| opendevreview | Ian Wienand proposed zuul/nodepool master: Require DIB 3.13.0 https://review.opendev.org/c/zuul/nodepool/+/801930 | 04:11 |
|---|---|---|
| *** bhagyashris__ is now known as bhagyashris | 04:41 | |
| *** marios is now known as marios|ruck | 05:20 | |
| *** marios|ruck is now known as marios | 07:01 | |
| *** marios is now known as marios|ruck | 07:05 | |
| *** rpittau|afk is now known as rpittau | 07:35 | |
| *** marios_ is now known as marios | 08:06 | |
| *** marios is now known as marios|ruck | 08:08 | |
| *** jcapitao is now known as jcapitao_afk | 12:14 | |
| *** mgoddard- is now known as mgoddard | 12:18 | |
| *** jcapitao_afk is now known as jcapitao | 12:44 | |
| opendevreview | Simon Westphahl proposed zuul/zuul master: Use a temp ZK config cache for tenant validation https://review.opendev.org/c/zuul/zuul/+/800800 | 13:17 |
| swest | clarkb ^ changed the interface as proposed | 13:19 |
| *** rpittau is now known as rpittau|afk | 14:13 | |
| clarkb | swest: thanks, change lgtm | 14:33 |
| *** jcapitao1 is now known as jcapitao | 14:43 | |
| *** holser is now known as holser_ | 15:24 | |
| opendevreview | Merged zuul/nodepool master: Require DIB 3.13.0 https://review.opendev.org/c/zuul/nodepool/+/801930 | 15:43 |
| opendevreview | Merged zuul/zuul master: Use a temp ZK config cache for tenant validation https://review.opendev.org/c/zuul/zuul/+/800800 | 15:48 |
| *** holser_ is now known as holser | 15:51 | |
| *** marios|ruck is now known as marios | 16:03 | |
| *** marios is now known as marios|out | 16:08 | |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Support imagePrefix and versions https://review.opendev.org/c/zuul/zuul-operator/+/785279 | 16:51 |
| corvus | tristanC: ^ that's fixed now and i replied to your comment | 17:26 |
| corvus | i'm about to patchbomb the rest of the stack | 17:26 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Support fingergw https://review.opendev.org/c/zuul/zuul-operator/+/785300 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Add docs https://review.opendev.org/c/zuul/zuul-operator/+/785083 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Support zuul-preview https://review.opendev.org/c/zuul/zuul-operator/+/785760 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Add support for zuul-registry https://review.opendev.org/c/zuul/zuul-operator/+/785761 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Remove extra 2 minute wait from tests https://review.opendev.org/c/zuul/zuul-operator/+/785762 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Add allowUnsafeConfig database setting https://review.opendev.org/c/zuul/zuul-operator/+/785764 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Pass through environment to scheduler, web and launcher https://review.opendev.org/c/zuul/zuul-operator/+/785988 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Allow terminationGracePeriodSeconds to be configurable https://review.opendev.org/c/zuul/zuul-operator/+/785989 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Flake8 cleanups https://review.opendev.org/c/zuul/zuul-operator/+/786349 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Fix error with multiple nodepool providers https://review.opendev.org/c/zuul/zuul-operator/+/799917 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Add instructions and tools for running tests with kind https://review.opendev.org/c/zuul/zuul-operator/+/785763 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Add static node to functional test https://review.opendev.org/c/zuul/zuul-operator/+/799874 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Mount connection sshkeys on executors and mergers https://review.opendev.org/c/zuul/zuul-operator/+/799998 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Set component command with args instead of command https://review.opendev.org/c/zuul/zuul-operator/+/800263 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Configure debug logs for merger https://review.opendev.org/c/zuul/zuul-operator/+/800264 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Make nodepool external_config mount more generic https://review.opendev.org/c/zuul/zuul-operator/+/800786 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Document externalConfig https://review.opendev.org/c/zuul/zuul-operator/+/800791 | 17:27 |
| opendevreview | James E. Blair proposed zuul/zuul-operator master: Fix config update detection https://review.opendev.org/c/zuul/zuul-operator/+/800991 | 17:27 |
| clarkb | corvus: I'm trying to put together a test for the rename repos playbook and notice that one of the steps is the move the zuul project secrets. If you change to put all of that in zk merges before we do the rename we'll need a process for that. | 20:01 |
| clarkb | corvus: but also you probably want to add a third tool to the two you are adding whcih is to rename a project in zk? | 20:01 |
| corvus | clarkb: well, as it stands, the situation would be the same before or after my change: it's relatively easy to copy the key from old to new, but difficult to delete old | 20:06 |
