| ianw | tonyb: remember https://bugzilla.redhat.com/show_bug.cgi?id=2208143 and kafs not working ... try with "setenforce 0" | 00:06 |
|---|---|---|
| opendevreview | Ian Wienand proposed opendev/system-config master: [wip] kafs support https://review.opendev.org/c/opendev/system-config/+/623974 | 01:12 |
| opendevreview | Ian Wienand proposed opendev/system-config master: [wip] kafs support https://review.opendev.org/c/opendev/system-config/+/623974 | 02:13 |
| opendevreview | Ian Wienand proposed opendev/system-config master: mirror: kafs support https://review.opendev.org/c/opendev/system-config/+/623974 | 03:29 |
| opendevreview | Ian Wienand proposed opendev/system-config master: mirror: install cachefilesd https://review.opendev.org/c/opendev/system-config/+/885671 | 03:42 |
| ianw | clarkb/tonyb: I think ^ is very seriously worth considering; especially as we're looking at updating mirror nodes. My suggestion would be to (post-review of that) bring up a test node in one of the regions and switch it in via a CNAME switch | 04:40 |
| opendevreview | Lajos Katona proposed openstack/project-config master: Add ovn-bgp-agent to openstack-neutron IRC channel https://review.opendev.org/c/openstack/project-config/+/885721 | 06:49 |
| *** cloudnull8 is now known as cloudnull | 09:18 | |
| *** cloudnull1 is now known as cloudnull | 10:42 | |
| *** amoralej is now known as amoralej|lunch | 12:47 | |
| opendevreview | Merged openstack/project-config master: Add ovn-bgp-agent to openstack-neutron IRC channel https://review.opendev.org/c/openstack/project-config/+/885721 | 12:53 |
| *** amoralej|lunch is now known as amoralej | 13:30 | |
| *** amoralej is now known as amoralej|off | 15:11 | |
| clarkb | ianw: I guess the upside is that for clients we don't need packaging for newer openafs so reduces the number of platforms we do that for? | 15:56 |
| fungi | yeah, we'd still need it for things like docs publishing from executors, or platform-specific wheel builders unless we start funneling those files back through the executors | 16:00 |
| fungi | but having to only worry about openafs packaging for executors and mirror builders would still be very nice | 16:00 |
| tonyb | ianw: Thanks for the update, hopefully the kafs package grows the correct policy change to fox that. | 17:15 |
| tonyb | In the meantime: | 17:56 |
| tonyb | [tony@thor ~]$ getenforce | 17:56 |
| tonyb | Enforcing | 17:56 |
| tonyb | [tony@thor ~]$ ls /afs/openstack.org/ | 17:56 |
| tonyb | developer-docs docs docs-old mirror project service user | 17:56 |
| tonyb | [tony@thor ~]$ | 17:56 |
| tonyb | For reasons I don't understand (yet) I can't log into RHBZ (or the VPN) but this is the TE file I created: https://paste.opendev.org/show/bLR6K446HPaYMuOAnmJ3/ | 18:06 |
| clarkb | tonyb is now confirmed to be a wizard after fixing selinux and openafs problems at the same time | 18:09 |
| clarkb | s/openafs/afs/ | 18:09 |
| tonyb | clarkb: I just can't fix zuul-registry on jammy :/ | 18:09 |
| clarkb | tonyb: really I think if crashing is the existing behavior we can just check that it tried ot start (and possibly crashed) then sort out better testing as a followup | 18:11 |
| tonyb | Okay, I was thnking I'd do it in the other order fix the testing on bionic and then show that we don't regress on jammy | 18:12 |
| tonyb | I guess we can show that jammy is no worse than bioinc, bring up a server double check and then do the switcheroo | 18:12 |
| clarkb | thats probably more correct if possible. I just don't think we need to focus too hard on it if not easy | 18:12 |
| clarkb | since it isn't a regression with jammy | 18:13 |
| tonyb | I actually think this has been a really helpful "first attempt" I've seen a bunch of new stuff. The downside is it's taking more time from you and fungi | 18:14 |
| clarkb | things being broken is a great way to learn | 18:14 |
| tonyb | \o/ | 18:15 |
| clarkb | and this week was always going to be a weird one pre summit so not a big deal | 18:15 |
| tonyb | okay | 18:15 |
| opendevreview | Clark Boylan proposed opendev/system-config master: Update acme.sh 3.0.6 https://review.opendev.org/c/opendev/system-config/+/885795 | 18:29 |
| fungi | yeah, things being broken is a good way to learn as long as it's not production things broken (which is an even quicker but far more stressful way to learn) | 20:07 |
