| openstackgerrit | Merged zuul/zuul-jobs master: test-upload-logs-swift: add download script as an artifact https://review.opendev.org/713998 | 00:01 |
|---|---|---|
| mordred | we're doing another ansible pulse - so gerrit change should hit on this one | 00:03 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [dnm] test unittests with base-test https://review.opendev.org/714006 | 00:07 |
| * mordred going to get food - back in a bit | 00:13 | |
| openstackgerrit | Merged zuul/zuul-jobs master: Replace occurences of '/' in nodename to something more path friendly https://review.opendev.org/713182 | 00:21 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: upload-logs-swift: Create a download script https://review.opendev.org/592341 | 00:25 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: upload-logs-swift: Add a unicode file https://review.opendev.org/592853 | 00:25 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "upload-logs-swift: test bulk download script change" https://review.opendev.org/714000 | 00:25 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: test-upload-logs-swift: move script artifact after upload https://review.opendev.org/714007 | 00:25 |
| ianw | fungi / clarkb: ^ sorry ... minor ordering error. you really realise how much you rely on the gate when you work on these protected jobs ... | 00:25 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: [dnm] test unittests with base-test https://review.opendev.org/714006 | 00:31 |
| *** tosky has quit IRC | 00:54 | |
| mordred | ok - ansible ran and gerrit is still running - so at least there's no immediate crisis | 01:04 |
| mordred | ianw: I'll look in the morning before we get going - but if you get bored: /home/gerrit2/review_site# diff -Nru etc etc.mt/ | 01:06 |
| mordred | ianw: is the difference between puppet and ansible created config files | 01:06 |
| mordred | the main things seem to be indentation (don't think that matters) and True vs true | 01:08 |
| mordred | there's a key missing that we should probably fix before restarting - that seems to have actual substance | 01:09 |
| mordred | but - like I said - I'll get to that in the morning if nobody else does | 01:09 |
| *** mlavalle has quit IRC | 01:10 | |
| mordred | Ah - GerritSiteHeader got put into etc/etc/ - should sort that out | 01:12 |
| ianw | mordred: yeah, i did a side-by-side and all i saw was the indentation & captialisation , along with addition of restTokenPrivateKey | 01:30 |
| openstackgerrit | Merged zuul/zuul-jobs master: test-upload-logs-swift: move script artifact after upload https://review.opendev.org/714007 | 01:36 |
| mordred | ianw: yeah - I think restTokenPrivateKey is the only important bit | 02:13 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: upload-logs-swift: Create a download script https://review.opendev.org/592341 | 02:16 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: upload-logs-swift: Add a unicode file https://review.opendev.org/592853 | 02:16 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: Revert "upload-logs-swift: test bulk download script change" https://review.opendev.org/714000 | 02:16 |
| openstackgerrit | Ian Wienand proposed zuul/zuul-jobs master: test-upload-logs-swift: fix download link https://review.opendev.org/714016 | 02:16 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add restTokenPrivateKey https://review.opendev.org/714017 | 02:20 |
| mordred | ianw, fungi: ^^ | 02:20 |
| mordred | I think that should sort that setting | 02:21 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add idleTimeout to sshd config section https://review.opendev.org/714018 | 02:24 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Copy the contents of static and etc https://review.opendev.org/714020 | 02:28 |
| mordred | ianw, fungi: ^^ and I think that takes care of the other bits | 02:29 |
| mordred | frickler: if you have a sec when you get up - it would be nice to get those reviewed/landed even so they're good to go in the morning - but also no worry if not | 02:29 |
| * mordred goes to watch tv things ... night everybody | 02:30 | |
| openstackgerrit | Merged zuul/zuul-jobs master: test-upload-logs-swift: fix download link https://review.opendev.org/714016 | 02:40 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: Add restTokenPrivateKey https://review.opendev.org/714017 | 03:22 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: Add idleTimeout to sshd config section https://review.opendev.org/714018 | 03:24 |
| openstackgerrit | Ian Wienand proposed opendev/system-config master: Copy the contents of static and etc https://review.opendev.org/714020 | 03:25 |
| ianw | mordred: minor update to the rest key one, i unstacked the others so they should pass too | 03:25 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/project-config master: Added new project openstack-tempest-skiplist https://review.opendev.org/713809 | 05:37 |
| openstackgerrit | Merged openstack/project-config master: Revise gerrit ACLs for devstack-plugin-open-cas https://review.opendev.org/713974 | 05:51 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/project-config master: Added new project openstack-tempest-skiplist https://review.opendev.org/713809 | 05:59 |
| openstackgerrit | Abhishek Kekane proposed opendev/irc-meetings master: Update glance meeting info https://review.opendev.org/714032 | 06:08 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/project-config master: Added new project openstack-tempest-skiplist https://review.opendev.org/713809 | 06:27 |
| *** DSpider has joined #opendev | 06:53 | |
| *** dpawlik has joined #opendev | 07:21 | |
| *** rpittau|afk is now known as rpittau | 07:57 | |
| *** ralonsoh has joined #opendev | 08:46 | |
| *** tosky has joined #opendev | 08:51 | |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Clean up infra gerritbot irc channel configs https://review.opendev.org/650083 | 09:22 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Clean up infra gerritbot irc channel configs https://review.opendev.org/650083 | 09:32 |
| openstackgerrit | Merged opendev/irc-meetings master: Update glance meeting info https://review.opendev.org/714032 | 10:03 |
| *** lpetrut has joined #opendev | 10:03 | |
| *** hashar has joined #opendev | 10:08 | |
| openstackgerrit | Thierry Carrez proposed openstack/project-config master: check-release-approval: handle no-review case https://review.opendev.org/714066 | 10:27 |
| *** roman_g has joined #opendev | 10:27 | |
| openstackgerrit | Thierry Carrez proposed openstack/project-config master: check-release-approval: handle no-review case https://review.opendev.org/714066 | 10:38 |
| openstackgerrit | Merged opendev/system-config master: Add idleTimeout to sshd config section https://review.opendev.org/714018 | 10:39 |
| openstackgerrit | Merged opendev/system-config master: Copy the contents of static and etc https://review.opendev.org/714020 | 10:39 |
| openstackgerrit | Thierry Carrez proposed openstack/project-config master: check-release-approval: handle no-review case https://review.opendev.org/714066 | 10:40 |
| *** roman_g has quit IRC | 11:03 | |
| *** hashar has quit IRC | 11:24 | |
| *** rpittau is now known as rpittau|bbl | 11:37 | |
| mordred | ianw: thanks! | 11:44 |
| *** dpawlik has quit IRC | 11:50 | |
| *** roman_g has joined #opendev | 12:09 | |
| openstackgerrit | Merged opendev/system-config master: Add restTokenPrivateKey https://review.opendev.org/714017 | 12:18 |
| *** dpawlik has joined #opendev | 12:21 | |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Update gerrit rename playbook for docker-compose https://review.opendev.org/714101 | 12:28 |
| mordred | fungi: whence you awaken ^^ | 12:29 |
| mordred | fungi: that's an update to the rename playbook that should, I think make our tasks slightly easier | 12:29 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds variable to toggle whether to revoke sudo https://review.opendev.org/706248 | 12:39 |
| fungi | mordred: do we expect that manage-projects is still going at the moment? | 12:57 |
| mordred | nope. manage-projects will get done as a followup | 12:58 |
| mordred | my brain just completely rebelled at thinking about it with the puppet stuff still in place | 12:58 |
| mordred | don't know why - it's a pretty simple thing to add :) | 12:59 |
| fungi | mordred: er, no i meant do we expect that manage projects is working right now (pre maintenance) | 13:00 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds variable to toggle whether to revoke sudo https://review.opendev.org/706248 | 13:00 |
| mordred | fungi: oh - yes. | 13:01 |
| mordred | oh - wait | 13:01 |
| fungi | 713974 merged at 05:51z (over 7 hours ago) and didn't seem to take effect | 13:01 |
| fungi | so i was about to start looking into it | 13:01 |
| mordred | no. | 13:01 |
| mordred | we merged the ansible change over 7 hours ago | 13:01 |
| fungi | okay, is it safe for me to manage-projects that one project for now? | 13:02 |
| mordred | at which point nothing will be running manage-projects anymore | 13:02 |
| mordred | fungi: totally | 13:02 |
| fungi | cool, thanks, easy enough for the moment | 13:02 |
| mordred | fungi: everything shoudl still work - just nothing is running it | 13:02 |
| fungi | oh, and also i suspect nothing is updating the copy of project-config on review.o.o now? | 13:04 |
