `git clone relengrepo` and `./scripts/some-script.sj`
<@patrikp:matrix.org>
15:15:52
That part is concerning, I've been thinking about it. How can we check for if some script clones the repo and runs a script using a path? I.e. `git clone relengrepo` and `./scripts/some-script.sh`
<@patrikp:matrix.org>
15:15:52
This will be tricky.
<@jnsamyak:matrix.org>
15:16:33
The ones here are mostly used with conjuction with qe/infra, for example removing a dist-git branch check it's under infrastructure but it is indeed used for releng and by releng
<@amedvede:fedora.im>
15:16:42
I mean folders in suggested structure
<@amedvede:fedora.im>
15:17:03
oh i see
<@jnsamyak:matrix.org>
15:17:46
that's something I'll be asking help from you both once we have a +1 from most of us for the proposal
<@jnsamyak:matrix.org>
15:17:53
divide and conquer :)
<@amedvede:fedora.im>
15:18:56
anyway, I think it would be nice to lets say devide those scripts into 3 parts so each of us investigate own portion of script and make conclusion which are up to date and which are just a copy
<@amedvede:fedora.im>
15:19:16
aaaah I'm writing so slow
<@jnsamyak:matrix.org>
15:19:25
that is exactly what i said hehe
<@jnsamyak:matrix.org>
15:19:27
nw
<@jnsamyak:matrix.org>
15:19:37
okay anything else?
<@patrikp:matrix.org>
15:19:51
It's not about the scripts themselves. It's about where those scripts are called FROM.
<@jnsamyak:matrix.org>
15:20:00
yeah
<@patrikp:matrix.org>
15:20:36
Fedora Infra is dark and full of terrors and we will inevitably break something important. But if we look out and patch it quickly... Let's go.
<@jnsamyak:matrix.org>
15:21:12
yeah dont be scare to break things, everything is fixable after a rant:P
<@jnsamyak:matrix.org>
15:21:18
yeah dont be scare to break things, everything is fixable after a rant :P
<@amedvede:fedora.im>
15:21:45
xd
<@patrikp:matrix.org>
15:21:49
I think the docs are in such a state that if someone has a big enough thing to complain about they will inevitably open a ticket on either fedora-infra or releng tracker and we can pick it up from there.
<@patrikp:matrix.org>
15:22:14
Someone will point the person to the repo where the ticket can be opened...
<@patrikp:matrix.org>
15:22:49
And if it's not a big enough deal to open a ticket then is it really an issue?
<@amedvede:fedora.im>
15:23:42
so for now, we will wait until +1 from Kevin and after that plan work regarding it?
<@jnsamyak:matrix.org>
15:25:08
whoever wants to review i'll open it for a week, otherwise i'll go ahead
<@patrikp:matrix.org>
15:25:34
You mean the ticket? Or you already have a PR?
<@jnsamyak:matrix.org>
15:26:06
ticket issue -> then we all do investigations -> finally after looking we will have a pr that's my plan
<@patrikp:matrix.org>
15:26:35
To double check, what is the deadline on this as a date in the calendar? Is there already a date set for the Forgejo migration?
<@jnsamyak:matrix.org>
15:27:15
I'm not sure I'll talk to tomas but it should happen this month
<@jnsamyak:matrix.org>
15:27:26
but there is no stone date
<@jnsamyak:matrix.org>
15:27:28
set
<@patrikp:matrix.org>
15:28:03
This month already? Soon there will be no more pagure?
<@jnsamyak:matrix.org>
15:28:25
no not everything i smoving i think it is interation
<@patrikp:matrix.org>
15:28:34
Or if there's a delay then next month? But so soon pagure will no longer exist as a part of the daily workflow?
<@jnsamyak:matrix.org>
15:28:40
of small events/repos we are gradually trying to test things that are working or not
<@jnsamyak:matrix.org>
15:28:41
iirc
<@patrikp:matrix.org>
15:29:13
So the migration will start and be gradual, not all at once, gotcha.
<@jnsamyak:matrix.org>
15:29:45
you are overthinking all the migration happens gradually releng is a good candidate for us to test the new forge since most of the people are the one working with or adjacent with forgejo folks
<@patrikp:matrix.org>
15:29:53
The thought of not having pagure, whenever it happens, struck me in a surprising way, it'll be like saying bye to an old friend.
<@jnsamyak:matrix.org>
15:30:01
:)
<@patrikp:matrix.org>
15:30:39
It works surprisingly well despite not having all the bells and whistles, I don't have all that many complaints.
<@amedvede:fedora.im>
15:30:44
:), hopefully Forgejo sometimes will also be slow which will remind us old days
<@jnsamyak:matrix.org>
15:30:50
Time check! We have two more topics
<@patrikp:matrix.org>
15:31:06
Next.
<@jnsamyak:matrix.org>
15:31:16
!topic Scheduled actions coming up in the next week.
<@jnsamyak:matrix.org>
15:31:21
!info Here we list/discuss anything about items that are due to be done in the next week.
