| clarkb | https://iscinumpy.dev/post/bound-version-constraints/ 2015 me is surprised it took so long for the rest of the python world to discover this si a problem | 00:05 |
|---|---|---|
| fungi | a frog cannot explain dry land to a tadpole | 00:06 |
| fungi | no matter how many times you try to convince someone they're going to run into problems you've experienced, they'll fail to comprehend or refuse to believe until it happens to them | 00:08 |
| fungi | and also conveniently forget you encountered and came up with solutions for the problem | 00:09 |
| clarkb | I'm happy someone took the time to write all that down though because ya its basically all the wisdom. Don't cap deps, security fixes aren't backported, dep solver gets sad, etc etc | 00:12 |
| clarkb | "Artificially limiting versions will always reduce the chances of it working in the future." This is well and simply said and something I think I've failed to articulate well | 00:13 |
| clarkb | "If you don’t cap your dependencies, you are immediately notified when a dependency releases a new version, probably by your CI, the first time you build with that new version." I love this post | 00:14 |
| fungi | by extension, your ci jobs breaking on a new version of something isn't a problem, it's a signal | 00:15 |
| opendevreview | Steve Baker proposed openstack/diskimage-builder master: Overwrite EFI grub.cfg when it exists https://review.opendev.org/c/openstack/diskimage-builder/+/916021 | 01:19 |
| opendevreview | Merged openstack/diskimage-builder master: fix(rocky): don't uninstall linux-firmware https://review.opendev.org/c/openstack/diskimage-builder/+/912816 | 01:21 |
| *** dmellado030 is now known as dmellado03 | 02:00 | |
| opendevreview | Steve Baker proposed openstack/diskimage-builder master: Overwrite EFI grub.cfg when it exists https://review.opendev.org/c/openstack/diskimage-builder/+/916021 | 02:21 |
| opendevreview | Alberto Gonzalez proposed openstack/project-config master: Add new repository powertrain-build https://review.opendev.org/c/openstack/project-config/+/916038 | 07:06 |
| opendevreview | Alberto Gonzalez proposed openstack/project-config master: Add new repository powertrain-build https://review.opendev.org/c/openstack/project-config/+/916038 | 07:29 |
| opendevreview | Alberto Gonzalez proposed openstack/project-config master: Add new repository powertrain-build https://review.opendev.org/c/openstack/project-config/+/916038 | 07:31 |
| opendevreview | Alberto Gonzalez proposed openstack/project-config master: Add new repository powertrain-build https://review.opendev.org/c/openstack/project-config/+/916038 | 07:43 |
| opendevreview | Alberto Gonzalez proposed openstack/project-config master: Add new repository powertrain-build https://review.opendev.org/c/openstack/project-config/+/916038 | 08:05 |
| opendevreview | Jan Marchel proposed openstack/project-config master: Add repository for NebulOuS scripts https://review.opendev.org/c/openstack/project-config/+/916048 | 08:29 |
| opendevreview | Radosław Piliszek proposed openstack/project-config master: Add repository for NebulOuS scripts https://review.opendev.org/c/openstack/project-config/+/916048 | 08:43 |
| opendevreview | Dr. Jens Harbott proposed openstack/project-config master: Support Ubuntu 24.04 in nodepool elements https://review.opendev.org/c/openstack/project-config/+/916050 | 09:19 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Add Ubuntu 24.04 (noble) build to testing https://review.opendev.org/c/openstack/diskimage-builder/+/915915 | 09:35 |
| opendevreview | Merged openstack/project-config master: Fix the ACL associated with charm-keystone-ldap-k8s https://review.opendev.org/c/openstack/project-config/+/903667 | 11:59 |