| mordred | oh - I just realized something ... once we're done with this rename and while manage-projects is off ... it might be a good time to move the projects.yaml content from openstack/project-config to opendev/project-config | 13:04 |
| mordred | fungi: yeah - that's almost certainly true | 13:04 |
| fungi | hrm, though the commit shows up as HEAD in /opt/project-config already | 13:05 |
| fungi | manage-projects is saying "openstack/devstack-plugin-open-cas has matching sha, skipping ACLs" | 13:05 |
| fungi | but for whatever reason the new devstack-plugin-open-cas-core it added in the acl isn't showing up in gerrit's groups list | 13:06 |
| fungi | the acl in gerrit hasn't been updated either | 13:07 |
| fungi | maybe i'll find and "corrupt" that sha to see what happens | 13:07 |
| mordred | fungi: oh - so | 13:08 |
| mordred | for hysterical raisins - manage-projects doesn't work ont he file in /opt/project-config | 13:08 |
| mordred | the puppet copies the projects.yaml file to /home/gerrit2 | 13:08 |
| mordred | and same with the acls dirs | 13:09 |
| fungi | ahh | 13:09 |
| fungi | checking there | 13:09 |
| mordred | fungi: (this is one of the reasons I bounced off of fixing manage-projects until we had the basic ansible in place - I kind of think this whole thing can be easier) | 13:09 |
| fungi | yeah, ~gerrit2/acls/openstack/devstack-plugin-open-cas.config has the old data | 13:10 |
| fungi | i'll copy the changed acl file over top that one for now | 13:11 |
| mordred | ++ | 13:11 |
| mordred | frickler, Shrews: mind looking at https://review.opendev.org/#/c/714101/ for this morning's rename fun? | 13:13 |
| Shrews | mordred: +2. didn't know if you wanted to wait for the +A | 13:17 |
| AJaeger | config-core, want to merge another new repo creation: https://review.opendev.org/#/c/713809/5 ? | 13:18 |
| fungi | mordred: manually replacing the acl file and then manually running manage-projects did ultimately update the acl in gerrit and add the new group, thanks | 13:19 |
| fungi | AJaeger: no repo creation changes are taking effect since at least the past 8 hours | 13:19 |
| AJaeger | fungi: oh, missed that - ok, will WIP | 13:20 |
| fungi | since the ansible to update the files in gerrit's homedir and to fire manage-projects isn't being called any longer | 13:20 |
| mordred | manage-projects should be back after not too long | 13:21 |
| mordred | Shrews: thanks! | 13:21 |
| AJaeger | fungi: I know that we couldn't create new repos after the rename - but missed that it's already active and has further implications. I read backscroll now... | 13:21 |
| fungi | AJaeger: yeah, sorry, i don't think any of us considered that either. no worries | 13:22 |
| AJaeger | ;) | 13:22 |
| *** roman_g has quit IRC | 13:22 | |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add an use a utility service for gerrit commands https://review.opendev.org/714111 | 13:30 |
| mordred | fungi: just figured out a trick | 13:30 |
| AJaeger | config-core, another update for the release repo needs review: https://review.opendev.org/714066 ; and please check also ianw's download-script work at https://review.opendev.org/592341 | 13:30 |
| openstackgerrit | Albin Vass proposed zuul/zuul-jobs master: Adds variable to toggle whether to revoke sudo https://review.opendev.org/706248 | 13:31 |
| *** rosmaita has joined #opendev | 13:34 | |
| *** rosmaita has left #opendev | 13:34 | |
| AJaeger | should we sent a pre-warning per IRC about the gerrit downtime? | 13:36 |
| AJaeger | or too late for that, infra-root? ^ I missed that it's already in 24 minutes... | 13:37 |
| mordred | yeah - also - let me go put gerrit into emergency | 13:37 |
| mordred | I'd like to wait for the updated rename playbook to land (should happen in a couple of minutes anyway) before stopping gerrit | 13:40 |
| AJaeger | what about #status notice Gerrit (review.opendev.org) will be down for maintenance starting at 14:00 (in less than 20 mins), probably until 15:00 UTC ß | 13:41 |
| mordred | how's this: #status notice Gerrit going down for scheduled maint in 20 minutes (14:00 UTC) | 13:41 |
| mordred | oh - yours is better | 13:41 |
| AJaeger | #status notice Gerrit (review.opendev.org) will be down for maintenance starting at 14:00 (in less than 20 mins), probably until 15:00 UTC | 13:41 |
| openstackstatus | AJaeger: sending notice | 13:41 |
| fungi | i'm still knocking out some other morning tasks in preparation | 13:41 |
| AJaeger | mordred: thanks - send ;) | 13:41 |
| fungi | but lgtm, thanks AJaeger! | 13:41 |
| -openstackstatus- NOTICE: Gerrit (review.opendev.org) will be down for maintenance starting at 14:00 (in less than 20 mins), probably until 15:00 UTC | 13:41 | |
| * AJaeger also replied to clarkb's email on the dicuss mailing list | 13:43 | |
| mordred | fungi: I thnk it's possible we have no gate jobs defined to be able to land https://review.opendev.org/#/c/713943/1 | 13:44 |
| openstackstatus | AJaeger: finished sending notice | 13:45 |
| fungi | mordred: i thought the etherpad said we were force merging those? | 13:45 |
| fungi | i can't check right this second | 13:45 |
| AJaeger | mordred: we have "noop" defined | 13:45 |
| AJaeger | mordred: I agree, strange that Zuul didn't pick it up | 13:46 |
| openstackgerrit | Merged opendev/system-config master: Update gerrit rename playbook for docker-compose https://review.opendev.org/714101 | 13:46 |
| AJaeger | mordred: https://opendev.org/opendev/project-config/src/branch/master/zuul.d/projects.yaml | 13:46 |
| openstackgerrit | Merged opendev/project-config master: March 20, 2020 project renames https://review.opendev.org/713943 | 13:47 |
| AJaeger | mordred: running now in gate... | 13:47 |
| mordred | neat | 13:47 |
| AJaeger | argh, too slow - already merged ;) | 13:47 |
| mordred | :) | 13:47 |
| AJaeger | mail is slower than IRC ;9 | 13:47 |
| mordred | fungi: when you're ready, I've updated the etherpad with strikethrough to account for the updated rename playbook | 13:49 |
| mordred | clarkb, corvus: ^^ if you happen to be awake, I wouldn't mind a double check on that | 13:50 |
| corvus | o/ | 13:51 |
| mordred | corvus: https://review.opendev.org/714101 is the updated rename playbook to go with https://etherpad.openstack.org/p/gerrit-2020-03-20 - but if we're not comfortable doing that we can also just edit it in place as per the original etherpad | 13:51 |
| *** rpittau|bbl is now known as rpittau | 13:52 | |
| AJaeger | the release team just approved a few releases, guess we need to get them through... | 13:52 |
| corvus | mordred: still need to systemctl stop it the first time, right? but that's easy | 13:52 |
| mordred | yes | 13:52 |
| corvus | mordred: what's gerrit_run_init for? | 13:53 |
| AJaeger | mmh, or is Zuul too busy and I misread the queue? | 13:53 |
| mordred | that's the flag that controls whether we run the init command before running the gerrit itself | 13:53 |
| corvus | mordred: right, but why should we do it here? | 13:53 |
| mordred | corvus: hrm. actually - good point - there's no good reason | 13:54 |
| corvus | i don't think the init script does that | 13:54 |
| mordred | yeah. let's change that to init_command false | 13:54 |
| mordred | it's really just for initial install or upgrade times | 13:54 |
| fungi | oh, right, we're not changing versions of gerrit, just changing where/how it's installed | 13:55 |
| corvus | mordred: other than that, looks good -- my main concern is if it doesn't come up right the first time, but there's a pause after that and we can abort there if necessary | 13:55 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add an use a utility service for gerrit commands https://review.opendev.org/714111 | 13:56 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Don't run gerrit init on restart https://review.opendev.org/714117 | 13:56 |
| mordred | corvus: yah. I agree | 13:56 |
| mordred | there's the patch to update it - but I think we can just do that in a local copy as well | 13:56 |
| mordred | I made a copy of the playbook and updated the etherpad to reflect | 13:58 |
| mordred | corvus: the old rename playbook executed "invoke-rc.d gerrit stop" - is that what we should run this time? | 13:59 |
| fungi | we'll run these from a root screen session on bridge.o.o? | 13:59 |
| fungi | or locally on review.o.o? | 13:59 |
| mordred | bridge | 13:59 |
| corvus | mordred: in that patch, is "gerrit_run_init: false" the default? i hope so, and maybe we could just omit it then? | 13:59 |
| fungi | ahh, yeah, etherpad says bridge is where the yaml is anyway | 14:00 |
| corvus | mordred: i think the usual systemctl commands should work too | 14:00 |
| mordred | yes - it is the default - I've been putting it in placees though just for explcitness :) | 14:00 |
| mordred | corvus: so systemctl stop gerrit? | 14:00 |
| corvus | mordred: ok; seeing it explicitly false makes me worry it might accidentally be true, so that may not be having the desired calming effect, but we can work that out later :) | 14:01 |
| corvus | mordred: ya | 14:01 |
| corvus | and i agree with fungi, a root screen would be good for this | 14:01 |
| mordred | k. there was already one - so I attached to it | 14:01 |
| *** lpetrut has quit IRC | 14:01 | |
| AJaeger | mordred: want to sent a #status alert first? | 14:01 |