<@jnsamyak:matrix.org>
15:31:26
<@jnsamyak:matrix.org>
15:31:26
!link https://fedorapeople.org/groups/schedule/f-43/f-43-releng-tasks.html
<@jnsamyak:matrix.org>
15:31:51
We are in beta freeze since last week, the qe folks had some of the stable requests filed, which were processed
<@jnsamyak:matrix.org>
15:32:12
currently we are dealing with https://qa.fedoraproject.org/blockerbugs/milestone/43/beta/buglist before they will request a rc beta compose
<@jnsamyak:matrix.org>
15:33:03
<@jnsamyak:matrix.org>
15:33:03
but I would expect you folks to be proactive and keep looking into the issue if we have candidate request filed by qe: https://docs.fedoraproject.org/en-US/infra/release_guide/sop_beta_RC_compose/
<@jnsamyak:matrix.org>
15:33:03
Here is the RC Compose SOP
<@jnsamyak:matrix.org>
15:33:50
go over this if you have questions discuss in releng or with me, and when the actual work will come we will try to do these.
<@jnsamyak:matrix.org>
15:34:16
The first Go/No-GO meeting is scheduled to happen on 11 september
<@patrikp:matrix.org>
15:34:51
I have a thing about this but I'm unsure how to word it.
<@jnsamyak:matrix.org>
15:35:09
?
<@patrikp:matrix.org>
15:38:20
I talked to Fabian about how CentOS Stream does these things but I haven't been able to extract all that much information. I know some things but I don't know the whole story, the CentOS Stream infra isn't as open as the Fedora infra, try to find their equivalent of our `ansible` repo. But I think it's something worth exploring. I don't want to use up too much time so I'll write down what I know and we can move to next section, just wanted to mention it.
<@jnsamyak:matrix.org>
15:39:14
Yeah if you have some to discuss either open a improvement ticket on releng and we can discuss there as well
<@jnsamyak:matrix.org>
15:39:19
!topic Tickets that need attention.
<@jnsamyak:matrix.org>
15:39:25
!info This is the time where you can bring up releng tickets to discuss if there are blockers, etc.
<@jnsamyak:matrix.org>
15:39:31
!link https://pagure.io/releng/issues
<@jnsamyak:matrix.org>
15:39:31
<@jnsamyak:matrix.org>
15:39:43
!releng 12888
<@zodbot:fedora.im>
15:39:46
<@zodbot:fedora.im>
15:39:46
**releng #12888** (https://pagure.io/releng/issue/12888):**shell-color-prompt has description "NA" in Bugzilla**
<@zodbot:fedora.im>
15:39:46
● **Opened:** 2 weeks ago by petersen
<@zodbot:fedora.im>
15:39:46
● **Last Updated:** a week ago
<@zodbot:fedora.im>
15:39:46
● **Assignee:** jnsamyak
<@jnsamyak:matrix.org>
15:40:20
Wanted ti bring this uo with kevin because i think i'm little lost with the component settings here because the access that i have I'm not able to look into the details
<@jnsamyak:matrix.org>
15:40:39
Wanted to bring this up with Kevin because I think I'm a little lost with the component settings here, because the access that I have, I'm not able to look into the details
<@jnsamyak:matrix.org>
15:40:55
So moving on
<@jnsamyak:matrix.org>
15:41:19
Do you folks want to bring up a ticket, you can use !releng NUM
<@jnsamyak:matrix.org>
15:41:59
BTW these are good to go
<@jnsamyak:matrix.org>
15:41:59
- 12880
<@jnsamyak:matrix.org>
15:41:59
- 12884
<@jnsamyak:matrix.org>
15:41:59
Once we are done with retrospective
<@jnsamyak:matrix.org>
15:41:59
<@jnsamyak:matrix.org>
15:41:59
- 12870
<@patrikp:matrix.org>
15:42:41
I have one thing.
<@patrikp:matrix.org>
15:42:53
!releng 12905
<@zodbot:fedora.im>
15:42:55
● **Assignee:** patrikp
<@zodbot:fedora.im>
15:42:55
● **Last Updated:** 3 hours ago
<@zodbot:fedora.im>
15:42:55
● **Opened:** a week ago by churchyard
<@zodbot:fedora.im>
15:42:55
<@zodbot:fedora.im>
15:42:55
**releng #12905** (https://pagure.io/releng/issue/12905):**When packages are unretired, why are old builds of them tagged?**
<@patrikp:matrix.org>
15:44:08
Me and Samyak lean one way, two senior members of Fedora ecosystem gave very compelling arguments for the other way that I generally agree with and Kevin said both sound very reasonable.
<@jnsamyak:matrix.org>
15:45:41
I think the argument you are mentioning to refer is untagging all the builds irrespective of reasons?
<@patrikp:matrix.org>
15:46:30
It's an open question. All approaches sound viable to me.
<@jnsamyak:matrix.org>
15:47:09
https://pagure.io/releng/issue/12905#comment-984075 :)
<@jnsamyak:matrix.org>
15:47:11
sorted
<@jnsamyak:matrix.org>
15:47:49
I just added a line and this should go into the change in template so WE ALWAYS proactively put this in ticket
<@patrikp:matrix.org>
15:47:53
I don't see a reason not to untag all builds IF AND ONLY IF the package is retired for FTBFS/FTI reasons. But when I asked Miro about that he said he is in favor of untagging ALL builds for ALL packages at the time of retirement, i.e. the retirement toddler way. But you as the releng lead don't think the toddler is the way to go. So we are at a conundrum.
