19:00:41 #startmeeting Fedora Infrastructure Ops Daily Standup Meeting 19:00:41 Meeting started Wed Nov 16 19:00:41 2022 UTC. 19:00:41 This meeting is logged and archived in a public location. 19:00:41 The chair is phsmoura. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 19:00:41 Useful Commands: #action #agreed #halp #info #idea #link #topic. 19:00:41 The meeting name has been set to 'fedora_infrastructure_ops_daily_standup_meeting' 19:00:42 #chair nirik aheath1992 smooge 19:00:42 Current chairs: aheath1992 nirik phsmoura smooge 19:00:42 #meetingname fedora_infrastructure_ops_daily_standup_meeting 19:00:42 #info meeting is 30 minutes MAX. At the end of 30, its stops 19:00:42 #info agenda is at https://board.net/p/fedora-infra-daily 19:00:42 The meeting name has been set to 'fedora_infrastructure_ops_daily_standup_meeting' 19:01:06 hello. good morning. 19:01:07 here 19:01:16 good evening 19:01:32 hello 19:01:45 #info reminder: speak up if you want to work on a ticket! 19:01:45 #topic Tickets needing review 19:01:58 #info https://pagure.io/fedora-infrastructure/issues?status=Open&priority=1 19:02:10 we dont have infra tickets today :) 19:02:35 we don't? oh they must have been dealt with earlier 19:02:49 yep 19:02:53 #info https://pagure.io/releng/issues?status=Open 19:02:58 I closed the outage ones, they seemed fixed to me. 19:03:01 neither releng tickets 19:03:30 #topic Planning, Upcoming work and Open floor 19:03:31 okie dokie.. 19:03:41 #info freeze is over! 19:03:42 nirik, any idea on how to mitigate the DB issue? 19:04:21 jednorozec: not sure. We may be able to look at throttling connections per ip? or perhaps koji developers have some ideas. 19:04:52 throttling connections seems like quick fix 19:04:57 they were doing rpminfo requests... so that takes a lot of queries. 19:05:18 yeah the selects were pretty complex 19:05:26 and a lot of them... 19:05:54 so basically things like: https://koji.fedoraproject.org/koji/rpminfo?rpmID=32489050 19:06:10 so it has to get all the files in the rpm, all the requires/provides, sizes, etc 19:07:01 anyhow, I am not sure apache can limit requests, we will need to look into it. There used to be a module, but I think it went away 19:08:02 mod_qos perhaps 19:08:34 anyhow, for plans: I have more meetings today, but aside that I want to start merging pr's soon... tomorrow I will be nuking the old openshift 3 clusters 19:10:15 hm, mod_qos-11.70-2.fc37.x86_64.rpm 19:10:16 about the outage tadoy. I tryed to help, but didnt had perms to ssh koji machines. Also found a sop that could help but seems its outdated. So, I could update that sop if you all agree. what do you all say? 19:12:06 sure! which sop? 19:12:24 it was under sysadmin guide 19:12:38 let me see if I can find again 19:13:02 https://docs.fedoraproject.org/en-US/infra/sysadmin_guide/koji/#_troubleshooting_and_resolution 19:13:34 nirik, I started to read through that and saw mentions of xen_builders and plague and did not think any of it could be trusted 19:13:52 yeah, that is ancient/out of date for sure. 19:14:03 yep. old old old 19:14:34 and phsmoura my apologies again for not giving you more play by play of what I was trying to help you learn what was going on 19:15:25 this actually brought up something I have been wanting to do for a paper.. mapping how the build system works backwards from a user downloading something to how a developer commits source code 19:15:55 I did a flock talk on that long ago... 'package life cycle' 19:16:21 np, couldnt ssh to them so not much I could do I guess :/ 19:16:22 it was my continual blind spot when I was in CPE because by the time I felt I had learned what was being done, nirik would have to clarify whatever i was saying because I forgot several subparts or still had something which wasn't there anymore 19:17:09 it would be nice to have a thing we keep up to date... 19:17:19 but we keep adding things... ostrees, containers, etc 19:17:25 nirik, yeah I remembered that but wasn't sure how it fit in with things like containers, flatpaks, ostrees, etc 19:17:52 so I was thinking of 'drawing' a picture for each 'flow' and seeing what things affected it and why 19:18:57 sounds like a great project 19:21:55 I don't have much else 19:22:26 ok, so maybe next week I could to one of you to know how to fix those outages and update the sop? 19:23:49 phsmoura: sure, happy to share what we learned in this case, and help you update it. 19:24:00 cool :) 19:24:31 is there anything else? 19:24:55 not from me. 19:24:58 not from me 19:25:49 thanks for joining :) 19:25:57 #endmeeting