16:12:39 <mboddu> #startmeeting RELENG (2020-11-03)
16:12:39 <zodbot> Meeting started Tue Nov  3 16:12:39 2020 UTC.
16:12:39 <zodbot> This meeting is logged and archived in a public location.
16:12:39 <zodbot> The chair is mboddu. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:12:39 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
16:12:39 <zodbot> The meeting name has been set to 'releng_(2020-11-03)'
16:12:39 <mboddu> #meetingname releng
16:12:39 <zodbot> The meeting name has been set to 'releng'
16:12:39 <mboddu> #chair nirik sharkcz pbrobinson pingou mboddu dustymabe ksinny jednorozec
16:12:39 <mboddu> #topic init process
16:12:39 <zodbot> Current chairs: dustymabe jednorozec ksinny mboddu nirik pbrobinson pingou sharkcz
16:12:55 <nirik> morning
16:13:15 <sharkcz> hi all
16:13:22 <mboddu> Sorry, that 1 min turned into 10 min
16:15:01 <mboddu> Okay, lets get started
16:15:24 <mboddu> #topic Alternate Arch Updates
16:15:31 <mboddu> sharkcz: Any updates?
16:16:13 <sharkcz> nothing for this week
16:17:10 <mboddu> nirik: How's s390x? I dont know why 16 is having the issues but not 15
16:17:17 <mboddu> Should we reprovision it?
16:17:22 <nirik> all of them did today
16:17:40 <nirik> no, it's unlikely anything to do with the instances... it's more likely network flakeyness
16:17:58 <mboddu> Oh, I didn't check today's compose, still in meetings (even now, I am attending another meeting)
16:18:09 <mboddu> Okay
16:18:19 * pingou in a meeting, won't be able to follow this one
16:18:22 <pingou> if you need me, ping me
16:18:27 <nirik> we will reprovision them soon anyhow when we move to f33, but I don't think it will help. ;(
16:19:17 <mboddu> Okay
16:19:25 <sharkcz> I agree, sounds more like network issue
16:20:24 <nirik> it may be related to the overloaded IAD2 upstream pipe...
16:20:37 * nirik doesn't know if it goes over those or not
16:20:45 <mboddu> ^ can you explain it?
16:20:59 <mboddu> "overloaded IAD2 upstream pipe"
16:21:29 <sharkcz> wouldn't traceroute show it?
16:21:46 <sharkcz> I mean the path
16:21:59 <nirik> yes, the thing that was causing our mirrors to not be able to sync last week? One of the 2 upstream internet connections we have in IAD2 was/is getting saturated. They are looking at increasing it's BW, but that hasn't happened yet.
16:22:24 <mboddu> Okay, thanks nirik for that info
16:22:30 <nirik> sharkcz: no, because it's all internal ip's... I have no idea where say 10.18.255.253 is
16:22:47 <sharkcz> ok
16:23:50 <mboddu> Okay, moving on
16:24:11 <mboddu> #topic Arm images rc1.3 recomposing
16:24:32 <mboddu> So, this work is done on Thu, and I synced them to mirrors on Friday
16:25:13 <mboddu> We updated gf.o and arm.fp.o but not alt.fp.o, there is a ticket on it - https://pagure.io/fedora-web/websites/issue/158
16:25:27 <mboddu> Once this is fixed, I will remove 1.2 versions of them
16:26:43 <mboddu> This is a very interesting release :)
16:26:45 <nirik> cool. thanks much mboddu
16:26:52 <nirik> yeah, we live in interesting times. ;)
16:27:03 <nirik> if we do one for shim it's going to be harder...
16:27:11 <mboddu> Not sure when we are going to get the shim update
16:27:30 <mboddu> Yeah, we will see what happens
16:28:42 <nirik> yeah, indeterminite, and also if we need to redo images we don't know yet either.