| mordred | probably not a bad idea | 14:02 |
| corvus | i forgot about the renames yaml files; we need to update our docs to mention that | 14:02 |
| * corvus makes patch | 14:03 | |
| fungi | status alert The Gerrit service on review.openstack.org is offline for maintenance until 15:00 UTC http://lists.opendev.org/pipermail/service-announce/2020-March/000001.html | 14:03 |
| mordred | like: #status alert Gerrit (review.opendev.org) going down for maintenance, probably until 15:00 UTC | 14:03 |
| *** lpetrut has joined #opendev | 14:03 | |
| mordred | fungi: yours is better | 14:03 |
| fungi | sending | 14:03 |
| fungi | er, opendev not openstack | 14:03 |
| fungi | will fix | 14:03 |
| mordred | corvus: also - the renames files go into opendev/project-config which isn't really anywhere on bridge - so we might want to follow up something something | 14:03 |
| fungi | #status alert The Gerrit service on review.opendev.org is offline for maintenance until 15:00 UTC http://lists.opendev.org/pipermail/service-announce/2020-March/000001.html | 14:03 |
| openstackstatus | fungi: sending alert | 14:03 |
| -openstackstatus- NOTICE: The Gerrit service on review.opendev.org is offline for maintenance until 15:00 UTC http://lists.opendev.org/pipermail/service-announce/2020-March/000001.html | 14:04 | |
| *** ChanServ changes topic to "The Gerrit service on review.opendev.org is offline for maintenance until 15:00 UTC http://lists.opendev.org/pipermail/service-announce/2020-March/000001.html" | 14:04 | |
| fungi | our playbook stops the zuul scheduler too? | 14:05 |
| fungi | or does that need to be done as an additional step so we can reenqueue running jobs which otherwise might complete and fail to report/submit? | 14:05 |
| mordred | uh. I think that might need to be done as an additional step | 14:06 |
| corvus | i think maybe it does not stop zuul because it's "probably fast enough" under normal conditions | 14:06 |
| corvus | those may not apply today | 14:06 |
| fungi | right, for just a rename we don't worry about it | 14:06 |
| corvus | so we may want to save zuul queues | 14:06 |
| corvus | but, maybe we don't actually have to stop it? | 14:07 |
| mordred | yeah. that's a good idea | 14:07 |
| fungi | that works | 14:07 |
| fungi | i can do that now | 14:07 |
| fungi | unless someone is already on it | 14:07 |
| corvus | maybe we can just save queues, and if it's still a reasonably short downtime, just re-enqueue the one or 2 changes that fell out | 14:07 |
| corvus | fungi: all yours | 14:07 |
| mordred | go for it | 14:07 |
| fungi | done | 14:08 |
| mordred | when you're done, I'll stop gerrit | 14:08 |
| mordred | stopping gerrit | 14:08 |
| mordred | sigh. ok - let's do it the other way | 14:08 |
| fungi | it's in ~root/queue.sh on zuul.o.o for when we need it later | 14:08 |
| corvus | oh, we probably need to restart zuul anyway | 14:08 |
| corvus | we're moving keys out from under it | 14:08 |
| fungi | aha | 14:08 |
| corvus | er | 14:08 |
| mordred | yeah. ok. well - I haven't stopped gerrit | 14:08 |
| corvus | actually, that should be handled by a reconfig | 14:08 |
| corvus | so nevermind, let's stick with the plan: do not stop zuul | 14:09 |
| fungi | yeah, we've done it before without a restart | 14:09 |
| mordred | because it turns out stopping gerrit should be done on ... gerrit | 14:09 |
| fungi | hah | 14:09 |
| mordred | doing that now | 14:09 |
| mordred | ( I suppose I could have done that with an ansible adhoc command come to think of it - oh well) | 14:10 |
| mordred | gerrit seems to be down | 14:10 |
| openstackstatus | fungi: finished sending alert | 14:10 |
| mordred | I will now run the playbook - that look good to y'all? | 14:10 |
| corvus | where is this root screen? | 14:10 |
| mordred | on bridge? | 14:11 |
| fungi | yeah, lgtm | 14:11 |
| corvus | oh, there, i found it | 14:11 |
| corvus | lgtm | 14:11 |
| mordred | well that's a good start | 14:11 |
| mordred | is that better? | 14:12 |
| corvus | i missed what changed | 14:12 |
| fungi | my terminal is too narrow to contain the solution, much like fermat's book margin | 14:13 |
| mordred | from -e@renames-20200320.yaml to -erepolist=renames-20200320.yaml | 14:13 |
| fungi | ahh | 14:13 |
| mordred | is that the right invocation? | 14:13 |
| *** roman_g has joined #opendev | 14:13 | |
| fungi | checking | 14:13 |
| corvus | the docs say: sudo ansible-playbook -f 10 /opt/system-config/playbooks/rename_repos.yaml -e repolist=ABSOLUTE_PATH_TO_VARS_FILE | 14:14 |
| corvus | mordred: maybe throw an f10 on there too? | 14:14 |
| mordred | ansible-playbook playbooks/rename_repos_mod.yaml -e repolist=$(pwd)/renames-20200320.yaml | 14:14 |
| fungi | https://docs.openstack.org/infra/system-config/gerrit.html#renaming-a-project says yes | 14:14 |
| corvus | for the gitea servers | 14:14 |
| mordred | ah | 14:14 |
| fungi | we have the command documented there | 14:14 |
| corvus | new cmd lgtm | 14:14 |
| fungi | and it also suggests -f10 | 14:15 |
| mordred | kk: ansible-playbook -f 10 playbooks/rename_repos_mod.yaml -e repolist=$(pwd)/renames-20200320.yaml | 14:15 |
| fungi | lgtm now | 14:15 |
| corvus | mordred: outdent vars | 14:15 |
| corvus | mordred: i'm unsure on tasks_from | 14:15 |
| corvus | mordred: and when you're done, can we see a diff? | 14:16 |
| corvus | mordred: looks like tasks_vars is at the right level, only vars needs to move | 14:16 |
| corvus | er tasks_from | 14:16 |
| corvus | mordred: looks like tasks_from is at the right level, only vars needs to move | 14:17 |
| mordred | uhm. - oh, gotit | 14:17 |
| corvus | yep | 14:17 |
| fungi | that looks better | 14:17 |
| corvus | i'll prepare that change | 14:18 |
| mordred | corvus: grab the start playbook too? (although maybe the start playbook isn't actually needed) | 14:18 |
| corvus | done | 14:19 |
| mordred | so maybe just delete the start playbook | 14:19 |
| corvus | so now we verify gerrit is working | 14:20 |
| mordred | docker-compose logs -f on review is logging things - so it seems to be starting | 14:20 |
| corvus | gertty is able to (very slowly) sync a change, so that's goood | 14:21 |
| mordred | web ui is back up | 14:22 |
| corvus | mordred: where are the rest of the logs? | 14:22 |
| corvus | usual place? | 14:22 |
| mordred | shall we hit returnto continue the playbook? | 14:22 |
| mordred | yup. usual place | 14:22 |
| corvus | hold | 14:22 |
| mordred | kk | 14:23 |
| corvus | [2020-03-20 14:22:24,747] [Gitweb-ErrorLogger] ERROR com.google.gerrit.httpd.gitweb.GitwebServlet : CGI: Can't locate CGI.pm in @INC (you may need to install the CGI module) (@INC contains: /etc/perl /usr/local/lib/x86_64-linux-gnu/perl/5.24.1 /usr/local/share/perl/5.24.1 /usr/lib/x86_64-linux-gnu/perl5/5.24 /usr/share/perl5 /usr/lib/x86_64-linux-gnu/perl/5.24 /usr/share/perl/5.24 /usr/local/lib/site_perl | 14:23 |
| corvus | /usr/lib/x86_64-linux-gnu/perl-base) at /usr/share/gitweb/gitweb.cgi line 13. | 14:23 |
| corvus | from error log | 14:23 |
| corvus | i think that's new | 14:23 |
| fungi | huh, we had disabled gitweb right? | 14:23 |
| fungi | or did we reenable it? | 14:23 |
| mordred | I also thought we had disalbed it | 14:23 |
| corvus | [2020-03-20 14:21:43,155] [HookQueue-1] INFO com.googlesource.gerrit.plugins.hooks.HookTask : hook[change-merged] output: timeout: failed to run command ‘/usr/local/bin/update-bug’: No such file or directory | 14:24 |
| fungi | i guess we need to mount the gerrit hooks in as well? | 14:24 |
| mordred | poo. that's worse. no - the gerrit hooks are mounted - but the hooks run a command which is missing | 14:24 |
| fungi | ahh, that right | 14:24 |
| fungi | not the hooks but the scripts the hooks call | 14:25 |
| mordred | yeah. but we install jeepyb into the container | 14:25 |
| fungi | which are entrypoints from jeepyb | 14:25 |
| corvus | we do have gitweb links; i just followed one to: https://review.opendev.org/gitweb?p=x%2Fwhitebox-tempest-plugin.git;a=commitdiff;h=80adc8c0ca1b2d54b9e567695dd363c4e30d66e8 | 14:25 |
| mordred | so - we do not have jeepyb in the 2.13 image (*wtf*) | 14:26 |
| mordred | I think we should stop gerrit from docker and start with init script | 14:26 |
| mordred | we'll want to sort out cgi and why jeepyb isn't there | 14:27 |
| mordred | and that'll take more than a few minutes | 14:27 |
| corvus | sounds good | 14:27 |
| mordred | blast | 14:27 |
| * mordred goes to run docker-compose down | 14:27 | |
| mordred | ok. it's down | 14:28 |
| mordred | systemctl start gerrit? | 14:28 |
| corvus | sounds good | 14:28 |
| corvus | i assume once we restart with init script, we'll continue the playbook | 14:28 |
| mordred | yeah | 14:28 |
| mordred | gerrit seems to be back up | 14:31 |
| mordred | shall we continue with the playbook? | 14:31 |
| fungi | yeah, looks like we should be ready. just to be clear, we're running from the old gerrit location again, not the container? | 14:32 |
| mordred | yes, that's right | 14:32 |
| corvus | mordred: lgtm.... gertty refreshed the change much faster this time. | 14:32 |
| mordred | I'll push up a patch to revert the puppet->ansible swap | 14:32 |
| mordred | corvus: I wonder what that's about | 14:33 |
| mordred | k. playbook is done | 14:33 |