| y2kenny | Before success/failure-url got deprecated, I set my failure-url to "http://zuul/t/{tenant.name}/build/{build.uuid}/console" for convenient access to the console. Is this still configurable in other ways after the deprecation? (I don't have to configure it per job... but having different success and failure-url would be useful.) | 20:06 |
| corvus | clarkb: (the only way to delete old now is manual zk surgery) | 20:07 |
| clarkb | corvus: you'd need to do manual surgery to copy old to new right? Possibly via editing of the export or direclty in zk | 20:07 |
| corvus | y2kenny: no, we're all-in on the build page. but the build page displays the text output near the failing task, so the hope is you can still see the initial error line with one click. | 20:08 |
| y2kenny | corvus: ok, thanks for the confirmation | 20:08 |
| corvus | clarkb: you can copy old to new by copying the file in the filesystem | 20:09 |
| corvus | it's deleting old from zk that isn't trivial | 20:09 |
| clarkb | corvus: today you can do that but your change removes the pull from the filesystem ? | 20:10 |
| corvus | correct, but adds an import | 20:10 |
| corvus | so it's equivalent (you just have to run a command) | 20:10 |
| clarkb | right ok | 20:11 |
| clarkb | we modify the filesystem then run the import | 20:11 |
| corvus | right | 20:11 |
| clarkb | corvus: what about the behavior before your change lands. Does zuul import what is on the filesystem? I know that it exports | 20:11 |
| corvus | yes, it's a fallback, so as soon as zuul requests a key for the project, it will pull it from the fs | 20:11 |
| clarkb | got it | 20:12 |
| corvus | clarkb: regardless -- can we improve the process? | 20:12 |
| corvus | i'm not sure a "rename-key" command would be straightforward -- when could you run that and guarantee that zuul will not request the wrong key and therefore generate a new one | 20:13 |
| corvus | (maybe that's okay because you just don't care anymore if it generates a throwaway key on an old project name) | 20:13 |
| clarkb | corvus: ya I guess you have to do an add, then remove later | 20:13 |
| corvus | that may be easier to understand and deal with | 20:13 |
| clarkb | and in this case the add would be a copy | 20:14 |
| corvus | so "copy-key" to copy old-> new and "delete-key" to delete old when unused | 20:14 |
| clarkb | ya | 20:16 |
| corvus | clarkb: i put a note on the import/export change, but it may make sense as a followup (for reviewability) | 20:16 |
| clarkb | I think that is fine as long as we land them together so that users get a full complement of tools when upgrading | 20:17 |
| avass[m] | corvus: y2kenny i think we talked about making some changes to the build dashboard a while ago though, since the summary page isn't super useful the way it is right now and could instead be replaced with the console page. But no one is really driving that | 20:17 |
| avass[m] | zbr: I think you were in on that discussion as well :) | 20:18 |
| clarkb | corvus: and for the remove later timeframe I think you'd be ok to run the delete-key command once all schedulers have had their configs reloaded? | 20:19 |
| clarkb | because at that point the jobs/builds/secrets should all be associated with the new name? | 20:19 |
| corvus | clarkb: yep. though i think the way it may actually shake out is that as soon as the first scheduler has its config reloaded, the rest will have the same view (i think it's going to end up being something like first scheduler copies tenant config file into zk, then all schedulers see it). so... "yes" but possibly faster than what one might expect :) | 20:20 |
| clarkb | ah nice | 20:21 |
| corvus | clarkb: i think that means the rename process would be pretty quick with zuul: copy-key; reload config (hopefully this is a synchronous call that returns once the new config is in zk); delete-key | 20:22 |
| clarkb | ++ | 20:23 |
| corvus | should be able to happen all in one shell session in a couple of seconds | 20:23 |
| corvus | modulo how long it takes to get the initial lock | 20:23 |
| opendevreview | James E. Blair proposed zuul/zuul master: Report all build[set] completions to DB https://review.opendev.org/c/zuul/zuul/+/802115 | 20:27 |
| opendevreview | James E. Blair proposed zuul/zuul master: Report all build[set] completions to DB https://review.opendev.org/c/zuul/zuul/+/802115 | 20:29 |
| corvus | clarkb, tobiash, mordred: i've noticed some edge cases where builds can be left "open" in the database; that should take care of that. | 20:31 |
| *** holser is now known as holser_ | 21:07 | |
| y2kenny | avass[m]: I see. Thanks for the info. | 21:09 |
| *** holser_ is now known as holser | 21:54 | |
| clarkb | corvus: some questions on https://review.opendev.org/c/zuul/zuul/+/802115 | 22:18 |
| opendevreview | James E. Blair proposed zuul/zuul master: Report all build[set] completions to DB https://review.opendev.org/c/zuul/zuul/+/802115 | 22:24 |
| corvus | clarkb: replied -- i went ahead and changed your nit about the list since the tests flaked out anyway | 22:25 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!