Friday, 2015-03-06

mordred	clarkb: however, if we upgrade novaclient and nodepool breaks, we can point our fingers at people and laugh	00:00
mordred	which is fun	00:00
clarkb	ah ya that just won't make them on a new node but won't kill them on an existing host mtreinish I will just disable them on this host	00:00
clarkb	mordred: your idea of fun is funny	00:00
clarkb	mordred: also I could not get that image to break on any of the three hpcloud AZs	00:00
clarkb	though az2 was consistently having scheduling errors	00:00
*** timcline has quit IRC		00:01
anteaya	gerrit is taking an inordinate amount of time to save my inline comment	00:01
*** ivar-lazzaro has quit IRC		00:01
mtreinish	clarkb: ok thanks. FWIW, I checked cacti and the memory growth issue we were seeing the last time we ran 4 doesn't seem to be happening	00:01
*** dmsimard is now known as dmsimard_away		00:01
*** MarkAtwood has quit IRC		00:02
*** ghostpl_ has quit IRC		00:02
openstackgerrit	Monty Taylor proposed openstack-infra/puppet-nodepool: Upgrade dependent libraries when we upgrade https://review.openstack.org/161951	00:02
mordred	jeblair, clarkb: I think we might want that ^^	00:02
mtreinish	mmedvede: hmm, ok. That differs from what I remember, but it's been almost a while so I'm probably just not remembering correctly	00:03
openstackgerrit	Merged openstack-infra/project-config: Add Big Switch Networking library to StackForge https://review.openstack.org/160339	00:03
mmedvede	mtreinish: I am not sure if that is the proper way to check though	00:03
clarkb	mtreinish: but we process quickly enough with one so might as well reduce the chance of having it explode	00:03
openstackgerrit	Ian Wienand proposed openstack-infra/nodepool: Add config validation option https://review.openstack.org/161952	00:03
*** achanda has quit IRC		00:04
mtreinish	clarkb: yeah, using 1 is the right way to do it. I was just pointing it out	00:04
clarkb	mtreinish: cool all done	00:04
*** ayoung has quit IRC		00:04
mtreinish	clarkb: cool, thanks	00:05
mtreinish	clarkb: if you're looking at subunit2sql stuff, I've got the patch up to do the backups before we run upgrades	00:06
openstackgerrit	Ian Wienand proposed openstack-infra/nodepool: Add config validation option https://review.openstack.org/161952	00:08
clarkb	ianw: thats a lot of yaml :)	00:09
devananda	jogo: in test runs of services, are clients being installed from trunk or pip?	00:09
clarkb	mtreinish: does the db-manage script not use the same my.conf?	00:10
devananda	jogo: and is it different between master branch of services, and their stable branch(es)?	00:10
clarkb	mtreinish: I guess not since the manage script is all python/alembic	00:10
mtreinish	clarkb: no the formats are different. One's an oslo.config thing the others the mysql thing	00:10
clarkb	devananda: pip for most jobs	00:10
mtreinish	they're close but different	00:10
clarkb	devananda: and thats not different for the branches	00:10
devananda	clarkb: that's what I thought, thanks. any plans to change that? (please say no)	00:10
*** ZZelle has quit IRC		00:10
*** markvoelker has quit IRC		00:11
*** ivar-lazzaro has joined #openstack-infra		00:11
*** ZZelle has joined #openstack-infra		00:11
clarkb	devananda: there are however jobs for the clients themselves that install from source	00:11
clarkb	devananda: I don't think there are plans.	00:11
devananda	clarkb: naturally	00:11
devananda	ok, great	00:11
devananda	that's the sanity I was hoping for :)	00:11
bkero	So what exactly is the Monty fact? http://puppetboard.openstack.org/fact/monty	00:11
ianw	clarkb: just copies of the real config file ... though some might argue that's a lot of yaml too	00:12
*** dangers is now known as dangers_away		00:12
mtreinish	mmedvede: that's probably as good as any other method. I assume that site just puts the inbound address in the response body	00:12
SpamapS	mordred: I'm thinking that with shade we might want to say if you don't pass cache_interval that we shouldn't do any caching. Thoughts?	00:13
mordred	SpamapS: that's an excellent idea	00:13
*** Sukhdev has quit IRC		00:13
openstackgerrit	Merged openstack-infra/subunit2sql: Add microsecond columns to the test_runs table https://review.openstack.org/160057	00:13
clarkb	ianw: it is I was mostly being silly, the initial diff numbers caught my eye	00:14
SpamapS	mordred: because otherwise we have a memory leak waiting for every casual user of shade. :-P	00:14
mordred	SpamapS: I would like to subscribe to your newsletter	00:14
clarkb	bkero: "blerg"	00:14
* SpamapS writes patch		00:14
bkero	clarkb: blerg, obviously	00:14
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Normalize Gerrit projects.yaml https://review.openstack.org/161954	00:15
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Wrap projects.yaml at 80 characters https://review.openstack.org/161955	00:15
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Remove upstreams from projects.yaml https://review.openstack.org/161956	00:15
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Propose projects.yaml normalization daily https://review.openstack.org/161957	00:15
devananda	gertty crashed. booo.	00:15
jeblair	devananda: current master?	00:15
clarkb	mtreinish: commented	00:15
devananda	jeblair: latest pip	00:15
devananda	jeblair: it's a rare enough event that I felt it worth comment	00:15
devananda	also, it lost a several-paragraph review that I was working on...	00:16
devananda	jeblair: possibly because I touched my screen (it's a touch screen)	00:16
jeblair	devananda: ok. lots of crashing bugs are fixed in latest master; a few are still documented at https://storyboard.openstack.org/#!/project/698	00:16
jogo	devananda:AFAIK pip	00:16
jogo	but its easy to check in pip-freeze	00:16
*** ivar-laz_ has joined #openstack-infra		00:16
*** patri394 has quit IRC		00:17
*** patri767 has quit IRC		00:17
*** patri755 has quit IRC		00:17
*** patri832 has quit IRC		00:17
*** patri460 has quit IRC		00:17
*** patri192 has quit IRC		00:17
jeblair	devananda: (and i'm going to release a new version tomorrow, so it'll be on pip then)	00:17
nibalizer	jeblair: awesome	00:17
devananda	jogo: cool, thx	00:17
*** ivar-lazzaro has quit IRC		00:17
devananda	jeblair: yay fixes!	00:18
mtreinish	mmedvede: I'm just really surprised they wouldn't be masquerading the outbound. It's not like it's routable from outside the firewall (unless you guys have changed how things are setup)	00:18
mtreinish	clarkb: thanks, I'll respin it quickly	00:18
mmedvede	mtreinish: I never asked that question until today	00:19
*** patrickeast_ has joined #openstack-infra		00:20
*** patrickeast__ has joined #openstack-infra		00:20
*** patri851 has joined #openstack-infra		00:20
*** patri601 has joined #openstack-infra		00:20
*** patri466 has joined #openstack-infra		00:20
*** pc_m has joined #openstack-infra		00:20
*** patri796 has joined #openstack-infra		00:20
*** dboik_ has quit IRC		00:21
*** dboik has joined #openstack-infra		00:22
openstackgerrit	Matthew Treinish proposed openstack-infra/puppet-subunit2sql: Add backup before running db migrations https://review.openstack.org/160514	00:23
mtreinish	clarkb: ^^^	00:23
*** bhunter71 has quit IRC		00:23
mmedvede	mtreinish: and yes, it is still not routable from outside the firewall	00:23
clarkb	mordred: can you maybe comment on 126747 with your ideas for how we will support that?	00:24
clarkb	mordred: I think it is heavily dependent on dib and shade things	00:24
*** bhunter71 has joined #openstack-infra		00:24
clarkb	woo for rereviewing your own code	00:26
openstackgerrit	Clark Boylan proposed openstack-infra/puppet-jenkins: Allow for plugin pinning https://review.openstack.org/160549	00:27
*** stevemar has quit IRC		00:27
*** ZZelle_ has quit IRC		00:28
SpamapS	mordred: why does shade.OpenStackCloud.__init__ take kwargs or things that are really clearly just more kwargs that could be static?	00:28
anteaya	mordred: does the gitorious information need to be updated in this patch? https://review.openstack.org/#/c/154132/	00:28
*** claudiub has quit IRC		00:29
mordred	clarkb: yes. thanks for that reminder	00:29
*** crc32 has quit IRC		00:29
*** liusheng has quit IRC		00:29
openstackgerrit	Spencer Krum proposed openstack-infra/project-config: Allow puppet-httpd to be publised to the forge https://review.openstack.org/161022	00:29
SpamapS	mordred: there are a kind of ridiculous number of kwargs used exactly as they'd be if they were just listed.	00:29
