Tuesday, 2019-03-05

« Monday, 2019-03-04 Index Wednesday, 2019-03-06 »
*** sdake has joined #zuul00:09
*** sdake has quit IRC00:12
*** sdake has joined #zuul00:14
*** openstackgerrit has joined #zuul00:30
openstackgerritJames E. Blair proposed openstack-infra/zuul-preview master: DNM: break dockerfile  https://review.openstack.org/64090100:30
openstackgerritJames E. Blair proposed openstack-infra/zuul-preview master: DNM: fix dockerfile  https://review.openstack.org/64090400:31
openstackgerritJosef Wells proposed openstack-infra/zuul master: quickstart: mount playbooks volume for web image  https://review.openstack.org/64090600:33
*** dkehn has joined #zuul00:33
*** sdake has quit IRC00:41
*** sdake has joined #zuul01:08
*** gouthamr has quit IRC01:48
*** dmellado has quit IRC01:50
*** jamesmcarthur has joined #zuul01:55
*** jamesmcarthur has quit IRC02:04
*** sdake has quit IRC02:05
*** jamesmcarthur has joined #zuul02:06
*** jamesmcarthur has quit IRC02:10
*** sdake has joined #zuul02:27
*** jamesmcarthur has joined #zuul02:37
*** gouthamr has joined #zuul02:50
*** jamesmcarthur has quit IRC02:51
*** jamesmcarthur has joined #zuul02:54
*** dmellado has joined #zuul02:55
*** eventingmonkey has quit IRC03:01
*** bhavikdbavishi has joined #zuul03:10
*** jamesmcarthur has quit IRC03:10
*** jamesmcarthur has joined #zuul03:10
*** sdake has quit IRC03:51
*** jamesmcarthur has quit IRC03:57
*** saneax has joined #zuul04:05
openstackgerritTristan Cacqueray proposed openstack-infra/zuul master: web: add triggers information to pipeline list  https://review.openstack.org/63767004:15
*** raukadah is now known as chandankumar04:34
*** jamesmcarthur has joined #zuul04:35
*** jamesmcarthur has quit IRC04:44
*** jamesmcarthur has joined #zuul04:48
*** jamesmcarthur has quit IRC04:52
*** jamesmcarthur has joined #zuul05:00
*** jamesmcarthur has quit IRC05:35
*** sdake has joined #zuul05:47
*** jhesketh has quit IRC05:47
*** jhesketh has joined #zuul05:48
*** sdake has quit IRC05:50
*** pcaruana has joined #zuul05:52
*** sdake has joined #zuul05:57
*** jamesmcarthur has joined #zuul06:03
*** pcaruana has quit IRC06:07
*** jamesmcarthur has quit IRC06:09
tristanCcorvus: thanks for the opendev-build-docker-image pointer. But it doesn't seems to match the documentation, is requires/provides supposed to work between two buildset? (e.g. between builder-project and final-project)06:10
tristanCjkt: thanks for sharing provides/requires examples, that confirms what i thought. It seems like we'll have to create 1-1 provides/requires link06:16
tristanCour use-case is a build-rpm job, currently each job build the packages for each project in zuul.items06:18
tristanCso i was hoping we could make that job provides/requires "rpm", and re-use what is build by the items ahead06:20
tristanCbut it seems like we'll have to have each rpm-build job provides and requires an unique resources...06:21
badboytobiash: in gerrit06:33
*** jamesmcarthur has joined #zuul06:40
*** jamesmcarthur has quit IRC06:44
tristanCcorvus: jkt: nevermind previous comments, it seems like we can make such build-rpm job. didn't realize artifacts was a list that can contains all the items ahead artifacts06:56
*** jamesmcarthur has joined #zuul06:56
tristanCi got confused with buildset vs item queue06:57
*** jamesmcarthur has quit IRC07:01
tristanCcorvus: however we got a couple of issue... is it expected that a child job (defined using the job.dependencies attribute) doesn't have the artifacts of the parent job?07:03
*** AJaeger has quit IRC07:05
tristanCcorvus: second issue was that when having C-[depends-on]->B-[depends-on]->A queue, for C the artifact list contained twice the B's artifact. Here is the test setup: http://paste.openstack.org/show/747270/07:06
