18:00:01 #startmeeting Fedora Infrastructure Ops Daily Standup Meeting 18:00:01 Meeting started Wed Jul 15 18:00:01 2020 UTC. 18:00:01 This meeting is logged and archived in a public location. 18:00:01 The chair is nirik. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:01 Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:00:01 The meeting name has been set to 'fedora_infrastructure_ops_daily_standup_meeting' 18:00:01 #chair pingou mboddu nirik smooge 18:00:01 #meetingname fedora_infrastructure_ops_daily_standup_meeting 18:00:01 Current chairs: mboddu nirik pingou smooge 18:00:01 The meeting name has been set to 'fedora_infrastructure_ops_daily_standup_meeting' 18:00:01 #info meeting is 30 minutes MAX. At the end of 30, its stops 18:00:01 #info agenda is at https://board.net/p/fedora-infra-daily 18:00:02 #topic Tickets needing review 18:00:03 #info https://pagure.io/fedora-infrastructure/issues?status=Open&priority=1 18:00:21 * mboddu is sorta here 18:00:35 hello I have mozilla open 18:00:56 * cverna waives 18:01:06 Moasiac? 18:01:39 so, only one in needs-review is a private one... I can answer that one. 18:01:49 nirik, eMosaic 18:02:17 I had answers for that one.. none of them printable 18:02:30 mboddu, anything for releng? 18:02:49 Yeah, just one 18:02:58 .releng 9602 18:02:59 mboddu: Issue #9602: Create detached signatures for the Ignition 2.4.0 release - releng - Pagure.io - https://pagure.io/releng/issue/9602 18:03:40 * pingou around 18:03:41 you hit a sigul bug on that one? might file it upstream and see if puiterwijk can fix it? or you know the fix already? 18:04:22 nirik: yeah, https://pagure.io/releng/issue/9602#comment-665588 is the latest update on it 18:05:04 nirik: I know I had to build py3 compatible sigul, which I did but then hit with something else :( 18:05:14 Which I have no idea how to fix 18:06:01 Not sure if its med trouble or high trouble 18:06:43 if you don't it's high? 18:07:31 I am going with: if you aren't Patrick, its high 18:07:43 Okay 18:07:50 Thats all I got 18:08:09 cool. any other tickets to discuss or blockers/other items? 18:08:20 None from me 18:08:27 I have 18:08:29 someone pinged about a fedscm-admin PR earlier in the day 18:08:46 * pingou not sure who is monitoring that project, if it's us or someone else 18:08:48 I have one too but can wait :) 18:09:07 I got just asked when would be a good time in the coming month for some network updates/fixes/reboots in IAD2. 18:09:29 It should be mostly invisible to us but could always have difficulties 18:09:30 related to that, I wonder if we should try and come up with SLE's for PR's... :) 18:10:05 smooge: I think most anytime, EXCEPT not during the mass rebuild or the beta/final freezes. 18:10:17 They need to schedule travel and Covid items 18:10:32 so the mass rebuild is next week? 18:10:57 the 21st iirc 18:11:12 Mass Rebuild: RPMs first, then modules Begins: Wed 2020-07-22 Ends: Tue 2020-08-11 18:11:17 22nd actually 18:11:24 pingou: I maintain fedscm-admin 18:11:39 mboddu: roger 18:11:46 * pingou should add it to the list 18:11:53 pingou: But what PR? 18:12:08 I didn't see a PR request in https://pagure.io/fedscm-admin/pull-requests 18:12:25 pemensik | hi. Is there any ETA, when might be https://pagure.io/fedscm-admin/issue/36 merged? 18:12:27 pemensik | is there anything blocking its review? 18:12:29 pemensik | what is blocking my requests now? 18:12:31 pemensik | hi. Is there any ETA, when might be https://pagure.io/fedscm-admin/issue/36 merged? 18:12:33 pemensik | is there anything blocking its review? 18:12:35 pemensik | what is blocking my requests now? 18:12:37 oups sorry for the double paste 18:12:48 pingou, still here :) 18:13:21 Thanks pingou 18:13:23 nirik, so I was thinking that after the 28th? would be a good time for the trip or would it need to wait til after the 11th? 18:13:45 the reason I ask about SLE's for pr's... is that it seems we currently have a workflow of: file pr, wait, ping people, get it merged. I would like to change the 'ping people' part to 'you will get a answer in 2 days' or something so we review them as they come and don't get pinged to merge. Also, I am thinking about being more agressive at closing them... 18:14:12 pemensik: I will build fedscm-admin with the fix and I have to distribute it to limb which will fix your issue, sorry about the delay 18:14:17 pingou: ^ FYI 18:14:37 mboddu: thanks 18:14:52 (I guess limb doesn't use a git clone then :() 18:15:02 nirik: this does sound good 18:15:04 mboddu: / pingou / pemensik: note that jednorozec answered that eariler 18:15:22 [07:43:04] pemensik, so its not failing with exceptions but failing. 18:15:22 [07:44:10] jednorozec, do you know what is the reason of the failure? 