mordred	SpamapS: we need to maybe sort out in shade how to list per-vendor info on image formats or something	00:29
*** liusheng has joined #openstack-infra		00:29
mordred	SpamapS: most of the time it's when it's passthrough -	00:29
mordred	SpamapS: but it also could just be because I'm a bad person	00:30
SpamapS	mordred: auth, auth_token, service_type, service_name, endpoint, api_version, publicURL, private ... all just used with kwargs.get	00:30
mordred	SpamapS: have you seen my "clearn up parameters" patch?	00:30
SpamapS	mordred: yeah	00:30
mordred	k	00:30
mordred	SpamapS: so yeah, I support cleaning that up	00:31
SpamapS	Oh!	00:31
SpamapS	some of htose are there	00:31
SpamapS	did not connect the dots	00:31
mordred	we should land that	00:31
mordred	it'll make reasoning easier	00:31
*** _shaps_ has quit IRC		00:32
mordred	clarkb: could you take a peek at 160184 and its parent 160183 ?	00:32
clarkb	mordred: sure	00:32
SpamapS	mordred: I'm going to rebase on that for my "don't cache if no interval is passed" patch.	00:32
*** sdake_ has quit IRC		00:33
SpamapS	we should also land unit tests and functional tests. :)	00:33
openstackgerrit	Merged openstack-infra/project-config: Add bashate job for release-tools https://review.openstack.org/160419	00:34
*** sdake_ has joined #openstack-infra		00:34
mordred	SpamapS: ++	00:35
mordred	clarkb: also, yes, patch #2 has a breaking interface change. to my knowledge nothing has landed anywhere that consumes the old interface and I'd like to clean it before we can't	00:35
clarkb	people that know apache better than I should read 161165	00:36
*** pc_m has quit IRC		00:36
openstackgerrit	Adam Gandelman proposed openstack-infra/project-config: Add expiremental plugin based ironic+ironicclient functional jobs https://review.openstack.org/161483	00:36
clarkb	its funny that spamaps always pokes mordred for tests expecting mordred to even consider writing them	00:40
openstackgerrit	Clint 'SpamapS' Byrum proposed openstack-infra/shade: Disable dogpile.cache if cache_interval is None https://review.openstack.org/161963	00:41
*** achanda has joined #openstack-infra		00:41
SpamapS	clarkb: actually I wrote them and want to land them. :)	00:41
clarkb	also I was told that shade is not being rigidly tested and as a reviewer I should assume the author has tested	00:41
SpamapS	clarkb: and rcarrillocruz also wrote some and wants to land them. :)	00:41
clarkb	SpamapS: those are orthogonal	00:41
SpamapS	clarkb: shade's growing up. :)	00:41
clarkb	er orthogonal to ^	00:42
SpamapS	clarkb: oh I thought we were talking about my comment that we should also land unit tests and functional tests. :)	00:42
clarkb	no reviewing changes and you are poking for tests :)	00:42
clarkb	I don't think its a bad thing, but I also know mordred won't write them :)	00:42
SpamapS	I was poking you indirectly.	00:42
clarkb	I know	00:43
SpamapS	poke poke	00:43
SpamapS	poke	00:43
clarkb	I thought I had reviewed rcarrillocruz's stuff	00:43
clarkb	and debugged d-g for him	00:43
mordred	clarkb: I am actually working on some os-client-config tests	00:43
clarkb	speaking of I should show that fix to sdague to see if he knows what is going on there	00:43
*** shivharis has quit IRC		00:43
*** shivharis has joined #openstack-infra		00:43
SpamapS	oohhhh burn	00:43
mordred	clarkb: but mostly just to confuse you	00:43
*** dims has quit IRC		00:44
*** shivharis has quit IRC		00:45
openstackgerrit	Merged openstack-infra/shade: Align cert, key, cacert and verify with requests https://review.openstack.org/160183	00:45
*** ivar-laz_ has quit IRC		00:45
*** MarkAtwood has joined #openstack-infra		00:45
*** ivar-lazzaro has joined #openstack-infra		00:45
*** achanda has quit IRC		00:46
devananda	jogo, clarkb: either of you have a moment to sanity-check my statements about how microversions should interact with the gate? https://etherpad.openstack.org/p/oHVxTVRFix	00:47
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Create a job template for tempest on devstack plugins https://review.openstack.org/161965	00:47
devananda	background is folks disagreeing on whether python-ironicclient should, by default, send the greatest API version that was known about at the time it was released -- or if it should not send ANY header at all unless explicitly told to	00:48
devananda	that will directly affect how we test the interaction between nova and Ironic. and, of course, users.	00:48
clarkb	shouldn't the client as the remote end instead of the other way around?	00:48
*** ivar-lazzaro has quit IRC		00:48
devananda	remote end supports a range of versions	00:49
clarkb	thats just me thinking out loud that it is better for clients to be forgiving to the servers	00:49
clarkb	devananda: ya so it should return that to the client and client picks best one	00:49
*** ivar-lazzaro has joined #openstack-infra		00:49
clarkb	so that we can avoid the glanceclient issue of "I'm gonna use v1 and only v1 and everything else is a user nightmare"	00:49
devananda	or the client says to the server "here's the version I want" and if the server supports that, it adheres to it	00:49
clarkb	devananda: ya thats what glanceclient does and it doesn't work	00:50
devananda	and if the server does not support that, then the client interprets the headers and tries to ask for something the server does support	00:50
mtreinish	clarkb: version discovery is broken in more than just glance fwiw :)	00:50
devananda	clarkb: so i'm talking about microversions, not /v1/ vs /v2/	00:50
*** signed8bit is now known as signed8bit_ZZZzz		00:50
clarkb	devananda: its the same difference really	00:50
devananda	clarkb: not really	00:51
clarkb	devananda: you are negotiating "a" version	00:51
clarkb	what type of version isn't really important	00:51
mordred	the biggest question is whether you're asking the developer using the library to request a version	00:51
mordred	if you are doing that, you are losing	00:51
devananda	though if the client defaults to always ask for v1.0, it is going to re-create the same problem	00:51
clarkb	mordred: yes that is what this would implicitly do	00:51
mordred	if you are not, and it's transparent - DONT CARE	00:51
clarkb	mordred: by installing a specific client you are requesting a specific version	00:51
mordred	wait - what?	00:51
clarkb	mordred: read scrollback	00:52
mordred	there is someone suggesting that we install a specific client version to talk to a cloud?	00:52
clarkb	mordred: not quite	00:52
devananda	mordred: that is definitely true. also, if the client defaults to oldest version, it means we can't effectively ever expect any other openstack project to use a newer version	00:52
devananda	like nova talking to ironic	00:52
mordred	devananda: let me suggest something different	00:52
devananda	which is where the present debate (in our team) arose from	00:52
*** rmcall has quit IRC		00:52
clarkb	mordred: but similar to how glanceclient works the idea here seems to be if you install fooclient version X it will use the corresponding api version Y to talk to server	00:52
clarkb	though at least there seems to be an attempt to negotiate	00:52
clarkb	but I think thats more work, client should say "what do you support" then pick	00:53
devananda	clarkb: right - that's basically what I'm outlining in the 'pad I linked	00:53
mordred	devananda: the client, by default, will always know the greatest API version that it knows about	00:53
mordred	because it can't not know that	00:53
devananda	mordred: right	00:53
devananda	link again -- https://etherpad.openstack.org/p/oHVxTVRFix :)	00:53
mordred	it should request that, but be able to accept all previous versions that the server sends back	00:53
devananda	mordred: if you disagree with what i'm saying three, I'd like to know why	00:53
mordred	then, if if the user tries to use a feature that the server does nto requests, there should be a "oh, hi, your server does not support this feature" error	00:53
mordred	but if it's a feature that just had a different rest api semantic, the client should do the right thing	00:54
adam_g	devananda, fwiw that testing plan in the pad looks sane to me. im actually putting together the barebones infra for the functional client testing now	00:54
devananda	also there is a lengthy spec up that descibes the different scenarios involved... since engineers want to describe every possible interaction :) https://review.openstack.org/#/c/161110/	00:54