*** quiquell|off is now known as quiquell07:08
*** AJaeger has joined #zuul07:09
*** quiquell is now known as quiquell|brb07:35
*** snapiri has joined #zuul07:57
*** quiquell|brb is now known as quiquell08:07
*** saneax has quit IRC08:11
*** pcaruana has joined #zuul08:19
*** pcaruana has quit IRC08:25
*** yolanda has joined #zuul08:25
*** sdake has quit IRC08:36
*** pcaruana has joined #zuul08:37
*** gtema has joined #zuul08:41
*** pcaruana has quit IRC08:44
*** jpena|off is now known as jpena08:52
*** pcaruana has joined #zuul09:01
*** hashar has joined #zuul09:01
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088409:10
*** panda|ruck|off is now known as panda|ruck|flu09:13
*** saneax has joined #zuul09:16
*** bhavikdbavishi has quit IRC09:27
*** bhavikdbavishi has joined #zuul09:27
tobiashachievement unlocked: self-rechecking check pipeline...09:47
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088409:55
*** sdake has joined #zuul09:55
mordredtobiash: aroo?09:58
tobiashmordred: we're load testing a new cloud so I thought a self-rechecking check pipeline would be idead for that ;)09:59
tobiashs/idead/ideal/09:59
mordredtobiash: hah. nice09:59
*** bhavikdbavishi has quit IRC10:00
*** bjackman has joined #zuul10:11
jkttristanC: there was an update to provides/requires very recently fixing a bug where only one artifact was ever being returned10:21
quiquelltobiash: Do you know if zuul_success is going to be false if the fail is at post ?10:22
jkttristanC: https://review.openstack.org/#/c/639472/10:22
quiquellfound it https://zuul-ci.org/docs/zuul/user/jobs.html?highlight=zuul_success#var-zuul_success10:24
*** dkehn has quit IRC10:45
*** sdake has quit IRC10:51
*** sdake has joined #zuul10:54
*** bhavikdbavishi has joined #zuul10:56
*** electrofelix has joined #zuul10:57
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088411:21
*** hashar has quit IRC11:28
*** bhavikdbavishi has quit IRC11:54
*** [GNU] has quit IRC12:21
*** jpena is now known as jpena|lunch12:31
*** [GNU] has joined #zuul12:37
bjackmanDo you guys think there's any scope for a nodepool driver that's like a static docker provider?12:37
bjackmanSo you just give it the environment/params to set up a docker client and it connects to a remote host to provision nodes like that?12:38
bjackmanNot exactly scalable, but could provide a middle ground between static nodes, and k8s/OpenStack etc.12:38
*** sdake has quit IRC12:39
bjackmanIf you're thinking "why the hell would anyone want that", it's because in my sector, management is generally too paranoid about IP to allow farming any compute out to ~~the cloud~~12:40
bjackmanBut managing an on-side Kubernetes tends to be overkill12:40
*** bjackman has quit IRC12:50
*** bjackman has joined #zuul12:51
*** hashar has joined #zuul12:53
*** panda|ruck|flu is now known as panda|ruck|lunch13:03
*** gtema has quit IRC13:09
*** bjackman has quit IRC13:14
*** quiquell is now known as quiquell|lunch13:16
*** jamesmcarthur has joined #zuul13:18
*** sdake has joined #zuul13:20
*** jesusaur has quit IRC13:22
*** jesusaur has joined #zuul13:25
*** gtema has joined #zuul13:27
*** jamesmcarthur has quit IRC13:31
*** rlandy has joined #zuul13:36
*** quiquell|lunch is now known as quiquell13:39
*** jpena|lunch is now known as jpena13:40
*** jamesmcarthur has joined #zuul13:46
*** panda|ruck|lunch is now known as panda|ruck|flu13:50
*** jamesmcarthur has quit IRC14:06
*** sdake has quit IRC14:11
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088414:17
*** sdake has joined #zuul14:20
*** sdake has quit IRC14:23
*** sdake has joined #zuul14:33
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088414:46