16:30:34 <mboddu> Right
16:30:37 <mboddu> Anyway, moving on
16:30:40 <mboddu> #topic #9308 F33 system wide change: Sqlite Rpmdb
16:30:46 <mboddu> #link https://pagure.io/releng/issue/9308
16:31:49 <nirik> this is on my list for this week.
16:31:59 <nirik> I need to test bootstrap in stg...
16:32:23 <mboddu> Okay, I am reading what we need to test actually :)
16:32:34 <mboddu> Its been a long time
16:33:55 <nirik> so, when we move the builders to f33... rpm is going to be moving to sqlite rpm db by default.
16:34:12 <nirik> which means that mock will use the host rpm to make chroots...
16:34:30 <nirik> which means that old fedoras and epels that have bdb rpm db will... all break
16:35:04 <nirik> so, we need to enable mock's bootstrap mode where it makes a chroot with the target rpm / dnf / yum in it and uses that to make the other stuff.
16:35:22 <nirik> last time I enabled it, it worked, but it broke epel7 at least...
16:35:32 <mboddu> Ahhh, okay, now I remembered, thanks nirik
16:35:42 <mboddu> Let me know if you need any help nirik
16:35:49 <nirik> because we have devtoolset enabled and a 'dnf install yum' pulls in some crazy lib from there, then yum doesn't run
16:36:06 <nirik> sure thing. Probibly will work on it tomorrow if you want to help out/play along.
16:36:19 <mboddu> Sure, let me know and I can help
16:40:03 <mboddu> #topic #9794 permissions for carlwgeorge to implement epel-next proposal
16:40:09 <mboddu> #link https://pagure.io/releng/issue/9794
16:41:18 <mboddu> I am planning to work on this probably this week or next week with Carl
16:41:21 <nirik> I am not sure what these might be.
16:41:50 <nirik> sounds good. Perhaps it would be good to pick a time and interested parties could follow along/help?
16:42:05 <mboddu> Okay
16:42:16 * carlwgeorge waves
16:42:33 <nirik> I find that scheduling also helps make sure you don't forget something when swamped with other items. :)
16:42:38 <mboddu> carlwgeorge: When are you available?
16:42:57 <mboddu> This week I am sorta busy, I prefer next week
16:44:33 <mboddu> carlwgeorge: Hello :)
16:45:39 <mboddu> Okay, I will check with carlwgeorge later and I will update the ticket with a date and time
16:45:45 <mboddu> #topic Open Floor
16:45:52 <mboddu> Anybody has anything to share?
16:45:53 <nirik> ok. I have a few items
16:46:17 <mboddu> Go ahead
16:46:39 <nirik> 1. After the bootstrap stuff is figured out, and after pbrobinson does some armv7 testing, we will want to move builders to fedora 33... so likely next week or the week after.
16:47:07 <nirik> in f33 hopefully the armv7 stuff moves to just direct uefi... which will be much nicer than the current setup we have. :)
16:47:52 <mboddu> Yeah, it would be nice :)
16:47:53 <nirik> 2. There's a koji update pending... it's just bugfixes, I don't think it even has a schema update (altough I would have to check).
16:47:59 <nirik> when do we want to schedule that?
16:48:27 <mboddu> How about scheduling it when we update the builders to f33?
16:48:50 <nirik> oh, there is a schema update... but it's not something that matters to us at all. :)
16:49:01 <nirik> sounds good... as long as it doesn't take too long.
16:49:05 <nirik> and finally...
16:49:48 <nirik> 3. Someone is requesting we sign repodata. I am going to close it WONTFIX. There are pungi and robosignatory issues on supporting that already and I think that there's better ways to spend out time...
16:50:33 <nirik> unless anyone objects?
16:51:26 <mboddu> Yeah, we can close it WONTFIX
16:52:25 <mboddu> Thats all I got
16:52:46 <mboddu> If nothing else, I will give back the 10 min back to you guys
16:53:06 <nirik> thats all for me
16:53:17 <mboddu> Okay, thanks nirik :)
16:53:21 <mboddu> #endmeeting