| ykarel | frickler, seeing Disk quota exceeded (122) again | 09:02 |
|---|---|---|
| ykarel | http://mirror.iad.rax.opendev.org/logs/rsync-mirrors/centos-stream.log | 09:02 |
| opendevreview | Bernhard Berg proposed zuul/zuul-jobs master: prepare-workspace-git: Add ability to define synced pojects https://review.opendev.org/c/zuul/zuul-jobs/+/887917 | 09:21 |
| apevec | how big is the quota? | 09:23 |
| tkajinam | I wonder if there is a quick way to check when the last sync between upstream repos and CI mirorrs was executed | 11:10 |
| tkajinam | I can check the packages in CI mirrors but there may be some job results (in zuul) I can check alternatively, I guess | 11:10 |
| ykarel | tkajinam, can check http://mirror.iad.rax.opendev.org/centos-stream/timestamp.txt | 11:18 |
| fungi | i can increase the quota on it some more shortly, just not quite yet. trying to catch up on triage before we start our meetings | 13:35 |
| ykarel | thx fungi | 13:39 |
| tkajinam | ykarel, ah. I didn't know that. thanks for the pointer ! | 13:47 |
| fungi | we also have a grafana dashboard for all the volumes | 13:47 |
| fungi | https://grafana.opendev.org/d/9871b26303/afs | 13:47 |
| fungi | that shows both quota utilization and how recently updates have occurred in the case of mirrors | 13:48 |
| fungi | says the mirror.centos-stream volume last updated 20 hours ago | 13:48 |
| Clark[m] | I might be a minute old two late. Just got the tea steeping | 13:56 |
| clarkb | made it just in time | 14:00 |
| clarkb | https://meetpad.opendev.org/opendev-preptg-202402 is where we'll be | 14:00 |
| *** amoralej_ is now known as amoralej | 14:02 | |
| ykarel | thanks fungi , so centos stream using all allocated 320 GB | 14:49 |
| ykarel | apevec, ^ | 14:49 |
| apevec | wut | 14:50 |
| apevec | amoralej_: ^ | 14:50 |
| apevec | we need to expand list of excludes then, I saw some new content like automotive SIG | 14:51 |
| *** d34dh0r5- is now known as d34dh0r53 | 15:02 | |
| fungi | #status log Increased mirror.centos-stream AFS volume from 320GB to 350GB while we work out how to trim down the subset of packages we rsync | 15:08 |
| opendevstatus | fungi: finished logging | 15:08 |
| fungi | i've got a manual rerun of the rsync underway now | 15:08 |
| fungi | apevec: and yes, it's pulling a bunch of stuff from the autosd tree | 15:11 |
| fungi | okay, it's in sync. reran it just to make sure the second time was a noop | 15:15 |
| fungi | if we manage to free up a bunch of space there we can easily reduce the quota again, of course | 15:16 |
| fungi | the graphite dashboard lags behind so isn't reflecting it just yet | 15:20 |
| fungi | `fs listquota /afs/openstack.org/mirror/centos-stream` now reports: | 15:21 |
| fungi | Volume Name Quota Used %Used Partition | 15:21 |
| fungi | mirror.centos-stream.read 350000000 321845167 92%<< 47% <<WARNING | 15:21 |
| fungi | so we have some breathing room | 15:21 |
| clarkb | apevec: one thing I noticed is that many packages have many versions of the package in the mirror | 15:52 |
| clarkb | and some are not small | 15:52 |
| clarkb | didn't mean to end so abruptly but I think we can all use a break and food and we are at the allotted time for the first session | 16:02 |
| clarkb | we're back againat 22:00 UTC today same location | 16:02 |
| clarkb | apevec: anyway I mention that beacuse I'm not sure if that is intentional from the centos side (it seemsweird to have like 5 versions of thunderbird available, latest and previous seems sufficient to me) and it is difficult to filter those things out though I guess theoretically possible on our end | 16:03 |
| fungi | i wonder if it's due to having a shared package pool for all the minor releases of centos? | 16:11 |
| fungi | so you have the thunderbird package for centos stream 9.0, 9.1, 9.2... | 16:12 |
| fungi | debian does that, but we're able to untangle it by using tools that can filter package versions based on which indices they are listed in | 16:14 |
| Clark[m] | I didn't think centos stream did releases like that. I thought it was a rolling release without checkpoints but I may be wrong | 16:14 |
| fungi | oh, maybe so. i'm not super familiar with the release model for that distro | 16:14 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Document adding Zuul WebUI admins https://review.opendev.org/c/opendev/system-config/+/908949 | 16:54 |
| fungi | corvus: ^ i tried to incorporate your additional settings and steps there | 16:55 |
| fungi | i also fixed my change for zuul's howto so that the missing scope addition is now covered | 17:06 |
| fungi | okay, heading up the island for a late lunch/early dinner and a bit of shopping, but should be back well before our next pre-ptg block at 22z | 17:21 |
| fungi | probably no more than 2 hours really, so should be around again by 19:30 | 17:22 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 17:26 |
| clarkb | enjoy! | 17:29 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 17:31 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 17:53 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 18:20 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:21 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 18:40 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:40 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Remove openshift + zuul registry testing https://review.opendev.org/c/zuul/zuul-jobs/+/909045 | 18:40 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Remove openshift + zuul registry testing https://review.opendev.org/c/zuul/zuul-jobs/+/909045 | 18:43 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Override DOCKER_MIN_API_VERSION for skopeo when installing docker https://review.opendev.org/c/zuul/zuul-jobs/+/909029 | 18:43 |