*** sdake has quit IRC14:48
*** sdake has joined #zuul14:50
*** pcaruana has quit IRC14:57
*** sdake has quit IRC15:04
*** sdake has joined #zuul15:06
*** dkehn has joined #zuul15:07
*** bhavikdbavishi has joined #zuul15:12
*** bjackman has joined #zuul15:27
*** openstackgerrit has quit IRC15:28
corvustristanC: yes, it's expected that a job using job.dependencies won't see the artifacts of its dependencies job.  that's because provides/requires was designed to support the use-case of (let's use your example) a buildset rpm repository.  so provides/requires is there so that at the start of the job graph, you can set up an rpm repo and import any rpms from any depends-on jobs into it.  then the idea is that15:30
corvusany jobs *within* that buildset would just push their own rpms into it -- you wouldn't need to pass artifact information between jobs in the same buildset, they would just use the same repo.15:30
corvustristanC: having said that, if there's another use case where we'd want to have job.dependencies set artifacts, i think that would probably be okay.  we could either do it by honoring provides/requires, or we could just always add all parent-job artifacts to child jobs (in the same way we already pass variables).  we'll have to make sure it doesn't screw up the first use case, but i'm pretty sure it won't.15:32
corvustristanC: re the second issue C->B->A -- that seems unexpected and buggy15:36
*** jamesmcarthur has joined #zuul15:37
*** pcaruana has joined #zuul15:42
*** chandankumar is now known as raukadah15:47
*** sdake has quit IRC15:49
*** openstackgerrit has joined #zuul15:59
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088415:59
*** sdake has joined #zuul16:00
*** raukadah is now known as chandankumar16:04
*** bhavikdbavishi has quit IRC16:05
*** bhavikdbavishi has joined #zuul16:10
*** bjackman has quit IRC16:33
*** quiquell is now known as quiquell|off16:37
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Web: plug the authorization engine  https://review.openstack.org/64088416:38
openstackgerritJames E. Blair proposed openstack-infra/zuul-preview master: Return NULL on all errors  https://review.openstack.org/64107816:44
*** quiquell|off is now known as quiquell16:45
*** jpena is now known as jpena|brb16:46
*** jamesmcarthur has quit IRC16:57
*** quiquell is now known as quiquell|off17:00
*** gtema has quit IRC17:03
*** bjackman has joined #zuul17:04
*** hashar has quit IRC17:04
*** jamesmcarthur has joined #zuul17:06
openstackgerritMerged openstack-infra/zuul-preview master: Return NULL on all errors  https://review.openstack.org/64107817:08
Shrewscorvus: so, are we switching to the rust version of that ^^? also, who besides mordred and SpamapS know rust? I thought tobiash did but seems not based on previous comments. I'm beginning to wonder if the switch is worth it if we don't have the expertise available17:11
*** bhavikdbavishi has quit IRC17:12
*** jpena|brb is now known as jpena17:12
-openstackstatus- NOTICE: Gerrit is being restarted for a configuration change, it will be briefly offline.17:12
*** bhavikdbavishi has joined #zuul17:12
*** chandankumar is now known as raukadah17:12
*** jamesmcarthur has quit IRC17:14
mordredShrews: I think we haven't made a decision to do that or not yet - the rust version is up as a basis for discussion17:15
*** bjackman has quit IRC17:16
Shrewsmordred: *nod*. it's just that when i made the comment last week that i could go either way with it, i thought there was more rust knowledge among us, but finding out that may not be the case.17:17
Shrewsnot that i mind learning it17:18
Shrewsit does seem fascinating17:18
clarkbfwiw that was one of the reasons I was asking for a bit more concrete criteria around why we might want to switch17:19
clarkbbecause it is entirely possible the benefits outweight the lack of knowledge and we'll get ther ein the process17:19
Shrewsyah17:20
mordredyah17:20
