15:01:40 #startmeeting RELENG (2022-06-14) 15:01:40 Meeting started Tue Jun 14 15:01:40 2022 UTC. 15:01:40 This meeting is logged and archived in a public location. 15:01:40 The chair is jednorozec. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 15:01:40 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:40 The meeting name has been set to 'releng_(2022-06-14)' 15:01:40 #meetingname releng 15:01:40 The meeting name has been set to 'releng' 15:01:40 #chair nirik sharkcz pbrobinson phsmoura dustymabe jednorozec 15:01:40 #topic init process 15:01:40 Current chairs: dustymabe jednorozec nirik pbrobinson phsmoura sharkcz 15:01:42 sorry 15:01:51 np 15:01:53 morning 15:01:54 no worries. morning 15:01:57 afternoon 15:02:53 so what do we have here for today? 15:03:24 I didn't have anything urgent. we could go thru tickets or discuss whatever folks would like 15:03:49 same don't have anything urgent just had some extra time 15:04:29 same here, just a note about ODCS being broken 15:04:58 we need to document our deployments little bit more, ansible playbook is not necessarly a documentation :) 15:05:21 yeah, sorry about that... but the f32 backend needed replacing. I didn't realize that the versions wouldn't be compatible 15:05:52 well that just had to be done 15:05:56 no questions about that 15:06:18 also it breaks so it breaks, that is normal. 15:06:24 I'm not sure why it was fedora in the first place rather than rhel8... 15:06:37 but I figured there was some reason, so I kept it fedora 15:06:45 But there is not much to find about the service deployment it self like 15:07:10 why it has multiple backends and what funcion they have 15:07:11 and so 15:07:39 The -releng backend was setup with more resources at some time when it was unclear who was going to be using it for what. 15:07:52 but yeah, it would be more clear if it just had one backend... 15:08:08 or 2 or 3 but with a few lines why 15:08:15 my understanding is that it's used by: 15:08:30 eln composes, fedora minimal composes? and flatpaks 15:08:52 (we are talking about odcs... on demand compose service) 15:09:47 I am not sure about the flatpaks, but now I know how to findout and document it :) 15:10:07 I suppose we have a SOP for it? lets see 15:10:13 yup 15:10:17 but not much info there 15:10:51 https://docs.fedoraproject.org/en-US/infra/sysadmin_guide/odcs/ 15:11:10 yeah. also very out of date. ;) 15:11:28 I am adding deployment section with info about how it is done 15:11:44 and package version mismatch into the what could go wrong section 15:12:39 yeah, and the releng-backend01 and what it's used for. 15:12:48 also, there's no factory2 team anymore 15:13:06 yup that would be covered in the deployment section 15:13:34 anyway that is all I had for this topic 15:13:41 and releng01 has a cron for eln composes 15:14:52 so lets look at the tickets 15:15:02 I think we can move with this one 15:15:03 .releng 10794 15:15:04 jednorozec: Issue #10794: Enable rpm signing for F-37 - releng - Pagure.io - https://pagure.io/releng/issue/10794 15:15:19 now we have info about the key that is needed 15:15:48 ah yeah. +1 15:16:00 +1 15:16:57 where was that docs pr again? 15:17:11 https://pagure.io/releng/pull-request/10825#request_diff 15:18:04 ok, I was looking on docs.pagure.org. ;) 15:18:12 huh 15:18:15 its is not there 15:18:18 hmm 15:18:22 Let me update it 15:18:23 yeah, thats manually updated. 15:18:32 docs.pagure.org is 404 15:18:35 you have to build it and check it in or something. 15:18:47 https://docs.pagure.org/releng/ 15:18:54 ah 15:19:20 we really should convert these over to docs.fedoraproject.org with/next to the infra ones. 15:20:03 we should, but the docs.fedoraproject.org is soo ugly 15:20:05 anyhow, we need to make a fedora-37-ima and a fedora-38-ima keys (since we want to stay 1 ahead). I can do that, or you can jednorozec 15:20:43 nirik, I probably cannot 15:21:18 or can I? 15:21:21 oh? I would think you could... 15:22:42 just make sure to add me for access and admin... to avoid the lottery factory 15:22:45 factor 15:24:05 ok, I will try to create them 15:24:22 cool. :) 15:25:46 pushing the new docs meanwhile 15:29:05 any more tickets? 15:30:11 .releng 10798 15:30:12 jednorozec: Issue #10798: impact of https://fedoraproject.org/wiki/Changes/JdkInTreeLibsAndStdclibStatic to rcm - releng - Pagure.io - https://pagure.io/releng/issue/10798 15:30:25 did any of you followed the discussions on devel thread? 