ianw | i guess so, or versions of ssh-keygen that include -D *also* ignore DSS/DSA | 00:00 |
---|---|---|
ianw | although, no, that's ssh-keygen -r running on a focal node (the new mirror node) | 00:01 |
ianw | yeah, ssh-keyscan -D shows only 1,3,4 .. so i guess it generates the dsa keys but doesn't use them or something | 00:05 |
*** ryohayakawa has joined #opendev | 00:05 | |
openstackgerrit | Ian Wienand proposed opendev/zone-opendev.org master: Add replacement OE mirror https://review.opendev.org/745000 | 00:06 |
openstackgerrit | Merged opendev/zone-opendev.org master: Add replacement OE mirror https://review.opendev.org/745000 | 00:11 |
ianw | mnaser: so we have bionic hosts in use on ca-ymq ... so i assume the old image is ok. is there anything i can do at this point? | 00:30 |
mnaser | Yeah I think we’re good ianw Thanks | 00:31 |
fungi | #status log updated review.openstack.org cname to refer to review.opendev.org rather than review01.openstack.org | 00:47 |
openstackstatus | fungi: finished logging | 00:47 |
fungi | hopefully that completes solving the sshfp regression users were hitting for the gerrit ssh api | 00:48 |
ianw | it seems like an oversight to not include ports in the record | 00:51 |
ianw | then again, it would also advertise your non-standard ssh port ... so | 00:52 |
openstackgerrit | Ian Wienand proposed opendev/system-config master: launch-node: get sshfp entries from the host https://review.opendev.org/744821 | 02:57 |
*** markmcclain has quit IRC | 03:31 | |
*** markmcclain has joined #opendev | 03:35 | |
*** sgw1 has quit IRC | 04:40 | |
*** ryohayakawa has quit IRC | 04:41 | |
*** ryohayakawa has joined #opendev | 04:41 | |
*** ryohayakawa is now known as rhayakaw | 04:41 | |
*** raukadah is now known as chkumar|rover | 05:06 | |
*** ysandeep|off is now known as ysandeep | 05:58 | |
openstackgerrit | Lajos Katona proposed openstack/project-config master: Import netowrking-l2gw & networking-l2gw-tempest-plugin to x/ https://review.opendev.org/744110 | 06:35 |
*** DSpider has joined #opendev | 06:37 | |
*** xiaolin has quit IRC | 06:49 | |
*** ysandeep is now known as ysandeep|lunch | 07:18 | |
*** fressi has joined #opendev | 07:28 | |
*** dtantsur|afk is now known as dtantsur | 07:30 | |
openstackgerrit | Michal Nasiadka proposed opendev/system-config master: Add focal to deb-docker mirror https://review.opendev.org/745084 | 07:45 |
*** tosky has joined #opendev | 07:46 | |
lourot | AJaeger o/ does https://review.opendev.org/#/c/744890/ now look ok? thanks! | 07:54 |
*** moppy has quit IRC | 08:01 | |
*** moppy has joined #opendev | 08:01 | |
*** hashar has joined #opendev | 08:14 | |
*** ysandeep|lunch is now known as ysandeep | 08:16 | |
*** jaicaa has quit IRC | 08:22 | |
*** jaicaa has joined #opendev | 08:25 | |
*** priteau has joined #opendev | 08:29 | |
*** osmanlicilegi has quit IRC | 08:32 | |
lourot | ^ thanks! it looks like I need more approvals. ttx, mordred maybe? | 08:41 |
*** iurygregory has quit IRC | 08:51 | |
*** iurygregory has joined #opendev | 09:02 | |
*** osmanlicilegi has joined #opendev | 09:10 | |
*** rhayakaw has quit IRC | 10:07 | |
*** tkajinam has quit IRC | 10:33 | |
*** fressi has quit IRC | 10:37 | |
*** ysandeep is now known as ysandeep|afk | 10:38 | |
*** fressi has joined #opendev | 10:41 | |
*** ysandeep|afk is now known as ysandeep | 11:14 | |
*** sshnaidm is now known as sshnaidm|afk | 11:33 | |
*** sshnaidm|afk is now known as sshnaidm | 11:49 | |
*** cloudnull has quit IRC | 12:43 | |
*** cloudnull has joined #opendev | 12:44 | |
*** fressi has quit IRC | 13:26 | |
openstackgerrit | Merged openstack/project-config master: Mirror keystone-kerberos and ceph-iscsi charms to GitHub https://review.opendev.org/744890 | 13:34 |
openstackgerrit | Merged zuul/zuul-jobs master: Avoid to use 'length' filter with null value https://review.opendev.org/742316 | 13:36 |
lourot | AJaeger, thanks for merging the mirroring of charm-keystone-kerberos to GitHub. I remember from last time that the next merge will trigger the mirroring for the first time. However from what I remember from last time I would expect https://github.com/openstack/charm-keystone-kerberos to now exist empty at this point | 13:48 |
lourot | but it's not there. Is it normal? | 13:48 |
AJaeger | lourot: best ask ttx ^ | 13:48 |