clarkb	my only request is that as a user I never have to tell my client to use a specific version or install a specific version of the client to get the previous result	00:54
devananda	adam_g: k, thanks	00:55
devananda	clarkb: ++	00:55
mordred	clarkb: ++	00:55
clarkb	as far as designing how that works I think you shove the logic in the client	00:55
*** ddieterly has joined #openstack-infra		00:55
mordred	other than taht, I think microversions are a mistake	00:55
mordred	and I thnk they are developers wanking themselves	00:55
devananda	clarkb: my goal, for a server and client within reasonable temporal distance (what ever we deem supportable) they will auto-negotiate, with as few round-trips as possible, the greatest common version	00:55
mordred	and not solving problems that users have	00:55
jogo	devananda: looking	00:56
mordred	API versions are bull	00:56
mordred	they solve nothing	00:56
mordred	they break users	00:56
mordred	they are a waste of time	00:56
mordred	as is, btw, semver	00:56
devananda	mordred: they allow applications that talk to a versioned service to choose when to upgrade to the next version of the API	00:56
clarkb	devananda: the only danger that immediately pops out with your test plan that I see is if you break that feature after merging it into the client but before releasing the client	00:56
mordred	meh	00:56
mordred	nobody wants to upgrade to the next version of the API	00:56
clarkb	devananda: there is a race there, I guess its up to ou to decide how dangerous it is	00:56
mordred	they want to do work	00:56
devananda	clarkb: good point. i think i'm willing to accept that	00:57
mordred	the question is "does my cloud support me uploading an image"	00:57
clarkb	devananda: I will note it on the etherpad for ease of findability	00:57
mordred	the question is not "does my cloud run v1 or v2 or v2.1.1"	00:57
devananda	clarkb: as it is, we aren't testing many things, because nova is effecetively pinned to the API that we had in stable/juno	00:57
mordred	literally no developer ahs ever cared about the verion	00:57
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Remove upstreams from projects.yaml https://review.openstack.org/161956	00:57
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Propose projects.yaml normalization daily https://review.openstack.org/161957	00:57
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Normalize Gerrit projects.yaml https://review.openstack.org/161954	00:57
openstackgerrit	James E. Blair proposed openstack-infra/project-config: Wrap projects.yaml at 80 characters https://review.openstack.org/161955	00:57
devananda	mordred: indeed	00:57
mordred	the version is a very very poor way to map capabilities	00:57
devananda	mordred: the problem is "developers feel the need to change something in the API but dont want to break support for existing applications, however old they are"	00:58
devananda	mordred: "while enabling new apps to use new feature"	00:58
mordred	devananda: the developers should just get over it	00:58
mordred	honestly	00:58
mordred	and shoyld not change that thing	00:58
* mordred points, once again, at libmysqlclient		00:58
devananda	mordred: so openstack should never grow new features? or we should only do fork-lift, wholesale changes to services. ...	00:58
mordred	grow new features is great	00:59
mordred	that doesn't break anything	00:59
mordred	changing existing things should just not happen	00:59
mordred	OR	00:59
mordred	if it des	00:59
mordred	know that it's the most painful thing that has ever happened	00:59
mordred	and no amount of increasing a version number will fix that	00:59
*** ddieterly has quit IRC		00:59
mordred	if nova addeda "create jellybean" api - I would not care at all	01:00
mordred	but when glance v2 _removes_ the ability to upload directly via the v1 api call	01:00
mordred	NOW I have to write a bunch of "if v1: elif v2: crap"	01:00
*** bhunter71 has quit IRC		01:01
*** patri466 has quit IRC		01:02
*** patri851 has quit IRC		01:02
*** patri601 has quit IRC		01:02
*** patri796 has quit IRC		01:02
*** patrickeast_ has quit IRC		01:02
*** patrickeast__ has quit IRC		01:02
*** otter768 has joined #openstack-infra		01:02
*** carl_baldwin has quit IRC		01:02
mordred	which leads to crap like this: http://git.openstack.org/cgit/openstack-infra/shade/tree/shade/__init__.py#n583	01:02
mordred	my fear is taht microversions are going to make me write 100x that ^^	01:02
mordred	because they're goign to open the door to devs thinking they have freedom to change stuff because they told people about it	01:03
anteaya	clarkb: any reason I should not +A enable swift logs for python-jobs? https://review.openstack.org/#/c/156521/7	01:03
*** mikedillion has joined #openstack-infra		01:03
anteaya	we were holding off on team agreement which we got at last meeting	01:03
anteaya	any technical reason to time the merge?	01:04
clarkb	anteaya: we need the parent to merge first, then rebuild images then merge the child	01:04
anteaya	okay so I can +A https://review.openstack.org/#/c/158514/7 ?	01:04
anteaya	you are +2	01:04
mordred	devananda: (reading etherpad)	01:05
clarkb	anteaya: yes that one should be ready	01:05
anteaya	very good, thanks	01:05
*** mikedillion has quit IRC		01:05
openstackgerrit	Adam Gandelman proposed openstack-infra/project-config: Add expiremental plugin based ironic+ironicclient functional jobs https://review.openstack.org/161483	01:05
devananda	mordred: if you have time, read the spec and the comments too	01:06
*** mikedillion has joined #openstack-infra		01:06
anteaya	so after images build tomorrow then look again at the python-jobs patch	01:06
anteaya	yes?	01:06
*** otter768 has quit IRC		01:06
*** MarkAtwood has quit IRC		01:07
anteaya	why would we add python2.6 jobs to anything? https://review.openstack.org/#/c/160928/	01:07
anteaya	I had thought we were trying to get rid of them	01:07
mordred	anteaya: I think it's temporary	01:08
mordred	anteaya: since oslo still supports 2.6 and that library is used by oslo	01:08
anteaya	yes that is what the commit message says	01:08
anteaya	we are okay with that?	01:08
*** teoleg has joined #openstack-infra		01:11
mordred	fine by me - but I'm only but one voice	01:12
jeblair	anteaya: i think we're okay with it until the last openstack release supporting 2.6 goes away	01:13
jeblair	(if it's needed)	01:13
jeblair	harlowja: btw, you don't have to wrap your commit messages at 50 chars, feel free to use 80 for anything after the first line :)	01:14
harlowja	lol	01:15
anteaya	mordred jeblair okay thanks	01:15
*** ivar-laz_ has joined #openstack-infra		01:15
harlowja	i'm striving for 10 char lines	01:15
harlowja	lol	01:15
*** ivar-laz_ has quit IRC		01:15
*** markvoelker has joined #openstack-infra		01:15
*** Jeba has quit IRC		01:16
*** ivar-laz_ has joined #openstack-infra		01:16
*** fandi has joined #openstack-infra		01:16
anteaya	have you tried kanji?	01:16
harlowja	hakiu u say	01:16
harlowja	u want me to write all my commit messages as haiku	01:17
harlowja	ok then	01:17
harlowja	lol	01:17
*** rwsu is now known as rwsu-afk		01:17
morganfainberg	harlowja, can i get a bot that rewrites all commit messages as haikus?	01:17
anteaya	asselin: your reply to the mailing list thread went to the third party announce list only from what I can see	01:17
morganfainberg	or is haiku also the plural	01:17
harlowja	probabably not impossible, lol	01:17
* morganfainberg should look		01:17
anteaya	asselin: the parent never should have included the -announce list	01:17
*** bhunter71 has joined #openstack-infra		01:18
morganfainberg	harlowja, can i also get it wired up to a speak and spell? ;)	01:18
harlowja	sureeee	01:18
morganfainberg	harlowja, awesome!	01:18
StevenK	morganfainberg: Pipe it to festival?	01:18
asselin	anteaya, hmmm...I just hit reply...	01:18
anteaya	asselin: oh no, it did go to -dev	01:18
anteaya	yeah, the author of teh original post included -announce	01:18
harlowja	just plug your bot into http://www.languageisavirus.com/interactive-haiku-generator.html and profit, lol	01:18
anteaya	so hitting reply would do the same	01:18
*** ivar-lazzaro has quit IRC		01:18
anteaya	asselin: I'm going to deny the post to the -announce list and thought I would let you know	01:19
asselin	anteaya, ok, I canceled the request to send to annouch	01:19
anteaya	thanks	01:19
openstackgerrit	Merged openstack-infra/project-config: zuul-swift-upload support recursive globs https://review.openstack.org/158514	01:19
