15:00:16 #startmeeting RELENG (2022-03-22) 15:00:16 Meeting started Tue Mar 22 15:00:16 2022 UTC. 15:00:16 This meeting is logged and archived in a public location. 15:00:16 The chair is jednorozec. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 15:00:16 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:16 The meeting name has been set to 'releng_(2022-03-22)' 15:00:16 #meetingname releng 15:00:16 #chair nirik sharkcz pbrobinson pingou phsmoura mboddu dustymabe ksinny jednorozec 15:00:16 #topic init process 15:00:16 The meeting name has been set to 'releng' 15:00:16 Current chairs: dustymabe jednorozec ksinny mboddu nirik pbrobinson phsmoura pingou sharkcz 15:00:27 morning 15:00:42 morning 15:00:45 I am here, but as usual for the last few weeks, I am in a meeting :) 15:01:04 mboddu, the meeting man. ;) 15:01:12 :) 15:01:17 master of meeting 15:01:24 s 15:02:12 * dustymabe lurking 15:02:31 this 15:02:33 .releng 10708 15:02:34 jednorozec: Issue #10708: More F36 Mass Rebuild / Mass Branching / Mass Tagging Problems + Race Conditions - releng - Pagure.io - https://pagure.io/releng/issue/10708 15:02:55 we should probably disable builds during branching? 15:03:13 or are there other options how to prevent it? 15:03:43 I think that would be best, but I am not sure how best to do that... 15:05:13 we want to disable just rawhide, so other releases can be build 15:05:19 perhaps there's a way to tell pagure to reject pushes except for admins. 15:05:41 hm we can make new hook 15:06:04 that will check if we are rebuilding and reject with some nice msg, like get back later 15:06:19 we still need to let us build fedora-release/repos/etc/ 15:06:57 preventing push to rawhide sounds good 15:07:19 but 15:07:28 hmm for 5 days? 15:07:31 well, and newly branched until we are ready. 15:07:40 right 15:07:50 5 days? I would think just during... less than 1 day? 15:08:50 right 15:10:33 anyhow, just need to implement and test that before next mass branching. :) 15:10:40 and these things we should cleanup after freeze 15:10:53 yeah 15:11:45 Sorry I am preocupied by our new friends from Ukraine, they are getting used to the place... 15:12:34 :( 15:13:01 oh you mean you are hosting some folks from there? good for you! 15:13:10 yeah family 15:13:16 jednorozec++ 15:13:16 dustymabe: Karma for humaton changed to 6 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 15:13:19 mom and 2 kids 15:13:31 the kids want to play all the time :D 15:13:34 thank you so much jednorozec 15:14:37 jednorozec++ 15:15:02 jednorozec++ 15:15:02 phsmoura: Karma for humaton changed to 7 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 15:15:47 so, I got them distracted with toys 15:16:42 .releng 10706 15:16:43 jednorozec: Issue #10706: Unretire python-certifi on epel7 - releng - Pagure.io - https://pagure.io/releng/issue/10706 15:17:14 toddler retired package in pdc and blocked it in koji 15:17:20 on it own 15:19:57 hm 15:20:11 jednorozec++ 15:21:56 Sorry, a bit late, but for previous ticket, blocking building is not the right thing to do, mass branching will be done in a day which is acceptable, but mass rebuild can take anywhere between 3 to 5 days which is not acceptable 15:23:05 hum? 15:23:36 mass rebuild isn't involved here is it? 15:23:43 mboddu, but we create all the commits and submit builds in one days 15:23:58 nirik, it is. 15:24:03 where? 15:24:30 * nirik re-reads the ticket 15:25:10 nirik: https://pagure.io/releng/issue/10687 15:25:26 specifically https://pagure.io/releng/issue/10687#comment-784704 15:25:28 if we are using the script to merge the rebuild tag in, how could it be failing ? 15:26:38 or are we? how is the rebuild tag merged? 