fungi | lourot: AJaeger: github project creation happens via a daily periodic job, i believe | 13:49 |
fungi | after that, mirroring is refreshed on each new change which merges to the repository | 13:50 |
AJaeger | fungi: so, should the URL be available from the daily run? Or did it fail | 13:52 |
fungi | i think the job to create them checks to see whether the mirror job has been added for them, so would in theory run this coming time around, but i'd have to go hunt down the job to confirm | 13:54 |
AJaeger | thanks | 13:56 |
openstackgerrit | Tristan Cacqueray proposed zuul/zuul-jobs master: ensure-pip: add instructions for RedHat system https://review.opendev.org/743750 | 14:00 |
*** icey has joined #opendev | 14:01 | |
*** ysandeep is now known as ysandeep|brb | 14:19 | |
*** ysandeep|brb is now known as ysandeep | 14:24 | |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use (upload|promote)-docker-image roles in periodic jobs https://review.opendev.org/740560 | 15:03 |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use (upload|promote)-docker-image roles in periodic jobs https://review.opendev.org/740560 | 15:04 |
*** hashar has quit IRC | 15:19 | |
*** mlavalle has joined #opendev | 15:19 | |
*** chkumar|rover is now known as raukadah | 15:20 | |
donnyd | fungi: ianw I noticed the mirror instance is running, but I cannot get to the web server to see if its working or not | 15:30 |
fungi | donnyd: i think we need a second infra-root review/approval on https://review.opendev.org/744999 | 15:32 |
donnyd | well I will throw my +1 in there for good measure | 15:34 |
donnyd | LOL | 15:34 |
fungi | or maybe i'll just single-core approve it if nobody else gets time to take a look here shortly | 15:35 |
clarkb | done | 15:35 |
fungi | thanks clarkb! | 15:36 |
openstackgerrit | Merged opendev/system-config master: Add OE mirror to inventory https://review.opendev.org/744999 | 15:57 |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use *-docker-image roles in periodic jobs https://review.opendev.org/740560 | 16:05 |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use *-docker-image roles in periodic jobs https://review.opendev.org/740560 | 16:10 |
*** iurygregory has quit IRC | 16:10 | |
*** iurygregory has joined #opendev | 16:11 | |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use *-docker-image roles in periodic jobs https://review.opendev.org/740560 | 16:12 |
openstackgerrit | Andrii Ostapenko proposed zuul/zuul-jobs master: Add ability to use *-docker-image roles in periodic jobs https://review.opendev.org/740560 | 16:34 |
openstackgerrit | Ghanshyam Mann proposed openstack/project-config master: Fix 'submit' permission for retired repo acl https://review.opendev.org/745187 | 16:52 |
openstackgerrit | Clark Boylan proposed openstack/project-config master: Cleanup old periodic pbr jobs https://review.opendev.org/745189 | 17:02 |
*** dtantsur is now known as dtantsur|afk | 17:03 | |
openstackgerrit | Merged openstack/project-config master: Fix 'submit' permission for retired repo acl https://review.opendev.org/745187 | 17:09 |
*** priteau has quit IRC | 17:27 | |
*** priteau has joined #opendev | 17:27 | |
*** priteau has quit IRC | 17:32 | |
clarkb | donnyd: fungi fyi http://mirror.us-east.openedge.opendev.org/ and https://mirror.us-east.openedge.opendev.org/ seem happy at first glance if we want to enable the nodepool provider | 17:37 |
clarkb | is there a change for that yet? | 17:37 |
fungi | i didn't see one yet, no | 17:37 |
clarkb | donnyd: ^ maybe you want to propose that now since you've got the best idea of what reasonable capacity will be | 17:39 |
fungi | basically start with a revert of the change where we removed oe, but tweak the max-servers line(s) according to current capacity | 17:43 |
fungi | oh, i guess we didn't remove it, just set max-servers to 0 | 17:46 |
fungi | :q | 17:46 |
fungi | hah, you're not my vi | 17:46 |
fungi | https://review.opendev.org/739605 (openstack/project-config da3f580) was the change | 17:47 |
openstackgerrit | Matt McEuen proposed openstack/project-config master: New Project Request: hostconfig-operator https://review.opendev.org/745193 | 17:48 |
donnyd | clarkb: fungi i am thinking of setting things pretty low and monitoring for a while | 17:50 |
donnyd | Patch inbound | 17:50 |
clarkb | donnyd: sounds good | 17:51 |