* morganfainberg goes back on the hunt for a portable laptop w/ 16GB of ram and real SSD space, and a usable screen. (portable means 13" or less)		01:19
*** mtanino has quit IRC		01:19
* morganfainberg wonders if anyone will make said laptop.		01:19
asselin	anteaya, it did go to the mailing list	01:20
anteaya	yes, I see it now	01:20
anteaya	just took my client a minute, so I checked -dev in the meantime	01:20
asselin	anteaya, i will be more careful next time. thanks	01:20
anteaya	no problem, the original issue didn't reside with you	01:20
anteaya	thanks for cancelling	01:20
*** markvoelker has quit IRC		01:20
StevenK	morganfainberg: That would be nice.	01:21
*** melwitt has quit IRC		01:22
morganfainberg	StevenK, right? i'm actually hoping the 12" MBA will have passible/usable hardware so i can use that whenever it comes out - use in this case means install linux as the main OS on it	01:22
clarkb	mtreinish: around? what is the name of devstack's imageid 42?	01:23
clarkb	mtreinish: is it m1.nano? and its the smallest one right?	01:23
clarkb	mtreinish: sorry not imageid, flavorid	01:23
*** crinkle has quit IRC		01:23
*** tsg has quit IRC		01:26
openstackgerrit	Merged openstack-infra/project-config: Add python26 jobs to debtcollector https://review.openstack.org/160928	01:26
*** bcourt has quit IRC		01:26
*** tsg has joined #openstack-infra		01:26
*** teoleg has quit IRC		01:27
*** Daisy_ has joined #openstack-infra		01:28
*** ghostpl_ has joined #openstack-infra		01:28
*** r-daneel has quit IRC		01:31
*** otter768 has joined #openstack-infra		01:32
*** bdpayne has quit IRC		01:33
*** fandi has quit IRC		01:33
*** fandi has joined #openstack-infra		01:33
*** crinkle has joined #openstack-infra		01:34
asselin	I'm getting this error using DIB: "Package ruby1.8 is not available, but is referred to by another package." http://paste.openstack.org/show/189875/	01:34
asselin	Wehre are those packages defined?	01:35
asselin	nevermind...I found the file: devstack-gate.d/dib-manifests/dib-manifest-dpkg-devstack-gate	01:36
*** ghostpl_ has quit IRC		01:36
anteaya	are you building on trusty?	01:36
asselin	I using precide node to build trusty image	01:37
anteaya	ah ha	01:37
*** tsg has quit IRC		01:37
anteaya	precise had ruby1.8	01:37
asselin	but ruby 1.8 isn't in that file....	01:37
openstackgerrit	Merged openstack-infra/project-config: Fix swiftclient func tests to run using both auth versions https://review.openstack.org/160844	01:37
anteaya	so that is what it would be looking for	01:37
*** dannywilson has quit IRC		01:38
*** ddieterly has joined #openstack-infra		01:38
asselin	but should dib be using the trusy packages?	01:38
anteaya	http://packages.ubuntu.com/trusty/ruby	01:38
anteaya	http://packages.ubuntu.com/precise/ruby	01:38
anteaya	so the clue in what it is calling it	01:38
anteaya	precise calls it ruby1.8 (even though it is ruby 1.9)	01:39
anteaya	and trusty calls it ruby	01:39
jesusaurus	asselin: i believe it bind-mounts the host deb cache. i dont think its possible to cross-build distros or distro versions	01:39
* jesusaurus is not well-versed in dib		01:39
greghaynes	it is	01:39
greghaynes	both things are possible for most combinations	01:39
*** ddieterl_ has joined #openstack-infra		01:40
*** ddieterly has quit IRC		01:40
openstackgerrit	Merged openstack-infra/project-config: add python26 jobs back for compass-core https://review.openstack.org/160251	01:40
asselin	is ruby the exception? anywya I see about printing some debug logging	01:40
greghaynes	asselin: youre building for a trusty target?	01:40
asselin	greghaynes, yes	01:40
greghaynes	asselin: huh, I cant find that 50-download-pkgs script, where did you find it?	01:41
greghaynes	oh nvm, found	01:41
*** MarkAtwood has joined #openstack-infra		01:44
greghaynes	asselin: so http://git.openstack.org/cgit/openstack-dev/devstack/tree/files/debs is where its getting the packages to install AIUI	01:44
greghaynes	its a rather large collection	01:45
greghaynes	but something in there obviously depends on ruby1.8	01:45
greghaynes	I do wonder how the existing image builds dont have this issue though	01:45
greghaynes	clarkb: ^	01:45
clarkb	we build on trusty?	01:46
greghaynes	its in the chroot so it matters what you target	01:46
greghaynes	I think	01:46
*** achanda has joined #openstack-infra		01:46
greghaynes	yea	01:46
clarkb	also its just downloading packages not installing anything	01:47
clarkb	so nothing should fail right?	01:47
greghaynes	apparently not	01:47
greghaynes	you do build trusty targets though?	01:47
clarkb	yes	01:47
clarkb	centos7, trusty, and precise	01:47
asselin	I see DIB is setting this up, not sure where it comes from: /home/stack/project-config/devstack-gate.d/dib-manifests	01:48
greghaynes	asselin: do you have a trusty vm handy?	01:48
asselin	greghaynes, this file has ruby in it: http://paste.openstack.org/show/189889/	01:48
asselin	greghaynes, yes	01:48
asselin	greghaynes, I think infra runs nodepool on trusty though. I'm running it on precise right now	01:49
greghaynes	asselin: If you edit openstack-infra/project-config/nodepool/elements/cache-devstack/install.d/50-download-pkgs you can echo out the list of packages it is attempting to install as it goes through the loop	01:50
greghaynes	that should tell you what is failing	01:50
greghaynes	er, attempting to download	01:50
asselin	greghaynes, yes I did that...just changed set +ex	01:50
asselin	running now	01:50
clarkb	SpamapS: ok I think I have reviewed all the shade test changes	01:51
*** MarkAtwood has quit IRC		01:52
openstackgerrit	Ramy Asselin proposed openstack-infra/project-config: DIB project-config scripts requires DIB_RELEASE https://review.openstack.org/161989	01:53
*** rkukura has quit IRC		01:57
mtreinish	clarkb: yeah it's m1.nano	01:59
mtreinish	clarkb: https://git.openstack.org/cgit/openstack-dev/devstack/tree/lib/tempest#n193	01:59
*** ashleighfarnham has quit IRC		02:00
clarkb	mtreinish thanks	02:00
*** mikedillion has quit IRC		02:02
*** ChuckC has quit IRC		02:04
*** tiswanso has joined #openstack-infra		02:04
*** ChuckC has joined #openstack-infra		02:05
*** tiswanso has quit IRC		02:05
*** tiswanso has joined #openstack-infra		02:05
*** stevemar has joined #openstack-infra		02:08
*** tiswanso has quit IRC		02:08
*** tiswanso has joined #openstack-infra		02:09
*** rkukura has joined #openstack-infra		02:09
*** tsg has joined #openstack-infra		02:14
*** oomichi has quit IRC		02:14
*** ivar-laz_ has quit IRC		02:16
anteaya	greghaynes: so I commented on https://review.openstack.org/#/c/157554/	02:16
anteaya	greghaynes: I am unable to see what benefit your patch is offering	02:16
*** ivar-lazzaro has joined #openstack-infra		02:17
*** markvoelker has joined #openstack-infra		02:17
*** crc32 has joined #openstack-infra		02:17
anteaya	but it is expanding	02:17
*** baoli has joined #openstack-infra		02:17
*** signed8bit_ZZZzz is now known as signed8bit		02:18
*** signed8bit has quit IRC		02:19
*** weshay has quit IRC		02:20
*** markvoelker has quit IRC		02:21
openstackgerrit	Merged openstack-infra/project-config: skip rerunning requirements changes in devstack https://review.openstack.org/156214	02:22
*** vilobhmm2 has quit IRC		02:23
*** vilobhmm1 has joined #openstack-infra		02:23
*** ivar-lazzaro has quit IRC		02:23
greghaynes	anteaya: If export NODEPOOL_STATIC_NAMESERVER='1.2.3.4' then that value will be set for your resolv.conf rather than 8.8.8.8	02:24
*** ivar-lazzaro has joined #openstack-infra		02:24
*** dmsimard_away is now known as dmsimard		02:25
anteaya	okay	02:25
*** ivar-lazzaro has quit IRC		02:25
greghaynes	Which is important if you want to, for example, use a mirror that only has dns entries on an internal dns server	02:26
*** ivar-lazzaro has joined #openstack-infra		02:26
anteaya	okay	02:27
*** yamamoto has joined #openstack-infra		02:27
anteaya	and 8.8.8.8 remains the default and can now be overridden	02:27
greghaynes	yep	02:28
anteaya	if you export a different value in the environment	02:28
anteaya	okay, i'm starting to see that	02:28
greghaynes	Yea, the bash for it is pretty obtuse, but ${VAR:-default} expands to $VAR if VAR is set else default	02:29
anteaya	okay thanks	02:29
anteaya	I've commented again saying you have answered my questions	02:30
greghaynes	:)	02:30
anteaya	but I don't know enough about nodepool internals to vote on the patch	02:30
anteaya	so thanks	02:30
anteaya	:)	02:30