| opendevreview | Clark Boylan proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 18:43 |
| *** dhill is now known as Guest2793 | 18:51 | |
| clarkb | infra-root heads up the zuul circular deps upgrade required a zk db clearing for zuul. This cleared out our autoholds in zuul but didn't remove them in nodepool | 18:56 |
| clarkb | this means we'll need to clean those up directly in nodepool. I've already done so for my etherpad 1.9.6 hold (keeping the 1.9.7 hold until we upgrade). fungi frickler jrosser and corvus all have holds otherwise (though I think corvus' has a valid authold as it was made after the db changes) | 18:57 |
| clarkb | if you don't need those anymore I'm happy to clean them up. fungi yours is related to keycloak stuff. Maybe we don't need them anymore? | 18:58 |
| jrosser | clarkb: mine is done with now - you can delete it | 19:02 |
| clarkb | jrosser: thanks for confirming I'll get that done | 19:02 |
| jrosser | for interest the outcome was that using one of the vexxhost nodes with nested virt acelleration that works was the only way i could make my magnum job run in a reasonable time | 19:04 |
| corvus | fungi: btw, firefox containers is a godsend for dealing with keycloak; one container for admin, one for user | 19:23 |
| corvus | fungi: both doc changes lgtm! | 19:25 |
| clarkb | oh good reminder for me to rereview those | 19:27 |
| clarkb | firefox containers are useful for other thinsg too. I try to keep all the google stuff isolated into one | 19:27 |
| opendevreview | Merged opendev/system-config master: Document adding Zuul WebUI admins https://review.opendev.org/c/opendev/system-config/+/908949 | 19:39 |
| fungi | clarkb: thanks for the autohold reminder, i remember thinking yesterday that our list of autoholds looked remarkably empty for a change | 20:02 |
| fungi | and yeah, i was thinking about using the held keycloak test to dry-run repeat some config steps for the docs, but i ended up not needing to so i'll clean it up now | 20:03 |
| fungi | corvus: yes, absolutely, i was using multiple ff account containers for it. very convenient | 20:04 |
| fungi | clarkb: i actually have three different google account containers. for someone who hates google i have to juggle three different accounts (one is my youtube account from before they were acquired, a second for being on openinfra staff, third is used for spi board of directors business) | 20:05 |
| fungi | pretty insane, now that i think about it | 20:06 |
| fungi | ouch: https://jitsi.debian.social/ | 20:23 |
| errcsool | Trivial question for setting up devstack and tempest. Where is the tempest config file? According to the docs, the `/opt/stack/tempest` directory is created, and the `.../etc/` directory should contain the config files. I see four files called `.../etc/<something>.yaml|conf.sample`, but no `tempest.conf`. What is the recommended move here, should I write the file myself? | 20:35 |
| fungi | errcsool: you probably want to ask in the #openstack-qa channel where the devstack and tempest maintainers hang out | 20:37 |
| errcsool | ok | 20:38 |
| errcsool | thank you | 20:38 |
| fungi | yw | 20:38 |
| errcsool | Is that channel still active? I can't load it or find it on oftc | 20:40 |
| fungi | errcsool: they just wrapped up their weekly meeting in there about 3 hours ago, and it has 70 people joined from what i can see | 20:42 |
| clarkb | I don't think the channel is publicly listed due to spam. but it is there if you attempt to join it | 20:43 |
| fungi | yes, if you're using the matrix oftc bridge, you have to enter the channel name directly and can't rely on channel listings | 20:43 |
| fungi | spammers like to scrape all listed channels on the network for targeting, so most (all?) of our projects' channels are "hidden" from the public network list as a resuly | 20:44 |
| fungi | result | 20:44 |
| fungi | as is often said, "this is why we can't have nice things" | 20:44 |
| errcsool | got it, im in | 20:55 |
| errcsool | ty | 20:55 |
| fungi | of course! | 20:56 |
| fungi | mtreinish: have you ever seen `stestr init` return "stestr: 'init' is not a stestr command." | 21:22 |
| fungi | thinking there's something pathologically broken with the environment/installation where that happened | 21:23 |
| tonyb | I may be a few mins late to the preptg as I'm dropping the kids at school | 21:25 |
| clarkb | noted | 21:28 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 21:52 |
| tonyb | and of course in a time crunch pulseaudio and pipewire are arguing about who owns the sound subsystem | 22:12 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: fix old stestr test https://review.opendev.org/c/zuul/zuul-jobs/+/909060 | 22:51 |
| opendevreview | James E. Blair proposed zuul/zuul-jobs master: Remove command.warn usage https://review.opendev.org/c/zuul/zuul-jobs/+/908671 | 23:19 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!