| opendevreview | Merged openstack/project-config master: Retire PowerVMStacker SIG: Remove Project from Infrastructure Systems https://review.opendev.org/c/openstack/project-config/+/909584 | 12:02 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Add Ubuntu 24.04 (noble) build to testing https://review.opendev.org/c/openstack/diskimage-builder/+/915915 | 12:14 |
| opendevreview | Merged openstack/project-config master: Ironic UM group access to bifrost UM branches https://review.opendev.org/c/openstack/project-config/+/910287 | 12:17 |
| opendevreview | Merged openstack/project-config master: Add warning to nodepool configs about changing cloud name https://review.opendev.org/c/openstack/project-config/+/911959 | 12:29 |
| opendevreview | Merged openstack/project-config master: Add openstack-helm-plugin git repository https://review.opendev.org/c/openstack/project-config/+/916007 | 12:29 |
| opendevreview | Merged openstack/project-config master: Add repository for NebulOuS scripts https://review.opendev.org/c/openstack/project-config/+/916048 | 12:33 |
| fungi | ugh, manage-projects is failing | 13:38 |
| fungi | TASK [gitea-git-repos : Create Gitea Repos and Org] | 13:39 |
| fungi | No usable temporary directory found in ['/tmp', '/var/tmp', '/usr/tmp', '/root'] | 13:39 |
| fungi | fatal: [gitea10.opendev.org]: FAILED! | 13:39 |
| fungi | /dev/vda1 155G 155G 0 100% / | 13:40 |
| fungi | 115G /var/gitea/data/gitea/repo-archive | 13:41 |
| fungi | https://meetings.opendev.org/irclogs/%23opendev/%23opendev.2024-01-05.log.html#t2024-01-05T14:45:33 | 13:43 |
| fungi | that's what we did to solve a similar situation on gitea09 in january | 13:43 |
| fungi | i'll give it a try | 13:44 |
| fungi | i'm deleting ~clarkb/giteaXY_transplant_db.sql from 2023-03-03 on gitea10 like i did on gitea09, and then rebooting it | 13:50 |
| fungi | then i'll proceed with the archive cleanup button clicking | 13:50 |
| Clark[m] | It looks like we did end up landing the updated robots.txt which was part of the upstream suggestion from matrix/discord. I guess insufficient for us though | 13:53 |
| Clark[m] | Or at least the file currently served looks new (even has a January 2024 timestamp in it) | 13:54 |
| fungi | Started Task: Delete all repositories' archives (ZIP, TAR.GZ, etc..) | 13:55 |
| fungi | /dev/vda1 155G 41G 115G 27% / | 13:57 |
| fungi | much better | 13:57 |
| fungi | #status log Rebooted and deleted repository archives on gitea10 after they filled up its rootfs | 13:58 |
| opendevstatus | fungi: finished logging | 13:58 |
| fungi | now we either need to re-enqueue a deploy to get manage-projects run again, or merge something else that will trigger it | 13:59 |
| Clark[m] | Don't forget to have Gerrit rereplicate to 10 and we should probably spot check the others. Thank you for fixing this | 13:59 |
| Clark[m] | Does manage projects run hourly? That may be another option if so. Simply wait for the next hourly run. I want to say it doesn't though | 14:00 |
| opendevreview | Dr. Jens Harbott proposed openstack/diskimage-builder master: Add Ubuntu 24.04 (noble) build to testing https://review.opendev.org/c/openstack/diskimage-builder/+/915915 | 14:00 |
| fungi | daily i think | 14:00 |
| fungi | Clark[m]: i think our only options are replicate all repos or specific repos, but we don't have a way to limit which remote servers get replicated to | 14:02 |
| fungi | i can kick off a full replication | 14:02 |
| Clark[m] | We do | 14:02 |
| fungi | any idea what that syntax is? | 14:02 |
| Clark[m] | You can specify a URL filter when you ask for replication | 14:02 |
| fungi | oh, so that's what "--url PATTERN : pattern to match URL on" means | 14:02 |
| fungi | any idea what that pattern would look like, do we have any examples? | 14:03 |
| Clark[m] | Ya I think if you do --url gitea10 that will work | 14:03 |
| fungi | or can i nab an example from the task queue i guess? | 14:03 |