*** dmsimard is now known as dmsimard_away		02:35
*** camunoz is now known as camunoz_afk		02:36
anteaya	okay if you haven't seen this one yet: http://i.imgur.com/9hRi2jN.gifv	02:38
openstackgerrit	Brian Hunter proposed openstack-infra/nodepool: Pretty Age in nodepool list https://review.openstack.org/161994	02:38
*** ddieterly has joined #openstack-infra		02:38
*** tsg has quit IRC		02:39
*** camunoz_afk is now known as camunoz		02:41
*** ddieterl_ has quit IRC		02:41
*** baoli has quit IRC		02:43
*** ddieterl_ has joined #openstack-infra		02:44
*** ddieterly has quit IRC		02:45
openstackgerrit	Joshua Hesketh proposed openstack-infra/project-config: zuul-swift-upload place mime icon into cell https://review.openstack.org/161995	02:48
*** jerryz_ has joined #openstack-infra		02:50
*** vilobhmm1 has quit IRC		02:55
*** ChuckC_ has joined #openstack-infra		02:56
*** rfolco has joined #openstack-infra		02:57
*** rmcall has joined #openstack-infra		02:58
*** ChuckC has quit IRC		02:59
*** ashleighfarnham has joined #openstack-infra		03:00
*** nelsnelson has joined #openstack-infra		03:01
*** baoli has joined #openstack-infra		03:02
*** radez is now known as radez_g0n3		03:05
*** unicell has quit IRC		03:07
*** rfolco has quit IRC		03:09
*** yamahata has quit IRC		03:12
*** ivar-lazzaro has quit IRC		03:16
*** markvoelker has joined #openstack-infra		03:18
*** ayoung has joined #openstack-infra		03:19
*** rmcall has quit IRC		03:20
*** ghostpl_ has joined #openstack-infra		03:21
*** markvoelker has quit IRC		03:22
*** ghostpl_ has quit IRC		03:25
*** ChuckC_ is now known as ChuckC		03:27
*** woodster_ has quit IRC		03:30
*** dimtruck is now known as zz_dimtruck		03:30
*** ashleighfarnham has quit IRC		03:34
*** rkukura has quit IRC		03:36
*** nelsnelson has quit IRC		03:39
*** baoli has quit IRC		03:42
*** david-lyle is now known as david-lyle_afk		03:44
*** hdd_ has joined #openstack-infra		03:44
*** ddieterl_ has quit IRC		03:45
*** ddieterly has joined #openstack-infra		03:45
*** Daisy_ has quit IRC		03:46
*** nelsnelson has joined #openstack-infra		03:46
*** ayoung has quit IRC		03:49
*** fandi has quit IRC		03:49
*** ddieterly has quit IRC		03:50
*** oomichi has joined #openstack-infra		03:51
*** baoli has joined #openstack-infra		03:53
*** nelsnelson has quit IRC		03:56
*** harlowja is now known as harlowja_away		03:57
*** julim has quit IRC		03:57
*** achanda has quit IRC		03:57
*** vilobhmm has joined #openstack-infra		03:58
*** esker has joined #openstack-infra		04:04
*** Longgeek has joined #openstack-infra		04:08
*** mriedem_away has quit IRC		04:10
*** mriedem has joined #openstack-infra		04:11
*** BharatK has joined #openstack-infra		04:15
*** nelsnelson has joined #openstack-infra		04:16
*** amotoki has joined #openstack-infra		04:16
*** Daisy_ has joined #openstack-infra		04:18
*** markvoelker has joined #openstack-infra		04:19
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Refactor sources out of triggers https://review.openstack.org/118993	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add gerrit reviews into patchset approvals https://review.openstack.org/97390	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add support for negative requirements https://review.openstack.org/102726	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Configure triggers dynamically https://review.openstack.org/119534	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add support for 'connection' concept https://review.openstack.org/121528	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add base class for triggers https://review.openstack.org/119532	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add base class for sources https://review.openstack.org/119531	04:21
openstackgerrit	Joshua Hesketh proposed openstack-infra/zuul: Add base class for reporters https://review.openstack.org/119530	04:21
*** alexpilotti has quit IRC		04:22
openstackgerrit	Joshua Hesketh proposed openstack-infra/project-config: zuul-swift-upload place mime icon into cell https://review.openstack.org/161995	04:22
*** gyee has quit IRC		04:23
*** markvoelker has quit IRC		04:25
*** tiswanso has quit IRC		04:27
*** patrickeast has quit IRC		04:28
*** otter768 has quit IRC		04:28
*** baoli has quit IRC		04:28
*** baoli has joined #openstack-infra		04:31
*** achanda has joined #openstack-infra		04:38
*** ddieterly has joined #openstack-infra		04:46
*** wenlock has quit IRC		04:48
*** koolhead17 has joined #openstack-infra		04:49
*** ddieterly has quit IRC		04:50
*** esker has quit IRC		04:53
*** achanda has quit IRC		04:54
*** esker has joined #openstack-infra		05:00
*** vilobhmm1 has joined #openstack-infra		05:01
openstackgerrit	Syed Ismail Faizan Barmawer proposed openstack-infra/project-config: Adding publish-to-pypi template to ironic-lib project https://review.openstack.org/162010	05:01
*** ujuc has joined #openstack-infra		05:01
*** vilobhmm has quit IRC		05:04
*** baoli has quit IRC		05:06
purp	c	05:08
*** ghostpl_ has joined #openstack-infra		05:09
*** garyh has quit IRC		05:09
*** deepakcs has joined #openstack-infra		05:11
*** nelsnelson has quit IRC		05:13
*** ghostpl_ has quit IRC		05:16
openstackgerrit	Timothy R. Chavez proposed openstack-infra/shade: Add 'rebuild' to shade https://review.openstack.org/161067	05:18
*** ashleighfarnham has joined #openstack-infra		05:20
*** oomichi has quit IRC		05:20
*** dannywilson has joined #openstack-infra		05:21
*** markvoelker has joined #openstack-infra		05:21
*** ashleighfarnham has quit IRC		05:24
*** dannywilson has quit IRC		05:25
*** markvoelker has quit IRC		05:26
*** vilobhmm1 has quit IRC		05:27
lifeless	I just did a stupid - sorry. (released unittest2 1.0.0 on my friday)	05:29
*** esker has quit IRC		05:29
lifeless	however its done. So if there is fallout, I'll make sure I can help.	05:30
*** sdake_ has quit IRC		05:32
*** esker has joined #openstack-infra		05:33
*** bhunter71 has quit IRC		05:41
*** ddieterly has joined #openstack-infra		05:47
*** ddieterly has quit IRC		05:51
*** achanda has joined #openstack-infra		05:56
*** yfried has joined #openstack-infra		05:56
*** esker has quit IRC		06:00
*** resker has joined #openstack-infra		06:02
*** _nadya_ has joined #openstack-infra		06:03
*** michchap has quit IRC		06:04
*** michchap has joined #openstack-infra		06:04
*** xyang1 has quit IRC		06:05
*** michchap has quit IRC		06:05
*** michchap has joined #openstack-infra		06:06
*** armax has quit IRC		06:09
*** coolsvap_ has quit IRC		06:11
*** Bsony has joined #openstack-infra		06:13
*** ghostpl_ has joined #openstack-infra		06:13
*** Bsony has quit IRC		06:17
*** ghostpl_ has quit IRC		06:18
*** HeOS has quit IRC		06:18
*** mfink_ has joined #openstack-infra		06:22
*** markvoelker has joined #openstack-infra		06:22
*** unicell has joined #openstack-infra		06:24
*** markvoelker has quit IRC		06:27
openstackgerrit	Merged openstack-infra/project-config: Remove devstack-plugins group https://review.openstack.org/161917	06:28
*** otter768 has joined #openstack-infra		06:29
*** mrmartin has joined #openstack-infra		06:32
*** otter768 has quit IRC		06:34
*** _nadya_ has quit IRC		06:34
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Only run manuals niceness checks as needed https://review.openstack.org/162032	06:39
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Make manual checklang target voting https://review.openstack.org/162033	06:39
*** Longgeek has quit IRC		06:39
*** mfink_ has quit IRC		06:41
*** jyuso1 has joined #openstack-infra		06:43
*** mrunge has joined #openstack-infra		06:47
*** tnovacik has joined #openstack-infra		06:48
openstackgerrit	yolanda.robla proposed openstack-infra/puppet-cgit: Add configurable Timeout settings on git vhost https://review.openstack.org/160871	06:48
*** ddieterly has joined #openstack-infra		06:48
openstackgerrit	yolanda.robla proposed openstack-infra/puppet-gerrit: Add logging to fetchremotes call https://review.openstack.org/161663	06:51