18:15:22 [07:46:42] pemensik, I am not sure. But the tool is most probably not using name-spaces and its able to create only rpms. 18:15:22 [07:49:14] * huzaifas_ (~huzaifas@2409:4042:2305:5392:ddb1:61f5:c508:29b5) has joined 18:15:23 [07:49:28] pemensik, I can look into it tomorrow and implement missing pieces into fedscm-admin 18:15:47 smooge: yeah, 28th - 11th would be ok. I think 18:15:51 oh indeed, I had missed jednorozec's reply, sorry jednorozec 18:15:51 Ahhh okay 18:16:35 smooge: we should also ourselves do a mass update/reboot cycle too sometime 18:17:07 nirik: this goes into the widder question of: how do we cope with ~42 apps 18:17:09 yeah.. I realized our last one for a lot of boxes was 180 days ago 18:17:21 esp when we have only 2 devs "free" 18:18:17 pingou: yeah. Although some are quicker/easier than others... like the releng ones are often just tweaking something... 18:19:31 but also... https://pagure.io/releng/pull-requests we have a pr from maxamillion from 4 years ago! 18:19:44 nirik: I think we should be able to covers PRs to releng and ansible in a timely fashion 18:19:55 so I think we shouldn't be afraid of saying 'no, make a new pr when you sort this out' 18:20:18 I'm more worried about bodhi, fedscm-admin, mdapi, pagure, robosignatory... 18:20:22 yeah, and fedora-comps/fedora-kickstarts/fedora-repos/fedora-release are all sometimes time critical 18:20:23 +1 18:20:28 yeah, agreed. 18:20:29 nirik, it would help having some tag to mark PR that cannot be satistified, because they are blocked by something else 18:20:29 PR can be re-opened in pagure 18:20:31 +1 to close old PRs 18:21:13 I think there was talk about some kind of blocked tag, but I don't recall exactly what 18:21:15 +1 to close old PR's more than 6 months old 18:21:23 it would be more polite to ask people to rebase their PR than closing it without explanation 18:21:44 Sure, we can ask them to reopen if its still needed with a rebase 18:21:54 pemensik: after 4 years I think it fine to close the PR saying that it is stale 18:21:56 is flr something that ever landed properly? 18:22:40 cverna, depends on what was the reason it got stale. I think each PR should be marked somehow, to show what it is waiting for 18:23:14 and should be closed only if original reported was unresponsive :) 18:23:31 * pingou notes that PR #62 has been pinged a year ago 18:24:06 well, often it's not just rebase... 18:24:16 especially anymore when you can rebase for them 18:24:43 that doesn't work for old PRs though, the checkbox needs to be checked and it's not for old PRs 18:24:46 it's discussion... or deciding what to do... which shouldn't be in a pr. Thats what lists and meetings are for. ;) 18:24:52 Just wanted to discuss the OSBS ticket quickly because I am a newbie with httpd mod_rewrite and I am not sure what to do :P 18:24:58 if you can rebase, rebase them without closing them. 18:25:06 cverna: mod_rewrite is a horror. ;) I hate it. 18:25:25 nirik: I feel a bit better :P 18:25:54 pemensik: the rebase is only part of the problem of stale PRs 18:26:02 nirik, why it should not be in PR which it is related to? 18:26:07 I was trying to understand what we currently have in the config but without much success 18:26:19 for context https://pagure.io/fedora-infrastructure/issue/9116#comment-664421 18:26:37 pemensik: PR and tickets aren't great discussion tools 18:27:05 pemensik: IMHO, tickets and pr's are bad places to decide things. You exclude all the rest of the community that might be interested in it, there's no threading or quoting, etc 18:27:12 ok, if it gets too talkative, it would be nice to include link to proper thread. 18:27:57 like you know. This is a way to tell people do not bother us. https://pagure.io/releng/fedora-scm-requests/issue/27135 18:28:21 cverna: so, thats if the via: header doesn't have cdn77 or cloudfront, or the server-name (what you are requesting from) isn't that no-cdn thing and it's not any of those request methods, then use the cdn, otherwise don't 18:28:37 Just close, without any explanation, reason, link. Please avoid it 18:29:02 pemensik: that queue is processed by a tool. A human never will see that. If you run into issues with that, you need to file a releng ticket. 18:29:12 pemensik: that repo is processed by a CLI tool, it has a lot of notifications so you should not expect discussions there 18:29:35 nirik, May I tell you I feel like a human and saw it? 18:29:50 pemensik: thats not what I meant... 18:30:08 nirik: ok so I think I we could use the `REMOTE_HOST` variable and say something like if osbs in REMOTE_HOST don't use the cdn 18:30:34 I agree this process is not ideal. BUt we can't magically fix it... I'm sorry you are hitting issues, we are trying to fix it for you. 18:30:41 nirik: I can open a PR and see how that works :P 18:30:43 cverna: yep. 18:30:58 ack 18:32:24 ok, if nothing else will close out in a min... 18:32:30 thanks all 18:32:38 thanks everyone. 18:32:41 #endmeeting