| Clark[m] | Any substring of the name in our replication config file | 14:03 |
| fungi | okay, so no wildcarding required | 14:03 |
| fungi | 2410 tasks | 14:04 |
| Clark[m] | Yes the docs imply it is substring matching only not a regex or glob | 14:04 |
| fungi | i see a bunch of retries in the queue for gitea11 | 14:05 |
| fungi | /dev/vda1 155G 155G 0 100% / | 14:05 |
| fungi | wax on, wax off | 14:05 |
| fungi | doing the same dance with that server now | 14:06 |
| Clark[m] | Didn't we also have a change to run the cron job to clear the archive daily? | 14:07 |
| fungi | i saw discussion of cron tasks in the irc log | 14:08 |
| Clark[m] | https://opendev.org/opendev/system-config/src/branch/master/playbooks/roles/gitea/templates/app.ini.j2#L130 weekly | 14:08 |
| fungi | gitea11 cleaned up and re-replication started for it as well | 14:09 |
| Clark[m] | And ya the disk usage is within the week. Maybe we want to run that daily | 14:09 |
| fungi | #status log Rebooted and deleted repository archives on gitea11 after they filled up its rootfs | 14:09 |
| opendevstatus | fungi: finished logging | 14:10 |
| Clark[m] | I think we just go ahead with daily deletion ... Unfortunately not very efficient but should keep things down to a more manageable state | 14:10 |
| fungi | working on that now | 14:11 |
| fungi | the cron timespec this uses doesn't seem aligned with vixie cron | 14:13 |
| fungi | oh, it's using optional seconds as the first field | 14:14 |
| Clark[m] | It's a golang specific cron lib iirc | 14:15 |
| opendevreview | Jeremy Stanley proposed opendev/system-config master: Switch Gitea archive cleanup from weekly to daily https://review.opendev.org/c/opendev/system-config/+/916116 | 14:17 |
| clarkb | 9, 12, 13, and 14 appear to not be suffering the same issue | 14:22 |
| fungi | their disks didn't look filled up when i checked | 14:22 |
| clarkb | ya sorry the way I worded that was not super claer. I'm saying they are fine | 14:23 |
| fungi | oh, right yes | 14:23 |
| fungi | after i saw 11 was also full i checked the others, should have said as much | 14:23 |
| fungi | my guess is that the crawler(s) got persisted to one or two backends | 14:25 |
| clarkb | the robots.txt rules seem to match for the archive download paths too | 14:26 |
| clarkb | so ya this must be bad crawlers? I feel like humans wouldn't be able to click buttons fast enough | 14:27 |
| fungi | right, i expect it's crawlers that gon't care about being polite or respecting our wishes | 14:27 |
| fungi | s/gon't/don't/ | 14:27 |
| fungi | i suppose we could try to find common user agent strings or something and see if they're unique enough to add to our filter | 14:28 |
| clarkb | fungi: ++ and you can start by grepping 'archive' out of the access logs to see what if anything is fetching those | 14:28 |
| fungi | going to try to get a quick walk in now that things have quieted down, and then we can work on landing upgrades i guess | 14:29 |
| clarkb | sounds good | 14:29 |
| clarkb | I'm still in hte process of waking up myself | 14:29 |
| Clark[m] | fungi: I suspect the cron update won't auto deploy. We can land it first then the 1.21.11 upgrade to have Ansible update the config in the running services for us | 14:36 |
| fungi | makes sense, agreed | 14:38 |
| *** blarnath is now known as d34dh0r53 | 15:04 | |
| frickler | clarkb: noble dib passing now https://zuul.opendev.org/t/openstack/build/53135a39694e4c2c8de0505a91263598 | 15:15 |
| clarkb | frickler: in the symlink change should we provide fully rooted paths for both target and link name? It isn't clear to me what gutsy is on the fs | 15:26 |
| frickler | clarkb: this is copying how the other symlinks look like. a relative path is relative to the target, this is where ln is different from cp | 15:27 |
| clarkb | TIL | 15:29 |
| clarkb | I +2'd it anyway. I also left a nit on the project-config change that I +2 as well. | 15:29 |
