opendevreview | Ian Wienand proposed opendev/lodgeit master: [DNM] Update to Python 3.9 https://review.opendev.org/c/opendev/lodgeit/+/819562 | 00:03 |
---|---|---|
opendevreview | Ian Wienand proposed opendev/lodgeit master: [DNM] Update new paste message https://review.opendev.org/c/opendev/lodgeit/+/819563 | 00:03 |
fungi | welcome back, ianw! | 00:07 |
ianw | :) trying to get things in order for LCA talk tomorrow | 00:10 |
ianw | i'm about 80% | 00:11 |
ianw | i would suggest, if you have the option of *not* getting covid, you take that :) | 00:11 |
Clark[m] | ianw: noted :) | 00:16 |
*** mazzy50988129295 is now known as mazzy5098812929 | 00:18 | |
fungi | um, yeah, sound advice | 00:18 |
corvus | infra-root: i'd like to restart zuul. this will be a disruptive restart with a full queue save/restore (can't do a rolling restart this time) | 00:30 |
clarkb | I'm fading out and need to help start dinner, but no objections from me. | 00:31 |
fungi | sounds good, is the expectation this may end up tagged 4.12.0? | 00:33 |
corvus | fungi: exactly | 00:34 |
fungi | i'll let #openstack-release know, though they're almost certainly done for the day | 00:34 |
fungi | they've been forewarned | 00:35 |
fungi | i'm around too if you need help with the restart | 00:35 |
corvus | all right. i've pulled and saved queues (using the new script from zuul/tools), so here we go. | 00:35 |
fungi | oh, right, new script! | 00:35 |
corvus | stopped and running `zuul delete-state` now | 00:37 |
corvus | hrm, i'm not sure it actually stopped | 00:40 |
corvus | running twice helped. maybe raced a deployment or something. | 00:40 |
corvus | starting now | 00:40 |
fungi | not sure it stopped as in you saw a rogue running process or new data appeared in zk after the delete-state? | 00:41 |
corvus | both :) | 00:41 |
fungi | oh, hrm | 00:41 |
corvus | delete-state errored, and then i checked zuul02 and it had restarted the docker container | 00:42 |
corvus | i have started both schedulers. | 00:43 |
corvus | amusingly, zuul01 is now idle as it waits for zuul02 to finish the initial openstack tenant configuration. it has already completed all the others. | 00:44 |
fungi | interesting | 00:51 |
corvus | yeah, the config is a wee bit lopsided | 00:57 |
corvus | especially since the number of branches seems to be growing | 00:57 |
corvus | zuul02 is up; waiting for zuul01 and zuul-web to catch up now | 01:00 |
corvus | zuul01 is ready, zuul-web is next. | 01:04 |
corvus | okay, everything is up and running the same version :) | 01:08 |
corvus | re-enqueing | 01:08 |
corvus | hrm, looks like there was a race condition with zuul web and an empty branch cache which has caused it to fail to load the zuul tenant. i'm going to restart zuul-web now. hopefully later i can track down what caused it. | 01:13 |
fungi | noted | 01:13 |
corvus | re-enqueue for openstack tenant is done | 01:14 |
corvus | zuul-web is back up, and re-enqueue for zuul tenant is done | 01:21 |
corvus | i think that concludes maintenance | 01:21 |
fungi | thanks! | 01:48 |
corvus | things are still looking good | 03:08 |
fungi | yep | 03:45 |
*** lajoskatona_ is now known as lajoskatona | 19:00 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!