15:00:24 <mboddu> #startmeeting RELENG (2020-07-21)
15:00:24 <zodbot> Meeting started Tue Jul 21 15:00:24 2020 UTC.
15:00:24 <zodbot> This meeting is logged and archived in a public location.
15:00:24 <zodbot> The chair is mboddu. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:24 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:24 <zodbot> The meeting name has been set to 'releng_(2020-07-21)'
15:00:24 <mboddu> #meetingname releng
15:00:24 <zodbot> The meeting name has been set to 'releng'
15:00:24 <mboddu> #chair nirik sharkcz pbrobinson pingou mboddu dustymabe ksinny jednorozec
15:00:24 <zodbot> Current chairs: dustymabe jednorozec ksinny mboddu nirik pbrobinson pingou sharkcz
15:00:25 <mboddu> #topic init process
15:00:33 <nirik> morning
15:01:30 <mboddu> Morning nirik
15:01:32 <mboddu> How goes?
15:01:47 <nirik> Not too bad so far.
15:03:06 <mboddu> Sorry, I am looking at mass rebuild stuff
15:03:16 * mboddu is kinda deviated
15:03:25 <jednorozec> hello
15:03:36 <nirik> hey jednorozec
15:03:46 <nirik> mboddu: yeah, how's it looking? might be a delay?
15:05:04 <mboddu> nirik: Dont know yet, but it seems probable
15:05:21 <mboddu> On that note
15:06:12 <mboddu> #topic  #9565 Releng assessment of change: Patches in Forge macros - Auto macros - Detached rpm changelogs
15:06:17 <mboddu> #link https://pagure.io/releng/issue/9565
15:07:16 <mboddu> So, seems like we should include it in mass rebuild
15:07:23 <mboddu> ?
15:07:36 <mboddu> But I am not sure whats the status from nim
15:07:58 <nirik> no.
15:08:03 <mboddu> .fasinfo nim
15:08:04 <zodbot> mboddu: User: nim, Name: None, email: nicolas.mailhot@laposte.net, Creation: 2005-05-21, IRC Nick: None, Timezone: None, Locale: None, GPG key ID: None, Status: active
15:08:04 <nirik> it's not been approved at all.
15:08:07 <zodbot> mboddu: Approved Groups: cla_done fedorabugs +packager ambassadors cla_fedora cla_fpca @gitfontpackages provenpackager @go-sig
15:08:31 <mboddu> nirik: Oh, thats good to know, I didn't check that
15:08:40 <nirik> also it would need work in koji or something...
15:08:45 <mboddu> So, closed, get back later?
15:09:16 <nirik> except... I might be thinking of the other one...
15:09:52 <nirik> in any case it's not approved yet...
15:09:58 <nirik> so I would just leave it open for now
15:10:03 <mboddu> Okay
15:10:05 <nirik> and not worry about it for the mass rebuild.
15:11:30 <mboddu> ack
15:11:33 <mboddu> #topic #9359 F33 system wide change: Introduce module Obsoletes and EOL
15:11:47 <mboddu> #link https://pagure.io/releng/issue/9359
15:12:34 * pingou has a question about the mass rebuild later
15:12:55 <mboddu> pingou: Sure
15:13:20 * mboddu also has question on mass rebuild :D
15:13:32 <nirik> On this no idea.. I guess it could be after the mass rebuild... no need to tie it to that.
15:14:33 <mboddu> Its definitely not mass rebuild related, but I also dont know whats the status on it
15:14:45 <mboddu> So, I will let it be as is
15:15:53 <nirik> Might ask on ticket for an update...
15:15:57 <pingou> do we have ways to ask for follow up on those or is it up to us to check them?
15:16:32 <mboddu> pingou: We generally ping them few days in advance to check the status on them
15:16:43 <mboddu> #topic #9308 F33 system wide change: Sqlite Rpmdb
15:16:49 <mboddu> #link https://pagure.io/releng/issue/9308
15:16:55 <mboddu> nirik's favorite ticket :D
15:17:01 * mboddu hides from nirik
15:18:01 <mboddu> nirik: So, how do you think we should handle this?
15:19:09 <nirik> I think we are fine and don't need bootstrap... at least thats the way it seems.
15:19:54 <nirik> I can ask if there's any reason we still need to in ticket.
15:21:09 <nirik> updated
15:21:26 * mboddu checking something
15:24:15 <mboddu> nirik: But it seems to be included in mass rebuild, right?
15:24:35 <mboddu> I am seeing rpm 4.16 but its still in beta
15:25:54 <nirik> yes, it landed a while back
15:26:18 <mboddu> Just go with the beta?
15:26:28 <mboddu> Or should we ask in the ticket?