| clarkb | The only other thing I notice si that we should probably plan to make a glean release after the glean change lands | 15:31 |
| clarkb | thank you for putting this together | 15:31 |
| clarkb | frickler: any interest in reviewing https://review.opendev.org/c/opendev/system-config/+/916116 to try and counter the gitea archive ballooning? | 16:12 |
| fungi | also i'm ready to start approving upgrades any time. should we start with the mailman mariadb container? | 16:17 |
| clarkb | fungi: wfm | 16:19 |
| fungi | fire in the hole! | 16:20 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Add more User Agent filters https://review.opendev.org/c/opendev/system-config/+/916125 | 16:35 |
| opendevreview | Merged opendev/system-config master: Upgrade Mailman's MariaDB to 10.11 https://review.opendev.org/c/opendev/system-config/+/915183 | 17:05 |
| fungi | deploy for that's going to be waiting a bit while the hourly builds complete | 17:05 |
| clarkb | fungi: the lists deploy job is running now | 17:20 |
| fungi | yep | 17:20 |
| fungi | looks like the mariadb container restarted a minute agio | 17:21 |
| fungi | or it was upgrading | 17:21 |
| fungi | database_1 | 2024-04-17 17:22:26 0 [Note] Starting MariaDB 10.11.7-MariaDB-1:10.11.7+maria~ubu2204 source revision 87e13722a95af5d9378d990caf48cb6874439347 | 17:23 |
| fungi | as process 1 | 17:23 |
| fungi | mailman services are up again, i can browse list details (postorius) and archives (hyperkitty) | 17:24 |
| clarkb | fungi: tehre are a few connection errors in the mariadb log that surprise me | 17:25 |
| clarkb | but maybe that is related to how we test if things are up in ansible? | 17:25 |
| fungi | i think we've seen those in the past too | 17:25 |
| clarkb | I think fi services are working then it isn'y a big deal | 17:25 |
| fungi | we only see them at startup, from what i recall | 17:25 |
| clarkb | and I too can confirm services seem to be working for me. Maybe we want to send a test email through? | 17:25 |
| clarkb | ya so perhaps related to how we do the db based startup checking | 17:26 |
| clarkb | we dont' appear to log these services to /var/log/containers though so we don't have the log from when the actual mariadb upgrade occured? | 17:26 |
| clarkb | I think because we must stop and start mariadb a couple times? | 17:26 |
| fungi | that's what it looked like when i was grepping ps output | 17:26 |
| clarkb | I do see the small backup in the database dir though which confirms the upgrade ran | 17:27 |
| fungi | and yeah, we could add syslog logging to the compose file, what we use now is almost identical to upstream mailman-docker's compose file | 17:27 |
| clarkb | ack | 17:28 |
| clarkb | I think the only other thing I can think of to test is sending an email through | 17:28 |
| clarkb | otherwise what I've checked lgtm | 17:28 |
| fungi | yeah, i figured i'd wait a bit to see if one came through naturally | 17:28 |
| fungi | what upgrade did you want to do next? merge the outstanding gitea patches and then upgrade it, or do etherpad? | 17:29 |
| clarkb | I'm thinking that maybe we should do gitea in order to mitigate disk consumption concerns | 17:30 |
| clarkb | that seems more ugent than etherpad | 17:30 |
| fungi | makes sense, yep | 17:30 |
| fungi | so we want to go ahead with the archive cleanup cron and ua filter additions, then the gitea version update? | 17:30 |
| clarkb | I think the ua filter addition can go in any order, but the cron update should go before the gitea version update | 17:31 |
| clarkb | that way we don't need to manually restart anything | 17:31 |
| fungi | clarkb: should i wait to approve the gitea upgrade change until we're sure the cron config has merged? | 17:42 |
| fungi | or just risk it? | 17:42 |
| clarkb | it is probably ok to risk it? | 17:46 |