| corvus | mordred: maybe just random cold cache issues? i can't think of any reason it should be, but it's a point of anecdata we may want to keep in mind. | 14:33 |
| mordred | corvus: I wonder if we're getting some default resource limiting | 14:33 |
| mordred | yeah. | 14:33 |
| corvus | mordred: according to the docker docs: 'By default, a container has no resource constraints and can use as much of a given resource as the host’s kernel scheduler allows." | 14:34 |
| corvus | we raise the file limits on review01; i think those apply to everyone | 14:35 |
| corvus | (so i would expect that to apply to containers too) | 14:36 |
| corvus | let's assume, for the moment, that i just waited longer to perform the test the second time | 14:36 |
| fungi | did we do it in the initscript though? | 14:36 |
| * fungi looks | 14:36 | |
| corvus | fungi: limits.d | 14:36 |
| corvus | oh | 14:37 |
| corvus | there are ulimit calls in the init script | 14:37 |
| mordred | do we need to bindmount that? | 14:37 |
| mordred | oh. hrm | 14:37 |
| corvus | so.... both? | 14:37 |
| mordred | let's put that on the list of things to investigate :) | 14:37 |
| fungi | we do set a bunch of ulimits in the initscript | 14:38 |
| fungi | ahh, you saw that as well | 14:38 |
| fungi | i had a vague memory that was the case, glad to have not imagined it | 14:38 |
| corvus | i made a list on line 25 in the etherpad | 14:38 |
| fungi | are we at the point where i should reenqueue things in zuul? | 14:38 |
| corvus | fungi: i think so | 14:39 |
| fungi | okay, doing that now before its internal state is too out of sync with what we recorded | 14:39 |
| mordred | corvus: we have a comment in the dockerfile: | 14:39 |
| mordred | # Ulimits should be set on command line or in docker-compose.yaml | 14:39 |
| mordred | so apparently at some point we did think about this | 14:39 |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Add a note about rename files to project renames doc https://review.opendev.org/714126 | 14:42 |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Correct vars indentation error in rename playbook https://review.opendev.org/714127 | 14:42 |
| fungi | also the renames are completed at this point, so we're ready to force merge the renames in project-config right? | 14:43 |
| mordred | fungi: yes, I believe so | 14:43 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add an use a utility service for gerrit commands https://review.opendev.org/714111 | 14:44 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Revert "Switch to running gerrit via ansible+containers" https://review.opendev.org/714129 | 14:44 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714130 | 14:44 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Revert "Revert "Switch to running gerrit via ansible+containers"" https://review.opendev.org/714131 | 14:44 |
| mordred | ok. there's a revert to put us back on puppet - and a change to fix the 2.13 container | 14:44 |
| fungi | AJaeger: care to un-wip 711834? | 14:45 |
| fungi | or i can delete your wip vote if you're busy | 14:45 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714130 | 14:46 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add an use a utility service for gerrit commands https://review.opendev.org/714111 | 14:46 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Revert "Revert "Switch to running gerrit via ansible+containers"" https://review.opendev.org/714131 | 14:46 |
| mordred | fungi, corvus: I think we're ok to send the ok alert now, yes? | 14:46 |
| AJaeger | fungi: on it... | 14:47 |
| fungi | mordred: yeah, i can do that now as well | 14:47 |
| corvus | i think so; i think the only remaining things is merging the p-c changes | 14:47 |
| mordred | yup | 14:47 |
| AJaeger | fungi: 711834 is in merge conflict, isn't it? | 14:47 |
| corvus | oh, and i guess the p-c changes trigger the full-reconfigure in zuul, so no manual action required there | 14:47 |
| fungi | status ok Gerrit maintenance is concluded at this time and requested renames have been performed; a follow-up maintenance will be scheduled to continue with further service improvements | 14:48 |
| fungi | something like that? | 14:48 |
| mordred | sounds good | 14:48 |
| corvus | fungi: maybe omit the last part? | 14:48 |
| AJaeger | I'll rebase 711834 on top of https://review.opendev.org/712149 ... | 14:48 |
| fungi | can do, so just the bits before the ; | 14:48 |
| corvus | ya | 14:48 |
| mordred | also - fwiw, once we're happy with the image updates we'll be in a pretty good place to do the docker swap and have it mostly be like a restart | 14:49 |
| corvus | i think it's at least plausible we might decide that we can retry with a simple unscheduled gerrit restart | 14:49 |
| mordred | yeah | 14:49 |
| fungi | we might be able to do the dockerdockergo in a non-scheduled maintenance, yeah | 14:49 |
| fungi | now that we have plenty of experimental results | 14:49 |
| corvus | dock...dock....dock...goose | 14:49 |
| fungi | seems very close anyway, and should be quick to try and unroll again if needed | 14:49 |
| fungi | thanks AJaeger | 14:50 |
| mordred | yeah. in fact, why don't I rebase the stack above to not include the revert - it's possible it might be a quick enough turn around that we just want to stay in emergency | 14:50 |
| fungi | AJaeger: once you push i'm ready to merge them both | 14:50 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714130 | 14:50 |
| AJaeger | fungi: again merge-conflict, I need a rebase as well ;( | 14:51 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add an use a utility service for gerrit commands https://review.opendev.org/714111 | 14:51 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Move openstack/infra-manuals to opendev https://review.opendev.org/712149 | 14:51 |
| AJaeger | fungi: give me a minute, please | 14:51 |
| mnaser | i know gerrit seems down but zuul is not responding | 14:52 |
| mnaser | i assume that might be a side effect of gerrit not responding | 14:52 |
| corvus | mnaser: looking | 14:52 |
| mordred | mnaser: gerrit should be back up | 14:52 |
| fungi | AJaeger: take your time, there's no rush | 14:53 |
| fungi | hrm, yeah my zuul enqueue script has hung as well | 14:53 |
| corvus | it may resume | 14:53 |
| corvus | and i think it just did | 14:54 |
| corvus | it was in an error backoff period because gerrit was returning errors | 14:54 |
| corvus | huh | 14:54 |
| corvus | it just got another 409 | 14:54 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Move openstack/infra-manuals to opendev https://review.opendev.org/712149 | 14:54 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Rename x/devstack-plugin-nfs to openstack/devstack-plugin-nfs https://review.opendev.org/711834 | 14:54 |
| AJaeger | fungi, I'm ready ^ | 14:54 |
| AJaeger | fungi: three minutes instead of 1 - no rush ;) | 14:55 |
| mordred | corvus: that doesn't seem awesome | 14:55 |
| mordred | corvus: oh. I know what it is | 14:55 |
| fungi | the enqueue script seems to be progressing, just very slowly | 14:55 |
| AJaeger | fungi: are you sending a change for .gitreviews - or shall I? | 14:56 |
| mordred | corvus: no - nevermind. I though it was the restTokenPrivateKey thing we fixed and maybe that having not gotten applied to the file - but it's there and has been for long enough to have been picked up :( | 14:56 |
| fungi | AJaeger: i hadn't thought about it, so feel free though it won't get tested until i merge 712149 and 711834 (which i'm doing now) and zuul sees those and reconfigures | 14:57 |
| AJaeger | fungi, will take care of them... | 14:58 |
| corvus | oh i think the 409 is because zuul is trying to merge a merged change | 14:58 |
| corvus | let's see if it eventually gets out of that | 14:58 |
| openstackgerrit | Merged openstack/project-config master: Move openstack/infra-manuals to opendev https://review.opendev.org/712149 | 14:59 |
| corvus | that's a yes | 14:59 |
| mordred | yeah. status page works now | 15:00 |
| corvus | mnaser: looks like it recovered | 15:00 |
| mnaser | cool :) just wanted to bring it up incase anyone hadn't noticed it yet :) | 15:00 |
| * mnaser is human nagios | 15:00 | |
| fungi | #status ok Gerrit maintenance is concluded at this time and requested renames have been performed | 15:00 |
| openstackstatus | fungi: sending ok | 15:00 |
| fungi | forgot to send that | 15:01 |
| openstackgerrit | Merged openstack/project-config master: Rename x/devstack-plugin-nfs to openstack/devstack-plugin-nfs https://review.opendev.org/711834 | 15:01 |
| *** ChanServ changes topic to "OpenDev is a space for collaborative Open Source software development | https://opendev.org/ | channel logs http://eavesdrop.openstack.org/irclogs/%23opendev/" | 15:01 | |
| -openstackstatus- NOTICE: Gerrit maintenance is concluded at this time and requested renames have been performed | 15:01 | |
| fungi | reenqueue script has finished | 15:05 |
| mordred | corvus: I did a quick check on review-dev - inside the gerrit container ulimit reports "unlimited" | 15:07 |
| openstackstatus | fungi: finished sending ok | 15:07 |
| *** hashar has joined #opendev | 15:09 | |
| AJaeger | fungi, https://review.opendev.org/714138 Update .gitreview after rename | 15:09 |
| mordred | ok - there are two settings in the init script that are different in the container on review-dev | 15:09 |