15:27:03 <nirik> you can ask, but I think the current one is ok.
15:27:13 <nirik> never hurts to ask and make sure
15:28:21 <mboddu> Yup
15:28:26 <mboddu> I just pinged in the ticket
15:29:19 <mboddu> F33 release seems crazy
15:29:25 <mboddu> https://pagure.io/releng/issues?status=Open&tags=mass+rebuild - for mass rebuild
15:29:32 <mboddu> And there are bunch of other changes
15:29:39 <nirik> yeah, indeed.
15:29:51 <nirik> wonder why (he says sarcastically)
15:30:40 <jednorozec> that is a lot of changes
15:31:08 <pingou> turns out a certain 3 years cadence is going to put pressure on us, who guessed? :D
15:31:15 <mboddu> nirik: Well, you are one of them to approve the changes?
15:31:28 <mboddu> pingou: Haha :)
15:31:46 <mboddu> #topic Mass rebuild
15:32:06 <mboddu> #info https://pagure.io/releng/issues?status=Open&tags=mass+rebuild are the list of changes that requires mass rebuild for F33
15:32:10 <mboddu> pingou: Your question?
15:34:16 <pingou> I was curious on who is going to be running point on it :)
15:34:23 <pingou> mboddu or jednorozec ? :)
15:34:36 <pingou> (who do we nag ? :))
15:34:55 <jednorozec> I am, and it seems i got the best for a first time
15:35:04 <pingou> \ó/
15:35:17 <pingou> jednorozec: okido, feel free to ping if I can help :)
15:35:20 <mboddu> pingou: Its going to be jednorozec and I will be on the side looking over it
15:35:52 <pingou> I'll read the logs of the meeting later, I need to step out
15:36:04 <mboddu> thanks pingou for joining
15:36:40 <mboddu> nirik: Saved the best for the last :D
15:36:45 <mboddu> So, branched-composer and compose-x86-01?
15:37:05 <mboddu> I guess compose-x86-01 is not that important, but we need another box to run mass rebuild
15:37:20 <mboddu> rawhide composer cant take both rawhide composes and mass rebuild
15:37:39 <mboddu> Historically, we used branched-composer for it
15:38:03 <nirik> we have a compose-branched01
15:38:14 <nirik> and a compose-x86-01
15:39:23 <mboddu> nirik: Oh, I didn't know that they are setup
15:39:28 <mboddu> When did that happened?
15:39:43 <nirik> I think they were in the minimal set... have always been there.
15:39:57 <nirik> they may need updating...
15:40:39 <mboddu> nirik: But compose-branched01 has the same mem as that of compose-rawhide01
15:41:10 <mboddu> Minimal set as in, running on less resources or they got setup last?
15:41:59 <nirik> minimal set as in "The list of vm's we need to be running to provide a minimal viable fedora infra" ie, we added them before we migrated to the new datacenter.
15:42:14 <nirik> let me know if you need any adjustments on them...
15:43:10 <mboddu> nirik: I thought branched composer was not one of them
15:43:15 <mboddu> Or did I read it wrong
15:43:22 <mboddu> I dont know, everything is a blur now
15:43:50 <nirik> yeah, for me too.
15:44:03 <mboddu> I can see the releng key tab in there, thats all that is needed for now
15:44:33 <mboddu> #topic Open Floor
15:44:41 <mboddu> Anybody got anything to share?
15:45:28 <nirik> yeah
15:46:26 <nirik> There seems to be a ongoing issue with a koji-gc (garbage collect) query taking up tons of cpu cycles and after a while dragging the database down into slowness. There's a koji issue on it I think... will try and get more visibility on it.
15:46:42 <nirik> The workaround right now is to restart postgresql. ;(
15:46:56 <mboddu> Okay, I will keep an eye on it
15:47:41 <mboddu> #info koji-gc seems to be having some problem, koji issue is filed and temporary solution is it restart postgresql
15:48:01 <nirik> hum, I guess I just mentioned it in a infra ticket, will refile upstream.
15:49:17 <mboddu> nirik: https://pagure.io/koji/issue/2361 ?
15:51:35 <nirik> nope... thats a different one. ;)
15:52:16 <nirik> filed https://pagure.io/koji/issue/2383
15:53:09 <mboddu> Okay
15:53:11 <mboddu> Thanks nirik
15:53:26 <mboddu> #info koji ticket to track this issue - https://pagure.io/koji/issue/2383
15:53:49 <mboddu> Anything else?
15:55:22 <nirik> nope... not that I can think of right now
15:55:32 <mboddu> Okay, thanks everyone for joining
15:55:38 <mboddu> Thats all I got as well
15:55:50 <mboddu> I will give back 5min
15:55:53 <mboddu> #endmeeting