| fungi | yeah, was trying to decide if the potential time savings outweighed the potential time loss if we hit an unexpected test failure | 17:49 |
| fungi | anyway, approved it now too | 17:49 |
| clarkb | we should have a window of time where we can -W the upgrade change if the parent fails | 17:49 |
| fungi | good point | 17:49 |
| clarkb | that way we can still rely on the upgrade to pick up the new config if we like | 17:49 |
| *** dmellado033 is now known as dmellado03 | 17:54 | |
| opendevreview | Merged opendev/system-config master: Cleanup lingering Mailman 2 playbook https://review.opendev.org/c/opendev/system-config/+/915184 | 17:54 |
| opendevreview | Dr. Jens Harbott proposed openstack/project-config master: Support Ubuntu 24.04 in nodepool elements https://review.opendev.org/c/openstack/project-config/+/916050 | 17:58 |
| clarkb | fungi: it occurred to me that manage-projects may not have run yet? I don't think that is a problem since it is a job distinct from the infra-prod-service-gitea job? | 18:23 |
| clarkb | two new emails arrived in openstack-discuss too and a third is in the archive that hasn't made it to my inbox so that lgtm | 18:26 |
| clarkb | its also my parents last day in town and they just asked us to pop over for lunch. I'll be about 5-10 minutes away from my computer should anything go wrong and keep an eye on this channe via matrix | 18:28 |
| clarkb | ya its taking about 7-8 minutes for new openstack-discuss emails to arrive in my inbox. But email is async so not a big deal | 18:32 |
| clarkb | I suspect this is not new either | 18:32 |
| fungi | yeah, fairly typical | 18:32 |
| fungi | there are a lot of subscribers | 18:32 |
| fungi | i'll be around for the gitea upgrade, go enjoy your lunch | 18:33 |
| fungi | and no, we haven't merged or retriggered any new manage-projects runs afaik, checking the builds list for confirmation | 18:33 |
| fungi | https://zuul.opendev.org/t/openstack/builds?job_name=infra-prod-manage-projects&project=openstack/project-config | 18:35 |
| fungi | last build started at 12:42 utc, no successful runs yet | 18:35 |
| clarkb | ack and ya I don't think that is a problem for the infra-prod-service-gitea job | 18:37 |
| clarkb | but getting that to run after wards is probably a good idea | 18:37 |
| fungi | it's unfortunate that i approved a slew of changes that ran the deploy job before noticing it was consistently failing, so there's a good chunk of work queued up for it the next time it runs | 18:39 |
| opendevreview | Merged opendev/system-config master: Switch Gitea archive cleanup from weekly to daily https://review.opendev.org/c/opendev/system-config/+/916116 | 18:55 |
| Clark[m] | The deploy job is running for that now. Should be able to check the updated config on the servers well before the next change lands | 18:58 |
| fungi | yep | 18:59 |
| fungi | fungi@gitea09:~$ ls -l /var/gitea/conf/app.ini | 19:01 |
| fungi | -rw-r--r-- 1 root root 3833 Apr 17 18:56 /var/gitea/conf/app.ini | 19:01 |
| fungi | fungi@gitea09:~$ grep SCHEDULE /var/gitea/conf/app.ini | 19:01 |
| fungi | SCHEDULE = 0 0 3 * * * | 19:01 |
| fungi | i'll make sure the other 5 match | 19:01 |
| fungi | all 6 lgtm | 19:02 |
| Clark[m] | Excellent | 19:02 |
| opendevreview | Merged opendev/system-config master: Add more User Agent filters https://review.opendev.org/c/opendev/system-config/+/916125 | 19:15 |
| opendevreview | Merged openstack/diskimage-builder master: Write fedora download redirect info to stderr https://review.opendev.org/c/openstack/diskimage-builder/+/908207 | 19:17 |
| opendevreview | Merged openstack/diskimage-builder master: Overwrite EFI grub.cfg when it exists https://review.opendev.org/c/openstack/diskimage-builder/+/916021 | 19:18 |
| opendevreview | Merged opendev/system-config master: Upgrade gitea to v1.21.11 https://review.opendev.org/c/opendev/system-config/+/916004 | 19:23 |
