16:00:12 #startmeeting RELENG (2023-01-31) 16:00:12 Meeting started Tue Jan 31 16:00:12 2023 UTC. 16:00:12 This meeting is logged and archived in a public location. 16:00:12 The chair is jednorozec. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 16:00:12 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:12 The meeting name has been set to 'releng_(2023-01-31)' 16:00:12 #meetingname releng 16:00:12 The meeting name has been set to 'releng' 16:00:12 #chair nirik sharkcz pbrobinson phsmoura dustymabe jednorozec 16:00:12 #topic init process 16:00:12 Current chairs: dustymabe jednorozec nirik pbrobinson phsmoura sharkcz 16:01:06 Morning 16:01:22 .hi 16:01:23 dherrera: dherrera 'Diego Herrera' 16:01:47 .hi 16:01:48 jonathanspw: jonathanspw 'Jonathan Wright' 16:01:53 just here to be a fly on the wall 16:02:52 .hi2 jnsamyak 16:03:40 hello and welcome 16:03:55 #topic current release cycle 16:04:20 welcome! 16:04:35 Mass rebuild will happen in one week on this day 16:04:41 uf 16:04:44 wrong line 16:04:58 Branching of Fedora 38 will happen next week 16:06:21 we have 2 relevant tickets for the branching event 16:06:23 .releng #startmeeting RELENG (2023-01-31) 16:06:23 jednorozec: Error: '#startmeeting RELENG (2023-01-31)' is not a valid integer. 16:06:23 #meetingname releng 16:06:23 The meeting name has been set to 'releng' 16:06:23 #chair nirik sharkcz pbrobinson phsmoura dustymabe jednorozec 16:06:23 #topic init process 16:06:23 Current chairs: dustymabe jednorozec nirik pbrobinson phsmoura sharkcz 16:06:25 Yeah. 16:06:25 uff 16:06:47 .releng 11188 16:06:48 jednorozec: Issue #11188: Update release process - releng - Pagure.io - https://pagure.io/releng/issue/11188 16:07:13 copy pasta. ;) 16:07:25 lots of pasta :D 16:07:57 anyhow, yes, we should use the 'frozen' state. :) 16:08:37 linguini 16:08:41 and 16:08:43 .releng 11192 16:08:44 jednorozec: Issue #11192: Ansible: vars/all/FedoraPreviousPrevious.yaml not being properly updated - releng - Pagure.io - https://pagure.io/releng/issue/11192 16:08:50 I think we can test this out a bit on the day. 16:09:19 ie, set frozen, then manually do a push and confirm that it doesn't do the frozen set 16:10:09 oh, I wonder. What happens if you freeze rawhide? does it not make auto updates? or does it do nothing? but I guess thats a sidenote 16:10:27 and yes, we need to make sure and update the variables. 16:11:30 nirik[m], I will runt the branching sooner that usually 16:11:57 around 7am UTC 16:12:38 ok. Fine with me. 16:12:53 it takes a while to run all the playbooks and we can either deal with unforeseen consequences during this meeting 16:12:55 I might still be awake as thats my evening. ;) 16:13:36 I will be on my way home from fosdem with stop in Brno to branch fedora :D 16:14:05 ha. hardcore. â˜ēī¸ 16:14:16 Anyway that is it for branching, all changes will be in PRs day before the event 16:14:43 there's a signing task on the docker for... today. 16:15:21 Yeah 16:15:39 I guess I will try and whip up something. I asked patrick and he said it would make sense for sigul to have a way to do this, but it doesn't currently. ☚ī¸ 16:16:16 so it's all brute force... use sign_unsigned to get a list of all packages in f38 tag, then feed them in a loop to robosignatory... after adding it to robosignatory config. 16:16:27 but I will try and get something working 16:18:30 I can help 16:18:53 * jednorozec was around into late nights past few nights 16:19:26 oh ad signing 16:19:51 .releng 10713 16:19:52 jednorozec: Issue #10713: Add ELN to the set of packages to be re-signed at each Branch - releng - Pagure.io - https://pagure.io/releng/issue/10713 16:20:07 I have added note about signing but not actual SOP on how to do that 16:20:52 ideally the SOP will become 'make this robosignatory change and then run this sigul command' 16:21:00 but it's not currently 16:21:13 right 16:21:36 so we can probably move to tickets 16:22:52 .releng 11240 16:22:54 jednorozec: Issue #11240: Rawhide branch not properly retired for libqb-epel and kronosnet-epel - releng - Pagure.io - https://pagure.io/releng/issue/11240 16:23:17 this happened multiple times, sometimes maintainers forget to retire rawhide 16:23:41 sometimes the PDC flaps and branches stay live 16:23:58 well, or retire packages is broken. ;) 16:24:05 or 16:24:27 could we detect these? ie, look at all dist-git with dead.package and confirm all of them are retired? 