| fungi | thanks AJaeger. if zuul runs tests on that then i think we're in the clear | 15:10 |
| openstackgerrit | Merged opendev/system-config master: Add #openvswitch to meetbot channels https://review.opendev.org/711756 | 15:11 |
| AJaeger | fungi: I'll monitor and ping ;) | 15:11 |
| fungi | it might need a recheck if it was pushed before zuul started a reconfigure | 15:11 |
| AJaeger | yes, that's what I expect | 15:12 |
| AJaeger | https://review.opendev.org/#/c/713929/ is an older review for openstack/infra-manuals - if I go to the URL I now get an internal error 500; do I need to wait a bit for the rename index to finish? | 15:14 |
| * corvus breakfasts; biab | 15:15 | |
| fungi | AJaeger: likely but i'll check | 15:16 |
| fungi | AJaeger: yeah, still backlogged by ~2k tasks | 15:17 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Set ulimits for gerrit https://review.opendev.org/714143 | 15:17 |
| AJaeger | fungi: what are you running? "gerrit show-queue" shows only 18... | 15:18 |
| mordred | corvus, fungi: ^^ ok - there were a couple of differences - nothing that should have made anything slow - but that should make us equiv to the init script | 15:18 |
| fungi | AJaeger: it may not show you all of them depending on your account permissions | 15:18 |
| fungi | AJaeger: and yeah, that's what i'm running as well | 15:18 |
| fungi | for me it reports "1911 tasks" as of a few seconds ago | 15:19 |
| fungi | and most of those are in a waiting state with description "Get changes to reindex caused by ..." | 15:19 |
| AJaeger | fungi: ok, different permissions | 15:22 |
| fungi | once we're upgraded it's probably time to revisit options for exposing gerrit stats somewhere public | 15:23 |
| fungi | hesitant to work on that right now since it may all change in 3.x | 15:23 |
| AJaeger | no worries | 15:23 |
| fungi | yeah, last time i looked at it was a few years ago and i think the best thing i came up with at the time was a custom snmp mib to hit javamelody locally with appropriate credentials | 15:25 |
| fungi | AJaeger: you can probably retry now, it's down to just reindexing nova, neutron and openstack-manuals | 15:26 |
| mordred | yeah. and ... this mornings hiccup notwithstanding - we're actually pretty darned close to being able to upgrade all the way to 3.1 now | 15:27 |
| mordred | probably not until after this release of course | 15:27 |
| AJaeger | fungi: yeah, see the patch - but we need the Zuul reconfigure first | 15:27 |
| *** lpetrut has quit IRC | 15:30 | |
| AJaeger | config-core, another update for the release repo needs review: https://review.opendev.org/714066 ; and please check also ianw's download-script work at https://review.opendev.org/592341 | 15:41 |
| corvus | mordred: i think i've reveiewed all the non-revert changes and they lgtm | 15:48 |
| AJaeger | infra-root, we had moved openstack/plugin-devstack-nfs out of openstack to x - and now moved back. github shows the repo as archived, see https://github.com/openstack/devstack-plugin-nfs - do we need to unarchive and fix ? | 15:48 |
| corvus | AJaeger: that sounds reasonable | 15:49 |
| fungi | AJaeger: yeah, i think ttx was the volunteer coordinator for openstack's github mirror so maybe he can just do the unarchiving and then gerrit will replicate to it again? | 15:49 |
| AJaeger | and what about https://github.com/openstack/infra-manual ? That one shouldn't be there, should it? | 15:50 |
| fungi | that can be either deleted or archived, i forget what was done for the others | 15:50 |
| corvus | i'd be okay with that being deleted | 15:50 |
| AJaeger | fungi: ok, let's ask ttx - if he's still around ^ | 15:51 |
| corvus | i had an idea i'd like to bat around, if folks have a minute: | 15:51 |
| ttx | AJaeger: I'm here | 15:51 |
| ttx | AJaeger: so.. unarchive https://github.com/openstack/devstack-plugin-nfs ? | 15:52 |
| AJaeger | ttx, yes | 15:52 |
| AJaeger | ttx, yes, please ;) | 15:52 |
| AJaeger | ttx and do the right thing for infra-manual, please | 15:52 |
| ttx | AJaeger: ok https://github.com/openstack/devstack-plugin-nfs unarchived now | 15:53 |
| ttx | please doublecheck | 15:53 |
| AJaeger | yeah, looks good - let's see when gerrit replicates... | 15:53 |
| mordred | corvus: cool. the ulimits patch and the docker image patch should be safe to land assuming they pass tests | 15:54 |
| mordred | corvus: then we can check them out on review-dev | 15:54 |
| AJaeger | argh, 26 errors now - http://zuul.opendev.org/t/openstack/config-errors | 15:54 |
| ttx | re: infra-manual, what do you want me to do ? | 15:54 |
| mordred | corvus: weren't we going to something something point gitweb links to gitea at some point? | 15:54 |
| corvus | mordred: maybe we were blocked on not having change refs, but maybe we have them now and could proceed? | 15:55 |
| ttx | trying to check what we did for the others in that situation | 15:55 |
| corvus | here's my idea: i was thinking, what's a design session at a ptg? it's people sitting around an etherpad. if we're looking at having more virtual events, maybe a conference system focused on etherpad might help. i think with very small changes, we could set up an instance of jitsi meet which is tied to our etherpad server. so you join, and the screen is mostly an etherpad, but also you have the normal | 15:55 |
| corvus | video conference stuff available (small video windows of participants, desktop sharing, etc). it's all in the web browser with webrtc. and the conference name is the etherpad name, so it's there and accessible both before and after the conference. | 15:55 |
| corvus | i prototyped this locally last night. jitsi meet is easy to run in docker-compose. | 15:56 |
| ttx | looks like for others we pushed a closing commit, like at https://github.com/openstack-infra/irc-meetings | 15:56 |
| ttx | hmm, did y'all create a rename YAML file for this recent renamefest? | 15:59 |
| ttx | If yes I can process it through the tooling we have for archiving things | 16:00 |
| AJaeger | ttx, https://opendev.org/opendev/project-config/src/branch/master/renames/20200320.yaml | 16:00 |
| mordred | corvus: I like the idea | 16:01 |
| AJaeger | mordred, http://zuul.opendev.org/t/openstack/config-errors shows errors in gerrit repo, could you check those, please? | 16:01 |
| ttx | heh short enough | 16:01 |
| *** mlavalle has joined #opendev | 16:01 | |
| * AJaeger fixes the devstack-plugin-nfs config-errors | 16:01 | |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update infra-manual links https://review.opendev.org/713925 | 16:03 |
| mordred | corvus: BWAHAHAHAHAHAHA | 16:03 |
| mordred | corvus: we haev track-upstream still runing for gerrit into opendev/gerrit | 16:03 |
| mordred | corvus: we just landed .zuul.yaml change sin upstream gerrit | 16:04 |
| corvus | mordred: HAHAHAHAHA | 16:04 |
| mordred | which means zuul just tried to load those jobs | 16:04 |
| mordred | right | 16:04 |
| mordred | I think we should a) turn off track-upstream | 16:04 |
| AJaeger | oh, fun ;( | 16:04 |
| corvus | victims of our own success! | 16:04 |
| mordred | and then b) just force-revert that change in our unused upstream/master branch | 16:04 |
| mordred | incidentally - that's the only repo using track-upstream | 16:05 |
| mordred | so we can also just not port that cronjob over to ansible | 16:05 |
| corvus | mordred: maybe instead we could tweak the include for opendev/gerrit? | 16:05 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Stop tracking gerrit upstream https://review.opendev.org/714160 | 16:05 |
| corvus | mordred: then we can continue to track upstream? | 16:06 |
| mordred | corvus: we could also do that - but I really don't think we need the upstream/ branches in our gerrit | 16:06 |
| mordred | I mean - we can't build 2.13 anymore | 16:06 |
| mordred | so we're actuallynot using the repo for any purpose | 16:06 |
| corvus | mordred: clarkb pointed out yesterday we still have local patches | 16:07 |
| mordred | only in 2.13 | 16:07 |
| mordred | which we can't build because the toolchain has rotted | 16:07 |
| mordred | (this is why the 2.13 image just wget's the war) | 16:07 |
| AJaeger | mordred: what about the two "upstream" lines? Can we remove them as well? I left a comment... | 16:07 |
| corvus | mordred: well, then there's no reason to even have a gerrit repo | 16:07 |
| mordred | corvus: I agree | 16:07 |
| corvus | mordred: i'm still not sure i'm on board here | 16:07 |
| corvus | we may yet need to patch gerrit | 16:08 |
| mordred | k. I'm also fine with editing the gerrit include line | 16:08 |
| corvus | why not leave everything the way it was so we don't have to figure out a new way to make a local modification? | 16:08 |
| mordred | sure - shall we just change the zuul config then? | 16:08 |
| corvus | yeah -- and i think maybe after we're running 3.1 with no patches, i'll be comfortable shutting down the local gerrit repo | 16:09 |
| mordred | kk | 16:09 |
| ttx | cleanup done at https://github.com/openstack/infra-manual | 16:09 |
| AJaeger | ttx, thanks! | 16:09 |
| mordred | corvus: maybe just load no objects from opendev/gerrit since we're doing gerrit build jobs in opendev/system-config anyway? | 16:09 |
| corvus | mordred: ++ | 16:11 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Don't load zuul objects from opendev/gerrit https://review.opendev.org/714160 | 16:12 |