*** hdd_ has quit IRC		06:51
*** ddieterly has quit IRC		06:52
*** unicell has quit IRC		06:54
*** mrmartin has quit IRC		06:54
*** vilobhmm has joined #openstack-infra		06:54
*** phschwartz has quit IRC		06:54
*** unicell has joined #openstack-infra		06:54
*** phschwartz has joined #openstack-infra		06:57
*** Bsony has joined #openstack-infra		06:57
*** yamahata has joined #openstack-infra		07:00
*** _nadya_ has joined #openstack-infra		07:15
*** jamielennox is now known as jamielennox\|away		07:16
*** _nadya_ has quit IRC		07:18
*** ivar-lazzaro has joined #openstack-infra		07:18
*** Longgeek has joined #openstack-infra		07:19
*** ivar-lazzaro has quit IRC		07:22
*** mrmartin has joined #openstack-infra		07:22
*** yamahata has quit IRC		07:23
*** markvoelker has joined #openstack-infra		07:24
*** scheuran has joined #openstack-infra		07:27
*** markvoelker has quit IRC		07:29
*** hichihara has quit IRC		07:33
*** Bsony has quit IRC		07:36
*** vilobhmm has quit IRC		07:39
*** Longgeek has quit IRC		07:43
*** teran_ has joined #openstack-infra		07:44
*** achanda has quit IRC		07:44
*** achanda has joined #openstack-infra		07:44
*** teran has quit IRC		07:46
*** mriedem has quit IRC		07:47
*** mriedem has joined #openstack-infra		07:47
GheRivero	morning all	07:47
*** achanda has quit IRC		07:48
*** ddieterly has joined #openstack-infra		07:49
*** liusheng has quit IRC		07:50
*** HeOS has joined #openstack-infra		07:50
*** liusheng has joined #openstack-infra		07:51
*** mpavone has joined #openstack-infra		07:52
*** hamzy has quit IRC		07:52
*** ddieterly has quit IRC		07:53
*** camunoz has quit IRC		07:54
*** mrmartin has quit IRC		07:58
*** jerryz_ has quit IRC		07:58
*** jerryz_ has joined #openstack-infra		07:59
*** hamzy has joined #openstack-infra		07:59
*** jcoufal has joined #openstack-infra		08:01
*** zul has quit IRC		08:01
*** markus_z has joined #openstack-infra		08:01
*** ghostpl_ has joined #openstack-infra		08:01
*** mrunge has quit IRC		08:03
*** MaxV has joined #openstack-infra		08:04
*** achanda has joined #openstack-infra		08:05
*** teran has joined #openstack-infra		08:07
*** ghostpl_ has quit IRC		08:09
*** teran_ has quit IRC		08:10
*** ashleighfarnham has joined #openstack-infra		08:11
*** chlong has quit IRC		08:11
*** mrunge has joined #openstack-infra		08:11
*** mrunge_ has joined #openstack-infra		08:11
*** MaxV has quit IRC		08:11
*** mrunge_ has quit IRC		08:12
*** mrunge has quit IRC		08:12
*** mrunge has joined #openstack-infra		08:13
*** SumitNaiksatam has quit IRC		08:13
*** zul has joined #openstack-infra		08:14
*** michchap has quit IRC		08:16
*** SumitNaiksatam has joined #openstack-infra		08:16
*** ashleighfarnham has quit IRC		08:16
*** luqas__ has joined #openstack-infra		08:16
*** MaxV has joined #openstack-infra		08:18
*** krtaylor has quit IRC		08:22
*** markvoelker has joined #openstack-infra		08:25
*** unicell1 has joined #openstack-infra		08:27
*** MaxV has quit IRC		08:28
*** MaxV has joined #openstack-infra		08:28
*** devvesa has joined #openstack-infra		08:29
*** unicell has quit IRC		08:29
*** notmyname has quit IRC		08:30
*** devvesa has quit IRC		08:30
*** otter768 has joined #openstack-infra		08:30
*** markvoelker has quit IRC		08:31
*** MaxV has quit IRC		08:33
*** otter768 has quit IRC		08:35
*** krtaylor has joined #openstack-infra		08:35
*** afazekas has joined #openstack-infra		08:39
*** mpaolino has joined #openstack-infra		08:39
*** mpavone_ has joined #openstack-infra		08:39
*** scheuran_ has joined #openstack-infra		08:40
*** arxcruz has joined #openstack-infra		08:40
*** stevemar has quit IRC		08:42
openstackgerrit	Spencer Krum proposed openstack-infra/puppet-logstash: Flesh out the readme https://review.openstack.org/153820	08:43
*** mpavone has quit IRC		08:43
*** scheuran has quit IRC		08:43
*** jlanoux has joined #openstack-infra		08:44
*** crc32 has quit IRC		08:46
*** dtantsur is now known as dtantsur\|bbl		08:48
*** jgrimm is now known as zz_jgrimm		08:49
*** ddieterly has joined #openstack-infra		08:49
*** doude has joined #openstack-infra		08:49
openstackgerrit	Lin Yang proposed openstack-infra/zuul: Doc: fix minor typo https://review.openstack.org/162060	08:51
*** Ala has joined #openstack-infra		08:53
*** ddieterly has quit IRC		08:54
*** achanda has quit IRC		08:54
*** che-arne has joined #openstack-infra		08:54
*** krtaylor has quit IRC		08:54
*** liusheng has quit IRC		08:55
*** liusheng has joined #openstack-infra		08:56
*** andreykurilin_ has joined #openstack-infra		08:56
*** Hal has joined #openstack-infra		08:57
*** achanda has joined #openstack-infra		08:57
*** Hal is now known as Guest16818		08:57
*** SumitNaiksatam has quit IRC		08:58
*** MaxV has joined #openstack-infra		08:59
*** bradm has quit IRC		09:00
*** jistr has joined #openstack-infra		09:02
*** hashar has joined #openstack-infra		09:06
*** hashar is now known as hasharConf		09:09
*** devvesa has joined #openstack-infra		09:10
*** yamamoto has quit IRC		09:10
*** krtaylor has joined #openstack-infra		09:10
*** achanda has quit IRC		09:12
*** achanda has joined #openstack-infra		09:13
*** achanda has quit IRC		09:14
*** achanda has joined #openstack-infra		09:14
openstackgerrit	Thomas Morin proposed openstack-infra/project-config: Add networking-bagpipe-l2 project https://review.openstack.org/161290	09:18
*** ghostpl_ has joined #openstack-infra		09:19
*** achanda has quit IRC		09:19
*** bradm has joined #openstack-infra		09:22
openstackgerrit	Ghe Rivero proposed openstack-infra/shade: Add cover to .gitignore https://review.openstack.org/162064	09:25
*** SumitNaiksatam has joined #openstack-infra		09:26
*** markvoelker has joined #openstack-infra		09:27
*** teran has quit IRC		09:31
*** Longgeek has joined #openstack-infra		09:32
*** dizquierdo has joined #openstack-infra		09:33
*** markvoelker has quit IRC		09:33
*** ihrachyshka has joined #openstack-infra		09:33
*** AJaeger has joined #openstack-infra		09:36
*** jistr is now known as jistr\|biab		09:36
*** AJaeger has quit IRC		09:37
*** jp_at_hp has joined #openstack-infra		09:40
*** mrmartin has joined #openstack-infra		09:40
*** koolhead17 has quit IRC		09:42
*** aarefiev has quit IRC		09:42
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Ensure that gerritbot channels are in accessbot https://review.openstack.org/162070	09:43
*** Daisy_ has quit IRC		09:45
*** aarefiev has joined #openstack-infra		09:45
*** teran has joined #openstack-infra		09:49
*** ddieterly has joined #openstack-infra		09:50
*** dizquierdo has quit IRC		09:51
*** teran has quit IRC		09:53
*** BobBall_AWOL is now known as BobBall		09:54
*** mpaolino has quit IRC		09:54
*** ssam2 has joined #openstack-infra		09:54
*** ddieterly has quit IRC		09:55
kevinbenton	anteaya: hey, how can i get added to the new networking-bigswitch-core group for the repo I requested?	09:56
*** ihrachyshka has quit IRC		09:57
*** mrmartin has quit IRC		09:59
*** hasharConf is now known as hashar		10:01
*** Longgeek has quit IRC		10:06
*** mpaolino has joined #openstack-infra		10:08
*** e0ne has joined #openstack-infra		10:09
*** ghostpl_ has quit IRC		10:11
*** ghostpl_ has joined #openstack-infra		10:14
*** Longgeek has joined #openstack-infra		10:14
*** andreykurilin_ has quit IRC		10:18
*** andreykurilin__ has joined #openstack-infra		10:18
*** skolekonov has joined #openstack-infra		10:21
*** e0ne is now known as e0ne_		10:22
*** liusheng has quit IRC		10:23
*** liusheng has joined #openstack-infra		10:24
openstackgerrit	Valeriy Ponomaryov proposed openstack-infra/system-config: Enable logging of #openstack-manila https://review.openstack.org/161200	10:25
*** markvoelker has joined #openstack-infra		10:29
*** otter768 has joined #openstack-infra		10:31
*** _nadya_ has joined #openstack-infra		10:32
*** e0ne_ has quit IRC		10:33
*** teran has joined #openstack-infra		10:33
*** markvoelker has quit IRC		10:35
*** otter768 has quit IRC		10:36
*** teran_ has joined #openstack-infra		10:36
*** ZZelle has quit IRC		10:37
*** andreykurilin__ has quit IRC		10:38
*** teran has quit IRC		10:39
*** e0ne has joined #openstack-infra		10:39
*** jistr\|biab is now known as jistr		10:39
*** liusheng has quit IRC		10:41
*** liusheng has joined #openstack-infra		10:42
*** Guest16818 has quit IRC		10:44
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Support translation of RST files https://review.openstack.org/162093	10:47
*** ddieterly has joined #openstack-infra		10:51
*** ddieterly has quit IRC		10:55
*** gmann_ has joined #openstack-infra		10:59
*** dims has joined #openstack-infra		11:00
*** hashar has quit IRC		11:01
*** mpaolino has quit IRC		11:07
*** e0ne is now known as e0ne_		11:07
*** BobBall has quit IRC		11:10
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Ensure that gerritbot channels are in accessbot https://review.openstack.org/162070	11:16