16:24:31 depends on the current constellation 16:24:51 yep 16:25:01 should be scriptable? 16:25:04 and there was a tracking ticket for that and we got them all I think 16:25:16 but there are new requests coming in 16:25:39 well, we got all that requested... but I worry there are more out there. 16:25:41 I have created this https://pagure.io/fedora-infra/toddlers/issue/151 16:26:10 all future requests should just disable the rawhide branch for such packages 16:26:14 yeah, I suppose that could work... 16:26:31 to your question about if we have them all 16:26:34 but there's probibly still packages that people marked dead.package but didn't get retired. 16:26:41 yes 16:26:51 I was specificaly talking about -epel packages 16:27:22 I don't think I want to commit to making such a script, but perhaps we should file a ticket and one of the newer folks could do it as a good way to learn things? 16:27:53 yup 16:28:08 want me to file the ticket? or you got it? 16:28:18 I can do it 16:28:28 and shall we close 11240 ? 16:28:45 yes 16:28:58 cool. 16:29:26 .releng 11254 16:29:28 jednorozec: Issue #11254: Allow setting of macros for side tag - releng - Pagure.io - https://pagure.io/releng/issue/11254 16:29:51 so, this came out of a devel list discussion. 16:30:02 I think we should get FESCo to approve before we just do it. ;) 16:30:20 I was really hoping we could only grant bootstrap 16:30:29 hmm 16:30:31 all macros is... a lot of power. 16:30:42 do we really want to allow any macro? 16:30:57 --rpm-macro dist=i-have-my-own-dist-tag 16:31:20 right, it's a ton of power... 16:31:55 it is only packagers tho... and we already trust them a fair bit. 16:32:14 well, I mean anyone could make a side tag, but only packagers can merge them back 16:33:26 are there some limitations to the param 16:33:27 ? 16:33:43 or can I just feed it with 200 lines of code? 16:34:02 I don't know. there was an upstream ticket, but I haven't read my mail yet, so I have no idea if they explained it more. 16:34:29 I think you can override/create any macro(s) you want. 16:35:21 ok should we tag the request? 16:35:21 ideally we could limit it to bootstrap and a allowlist of others. 16:35:35 Sure, it needs more investigation. 16:35:38 IMHO 16:35:51 ack 16:36:39 and possibly fesco approval. If we can limit it, I would think it would be fine... if not I would prefer FESCo know/ack it if we do it. 16:37:44 * jednorozec writing comment 16:43:37 I will close the mass rebuild f38 changes 16:43:48 probably also the tracker ticket? 16:44:56 yep. +1 16:45:04 That should clear out some tickets. 16:45:11 How is the docs stuff looking? any progress? 16:47:50 yeah I had a bit of git hell the other week 16:47:56 but got it sorted out 16:48:35 https://pagure.io/fork/humaton/infra-docs-fpo/blob/release_guide/f/modules/release_guide/pages 16:48:46 I should opne PR the the docs page 16:48:57 ice 16:48:58 nice even 16:49:46 but the main thing is not finished yet 16:49:48 https://pagure.io/fork/humaton/infra-docs-fpo/blob/release_guide/f/modules/release_guide/pages/release_process.adoc 16:50:06 I want this to be the only user facing page compiled from the SOPs 16:50:33 so its one page chronologically ordered 16:51:01 nice. 16:51:16 if will be very good to get that in place and have 1 thing to update. 16:52:29 and use GPT3 to rewrite it into ansible playbook :) 16:53:08 * nirik[m] shudders 16:54:31 #topic Open Floor 16:55:32 I'm not sure I had much... 16:56:05 Have you been showing the new person the ropes (I forgot their nick/handle) 16:56:15 yup 16:56:27 jnsamyak it is 16:56:37 he took over stalled epel packages 16:56:43 ah yea, right. 👋 16:56:52 and helped with changes for the rebuild 16:57:18 now we will go through the branching, also reason why I would like to start earlier 16:57:29 sounds good. 16:58:27 it is really refreshing to open inbox to closed tickets notifications :) 16:59:18 s/to/and see 16:59:39 😀 17:00:25 * nirik[m] has to go try and feed cats before fesco meeting. 17:00:29 time is up, have a nice rest of the day 17:00:33 #endmeeting