15:30:31 will this be real? 15:30:39 I stopped afte rfew days 15:30:44 * bittin did not 15:31:09 * nirik did 15:31:35 this part was approved... 15:31:46 but if doesn't affect releng any that I can think of. 15:31:53 yeah 15:31:57 so we can close it 15:31:59 :) 15:32:16 the other parts of this seem unlikely to pass fesco... 15:32:38 one part of which has a lot of impact on releng. The 'build once and tag into other releases' thing. 15:32:55 yeah 15:33:03 there was a heated discussion about that 15:33:07 on devel list 15:33:41 also noreleng change. 15:33:44 .releng 10836 15:33:45 jednorozec: Issue #10836: F37 Change: Gettext Runtime Subpackage - releng - Pagure.io - https://pagure.io/releng/issue/10836 15:35:08 yeah, no releng changes here I can see either 15:35:16 sounds good 15:35:31 another one 15:35:35 .releng 10720 15:35:36 jednorozec: Issue #10720: F37: Self contained change: Support FIDO Device Onboarding - releng - Pagure.io - https://pagure.io/releng/issue/10720 15:36:12 yeah, same here, no releng impact I don't think. 15:37:44 yeah i tried to tag them no-releng when I was sure they have 0 impact on us 15:37:51 but want to have it checked 15:38:37 another java one 15:38:39 .releng 10686 15:38:40 jednorozec: Issue #10686: Drop i686 builds of jdk8,11,17 and latest (18) rpms from f37 onwards - releng - Pagure.io - https://pagure.io/releng/issue/10686 15:39:59 well, I would say no releng impact, but not sure about their question... 15:41:04 they dont want to include excludearch in java packages 15:41:29 and want us to stop building x86 java packages 15:42:18 yeah, but I don't get the noarch part. 15:42:32 they don't want to build a noarch subpackage on i686. 15:42:39 right 15:42:41 but... if it's noarch, it shouldn't matter? 15:43:00 well 15:43:22 also, koji builds noarch packages for every arch... and checks to make sure they are the same... 15:43:30 yeah 15:44:30 so, I guess we should ask them to clarify... 15:44:56 https://kojipkgs.fedoraproject.org//packages/java-1.8.0-openjdk/1.8.0.332.b09/1.fc36/data/logs/i686/noarch_rpmdiff.json 15:44:58 for example 15:45:43 or perhaps they mean _other_ java packages? 15:46:57 I think they mean alll java packages but jdk 15:46:59 yeah, re-reading that seems like it. They exclude i686 from the openjdk, but want to know how to make all noarch java packages not try and build on i686 where there's no openjdk 15:47:18 that is what i read 15:47:29 I'm afraid we don't have any way to do that off hand. 15:50:10 ok moving on 15:50:13 this one on the top 15:50:16 .releng 10832 15:50:17 jednorozec: Issue #10832: I can't push FEDORA-2022-0feca8b2c9 to stable - releng - Pagure.io - https://pagure.io/releng/issue/10832 15:50:36 I tried to disable all checks for that update, but nothing 15:50:39 yeah, I poked at it, but failed to get it to do anything. ;( I was hoping that mattia could look 15:50:57 did you want to update 10686? or want me to? 15:51:17 jednorozec did it 15:51:50 .nextmeetings 15:51:51 x3mboy: One moment, please... Looking up the channel list. 15:51:54 x3mboy: b'In #fedora-i3 is Fedora i3wm SIG (starting in 8 minutes)' 15:51:57 x3mboy: b'In #fedora-meeting is FESCo Meeting (starting in an hour)' 15:52:00 x3mboy: b'In #fedora-admin is Fedora Infra Ops Daily Standup (starting in 2 hours)' 15:52:03 x3mboy: b'In #fedora-meeting-1 is FPgM office hours (starting in 21 hours)' 15:52:06 x3mboy: b'In #fedora-meeting-1 is Prioritized bugs and issues (starting in a day)' 15:52:13 x3mboy, that is little rude to interupt our meeting 15:52:28 Oh sorry, I didn't notice you were at meeting 15:52:33 I'm so so sorry 15:52:39 no biggie 15:52:39 My very sincere apologize 15:52:46 its ok :) 15:53:03 this chanell is probably used only for 2 meetings 15:53:38 anyway the update 15:54:02 it seems to me that it got cought in some random state because it had -1 karma before the update 15:54:12 and we updated bodhi to v 6.0 15:54:17 and this hapenned 15:54:44 might be. 15:55:07 I think it might be because someone pushed it stable... and in rawhide you normally have bodhi decide when things should go stable 15:56:30 hm 15:57:04 but not sure. I looked at logs... 15:57:20 and it shows the edits, but nothing else. it just ignores it when it's processing otherwise. 16:00:24 they are asking if rebuild in different side-tag would help 16:00:45 it probably wont because there is already update for that NVR 16:01:13 yeah it would work if they bumped release again. 16:02:36 wrote a comment there 16:02:45 and we are at the end 16:02:51 oh yeah, out of time. ;) 16:02:58 #endmeeting