| corvus | mordred, fungi, ttx, AJaeger: here's a screenshot of my local prototypo: https://imgur.com/a/7OqtgbV (openstack etherpad corvustest1) | 16:12 |
| corvus | oh, i did save full page, not visible, so the bottom looks weird, but you get the idea | 16:14 |
| fungi | corvus: local prototype of deploying etherpad in a container? | 16:14 |
| corvus | fungi: no, local prototype of deploying jitsi-meet bound to our etherpad server | 16:14 |
| fungi | ahh! | 16:14 |
| fungi | i was missing context | 16:14 |
| corvus | etherpad-oriented meetings | 16:14 |
| corvus | sorry, i thought i picked a quiet moment to ask for feedback | 16:15 |
| fungi | now i see that in scrollback | 16:15 |
| corvus | but that was 80 lines ago | 16:15 |
| mordred | corvus: there are no quiet moments | 16:15 |
| mordred | fungi: if I could bother you for reviews: https://review.opendev.org/#/c/714117/ | 16:16 |
| fungi | corvus: yeah, you picked the same quiet moment i picked to call the car rental company to find out why they listed my car return a day later than i had dropped it off, sorry :/ | 16:16 |
| AJaeger | corvus: nice! | 16:16 |
| mordred | fungi: and https://review.opendev.org/#/c/714143/ | 16:16 |
| mordred | (both from this morning's fun) | 16:16 |
| ttx | corvus: that sounds interesting... Would there still be the possibility to project slides ? | 16:16 |
| ttx | or share screen ? | 16:16 |
| corvus | ttx: yes, i think you can still share desktop | 16:17 |
| AJaeger | corvus: can you make the URL configurable? So, I could use another wiki instance or something similar? | 16:17 |
| ttx | My attention is now drawn to imgur's most viral images | 16:17 |
| corvus | ttx: all images are viral now | 16:18 |
| ttx | wow dad | 16:18 |
| ttx | too soon | 16:18 |
| fungi | my experience with teleconferencing tools is limited since i'm pretty much always just doing voice-only, so i really don't know if this would satisfy what folks expect. i had assumed everyone was addicted to the "brady bunch" webcam view for video conference calls these days | 16:19 |
| corvus | AJaeger: not really -- the etherpad integration is pretty specific | 16:19 |
| corvus | fungi: that's there | 16:19 |
| AJaeger | corvus: ok | 16:19 |
| corvus | fungi: the little letters in circles on the right are the brady cams | 16:19 |
| ttx | fungi: I find the video link helpful, as a non-native speaker. | 16:19 |
| fungi | ahh | 16:19 |
| AJaeger | mordred: still the same question on your review ;) | 16:19 |
| corvus | fungi: i have no cameras locally | 16:19 |
| fungi | yeah, it's mostly just complete lack of familiarity with the ui on my part that i didn't know what i was looking at | 16:20 |
| ttx | Body language helps me detect clues like tension that you might be able to pick as a native speaker | 16:20 |
| fungi | i basically saw etherpad in that screenshot, and some random stuff surrounding it i didn't recognize ;) | 16:20 |
| mordred | AJaeger: oh - whoops | 16:20 |
| AJaeger | can I get a second +2A on https://review.opendev.org/714157 to fix jobs after the devstack-plugin-nfs rename, please? | 16:20 |
| corvus | basically, the etherpad would, by default, take center stage, but it can be hidden and the normal web conf paradigm of dominant speaker taking center stage (or the other choices, like filmstrep, etc) would prevail. | 16:20 |
| corvus | ttx: so you could click the etherpad off, and click a talking head and watch them | 16:21 |
| openstackgerrit | Monty Taylor proposed openstack/project-config master: Don't load zuul objects from opendev/gerrit https://review.opendev.org/714160 | 16:21 |
| mordred | AJaeger: thanks | 16:21 |
| AJaeger | you're welcome, mordred | 16:21 |
| ttx | corvus: yeah -- was answering to fungi's disinterest for video data | 16:21 |
| corvus | (also, i have done no research or experimentation with this, but i've seen configuration settings related to captioning; so, maybe, at some point in the future, auto-captioning might be an option) | 16:22 |
| fungi | oh, it's not so much disinterest as i've not yet been sufficiently motivated to hunt down camera and microphone hardware | 16:22 |
| fungi | on my to do list though | 16:22 |
| fungi | i think i have a 4vl-supported usb webcam in a drawer | 16:22 |
| fungi | just need to work out where to mount it | 16:22 |
| corvus | sounds like there is >0 interest, thanks! i'll write up a spec and an implementing change | 16:24 |
| ttx | corvus: yes, I'm very excited | 16:24 |
| ttx | like you can see on my video feed | 16:24 |
| ttx | I need to play with hand signals and see if they could work to avoid the "everyone talking at the same time" issue | 16:25 |
| corvus | ttx: jitsi-meet has a "raise your hand" button | 16:25 |
| ttx | yeah, I know, just never tested it and see how practical that would be | 16:26 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update jobs for infra-manual https://review.opendev.org/714166 | 16:30 |
| *** hashar has quit IRC | 16:32 | |
| *** hashar has joined #opendev | 16:32 | |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714130 | 16:33 |
| *** hashar has quit IRC | 16:35 | |
| *** hashar has joined #opendev | 16:35 | |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Update gerrit-base to buster https://review.opendev.org/714171 | 16:37 |
| AJaeger | config-core, three fixes from the repo renames, please review https://review.opendev.org/714160 https://review.opendev.org/714166 and https://review.opendev.org/714157 | 16:39 |
| mordred | ttx: we could get everyone to learn scuba hand signals | 16:39 |
| mordred | ttx: I mean, they don't cover everything, but they will let us know when to look around the corner at the interesting shark, or when someone is running low on air | 16:40 |
| corvus | mordred: o/ | 16:40 |
| ttx | [o] | 16:41 |
| openstackgerrit | Jeremy Stanley proposed openstack/project-config master: Add #openvswitch channel to accessbot config https://review.opendev.org/714174 | 16:43 |
| *** openstack has joined #opendev | 16:57 | |
| *** ChanServ sets mode: +o openstack | 16:57 | |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Set ulimits for gerrit https://review.opendev.org/714143 | 16:57 |
| mordred | corvus, fungi: ^^ it's ulimits. not ulimit | 16:57 |
| fungi | d'oh! | 16:59 |
| AJaeger | fungi, mordred, we need three fixes from the repo renames, please review https://review.opendev.org/714160 https://review.opendev.org/714166 and https://review.opendev.org/714157 | 16:59 |
| corvus | those are all approved now | 17:13 |
| AJaeger | thanks, mordred and corvus ! | 17:13 |
| AJaeger | and now one change for infra-manual itself for review, please: https://review.opendev.org/#/c/713929/ | 17:14 |
| openstackgerrit | Merged openstack/project-config master: Don't load zuul objects from opendev/gerrit https://review.opendev.org/714160 | 17:29 |
| openstackgerrit | Merged openstack/project-config master: Update jobs for infra-manual https://review.opendev.org/714166 | 17:29 |
| AJaeger | infra-root, and now one change for infra-manual itself for review, please: https://review.opendev.org/#/c/713929/ | 17:48 |
| corvus | i had to restart gertty for that one | 18:01 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 18:07 |
| mordred | so - things I learned (or relearned) about gerrit | 18:08 |
| mordred | gerrit init extracts plugins from the war file into the plugins dir - and also downloads the appropriate database jar | 18:08 |
| *** hashar has quit IRC | 18:08 | |
| corvus | oh neat | 18:09 |
| mordred | neither of these things are things we want a containerized gerrit to do at runtime - since the invocation of the init command will be in an ephemeral container ... and the results of that will always be the same for a given war file | 18:09 |
| *** hashar has joined #opendev | 18:09 | |
| mordred | so the patch above does the extraction and download in the build - and then tells init to not do that please | 18:09 |
| mordred | corvus: incidentally, this means we don't have to worry about the stretch/buster mysql/mariadb thing | 18:10 |
| mordred | oh - I should rebase that | 18:10 |
| corvus | mordred: on it | 18:11 |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 18:11 |
| corvus | mordred: sorry, i was doing that while you were typing your gerrit lessons; i figured i could do it without distracting you | 18:11 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 18:12 |
| corvus | oh well | 18:12 |
| mordred | oh. | 18:12 |
| mordred | jinx | 18:12 |
| mordred | corvus: whatcha wanna bet we fixed it the same way? | 18:12 |
| corvus | no bet | 18:12 |
| corvus | mordred: iirc the upstream gerrit dockerfile runs init during the build? or is it on launch? at any rate, your solution of explicitly downloading and unzipping sounds like the best | 18:13 |
| mordred | I think on launch? like - I think they just run init every time? | 18:14 |
| mordred | oh - no - they run it at build | 18:16 |
| mordred | and then remove some stuff | 18:16 |
| mordred | https://github.com/GerritCodeReview/docker-gerrit/commit/a631a208a578a070bd953a1b61372679e3bb0717 | 18:16 |
| corvus | oh hrm | 18:16 |
| mordred | corvus: thing is - that won't work for us right now - because gerrit needs to see mysql in the config to know it needs to download the mysql library | 18:16 |
