Tuesday, 2022-03-15

-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 833676: Clear branch cache on full reconfiguration https://review.opendev.org/c/zuul/zuul/+/83367600:26
-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com:07:07
- [zuul/zuul] 830707: Use a transaction for BuildCompletedEvent https://review.opendev.org/c/zuul/zuul/+/830707
- [zuul/zuul] 830896: Use kazoo.retry in zkobject https://review.opendev.org/c/zuul/zuul/+/830896
@q:fricklercloud.defungi: tobiash the links to your keys in https://zuul-ci.org/docs/zuul/latest/vulnerabilities.html seem to be non-functional07:47
-@gerrit:opendev.org- Tobias Henkel proposed: [zuul/zuul] 833726: Fix links to pgp keys https://review.opendev.org/c/zuul/zuul/+/83372607:57
-@gerrit:opendev.org- Albin Vass proposed: [zuul/zuul-jobs] 833755: Add flag to reset submodules https://review.opendev.org/c/zuul/zuul-jobs/+/83375509:53
-@gerrit:opendev.org- Albin Vass proposed: [zuul/zuul-jobs] 833755: Add flag to reset submodules https://review.opendev.org/c/zuul/zuul-jobs/+/83375510:09
-@gerrit:opendev.org- Albin Vass proposed: [zuul/zuul-jobs] 833755: Add flag to reset submodules https://review.opendev.org/c/zuul/zuul-jobs/+/83375510:09
-@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 833760: Fix ZKObject compressed data size estimate https://review.opendev.org/c/zuul/zuul/+/83376010:17
@fungicide:matrix.org> <@q:fricklercloud.de> fungi: tobiash the links to your keys in https://zuul-ci.org/docs/zuul/latest/vulnerabilities.html seem to be non-functional13:17
thanks for the heads up, looks like that was broken in the transition to versioned documentation. i'll get a change up to correct that, but i'm also probably overdue for refreshing the copy of my key there with an updated self-sig expiration, and we need to drop the sks keyservers links now that their network has collapsed. series of changes for all of that up shortly
@tobias.henkel:matrix.orgfungi: see https://review.opendev.org/c/zuul/zuul/+/83372613:19
@tobias.henkel:matrix.org * fungi: : see https://review.opendev.org/c/zuul/zuul/+/83372613:20
@tobias.henkel:matrix.org * fungi: see https://review.opendev.org/c/zuul/zuul/+/83372613:20
@fungicide:matrix.orgoh, thanks tobiash! i'll add my other fixes atop that, then13:20
-@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed:13:33
- [zuul/zuul] 833790: Drop SKS Keyserver links from vulnerabilities doc https://review.opendev.org/c/zuul/zuul/+/833790
- [zuul/zuul] 833791: Refresh self-keysig expiration on my OpenPGP key https://review.opendev.org/c/zuul/zuul/+/833791
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/nodepool] 821711: Add IBM Cloud VPC driver https://review.opendev.org/c/zuul/nodepool/+/82171113:50
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 833794: Log debug messages when remote ref is missing https://review.opendev.org/c/zuul/zuul/+/83379413:50
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 833794: Log debug messages when remote ref is missing https://review.opendev.org/c/zuul/zuul/+/83379414:58
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 833794: Log debug messages when remote ref is missing https://review.opendev.org/c/zuul/zuul/+/83379415:07
-@gerrit:opendev.org- Tristan Cacqueray proposed: [zuul/zuul] 833841: WIP: web: add builds order_by attribute https://review.opendev.org/c/zuul/zuul/+/83384115:24
@clarkb:matrix.orgcorvus: https://review.opendev.org/c/zuul/zuul/+/833691 is a first step at trying to make the nodejs memory problems happier. If that doesn't help or makes it worse then I think we may need to find a way to pass in nodejs memory settings to the react-scripts build command15:37
@jim:acmegating.comClark: lgtm; i'm unaware of any reason we might need to stay on an older v.15:39
-@gerrit:opendev.org- Zuul merged on behalf of Tobias Henkel: [zuul/zuul] 833726: Fix links to pgp keys https://review.opendev.org/c/zuul/zuul/+/83372616:04
-@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org:16:21
- [zuul/zuul] 833790: Drop SKS Keyserver links from vulnerabilities doc https://review.opendev.org/c/zuul/zuul/+/833790
- [zuul/zuul] 833791: Refresh self-keysig expiration on my OpenPGP key https://review.opendev.org/c/zuul/zuul/+/833791
-@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 821711: Add IBM Cloud VPC driver https://review.opendev.org/c/zuul/nodepool/+/82171116:23
-@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 833691: Update nodejs runtime to version 16 https://review.opendev.org/c/zuul/zuul/+/83369117:01
@q:fricklercloud.deshould changing from one (github) connection to another in zuul.conf be supported? I kept getting zuul crashes because it still tried to do things with the old connection. solved for now by dumping zookeeper data completely, as this was still a wip deployment. but maybe people would like to migrate from github to gerrit at some stage19:52
@clarkb:matrix.org> <@q:fricklercloud.de> should changing from one (github) connection to another in zuul.conf be supported? I kept getting zuul crashes because it still tried to do things with the old connection. solved for now by dumping zookeeper data completely, as this was still a wip deployment. but maybe people would like to migrate from github to gerrit at some stage19:53
The OpenDev zuul is currently connected to the opendev gerrit, gerrit's gerrit, and one other gerrit iirc. As well as github.com. It should be supported
@clarkb:matrix.orgYou probably need to change the connection names though19:53
@q:fricklercloud.debut we never dropped a connection from opendev zuul so far, did we?20:04