Shrewsclarkb: you were one of the folks i thought knew the language  :)17:20
clarkbShrews: sort of. a few years back I dabbled then rust went on to chagne literally everything17:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Add perf testing framework  https://review.openstack.org/63679217:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Reimplement in Rust  https://review.openstack.org/63679317:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Remove C++ version  https://review.openstack.org/63679417:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Use rust:slim base image  https://review.openstack.org/63679517:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Use slice matching for hostname unpacking  https://review.openstack.org/63679617:21
openstackgerritMonty Taylor proposed openstack-infra/zuul-preview master: Port in changes from the c++ version  https://review.openstack.org/64108917:21
mordredShrews: there's a stab at porting in the recent c++ changes - might be total crap though17:21
clarkbShrews: that said I think I have a reasonably ok understanding of ownership and borrowing and I do think that memory safety is a worthwhile feature17:21
pabelangerfwiw: I am not strong in either c++ or rust, so both are learning process for me17:23
Shrewsit's such a small utility, memory safety may not be such a huge concern here17:24
*** jamesmcarthur has joined #zuul17:27
*** panda|ruck|flu is now known as panda|ruck|off17:27
*** bhavikdbavishi1 has joined #zuul17:32
*** bhavikdbavishi has quit IRC17:33
*** bhavikdbavishi1 is now known as bhavikdbavishi17:33
corvusShrews: my thoughts on the subject: http://lists.zuul-ci.org/pipermail/zuul-discuss/2019-February/000749.html17:35
*** sdake has quit IRC17:35
corvusShrews: i think the key things to keep in mind: 1) this thing is so trivial it doesn't matter what language we write it in, so let's explore :)17:36
Shrewsoh, i forgot about your etherpad17:37
corvusShrews: and 2) even if we merge the rust version, i don't think we're committing to anything -- we can revert to c++ or python or whatever.  it just lets us poke at it.17:37
Shrewscorvus: ok. i thought you mentioned last week switching this week. that may have been the pain killers clouding my head17:39
Shrewsbut i agree with your words17:39
corvusShrews: i did, but i'm not in a rush.... if.....17:40
corvusShrews: did mordred just take your task to forward-port the artifact thing? :)17:40
mordredcorvus: oh - crappit17:40
Shrewss'all good. he's faster than i am, fo' so'17:40
* mordred is a bad puppy17:41
mordredShrews: well - I think it was supposed to be an opportunity for learning17:41
Shrewsmordred: no, all good. i don't feel so rushed to learn it now17:41
mordredso maybe I should just delete my patch17:41
Shrewsno, don't do that.17:41
Shrewsthere are many things in your patch i have NO idea what they do, so learning opportunites abound17:44
clarkbmordred: fwiw there is a bug on https://review.openstack.org/#/c/641089/1 I also had some suggestions that may or may not be valid17:45
mordredclarkb: ++17:46
pabelangerSo, I seems to have gotten another stuck semaphore job, which blocks all others from running. This time was able to get a executor log with help from jpena.17:46
pabelangerhttp://paste.openstack.org/show/747299/ contains the traceback of the issue we hit when updating repos for executor server17:47
pabelangerI guess the issue is, we don't try the update again if we hit the exception17:47
pabelangertobiash: does ^ ring a bell for you17:48
*** hashar has joined #zuul17:56
SpamapSmordred: I'll peek at that stack later today17:56
pabelangerso, if I understand correct, we don't see to retry repo updates.17:56
*** hashar is now known as hasharDinner17:56
pabelangerwhich in this case, has blocked the job from running as the updated failed17:57