*** e0ne_ has quit IRC		11:18
*** deepakcs has quit IRC		11:19
*** yamamoto has joined #openstack-infra		11:20
*** mpaolino has joined #openstack-infra		11:21
*** chlong has joined #openstack-infra		11:22
*** _nadya_ has quit IRC		11:23
*** yamamoto has quit IRC		11:24
*** ldnunes has joined #openstack-infra		11:26
*** jcoufal_ has joined #openstack-infra		11:28
*** markvoelker has joined #openstack-infra		11:31
*** jcoufal has quit IRC		11:31
openstackgerrit	Igor Gnatenko proposed openstack-infra/jenkins-job-builder: Add support of multiple credentials for ssh-agent-credentials wrapper https://review.openstack.org/156681	11:32
*** markvoelker has quit IRC		11:36
*** dtantsur\|bbl is now known as dtantsur		11:39
*** ujuc has quit IRC		11:39
*** BharatK has quit IRC		11:42
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Expose jenkins_gitfullname and jenkins_gitemail https://review.openstack.org/161695	11:44
*** hashar has joined #openstack-infra		11:46
*** e0ne has joined #openstack-infra		11:47
*** flaper87 has quit IRC		11:47
*** flaper87 has joined #openstack-infra		11:47
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Ensure that gerritbot channels are in accessbot https://review.openstack.org/162070	11:51
*** ddieterly has joined #openstack-infra		11:52
*** luqas__ has quit IRC		11:52
*** BharatK has joined #openstack-infra		11:54
*** ddieterly has quit IRC		11:56
*** yamamoto has joined #openstack-infra		11:58
*** pc_m has joined #openstack-infra		12:01
openstackgerrit	yolanda.robla proposed openstack-infra/puppet-jenkins: Add configurable timeout for jjb update https://review.openstack.org/162121	12:01
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Install proper ruby1_9_1 packages for Debian Jessie https://review.openstack.org/162124	12:11
*** alexpilotti has joined #openstack-infra		12:13
openstackgerrit	Aleksey Ripinen proposed openstack-infra/storyboard: Tasks branches, milestones and projects ids https://review.openstack.org/162126	12:16
*** baoli has joined #openstack-infra		12:16
*** sambetts has quit IRC		12:17
*** sambetts has joined #openstack-infra		12:17
*** adalbas has joined #openstack-infra		12:17
*** gmann_ has quit IRC		12:20
*** ZZelle has joined #openstack-infra		12:25
*** rfolco has joined #openstack-infra		12:26
*** BobBall has joined #openstack-infra		12:26
*** Hal has joined #openstack-infra		12:29
*** Hal is now known as Guest94194		12:30
*** baoli_ has joined #openstack-infra		12:31
*** markvoelker has joined #openstack-infra		12:32
*** otter768 has joined #openstack-infra		12:32
openstackgerrit	Timothy R. Chavez proposed openstack-infra/shade: Add 'rebuild' to shade https://review.openstack.org/161067	12:32
HenryG	anteaya: ping	12:33
*** baoli has quit IRC		12:34
HenryG	I am wondering why this change https://review.openstack.org/161569 does not seem to be taking effect in the gate? cc: clarkb	12:35
HenryG	http://logs.openstack.org/84/160884/2/check/check-neutron-dsvm-functional/590e659/console.html#_2015-03-05_22_48_03_122	12:36
*** cdent has joined #openstack-infra		12:36
*** markvoelker has quit IRC		12:36
*** otter768 has quit IRC		12:36
*** MaxV has quit IRC		12:37
*** BobBall has quit IRC		12:37
*** koolhead17 has joined #openstack-infra		12:38
*** yamamoto has quit IRC		12:38
*** aysyd has joined #openstack-infra		12:39
*** ja has joined #openstack-infra		12:39
*** yamamoto has joined #openstack-infra		12:40
*** dhritishikhar has joined #openstack-infra		12:41
openstackgerrit	Paul Michali proposed openstack-infra/project-config: Partial revert of func job setup for VPNaaS repo https://review.openstack.org/162135	12:44
*** cdent has quit IRC		12:44
*** ssam2 has quit IRC		12:45
*** ssam2 has joined #openstack-infra		12:45
*** MaxV has joined #openstack-infra		12:46
*** yamamoto has quit IRC		12:46
ja	which distribution is infra using right now for zuul/jenkins/etc? I'm in the process of planning a 3rd party CI system and would prefer using something known to work with the infra components we'd be reusing. the only hint I've found so far is an example in the Ci automation pub at #link http://docs.openstack.org/infra/publications/ci-automation/#%2818%29	12:49
*** yamamoto has joined #openstack-infra		12:51
*** ddieterly has joined #openstack-infra		12:52
*** ujuc has joined #openstack-infra		12:54
*** ddieterly has quit IRC		12:57
nibalizer	most services run on ubuntu 14.04	12:58
mordred	nibalizer: why are you awake?	12:59
mordred	nibalizer: or are you _still_ awake?	12:59
* mordred is impressed either way		12:59
*** cdent has joined #openstack-infra		13:03
*** mpavone_ has quit IRC		13:04
nibalizer	because gsl	13:04
nibalizer	i have to sleep before the gsl comes on	13:04
nibalizer	other wise this ends up happening	13:04
nibalizer	http://wiki.teamliquid.net/starcraft2/2015_Global_StarCraft_II_League_Season_1/Code_S	13:04
*** markvoelker has joined #openstack-infra		13:06
*** woodster_ has joined #openstack-infra		13:06
*** yamamoto has quit IRC		13:07
*** dizquierdo has joined #openstack-infra		13:07
*** dprince has joined #openstack-infra		13:07
ja	nibalizer, thanks. is there anything known to have problems on 14.04? on 3rd party CI owner mentioned an indistinct problem "a while ago" that led them to use 12.04, be he was "pretty sure" it had been fixed by now hence my q here.	13:08
ja	if no known problems, I'll just plan on using 14.04 for the whole zuul/etc smash	13:08
nibalizer	ja: well actaully	13:09
nibalizer	i would say most of it is going to 14.04	13:09
nibalizer	a lot of it is still on 12.04	13:09
nibalizer	sorry	13:09
nibalizer	ja: check this out http://puppetboard.openstack.org/fact/lsbdistcodename	13:10
nibalizer	that gives you an idea of where things are	13:10
openstackgerrit	Monty Taylor proposed openstack-infra/shade: Small fixes found working on ansible modules https://review.openstack.org/162140	13:11
*** e0ne is now known as e0ne_		13:12
openstackgerrit	LiKai proposed openstack/requirements: Update xstatic-font-awesome to 4.3.0 https://review.openstack.org/162142	13:13
*** bobba has joined #openstack-infra		13:13
*** bobba has quit IRC		13:14
ja	nibalizer, if you know offhand what is "Final" on that puppetboard? clicking through its link isn't informative. but thanks for the trusty vs precise answer. so nodepool and puppetmaster can definitely cope with trusty from that list. any of the others we're in less-charted territory with.	13:15
*** bobba has joined #openstack-infra		13:15
*** bobba is now known as Guest70641		13:15
*** Guest70641 is now known as BobBall		13:15
Kiall	nibalizer: I'm intrigued - what's the difference between the two pie charts, just the orientation?	13:15
Kiall	ja: lsbdistcodename Final / lsbdistdescription CentOS release 6.6 (Final)	13:16
Kiall	http://puppetboard.openstack.org/fact/lsbdistrelease might be a tad clearer ;)	13:16
ja	thanks Kiall	13:16
nibalizer	Kiall: kindof, the two charts got removed in later puppetboard	13:17
nibalizer	one orders by value and one orders by number of things in the slice	13:17
*** mpavone_ has joined #openstack-infra		13:17
*** ujuc has quit IRC		13:18
nibalizer	http://puppetboard.openstack.org/fact/physicalprocessorcount might show it better	13:18
ja	right, I see it; starting point is straight up; left one is by frequency, right alpha by name	13:18
*** BobBall has quit IRC		13:21
openstackgerrit	Julien Danjou proposed openstack-infra/project-config: Move Gnocchi from Stackforge to OpenStack https://review.openstack.org/162146	13:22
*** BobBall has joined #openstack-infra		13:22
*** mbacchi has joined #openstack-infra		13:22
*** e0ne_ is now known as e0ne		13:22
*** kgiusti has joined #openstack-infra		13:23
openstackgerrit	Monty Taylor proposed openstack-infra/shade: Allow for passing cache class in as a parameter https://review.openstack.org/162148	13:24
openstackgerrit	Julien Danjou proposed openstack-infra/project-config: Move Gnocchi from Stackforge to OpenStack https://review.openstack.org/162146	13:25
openstackgerrit	Syed Ismail Faizan Barmawer proposed openstack-infra/project-config: Adding publish-to-pypi template to ironic-lib project https://review.openstack.org/162010	13:25
openstackgerrit	Julien Danjou proposed openstack-infra/project-config: Move Gnocchi from Stackforge to OpenStack https://review.openstack.org/162146	13:25
*** dkliban_afk is now known as dkliban		13:27
*** dhritishikhar has quit IRC		13:30
*** mwagner_lap has joined #openstack-infra		13:30
nibalizer	Kiall: you can probably hit our internal puppetboard to see how far its come in the year or so since the upstream one went up	13:34