| Clark[m] | The job that should upgrade things has just started | 19:28 |
| Clark[m] | Looks like gitea09 is currently down for the update | 19:30 |
| Clark[m] | And its up. System-config renders for me | 19:31 |
| fungi | yep, looking right to me | 19:38 |
| fungi | Powered by Gitea Version: v1.21.11 | 19:38 |
| fungi | and my test clones have completed successfully | 19:39 |
| fungi | i guess etherpad is next to look at? do we want to do that today or tomorrow? | 19:39 |
| fungi | i'm flexible | 19:39 |
| Clark[m] | I don't mind proceeding with etherpad I'm just not around for a bit longer | 19:40 |
| Clark[m] | Tomorrow would be fine too if you prefer | 19:41 |
| fungi | nah, the sooner the better in my opinion | 19:41 |
| Clark[m] | The gitea job was successful and all 6 back ends lgtm | 19:42 |
| fungi | i'm just mowing in short bursts | 19:42 |
| fungi | no real scheduled plans for a couple more hours and then i need to cook dinner | 19:42 |
| fungi | should i approve it now so we get the gate jobs underway, or wait a bit for you to free up? | 19:43 |
| Clark[m] | I think it's fine if you go for it now. Worst case I'm not far from home | 19:43 |
| fungi | oh, or did we want to wait for an actual etherpad release? | 19:44 |
| fungi | oh, it's the etherpad mariadb upgrade we wanted to do first | 19:44 |
| Clark[m] | The database update change doesn't depend on the etherpad release. But we shouldn't land the etherpad upgrade until etherpad has a release just in case they change stuff in non compatible ways | 19:44 |
| fungi | https://review.opendev.org/911000 | 19:44 |
| fungi | yeah, that just dawned on me | 19:45 |
| Clark[m] | Ya the DB upgrade is the one fine to do now if we are happy with the change | 19:45 |
| fungi | after several of these i'm not too worried. went ahead and approved it | 19:45 |
| Clark[m] | The etherpad proper upgrade needs to wait | 19:45 |
| fungi | agreed | 19:45 |
| fungi | so many upgrades | 19:46 |
| Clark[m] | Maybe double check that flag is correct against the other mariadb updates but otherwise it should be very similar to the other updates. Services go down. DB comes up and upgrades then etherpad connects to it | 19:46 |
| fungi | looks the same as the others | 19:51 |
| fungi | probably at least another half hour until it merges | 20:14 |
| fungi | but also the ducks don't want me to keep mowing | 20:15 |
| fungi | there's this mated pair of mallards that comes to hang out in the yard a few hours a day and scavenge the area below the deck where birds drop seed from the feeders upstairs | 20:17 |
| clarkb | smart | 20:18 |
| fungi | they tolerate me sitting here a few feet away, but are not fans of the lawn mower | 20:18 |
| fungi | though now they're getting into a territorial dispute with the canada geese hanging out a few lots up the water from us | 20:19 |
| fungi | so maybe i can finish up after all | 20:20 |
| fungi | also i don't think either the ducks or the geese are fans of the ospreys which are nesting out near the end of the dock | 20:21 |
| opendevreview | Merged opendev/system-config master: Upgrade Etherpad's MariaDB to 10.11 https://review.opendev.org/c/opendev/system-config/+/911000 | 20:23 |
| JayF | I'm trying to paste a large amount of preexisting text into an etherpad, and it's just spamming "connected" over and over, and I can confirm via another browser the text is not getting uploaded | 20:25 |
| JayF | I assume this is some kinda abuse prevention trick? but I'd like to know how you all get around it when you need to | 20:26 |
| JayF | I got it to work via batching, but if there's a neat trick (or a concrete line/character count), I'd still be interested | 20:29 |
| clarkb | JayF: depending on the timing you may have run into the db ugprade behind etherpad that just happened a few minutes ago | 20:29 |