| fungi | and yes, sorry i'm probably not as attentive as usual to communication channels.. i'm currently sitting in an airport on my way to vancouver | 20:08 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Add debbugging to docker config https://review.opendev.org/c/opendev/system-config/+/885659 | 22:00 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Test insecure-ci-registry.opendev.org on jammy https://review.opendev.org/c/opendev/system-config/+/885421 | 22:00 |
| tonyb | I think I have finally worked out why the existing testing is working with bionic. | 22:17 |
| fungi | oh? | 22:18 |
| fungi | the usual answer is: because we weren't actually testing what we thought we were testing | 22:19 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Add debbugging to docker config https://review.opendev.org/c/opendev/system-config/+/885659 | 22:26 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Test insecure-ci-registry.opendev.org on jammy https://review.opendev.org/c/opendev/system-config/+/885421 | 22:26 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Actually test the registry node https://review.opendev.org/c/opendev/system-config/+/885804 | 22:26 |
| tonyb | fungi: bingo | 22:28 |
| fungi | i take it 885804 is expected to fail then? | 22:29 |
| tonyb | Yup | 22:29 |
| fungi | exciting | 22:29 |
| tonyb | Thrilling :) | 22:29 |
| tonyb | It does mean I'm more inclined to go with clarkb's idea that we switch to jammy and then actually fix the testing | 22:30 |
| fungi | i'm on the edge of my seat, but mainly because i've been sitting in these airport seats so long my butt aches | 22:30 |
| fungi | yeah, i have no objection to that sequence | 22:30 |
| tonyb | LOL | 22:30 |
| tonyb | How long befoer you can get out of the airport? | 22:32 |
| fungi | finally boarding in an hour, but that's just for the flight to atlanta | 22:32 |
| fungi | i'm going to need to find somewhere inconspicuous to doze off for a while in atl | 22:33 |
| tonyb | Ergh, you get to hotel in atlanta I hope? | 22:33 |
| fungi | not really enough time to make that worthwhile | 22:33 |
| tonyb | Oh dear | 22:33 |
| fungi | there are no attached hotels, so factor in shuttle time and going back through security... just not worth it | 22:34 |
| tonyb | Yeah. I get that, it does make for a very long transit :( | 22:35 |
| fungi | i've spent the night in far worse airports, it won't be bad | 22:35 |
| tonyb | Good luck | 22:36 |
| fungi | thanks! i've had longer transits too, it's just disappointing for one where i'm not even leaving the continent | 22:37 |
| tonyb | Yeah. | 22:37 |
| tonyb | I've had to do similar for domestc flights in AU. | 22:37 |
| fungi | one of the tradeoffs for choosing to live at the edge of the world | 22:38 |
| tonyb | Is that me or you? | 22:39 |
| fungi | can't it be both? | 22:41 |
| tonyb | Hmmm I s'pose | 22:42 |
| tonyb | Okay 885804 behaved as exppected, now to update it a little. | 23:13 |
| tonyb | After next week I think we can probably launch the new jammy insecure-ci-registry02 for manually testing | 23:14 |
| opendevreview | Tony Breeds proposed opendev/system-config master: Update testing for insecure-ci-registry https://review.opendev.org/c/opendev/system-config/+/885804 | 23:15 |
| opendevreview | Tony Breeds proposed opendev/system-config master: [DNM] Test insecure-ci-registry.opendev.org on jammy https://review.opendev.org/c/opendev/system-config/+/885421 | 23:15 |
| ianw | clarkb: one thing on the came update to double check is https://github.com/acmesh-official/acme.sh/issues/4416 which i have to admit i've forgotten about checking if it's fixed... | 23:42 |
| ianw | s/came/acme/ even | 23:42 |
| ianw | yeah; using kafs on the r/o path first seems like a logical first step; then we can look at the publishing path including executors -- but the nice thing is we can pretty much take on things one-at-a-time from least impactful (probably say 9-stream wheel publishing) to most (executors) | 23:44 |
| opendevreview | Merged opendev/system-config master: Update acme.sh 3.0.6 https://review.opendev.org/c/opendev/system-config/+/885795 | 23:46 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!