15:28:33 this script is used 15:28:41 nirik: We use mass-tag.py 15:28:43 jednorozec++ 15:28:43 so, we should use mass-tag.py and it should not tag things in where there is a newer build already tagged. 15:28:52 https://pagure.io/releng/blob/main/f/scripts/mass-tag.py 15:28:54 huh then I am unclear how it's not working 15:29:18 ah... except it was going into f36-pending? 15:29:31 Yeah, that is true but ... ^ 15:29:40 hm 15:29:52 so, thats why 15:30:54 We can skip tagging into -pending tag, but it is adviced to use it even if autosigning is enabled, but autosigning has been stable. So, skip it and tag into f36 tag? 15:31:03 it was adviced* 15:31:59 yeah, I think we should tag into fN directly... that will fix this issue and also save a ton of time 15:32:04 I have the docs opened on another workspace so i can update it right now and drop the -pending 15:32:34 the problem is that autosign is paranoid and even when it sees something is signed it tells koji to write out signed rpms again... 15:32:50 which causes koji to do a bunch of i/o... so it takes a long time 15:34:41 if something isn't signed right in the end, pungi will let us know and we can fix it, but I think we can make sure the entire tag is signed before merging. 15:34:59 ack 15:35:22 Sorry, side tracked the meeting 15:38:26 moving on then 15:38:53 .releng 10686 15:38:54 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 15:39:02 but the comment 15:39:03 https://pagure.io/releng/issue/10686#comment-786124 15:39:51 I am not sure what he wants, to remove i686 without excludearch in the specs? 15:39:58 I'm not sure what they mean... a noarch package building on a i686 builder? 15:40:07 nirik, probably 15:41:16 well, for noarch subpackages koji builds them on all arches and makes sure they are the same... for purely noarch packages it just builds them on whatever arch, but you should be able to Exclude 15:42:17 right 15:43:18 let me ask them what is they actually want. 15:43:50 because I think they should excludearch in the spec 15:44:26 yeah, I am not sure what they want/are asking for. 15:44:37 * jednorozec writing commnet 15:44:52 just reading back through the chat 15:44:56 jednorozec++ 15:44:56 jrichardson: Karma for humaton changed to 8 (for the current release cycle): https://badges.fedoraproject.org/tags/cookie/any 15:45:00 fair play man 15:48:53 the right thing to do. 15:49:00 anyhow 15:49:12 what do we have in the tracker next 15:50:37 huh 15:50:41 I missed this 15:50:43 .releng 10622 15:50:44 jednorozec: Issue #10622: armhfp disk images are often corrupted during transfer from builder to hub - releng - Pagure.io - https://pagure.io/releng/issue/10622 15:51:56 yeah, I think this is no longer happening... but we should check with pwhalen about it. 15:52:37 Havent seen it in a while, I think we can close and I'll reopen if needed 15:52:44 ack 15:52:45 sounds good! thanks 15:54:39 .releng 10369 15:54:40 jednorozec: Issue #10369: Q: Why do we install epel-release and fedpkg-minimal into every buildroot in Koji? - releng - Pagure.io - https://pagure.io/releng/issue/10369 15:54:52 huh I dont recall testing it 15:55:33 I've not tested it yet... 15:56:53 ok 15:56:58 4 mins left 15:57:04 #Open floor 15:57:15 #topic Open floor 15:58:12 oh, I had one thing. 15:58:25 I submitted a PR that checks in announcement emails... 15:58:39 https://pagure.io/releng/pull-request/10704 15:59:33 nirik++ 15:59:36 thanks for this! 15:59:51 so we can improve them from there... just remember to use them! :) 16:00:40 I definately will use them. 16:00:59 copy/paste is one of my weak skills 16:01:06 always some errors in it 16:01:10 oh, we should also add a url at the bottom of them pointing to where they are for PR's. 16:01:21 good idea 16:03:04 and the time is up 16:03:08 #endmeeting