openstackgerrit | Donny Davis proposed openstack/project-config master: revert https://review.opendev.org/739605 https://review.opendev.org/745195 | 17:55 |
*** priteau has joined #opendev | 18:06 | |
*** priteau has quit IRC | 18:11 | |
openstackgerrit | Merged openstack/project-config master: revert https://review.opendev.org/739605 https://review.opendev.org/745195 | 18:17 |
*** gmann is now known as gmann_afk | 18:35 | |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 18:36 |
clarkb | fungi: ianw ^ ok I've crippled the connection details in testing to avoid making gerrit or freenode sad | 18:36 |
clarkb | but that should actualyl start the container | 18:37 |
fungi | awesome | 18:37 |
openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Fix partial subunit stream logging https://review.opendev.org/744565 | 18:40 |
*** priteau has joined #opendev | 18:46 | |
clarkb | ianw: I also updated the lack of restarts when the config changed. That was a good catch. I intended to do that and simply missed it in the original config | 18:46 |
clarkb | also I've not updated ansible vars on bridge to make that work in production yet. Figured we'd want to coordinate that with a disabling of gerritbot on review anyway | 18:47 |
clarkb | but I think it is really close now | 18:47 |
*** sshnaidm is now known as sshnaidm|afk | 18:47 | |
fungi | yeah, we'll want to make sure we don't wind up with two processes fighting over the account and ghosting each other | 18:49 |
*** priteau has quit IRC | 18:51 | |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 18:58 |
clarkb | it is weird that linting didn't catch that on ps1 | 18:58 |
* clarkb pops out for lunch | 18:59 | |
*** tosky has quit IRC | 19:20 | |
donnyd | hey for once I didn't start off with a bunch of errors when I turned things back on | 20:22 |
donnyd | it usually blows up on my immediately | 20:22 |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 20:22 |
clarkb | does no get parsed to a boolean in yaml? testing complained my no value above was not a string ... /me forces it to be a string | 20:23 |
clarkb | donnyd: we're getting better at this :) | 20:23 |
donnyd | I need to let it run for a day or so to gather some average metrics and I will be back to full operations | 20:23 |
clarkb | fungi: https://zuul.opendev.org/t/zuul/build/844437ca7d9a4a9e9ae6fce8a8148076/console#1/3/12/opensuse-15 we're hitting the suse/ package prefix bug on zuul-jobs jobs now | 20:32 |
clarkb | that tells me the issue remains persistent and that it is unlikely an issue with ironic | 20:33 |
* clarkb looks into this again and tries to figure it out | 20:33 | |
clarkb | I'm going to put a hold on the zuul-jobs job as that is a smaller problem space than ironic and try and igure it out from the host | 20:34 |
clarkb | doing a package install manually and ignoring the failures shows that while most packages have this issue not all do | 20:48 |
clarkb | repo-oss seems to have the problem but not repo-update | 20:49 |
fungi | you have at least some familiarity with opensuse compared to the none i possess | 20:49 |
clarkb | ok I removed the repo we have configured, added it back in but let zypper autodetect things like repo type and that works | 20:56 |
clarkb | I think the fix here is to minimize our config and let zypper figure things out automatically | 20:56 |
clarkb | I expect the upstream repo layout/type changed on us and now our old configs don't map to that | 20:57 |
fungi | any way for you to diff the configs and figure out what changed | 20:59 |
openstackgerrit | Clark Boylan proposed zuul/zuul-jobs master: Drop suse mirror types in our zypper repo configs https://review.opendev.org/745225 | 21:00 |
clarkb | fungi: its basically ^ | 21:00 |
clarkb | when I zypper addrepo mirror-url repo-oss-test I don't get the type set | 21:00 |
clarkb | and reading manpages it says let zypper autodetect | 21:00 |
clarkb | that will be self testing so we can wait and watch, but I expect suse upstream of us chagned the repo type | 21:01 |
clarkb | and now our hard set value is wrong and we fail | 21:01 |
fungi | awesome, thanks! | 21:01 |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 21:07 |
clarkb | fungi: the two suse jobs have passed on https://review.opendev.org/745225 | 21:15 |
clarkb | corvus: ^ fyi that would be a good one to land | 21:15 |
corvus | clarkb: i agree that looks like a change that makes things turn green. +2 :) | 21:22 |