| corvus | mordred: i guess that rm would take care of it. do you think that's a better app...nevermind | 18:16 |
| mordred | I think we could adopt that once we're post-db with 3.1 | 18:16 |
| corvus | k | 18:17 |
| mordred | corvus: container.javaOptions "-Djava.security.egd=file:/dev/./urandom" ? should we do something similar? | 18:17 |
| *** hashar has quit IRC | 18:18 | |
| *** hashar has joined #opendev | 18:18 | |
| corvus | mordred: WOW https://stackoverflow.com/questions/58991966/what-java-security-egd-option-is-for | 18:19 |
| corvus | the /./ is a thing | 18:19 |
| mordred | yeah | 18:19 |
| corvus | mordred: i'd go with "yes, as written" | 18:19 |
| *** roman_g has quit IRC | 18:20 | |
| corvus | with maybe a link to the SO answer https://stackoverflow.com/a/59097932 so we don't optimize out the /./ | 18:20 |
| corvus | should we call it "meet.opendev.org" or "meetpad.openstack.org" ? | 18:22 |
| mordred | I kindof like meetpad :) | 18:23 |
| AJaeger | covid-meet? ;) | 18:23 |
| corvus | AJaeger: sadface | 18:23 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add java.security.egd setting to java invocation https://review.opendev.org/714183 | 18:24 |
| corvus | i was thinking that a special name might let people know know it's intended to be used, and how it's different than just "self-hosted conference server". but also, it could be distracting and weird. | 18:24 |
| AJaeger | corvus: both those names work, and the way it's implemented, meetpad is the better fit | 18:26 |
| corvus | (also, we could have a bbq themed meatpad on april 1) | 18:27 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add support for setting heap limit properly https://review.opendev.org/714184 | 18:29 |
| AJaeger | corvus: clarkb's best smoker recipes? ;) | 18:29 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 18:33 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add java.security.egd setting to java invocation https://review.opendev.org/714183 | 18:33 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add support for setting heap limit properly https://review.opendev.org/714184 | 18:33 |
| mordred | corvus: sometimes you need a \ at the end of a line | 18:33 |
| corvus | mordred: && sometimes you don't | 18:34 |
| AJaeger | fungi: https://review.opendev.org/#/c/706599/ merged for airship but I don't see content at https://static.opendev.org/project/airshipit.org/ - the docs directory is missing. The promote job succeeded. ianw had a comment on missing docs dir at https://review.opendev.org/#/c/706598/3/zuul.d/jobs.yaml . Could you check this, please? | 18:36 |
| *** hashar is now known as hasharAway | 18:49 | |
| fungi | AJaeger: /afs/.openstack.org/project/airshipit.org/docs/ exists for me and has content | 18:56 |
| fungi | but | 18:57 |
| fungi | ls: cannot access '/afs/openstack.org/project/airshipit.org/docs': No such file or directory | 18:57 |
| fungi | so it's in the rw volume but not the ro replica | 18:57 |
| openstackgerrit | James E. Blair proposed opendev/infra-specs master: Add a spec for meetpad https://review.opendev.org/714189 | 18:59 |
| fungi | AJaeger: i'll check server logs in a bit, but my first guess is that something has broken our periodic vos release | 18:59 |
| fungi | or maybe it's not covering that volume | 18:59 |
| fungi | yep, that's the problem | 19:01 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Put in image depends between run-review and gerrit builds https://review.opendev.org/714192 | 19:03 |
| mordred | corvus: ^^ just noticed we were missing that. does that look right to you? if so, I'll rebase it to the bottom of the stack so we can actually see the 2.13 patch properly tested | 19:04 |
| fungi | AJaeger: yeah, that's what it was, i'll add it to the list of volumes to periodically release with a revision on playbooks/roles/afs-release/files/release-volumes.py: | 19:04 |
| fungi | er, a revision on 706601 | 19:05 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 19:09 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add java.security.egd setting to java invocation https://review.opendev.org/714183 | 19:09 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add support for setting heap limit properly https://review.opendev.org/714184 | 19:09 |
| mordred | yeah - it's got the right behavior in the gate - squashing wtih another thing | 19:09 |
| openstackgerrit | Jeremy Stanley proposed opendev/system-config master: Add a new docs.airshipit.org vhost on static01 https://review.opendev.org/706601 | 19:09 |
| fungi | AJaeger: ^ | 19:09 |
| *** hasharAway has quit IRC | 19:10 | |
| fungi | i'll manually vos release it now so that can merge in one go rather than needing to wait for a separate change to merge | 19:10 |
| AJaeger | fungi, thanks! | 19:12 |
| AJaeger | corvus: your infra-specs change forgot to git add the meatbot.rst file | 19:13 |
| fungi | and /afs/openstack.org/project/airshipit.org/docs now has content as well | 19:14 |
| AJaeger | fungi: indeed, that looks fine | 19:14 |
| AJaeger | fungi, could I trouble you with https://review.opendev.org/#/c/713929/ , please? Still untangling the infra-manual rename... | 19:16 |
| *** ralonsoh has quit IRC | 19:17 | |
| fungi | sure thing | 19:17 |
| AJaeger | thanks | 19:23 |
| openstackgerrit | James E. Blair proposed opendev/infra-specs master: Add a spec for meetpad https://review.opendev.org/714189 | 19:40 |
| AJaeger | infra-root, we publish the manual now to https://docs.opendev.org/opendev/infra-manual/latest/ - are we ok with the /latest/ ? Same for infra-specs... | 19:43 |
| fungi | i don't think we intend to version either of those | 19:43 |
| mordred | AJaeger: it doesn't bother me - but it's also maybe not needed. those are both things without a release series - so while latest *is* correct, it might also imply that there would be something that != latest | 19:43 |
| mordred | what fungi said | 19:43 |
| fungi | if it's easier to keep them consistent with versioned docs jobs, then it doesn't bother me | 19:44 |
| AJaeger | we had so far special infra-docs publish jobs that did not use latest - I just noticed it. | 19:45 |
| AJaeger | if we have latest, I need to change some redirects in flight ;) But can easily do so... | 19:45 |
| AJaeger | Just want to avoid doing it twice :) | 19:45 |
| fungi | i'm okay with not having "special" jobs for our docs | 19:46 |
| fungi | it's one place fewer to have to fiddle additional bits | 19:46 |
| * AJaeger has to leave now, have a great weekend and stay healthy | 19:48 | |
| corvus | AJaeger: thanks, you too! | 19:48 |
| * AJaeger will read backscroll and update changes accordingly | 19:48 | |
| fungi | thanks AJaeger! | 19:48 |
| AJaeger | yeah, openstack has no further config errors - really time to sign off ;) | 19:49 |
| AJaeger | thanks & bye | 19:49 |
| mordred | corvus: if you have a sec - check out zuul status for system-config atm | 19:49 |
| mordred | AJaeger: have a great weekend and stay healthy yourself! | 19:49 |
| mordred | corvus: note that opendev-buildset-registry is not paused - but service-review is still running - I feel liek I didn't do something good | 19:50 |
| mordred | corvus: (it's also on 2nd attempt - oh, change 714171,6 is what I'm watching) | 19:51 |
| corvus | mordred: looking | 19:52 |
| mordred | corvus: oh - I need to add opendev-buildset-registry to its depends don't I? | 19:52 |
| mordred | hrm. it's actually an error "Destination directory /run/user/1000 does not exist" | 19:57 |
| fungi | that seems like a systemd (maybe logind specifically?) error | 19:58 |
| fungi | or at least that's the only thing i'm familiar with which wants to write stuff in /run/user/$UID | 19:58 |
| mordred | yeah - that was from TASK [use-buildset-registry : Write containers auth configuration] | 19:58 |
| corvus | mordred: yes, the run job needs a dependency on the registry, but also, i think we have more requires/provides than necessary, gimme a min to review | 19:59 |
| mordred | corvus: kk. thanks | 20:00 |
| corvus | mordred: actually, i think the provides/requires are ok. so just the dep issue. | 20:00 |
| mordred | kk. I'll try that | 20:00 |
| corvus | mordred: so what's the deal with run/user? | 20:01 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 20:01 |
| mordred | corvus: that's an error I saw when streaming the logs | 20:01 |
| mordred | but it might be irrelevant | 20:01 |
| corvus | i guess it's too late to ask that you hold off on uploading for a second so we can get logs :) | 20:02 |
| corvus | if it is an issue, i was hoping we could debug it now | 20:02 |
| mordred | yeah - sorry - for some reason I thought it had reported already | 20:03 |
| corvus | well, there is a 6mb json file i can comb through | 20:04 |
| mordred | corvus: dependencies are additive, yeah? I just noticed that like a noob I have dependencies for that job set in two places | 20:05 |
| corvus | mordred: yes | 20:05 |
| mordred | awesome | 20:05 |
| mordred | this time it's going to work. I can feel it | 20:06 |
| corvus | mordred: i think that error is an error | 20:06 |
| mordred | oh no | 20:06 |
| corvus | i don't know why it doesn't work, but i think it's expected to | 20:07 |
| mordred | corvus: well - I will avoid uploading any new patchsets so that we can see the error in all of its rendered glory | 20:08 |