| clarkb | but yes I do believe there is a per connection rate limit | 20:29 |
| JayF | I think the answer is "I hit both" | 20:29 |
| JayF | which made the behavior extra weird :D | 20:29 |
| fungi | JayF: have you tried the import feature? | 20:29 |
| JayF | *blink* | 20:29 |
| JayF | :-O | 20:29 |
| fungi | check the menu icons at the top-right | 20:30 |
| fungi | it has import/export features | 20:30 |
| JayF | that is what I was looking for :D | 20:30 |
| JayF | thank you | 20:30 |
| fungi | no guarantees it will work better, but probably more efficient than pasting ginormous amounts of data | 20:30 |
| clarkb | fungi: fwiw the db upgrade seems to have gone well. We don't appear to set etherpad having a dependency on the db so we didn't restart it? | 20:31 |
| JayF | Given how much I had to break it up, I think it's technically 3 ginormous datas :P | 20:31 |
| clarkb | but it seems to have reconnected after the db was happy again so nto a huge deal | 20:31 |
| fungi | yeah, it's working for me now | 20:32 |
| clarkb | https://opendev.org/opendev/system-config/src/branch/master/playbooks/roles/etherpad/templates/settings.json.j2#L574-L589 and https://opendev.org/opendev/system-config/src/branch/master/playbooks/roles/etherpad/templates/settings.json.j2#L599-L613 are the rate limit configs | 20:33 |
| clarkb | JayF: ^ basically its 10 edits per second per IP | 20:33 |
| clarkb | I'm guessing that your copy paste buffer gets flushed multiple times pushing it over that limit maybe? | 20:33 |
| JayF | that's what I'd suspect, yeah | 20:34 |
| JayF | it also acted different with middle-click vs right-click paste | 20:34 |
| fungi | that's good to know, but probably browser-specific as to how it treats the primary and selection buffers | 20:39 |
| fungi | clarkb: so what was the plan for the gerrit upgrade? yolo this week or hold until after the rename maintenance is knocked out? | 20:40 |
| clarkb | fungi: I'm leaning towards after the rename | 20:40 |
| fungi | i'm good with either | 20:40 |
| clarkb | but I wonder stop you if you want to get it done sooner | 20:40 |
| clarkb | *I won't stop you | 20:40 |
| fungi | it's nice to be nearing the end of this tunnel at least | 20:41 |
| clarkb | ya huge progress today, thanks for helping drive it along | 20:41 |
| fungi | yeah, i could plan to do the gerrit upgrade tomorrow so we have the weekend to shake out any unexpected behavior changes | 20:41 |
| clarkb | I don't think there is a change for gerrit yet and it would have to be manually restarted too | 20:47 |
| clarkb | since the ansible for gerrit doesnt' auto restart things iirc | 20:47 |
| clarkb | fungi: https://review.opendev.org/c/opendev/system-config/+/911622 is the rename prep change we should consider landing soonish too | 20:49 |
| clarkb | maybe we should recheck to get up to date test job logs and we can check that things get copied to the correct location? | 20:50 |
| clarkb | that said it should have fairly low impact even if it doesn't work, we'll just end up with files copied somewhere we don't expect and maybe extra errors in the error log | 20:50 |
| fungi | yeah, i'll get that merged after dinner | 21:12 |
| fungi | er, rechecked i mean | 21:12 |
| fungi | we don't merge it until the end of the maintenance if memory serves | 21:13 |
| fungi | oh, wait, *that* prep change | 21:13 |
| fungi | yeah, approved now | 21:13 |
| fungi | we still need the opendev/project-config change to reflect the repo redirects, but that comes later | 21:14 |
| Clark[m] | Yup, I stepped away again but it is on my to-do list to write that change. | 21:18 |
| opendevreview | Merged opendev/system-config master: Move gerrit replication waiting queue aside during project renames https://review.opendev.org/c/opendev/system-config/+/911622 | 22:48 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!