fungi | green is soothing | 21:27 |
clarkb | I'll respond to dtantsur once that lands | 21:28 |
fungi | thanks for running that down the rest of the way! | 21:31 |
*** DSpider has quit IRC | 21:36 | |
openstackgerrit | Merged zuul/zuul-jobs master: Drop suse mirror types in our zypper repo configs https://review.opendev.org/745225 | 21:41 |
openstackgerrit | Merged zuul/zuul-jobs master: Fix partial subunit stream logging https://review.opendev.org/744565 | 21:56 |
clarkb | anyone know where/how we configure docker to log to journald? (I think it is what we want because that means we get auto rotation I htink) | 21:57 |
clarkb | but I can' see how we've done that | 21:57 |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 22:04 |
clarkb | fungi: ianw ^ I'm beginning to suspect that gerritbot's default double forking is a problem there but wanted to ensure logging supports that | 22:04 |
clarkb | docker will exit with failure if the first fork exits right? | 22:06 |
clarkb | we can address that either by not forking or using simple-init? | 22:06 |
clarkb | do we have a preference? | 22:06 |
fungi | i think "the docker way" (much like "the systemd way") is to run in the foreground without forking and direct all logging to stdout, so i guess if we're going to run it under docker then that gets my vote | 22:18 |
ianw | clarkb: daemonizing i guess would make it look like init died ... it's not running an init? | 22:18 |
fungi | the days of the traditional unix daemon are apparently numbered | 22:18 |
openstackgerrit | Clark Boylan proposed opendev/gerritbot master: Add option to disable daemonization https://review.opendev.org/745240 | 22:21 |
clarkb | ianw: no init, the cmd is gerritbot itself | 22:21 |
*** qchris has quit IRC | 22:22 | |
clarkb | I've added an option to --no-daemon the process in ^ though I've not done any testing of that | 22:22 |
clarkb | it may be worth letting taht build an imgae, then running the image locally against review.o.o and freenode with a minimal config just to sanity check it | 22:22 |
clarkb | I guess alternatively I can figure out how to depeonds on that in system-config | 22:24 |
ianw | it should "just work" from the intermediate registry? | 22:26 |
clarkb | I think we need to do the requires and provides things | 22:28 |
clarkb | I've got chagnes for thta and will add depends-on | 22:28 |
openstackgerrit | Clark Boylan proposed opendev/gerritbot master: Add option to disable daemonization https://review.opendev.org/745240 | 22:32 |
openstackgerrit | Clark Boylan proposed opendev/gerritbot master: Update docker image jobs to advertise the image they build https://review.opendev.org/745241 | 22:32 |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 22:32 |
clarkb | that may work? | 22:33 |
*** qchris has joined #opendev | 22:34 | |
clarkb | hrm the buildset registry job ahs started before the image build completed. Will it do the right thing in that case? | 22:34 |
clarkb | corvus: ^ | 22:34 |
*** gmann_afk is now known as gmann | 22:35 | |
clarkb | the sun finally came out so I'm going to go for a walk or something. Feel free to update if I got the tying together of those jobs wrong | 22:38 |
donnyd | seems like the expanded flavors are quite popular | 22:55 |
*** tkajinam has joined #opendev | 22:55 | |
donnyd | or there has just been a bunch of jobs waiting on OE to be back up | 22:55 |
*** mlavalle has quit IRC | 22:56 | |
fungi | probably a bit of both | 22:58 |
donnyd | if the custom jobs are popular, I am thinking with my reduced capacity maybe OE is best utilized to support these types of workloads. curious to know what everyone thinks about that | 23:00 |
donnyd | i also am curious if there is a way to find out if it curns through jobs faster with these processors or the old processors | 23:01 |
donnyd | churns* | 23:01 |
clarkb | so we do actually use dumb-init on all our python images | 23:48 |
clarkb | which makes me think the double fork wasn't a problem and intead it was just the logging config? | 23:48 |
clarkb | I'm going to force it to fail testing one more time to review logs and see what it looks like | 23:49 |
openstackgerrit | Clark Boylan proposed opendev/system-config master: Add ansible role to manage gerritbot https://review.opendev.org/744795 | 23:49 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!