| corvus | mordred: assuming these jobs worked before, i don't see anything in 714171 that should affect the ability to write to /run/user | 20:08 |
| mordred | me either - and the job definitely worked before | 20:13 |
| mordred | but - you know - let's hold the breath | 20:13 |
| corvus | mordred: use-buildset-registry is only run if a buildset-registry is running; perhaps in the past we have not run both an image build and the run-review job in the same buildset. but we did here, and since run-review depends on the image build job which depends on the buildset registry, it got the registry info (but maybe since it's a direct dependency, it didn't cause the registry to remain paused?), | 20:19 |
| corvus | and therefore ran the role for the first time. | 20:19 |
| corvus | mordred: if that's all correct then the current ps is both correct and will fail | 20:20 |
| corvus | possibly because we've never tried running use-buildset-registry on a xenial node before | 20:20 |
| mordred | ahhh | 20:20 |
| corvus | so we'll need to update that in zuul jobs and depend on it | 20:20 |
| mordred | that would be a good explanation | 20:20 |
| mordred | yay we're helping! | 20:20 |
| corvus | ( http://zuul.opendev.org/t/openstack/buildset/b08da9d7482c451ab26e8aceef1c8c04 is the most recent successful build, and that buildset does not have any image build jobs) | 20:21 |
| mordred | corvus: (and yes, in fact, only recently did that job become a xenial node - and I'm pretty sure ... yeah - I don't thing we've built images since we updated the run-review job to be on xenial) | 20:21 |
| mordred | so we're making rhings more correct yay! | 20:22 |
| corvus | the error in the role is about setting up a path needed for k8s | 20:22 |
| corvus | it's not that important, maybe we just wrap it in a "does dir exist?" check | 20:22 |
| mordred | yeah. becuase things we're not really concerned with - supporting k8s on xenial | 20:22 |
| corvus | oh that might actually be for crio | 20:23 |
| corvus | but still | 20:23 |
| mordred | yeah | 20:23 |
| corvus | if it's used, it should fail, so this should be a safe change to make | 20:23 |
| mordred | ++ | 20:23 |
| corvus | (this may be another reason to stick with docker until we're more upgraded) | 20:24 |
| openstackgerrit | sebastian marcet proposed opendev/system-config master: OpenStackId production release 3.0.5 https://review.opendev.org/714215 | 20:28 |
| openstackgerrit | James E. Blair proposed zuul/zuul-jobs master: use-buildset-registry: protect against /var/user/1000 not existing https://review.opendev.org/714216 | 20:28 |
| corvus | mordred: ^ i don't think we have a xenial test for that, so it might be good to go ahead and depends-on that | 20:30 |
| mordred | corvus: want me to wait for the current PS to fail? | 20:30 |
| mordred | or just update it? | 20:30 |
| corvus | mordred: i'm 90% confident enough that i read the json correctly and the current one will fail and we understand why, so i'd say go for it. | 20:31 |
| mordred | kk | 20:33 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 20:33 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update infra-manual links https://review.opendev.org/713925 | 20:42 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add java.security.egd setting to java invocation https://review.opendev.org/714183 | 20:43 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Add support for setting heap limit properly https://review.opendev.org/714184 | 20:43 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Only download bouncycastle for 2.13 https://review.opendev.org/714223 | 20:43 |
| openstackgerrit | Andreas Jaeger proposed opendev/system-config master: Update infra-manual location https://review.opendev.org/714229 | 21:03 |
| mordred | corvus: so - buildset registy shows success - not paused, even though there are image builds still running | 21:04 |
| mordred | corvus: but system-config-build-image-gerrit-base shows paused | 21:04 |
| mordred | corvus: that seems weird to my brianhole | 21:04 |
| mordred | I'm I over thinking it? | 21:04 |
| corvus | mordred: the registry should be the last to finish | 21:05 |
| mordred | yeah. that would be my expectation | 21:06 |
| corvus | mordred: https://236fa39aba5b694d3ffa-c578cacb39dd1edf606b634ec77d1998.ssl.cf1.rackcdn.com/714171/8/check/opendev-buildset-registry/7880f49/zuul-info/inventory.yaml | 21:06 |
| corvus | that shows child_jobs: [] | 21:06 |
| openstackgerrit | Andreas Jaeger proposed openstack/project-config master: Update infra-manual links https://review.opendev.org/713925 | 21:08 |
| mordred | well - run-review has it directly on the job, and build-gerrit-2.13 has if via system-config-build-image ... but in both of those cases there is both the dependency on the job and dependencies additionally in the pipeline config ... the pipeline config depends seem to be working right | 21:08 |
| corvus | mordred: that's pretty weird and i don't understand that. | 21:08 |
| mordred | so only thing I could thikn is that maybe there is some edge case where the combo of pipeline depends and direct job depends for the same job are overwriting each other? | 21:09 |
| corvus | mordred: maybe dependencies are not combined but overwritten | 21:11 |
| mordred | yeah | 21:11 |
| mordred | I'm gonna push upa dnm to test the theory | 21:12 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: DNM - Testing dependency https://review.opendev.org/714230 | 21:12 |
| mordred | corvus: that adds the depend directly to the pipeline config | 21:12 |
| corvus | mordred: the docs do not say that they are combined, which should mean that they are overidden | 21:13 |
| corvus | mordred: and i think the code backs that up. sorry for my misinformation earlier. | 21:14 |
| mordred | corvus: no worries | 21:14 |
| mordred | corvus: that said - opendev-buildset-registry in status output is currently showing no dependencies | 21:14 |
| mordred | oh - wait | 21:14 |
| mordred | child_jobs | 21:14 |
| corvus | mordred: 714230 has image jobs waiting while the registry is queued; that's a good sign | 21:16 |
| openstackgerrit | Merged opendev/system-config master: OpenStackId production release 3.0.5 https://review.opendev.org/714215 | 21:18 |
| mordred | corvus: so next I'd expect to see base build with registry paused, then 2.13, then run-review | 21:18 |
| corvus | so far so good | 21:20 |
| mordred | yup | 21:20 |
| mordred | corvus: you know - part of my brain really wants to see the dag here ... the other part of my brain knows that's it's a full dag with some complicated relationship possibilities and doesn't want to make the view worse | 21:21 |
| corvus | mordred: ayup. that about sums it up. :) | 21:26 |
| corvus | mordred: might be worth some experiments on the buildset page | 21:27 |
| corvus | it would still be useful debugging info after everything has completed, and we've got a bit more room to work with there | 21:27 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed opendev/lodgeit master: [WIP] Ad options to delete and find pastes by date https://review.opendev.org/714233 | 21:29 |
| mordred | corvus: yeah - I was thinking a similar thing - or even on the expanded single-change status page? | 21:29 |
| mordred | corvus: https://zuul.opendev.org/t/openstack/status/change/714230,1 has plenty of whitespace :) | 21:29 |
| mordred | corvus: we are still in a valid combination | 21:30 |
| corvus | mordred: i think it's progressed enough it's worth it to apply that to your real change | 21:32 |
| mordred | corvus: yup. working on that now | 21:36 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Base 2.13 image on gerrit-base https://review.opendev.org/714171 | 21:38 |
| mordred | corvus: I squashed the followup changes in | 21:38 |
| mordred | because they're not that big and I don't think the churn is that worth it | 21:38 |
| mordred | so this one should work and produce the correct content this time | 21:38 |
| mordred | corvus: I didn't fix the ones that aren't related to gerrit yet - I'll do that in a follow up | 21:42 |
| fungi | #status log removed openstackid01.openstack.org from ansible emergency disable list to apply https://review.opendev.org/714215 | 21:44 |
| openstackstatus | fungi: finished logging | 21:44 |
| openstackgerrit | Merged opendev/system-config master: Add a new docs.airshipit.org vhost on static01 https://review.opendev.org/706601 | 22:07 |
| *** DSpider has quit IRC | 22:12 | |
| mordred | fungi, corvus: ZOMG https://review.opendev.org/#/c/714171/ is green | 22:19 |
| mordred | it should be fairly safe to approve regardless - it'll only apply to review-dev | 22:19 |
| mordred | (and also it ran with at least _some_ functional testing) | 22:20 |
| corvus | mordred: yaaay! | 22:20 |
| openstackgerrit | Merged zuul/zuul-jobs master: use-buildset-registry: protect against /var/user/1000 not existing https://review.opendev.org/714216 | 22:21 |
| openstackgerrit | Monty Taylor proposed opendev/system-config master: Re-enable gerrit master image job https://review.opendev.org/714237 | 22:31 |
| *** diablo_rojo__ has joined #opendev | 23:25 | |
| *** diablo_rojo has quit IRC | 23:25 | |
| openstackgerrit | James E. Blair proposed opendev/system-config master: Add meetpad server https://review.opendev.org/714238 | 23:31 |
| *** tosky has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!