15:00:52 #startmeeting RELENG (2021-09-21) 15:00:52 Meeting started Tue Sep 21 15:00:52 2021 UTC. 15:00:52 This meeting is logged and archived in a public location. 15:00:52 The chair is jednorozec. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 15:00:52 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:52 The meeting name has been set to 'releng_(2021-09-21)' 15:00:52 #meetingname releng 15:00:52 The meeting name has been set to 'releng' 15:00:52 #chair nirik sharkcz pbrobinson pingou mboddu dustymabe ksinny jednorozec 15:00:52 Current chairs: dustymabe jednorozec ksinny mboddu nirik pbrobinson pingou sharkcz 15:00:58 #topic init process 15:01:12 * nirik is sort of here, but I am dealing with proxy outages... so very distracted. 15:01:22 * sharkcz is here 15:06:34 Hi guys, quick question. I made a PR but the code still needed to be fixed, so fixed and now I'd like to know if I cancel PR without merging it to make another one and if I overwrite previous commit 15:07:21 You can always add new commit to the existing branch. Or force push. Both will update current PR. 15:07:59 ok, thanks! :) 15:08:48 #info there si no Beta RC request yet. Currently there are 2 blocker bugs in new state. 15:09:28 Hello 15:09:45 welcome mboddu 15:11:13 I just checked last F35 compose and its doomed 15:11:20 :/ 15:11:28 really hope we get a rc today. 15:12:02 * mboddu hopes the same 15:12:21 That requires to have all blockers in it right? 15:12:47 jednorozec: Not exactly, but better it has all the blocker fixes 15:13:26 todays branched failed due to the proxy issues I think 15:13:51 DEBUG util.py:444: Cannot download toplink/packages/dbus/1.12.20/4.fc35/aarch64/dbus-1.12.20-4.fc35.aarch64.rpm: All mirrors were tried 15:13:53 yeah 15:14:50 ok that is about it for f35 15:15:04 I have couple of tickets do discuss 15:15:14 .releng 10305 15:15:31 jednorozec: Issue #10305: F35 Mass Branching Fallout: Koji builds stuck in weird states - releng - Pagure.io - https://pagure.io/releng/issue/10305 15:18:36 Close it and ask them to reopen if they have any more issues 15:20:19 so, I think we can unstick that one. 15:21:25 oh, I see... it went only to f36 15:21:31 yup 15:21:41 so, we could fix it by just tagging it into f35, but we are in freeze. so, wait until after? 15:21:53 Sorry, I opened a wrong ticket 15:22:00 so after the freeze jsut tag it in f35? 15:22:34 yeah... that would be one way 15:22:48 jednorozec: Yeah, or better ask them to unpush the update and bump and build another one for f35 15:23:21 that might be better as then we don't forget 15:23:53 but it does bring up another thought: should we disable builds during mass branching? 15:24:55 hmm 15:24:58 nirik: Since its about a day of work, I dont think its a good idea to disable builds for almost an entire day 15:25:28 yeah. 15:25:39 But how do we prevent this? 15:27:17 hmm, can we disable build per package? 15:27:41 Disabling it jsut before the script create branches and enabling it back afterwards 15:28:47 well, we could monkey with pdc... ie, have it block for a short time 15:29:19 but that will disable pushes 15:29:51 ell, yeah, but they have to push a commit(s) before building to update the nvr right? 15:30:16 and we have to push the new branch 15:30:25 but yeah, we can disable just rawhide 15:30:27 hmmm 15:30:58 I will think about it a little bit more, and prepare something for next week. 15:30:59 perhaps we should give it more thought. 15:31:01 yeah 15:32:03 .releng 10294 15:32:04 jednorozec: Issue #10294: Please send openh264-2.1.1-3.fc35 and openh264-2.1.1-3.fc36 to Cisco - releng - Pagure.io - https://pagure.io/releng/issue/10294 15:32:16 how is this done? 15:33:08 https://docs.pagure.org/releng/sop_generating_openh264_composes.html 15:33:30 thats out of date tho. needs fixing 15:33:51 well, the signing of repodata part should be dropped. 15:35:25 "We need to send this tar file to Cisco" 15:35:28 how? 15:35:47 lovely high tech email I think it is. :) mboddu ^ 15:36:24 mboddu, if you could fwd me last message you send them. I will take care about it tomorrow 15:39:26 it's important to rememver to copy it to mm-backend01 (so mirrormanager can update) 15:40:12 jednorozec: Yeah, I already started working on it, but I will CC you in the email 15:41:11 mboddu, ack. If there are any steps missing from the SOP would you please update it? SO I can take care about it next release... 15:41:20 jednorozec: Sure 15:42:21 So I have one more thing. Regarding changes tracker tickets. When do we close them? After reelase or when the change is done or does not require any more releng work? 15:42:38 this one for example is just sitting there for no reason 15:42:44 .releng 9952 15:42:45 jednorozec: Issue #9952: F35 Change: Fedora Kinoite - releng - Pagure.io - https://pagure.io/releng/issue/9952 15:43:07 IMHO we should close them when they no longer need any releng work. 15:44:00 I am of the same opinion 15:44:34 so I will go through the changes and close finished from releng point of view 15:45:25 I think mboddu was closing them after the GA? we could do that too... 15:46:48 But some like the kinoite one are obviously done. And our tracker looks nicer :) 15:47:37 I'm fine either way if mboddu is. ;) 15:48:19 Yeah, I am fine either way, its just that I generally go through the changes once after GA and verify whats done and whats not done and what moved to next release 15:48:42 And update the tickets as needed 15:51:57 Ok I dont have anything more for today 15:52:47 #topic Open Floor 15:54:59 * nirik has nothing off hand 15:55:03 oh wait 15:55:33 So, we got some more kvm lpar resources a while back for s390x. I was going to reinstall/rebalance them after freeze... 15:56:00 but it turns out ELN (and centos9 stream) have moved to a z14 baseline and we are on z13 15:56:32 So, there's a request to move us to the new z15 mainframe. 15:56:45 not sure when, but might be in the next few months. 15:56:49 Just a heads up on it. 15:58:04 thanks for the update 15:59:33 Thanks for the update nirik 15:59:49 This should have been done in Fedora first, oh well... 16:00:28 time is up 16:00:34 #endmeeting