@q:fricklercloud.deguess I can do a testcase with that in the quickstart setup20:05
@jim:acmegating.comq: there's a lot of stuff in zuul that assumes connection names are synonymous with their corresponding systems, so yeah, changing the target of a connection without changing the name would require a downtime and 'zuul delete-state'.20:28
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 833918: Fix gitlab timer events https://review.opendev.org/c/zuul/zuul/+/83391820:44
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 833918: Fix gitlab timer events https://review.opendev.org/c/zuul/zuul/+/83391820:44
@clarkb:matrix.orgianw: I'm looking at https://review.opendev.org/c/zuul/zuul/+/832486/5/zuul/web/__init__.py and two things pop out at me. First is that HEAD seems like a valid request against those URLS? In fact it seems we use HEAD for the opendev haproxy http checks. Second is that cherrypy seems to have an @cherrypy.tools.handle_options(allowed_methods=['GET', ]) decorator that we use in some places. I wonder if that might be a better way to accomplish this?20:48
@clarkb:matrix.orgLeft the notes inlien as well. I think I'll hold off on reviewing the rest of the stack until you can take a look at that as I expect this may impact the whole stack20:50
@clarkb:matrix.orgcorvus: https://review.opendev.org/c/zuul/zuul-registry/+/831440/ is the point in ianw's registry stack that you haven't reviewed yet if you have time to take a look at the rest of it. I think you've already reviewed a good chunk of what was going on there.20:53
@jim:acmegating.comClark: thx, i left notes for myself, so some easy +3s there. cc ianw 20:55
-@gerrit:opendev.org- Zuul merged on behalf of lotorev vitaly: [zuul/zuul] 833514: Switch to 8005 nodepool webapp port in example docs https://review.opendev.org/c/zuul/zuul/+/83351421:05
-@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand: [zuul/zuul-registry] 831319: podman buildset testing: dump image list https://review.opendev.org/c/zuul/zuul-registry/+/83131921:39
-@gerrit:opendev.org- Zuul merged on behalf of Ian Wienand:21:41
- [zuul/zuul-registry] 831339: podman: make sure we remove the pulled image https://review.opendev.org/c/zuul/zuul-registry/+/831339
- [zuul/zuul-registry] 831440: Fix and/or matching for image pre-conditions https://review.opendev.org/c/zuul/zuul-registry/+/831440
@iwienand:matrix.orgClark: to the first, I guess we should support HEAD; i imagine currently we just return the whole body22:24
@clarkb:matrix.orgianw: ya I suspect HEAD is equivalent to GET in this case22:25
@clarkb:matrix.orgbut at least that makes things like haproxy happy (though I have now discovered that we HEAD / not /api so not affected?)22:25
@iwienand:matrix.orgwith the second about the decorator, that was suggested but i thought it was worth considering separately; https://review.opendev.org/c/zuul/zuul/+/83273122:25
@iwienand:matrix.orgcurl --head https://zuul.opendev.org is doing the right thing22:27
@clarkb:matrix.orgoh interesting I didn't realize that was our own implementation of a decorator22:27
@jim:acmegating.comand so does /api/info22:27
@iwienand:matrix.orgi wonder if cherrypy intercepts that22:28
@jim:acmegating.comi would characterize it as "handling correctly" rather than "intercepting", but i do think cherrypy is responsible for the correct behavior.22:29
@iwienand:matrix.orgwell yeah, it must pass it on to the handlers as a GET?  because there's no specific HEAD handling I can see22:30
@iwienand:matrix.orgi guess currently with the @expose decorator like https://opendev.org/zuul/zuul/src/branch/master/zuul/web/__init__.py#L822 it is returning the full response22:34
@iwienand:matrix.orgpresumably cherry-py then lops off the content if it's a HEAD request22:35
@iwienand:matrix.orgbut you might be right that if we do a if request == GET in there, we won't respond correctly to HEAD requests now22:36
@clarkb:matrix.orgya I suspect the request would nknow it is a head request and raise the error22:36
@iwienand:matrix.orgso presumably we could do the same, and just treat GET==HEAD22:36
@clarkb:matrix.org++22:37
@iwienand:matrix.orga good unit test would be to GET and then HEAD and make sure the content-length is the same i guess22:38
@jim:acmegating.comit looks like you're trying to make DELETE not work -- how about updating the route dispatcher instead?22:40
@jim:acmegating.comthat would be the more framework-native thing to do in cherrypy22:40
@iwienand:matrix.orgwell more than DELETE just make it so that endpoints that only make sense to GET from fail with anything else22:42
@iwienand:matrix.orgit's a minor thing, but the haproxy did highlight that monitoring things do probe apis in ways like that22:42
@jim:acmegating.comsure -- you can specify with the route dispatcher what methods are valid22:43
@jim:acmegating.comthat's a better place for that rather than in the handler itself22:43
@jim:acmegating.comianw: https://docs.cherrypy.dev/en/3.2.6/concepts/dispatching.html#other-dispatchers22:47
@iwienand:matrix.orgahh, i guess "conditions=" in there22:47
@jim:acmegating.comyeah i think that should do it22:48
@iwienand:matrix.orghttps://docs.cherrypy.dev/en/3.2.6/refman/_cpdispatch.html#cherrypy._cpdispatch.Dispatcher is ... interesting22:53
@iwienand:matrix.orgsome things it seems autogenerated docs are not good for22:54
@iwienand:matrix.orgi'll rework the changes, probably not today23:02
@clarkb:matrix.orgno rush, just wanted to make sure we didn't do something silly there23:02

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!