SpamapSIMO Rust is a worthwhile thing for us all to learn. Even if you don't use it, it will teach you things about efficient, safe memory management. Learning Rust *should* make you a better C programmer.17:57
SpamapSand IMO it has made me a better python coder too17:57
SpamapS(but I owe that mostly to the functional-ish nature... so any functional language can do that for you ;)17:57
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Zuul Web: add /api/user/actions endpoint  https://review.openstack.org/64109918:07
josefwellsTo add a new repository/project to be tracked by zuul do I need to edit /etc/zuul/main.conf and restart zuul, is there a way to get it to just reconfig18:07
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Zuul Web: add /api/user/actions endpoint  https://review.openstack.org/64109918:13
josefwellsI see I can zuul-scheduler full-reconfigure18:16
clarkbjosefwells: yes the reconfigure command is how you do it withouta restart18:25
*** saneax has quit IRC18:31
*** jpena is now known as jpena|off18:33
openstackgerritMatthieu Huin proposed openstack-infra/zuul master: [WIP] Zuul Web: add /api/user/actions endpoint  https://review.openstack.org/64109918:44
*** jamesmcarthur has quit IRC18:52
*** electrofelix has quit IRC19:02
*** jamesmcarthur has joined #zuul19:21
*** jamesmcarthur_ has joined #zuul19:24
*** jamesmcarthur has quit IRC19:25
*** hasharDinner is now known as hashar19:27
*** sdake has joined #zuul19:33
*** sdake has quit IRC20:17
*** sdake has joined #zuul20:18
*** bhavikdbavishi has quit IRC20:23
*** dkehn has quit IRC20:28
fungiosf board meeting is happening in 3 minutes; the open infrastructure projects confirmation process is on their agenda; call details are in http://lists.openstack.org/pipermail/foundation/2019-February/002714.html20:56
*** jamesmcarthur has joined #zuul20:57
*** jamesmcarthur_ has quit IRC21:00
*** pcaruana has quit IRC21:10
dmsimardThe diagonal bars going across the progress bar when the a job is launching is a nice touch :)21:15
*** pwhalen has quit IRC21:26
*** hashar has quit IRC21:58
mhuanybody knows Morgan Fainberg's nick on IRC? Is he around these parts?22:07
clarkbkmalloc: ^22:08
mhuclarkb, thanks!22:08
mhukmalloc, when you got time, could you have a look at https://review.openstack.org/#/c/562321/ (JWT support for tenant-scoped admin web API), I've reworked the authZ part and I want to make sure it's not completely insane22:11
mhualso given that I have a growing PoC patch chain based on the current spec proposal, I'd rather not put more effort if everything needs to be redone22:11
mhuactually anybody else willing to have a look and chime in, you're welcome22:12
*** dkehn has joined #zuul22:20
*** sdake has quit IRC22:28
*** sdake has joined #zuul22:37
*** jamesmcarthur has quit IRC22:41
*** sdake has quit IRC22:45
*** sdake has joined #zuul22:51
*** rlandy is now known as rlandy|bbl22:54
*** pwhalen has joined #zuul22:55
*** jamesmcarthur has joined #zuul22:57
*** jamesmcarthur has quit IRC22:58
*** jamesmcarthur has joined #zuul22:58
*** sdake has quit IRC23:02
*** sdake has joined #zuul23:33
kmallocmhu: yeah it's on my list23:36
kmallocmhu: i've been looking into some other stuff, but i think you were headed the right way with the last iteration(s)23:36
*** jamesmcarthur has quit IRC23:37
*** jamesmcarthur has joined #zuul23:38
*** jamesmcarthur has quit IRC23:40
openstackgerritJames E. Blair proposed openstack-infra/zuul-preview master: DNM: break dockerfile  https://review.openstack.org/64090123:57
openstackgerritJames E. Blair proposed openstack-infra/zuul-preview master: DNM: fix dockerfile  https://review.openstack.org/64090423:59
« Monday, 2019-03-04 Index Wednesday, 2019-03-06 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!