<@patrikp:matrix.org>
15:48:55
So what Miro said then? Make it so that the retirement toddler automatically untags ALL builds for the newly retired package?
<@patrikp:matrix.org>
15:49:47
Regardless of retirement reason?
<@jnsamyak:matrix.org>
15:49:48
Can you point out the specifc area when he said the retirement toddler can be used for this?
<@jnsamyak:matrix.org>
15:50:27
this is where you proposed https://pagure.io/releng/issue/12905#comment-983596 doing at unretirement and in the next comment he agreed
<@jnsamyak:matrix.org>
15:50:34
sorry if I'm missing a context
<@patrikp:matrix.org>
15:50:43
I think blanket untagging of ALL builds for EACH (un)retired package is the way to go.
<@patrikp:matrix.org>
15:50:43
How else do you do it if not through the toddler?
<@patrikp:matrix.org>
15:50:43
<@patrikp:matrix.org>
15:50:43
It's implicit. He said:
<@patrikp:matrix.org>
15:51:13
Sounds like the untagging should be part of the retirement process itself?
<@jnsamyak:matrix.org>
15:51:25
And i see kevin gotmax23 also is okay with doing at the time of unretirement
<@jnsamyak:matrix.org>
15:51:59
manually at the time of unretirement how you unblock a package untag build too
<@patrikp:matrix.org>
15:52:16
But that's in stark contrast to what gotmax and Miro are saying.
<@patrikp:matrix.org>
15:52:43
Untagging builds at unretirement time sounds feasible, ***but I would suggest always untagging builds — not just for certain packages.***
<@patrikp:matrix.org>
15:52:49
There's a but in there.
<@jnsamyak:matrix.org>
15:53:03
Okay you are confused here
<@jnsamyak:matrix.org>
15:53:30
they meant untagging not the packages for just fti/ftbfs but for all the packages where unretirement are filed for
<@patrikp:matrix.org>
15:53:33
I think when he said unretirement later it was a typo and he meant retirement.
<@jnsamyak:matrix.org>
15:53:35
does it make sense now?
<@patrikp:matrix.org>
15:53:40
Ah.
<@patrikp:matrix.org>
15:53:57
OK, I completely missed this.
<@patrikp:matrix.org>
15:54:35
I will post a comment in there to double check with everyone. 5 minutes to go.
<@jnsamyak:matrix.org>
15:54:49
no worries, I think things are clear now, and my last comment is with aggreement with folks, let's just document it, make change to template onwards in releng settings, and then just close it.
<@jnsamyak:matrix.org>
15:55:06
That reminds me what's the status of unretirement toddler Anton Medvedev ?
<@amedvede:fedora.im>
15:56:58
its on staging, ready to use, I triedn to run it manually and up to revert time it worked
<@jnsamyak:matrix.org>
15:57:11
!topic Choose next chair.
<@amedvede:fedora.im>
15:57:29
wait
<@amedvede:fedora.im>
15:57:30
xd
<@amedvede:fedora.im>
15:57:54
I have a ticket to add
<@jnsamyak:matrix.org>
15:58:06
What do you mean by upto reverted? Does it not unblock the package and add owner?
<@jnsamyak:matrix.org>
15:58:45
We are on the time, put your ticket in releng channel we will take it there
<@jnsamyak:matrix.org>
15:58:55
meanwhile any volunteers to chair next weekly call?
<@amedvede:fedora.im>
15:59:10
I can chair
<@jnsamyak:matrix.org>
15:59:21
!info Next chair September 08th: Anton Medvedev
<@jnsamyak:matrix.org>
15:59:29
!topic Discussion / Any other business / Open floor.
<@jnsamyak:matrix.org>
15:59:29
<@amedvede:fedora.im>
15:59:37
https://pagure.io/releng/issue/12810 jnsamyak can you please take a look
<@jnsamyak:matrix.org>
15:59:38
okay one min to answer this :P
<@jnsamyak:matrix.org>
15:59:51
and then your ticket at releng
<@patrikp:matrix.org>
16:00:08
Good meeting, thanks for hosting and have a good rest of the day.
<@amedvede:fedora.im>
16:00:38
when I was testing I t made a commit revert and koji was down and it failed on it
<@jnsamyak:matrix.org>
16:00:55
so it was an outage issue, test it again
<@jnsamyak:matrix.org>
16:01:01
and let us know what's the status
<@amedvede:fedora.im>
16:01:14
I will update the ticket
<@jnsamyak:matrix.org>
16:01:15
because to be honest it has been long due for a very long time now
<@jnsamyak:matrix.org>
16:01:35
half of these work will be not required if we have that toddler in picture
<@jnsamyak:matrix.org>
16:01:48
That's all!
<@jnsamyak:matrix.org>
16:01:51
!info Thank you all for coming!
<@jnsamyak:matrix.org>
16:01:56
!endmeeting