*** vhoward has left #openstack-infra		13:35
*** weshay has joined #openstack-infra		13:35
*** HeOS has quit IRC		13:37
Kiall	nibalizer: If I knew the URL, I'd be able to! But I dont ;)	13:37
nibalizer	hipchat'd	13:38
*** ashleighfarnham has joined #openstack-infra		13:39
openstackgerrit	yolanda.robla proposed openstack-infra/puppet-jenkins: Add configuration for HipChat in jjb https://review.openstack.org/162152	13:39
*** dboik has quit IRC		13:43
*** gordc has joined #openstack-infra		13:43
*** ashleighfarnham has quit IRC		13:44
*** radez_g0n3 is now known as radez		13:46
*** nelsnelson has joined #openstack-infra		13:47
*** SumitNaiksatam has quit IRC		13:48
*** dustins has joined #openstack-infra		13:49
*** mpavone_ has quit IRC		13:49
openstackgerrit	Syed Ismail Faizan Barmawer proposed openstack/requirements: Add ironic-lib to project.txt and global-requirements.txt https://review.openstack.org/161603	13:52
*** julim has joined #openstack-infra		13:52
*** sigmavirus24_awa is now known as sigmavirus24		13:54
*** yamamoto has joined #openstack-infra		13:54
*** HeOS__ has joined #openstack-infra		13:56
sigmavirus24	Is there an issue with nodes in zuul's pool using old versions of project-config? os-ansible-deployment keeps seeing http://logs.openstack.org/75/161975/3/check/os-ansible-deployment-dsvm-check-commit/06ecd42/console.html#_2015-03-06_10_01_58_615 pop up every once and a while which is the result of a version of project config from several days ago	13:56
*** miqui has joined #openstack-infra		13:56
*** HeOS_ has quit IRC		13:56
*** mpavone has joined #openstack-infra		13:57
*** BharatK has quit IRC		13:57
*** gibi_ has joined #openstack-infra		13:57
*** AJaeger has joined #openstack-infra		13:59
*** luqas__ has joined #openstack-infra		13:59
*** annegentle has joined #openstack-infra		13:59
*** mattfarina has joined #openstack-infra		14:00
*** mpavone has quit IRC		14:00
*** annegentle has quit IRC		14:00
*** annegentle has joined #openstack-infra		14:01
*** ihrachyshka has joined #openstack-infra		14:01
*** gibi has quit IRC		14:01
*** Longgeek has quit IRC		14:05
*** fifieldt_ has quit IRC		14:05
Shrews	mordred: should shade's caching be 'opt-in' as well?	14:06
*** zz_jgrimm is now known as jgrimm		14:07
*** dboik has joined #openstack-infra		14:07
Shrews	if os-client-config is used, the behaviour will change to 'opt-in' anyway with your recent o-c-c change if no cache is defined	14:07
*** sc68cal_ has joined #openstack-infra		14:07
*** sc68cal_ has quit IRC		14:07
*** sc68cal_ has joined #openstack-infra		14:07
*** dboik has quit IRC		14:07
*** dboik_ has joined #openstack-infra		14:08
*** bcourt has joined #openstack-infra		14:08
*** sc68cal has quit IRC		14:09
*** salv-orl_ has joined #openstack-infra		14:09
*** salv-orlando has quit IRC		14:09
*** bcourt has quit IRC		14:10
Shrews	left comment on the review	14:11
mordred	Shrews: it should be opt-in with SpamapS's change	14:12
Shrews	review #?	14:12
*** tnovacik has quit IRC		14:13
Shrews	oh, the dependency	14:13
*** dizquierdo has quit IRC		14:14
*** ddieterly has joined #openstack-infra		14:15
*** resker has quit IRC		14:16
openstackgerrit	gordon chung proposed openstack-infra/project-config: add experimental ceilometer elasticsearch db jobs https://review.openstack.org/160824	14:16
*** garyh has joined #openstack-infra		14:16
*** markus_z has quit IRC		14:17
*** Longgeek has joined #openstack-infra		14:20
*** signed8bit has joined #openstack-infra		14:21
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Add configurable url for OpenStack status https://review.openstack.org/162169	14:25
*** SumitNaiksatam has joined #openstack-infra		14:28
*** peristeri has joined #openstack-infra		14:29
*** Ala has quit IRC		14:29
*** dansmith is now known as superdan		14:30
*** annegentle is now known as superanne		14:31
*** otter768 has joined #openstack-infra		14:33
*** chlong has quit IRC		14:34
*** dims has quit IRC		14:37
*** otter768 has quit IRC		14:37
*** dims has joined #openstack-infra		14:38
*** bhunter71 has joined #openstack-infra		14:38
*** tiswanso has joined #openstack-infra		14:38
openstackgerrit	David Lyle proposed openstack-infra/project-config: Adding xstatic notifications to openstack-horizon https://review.openstack.org/162174	14:38
*** Ala has joined #openstack-infra		14:40
openstackgerrit	Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for cinder CLI race bug 1429151 https://review.openstack.org/162176	14:40
openstack	bug 1429151 in Cinder "SimpleReadOnlyCinderClientTest.test_cinder_encryption_type_list fails randomly" [Undecided,New] https://launchpad.net/bugs/1429151	14:40
*** ddieterly has quit IRC		14:41
*** nelsnelson has quit IRC		14:41
*** ddieterl_ has joined #openstack-infra		14:41
*** ZZelle has quit IRC		14:43
*** sdake_ has joined #openstack-infra		14:43
dhellmann	anteaya: I'm confused by your comment about the tarball site on https://review.openstack.org/#/c/161447/4 -- I thought that was used by publish-to-pypi?	14:43
*** tonytan4ever has joined #openstack-infra		14:44
*** alaski is now known as lascii		14:44
jroll	is there a jenkins thing to upload to swift?	14:46
jroll	or did y'all just write code to do things like that	14:46
sigmavirus24	jroll: code was written to do that	14:47
sigmavirus24	I believe clarkb wrote it	14:47
*** sdake__ has joined #openstack-infra		14:47
jroll	ok, thanks	14:47
* jroll finds his swift client		14:48
*** mrmartin has joined #openstack-infra		14:48
*** sdake_ has quit IRC		14:48
sigmavirus24	jroll: I believe https://review.openstack.org/#/c/156788/ added it	14:49
mordred	jroll: we try to not have jenkins itself do things	14:49
mordred	jroll: because two reasons a) jenkins breaks b) we're working on getting rid of it	14:50
jroll	o.o	14:50
jroll	mordred: getting rid of jenkins in favor of?	14:50
mordred	jroll: b) follows from a) in a non-insignificant way	14:50
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Add configurable graphite url on zuul status https://review.openstack.org/162177	14:50
jroll	right	14:50
mordred	jroll: zuul and nodepool directly orchestrating work	14:50
jroll	interesting	14:50
mordred	jroll: you should check out jeblair's zuul v3 email if you haven't already	14:51
*** mtanino has joined #openstack-infra		14:51
jroll	mordred: neat, thanks	14:51
jroll	I'm setting up a jenkins for our team to just build some containers and whatnot, so trying to keep it simple :)	14:51
jroll	vastly different scale :P	14:51
mordred	jroll: but before zuul v3 was hatched, the plan for getting rid of jenkins involved just running gearman workers on the nodes like turbohipster does	14:51
mordred	jroll: yah - I mean, jenkins is fine for the normal workgroup size stuff	14:52
*** boris-42 has quit IRC		14:52
mordred	although we do hope that once we get zuul v3 going, it'll be easy to scale it _down_ to do simple tasks like that too	14:52
*** rmcall has joined #openstack-infra		14:52
openstackgerrit	Doug Hellmann proposed openstack-infra/project-config: import aeromancer to stackforge https://review.openstack.org/161447	14:52
openstackgerrit	Doug Hellmann proposed openstack-infra/project-config: Add detail to gerrit ACL normalization errors https://review.openstack.org/161914	14:52
*** amitgandhinz has joined #openstack-infra		14:52
jroll	mordred: cool	14:53
dhellmann	anteaya: let me know if ^^ looks more like what I should have if what I want is for the CI system to manage my releases	14:53
jroll	thanks for the info	14:53
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Adding publish-to-pypi template to ironic-lib project https://review.openstack.org/162010	14:53
AJaeger	ihrachyshka: what's the problem with https://review.openstack.org/161569 ?	14:54
ihrachyshka	AJaeger, I still see failures in e.g. neutron/juno. But need to recheck to make sure it's still the case.	14:54
AJaeger	ihrachyshka: do you have a recent run as reference?	14:55
*** e0ne is now known as e0ne_		14:55
AJaeger	dhellmann, jd__ : Could you review https://review.openstack.org/157472 and https://review.openstack.org/158379, please?	14:55
ihrachyshka	AJaeger, https://review.openstack.org/#/c/161930/ ?	14:56
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Add configurable graphite url on zuul status https://review.openstack.org/162177	14:56
openstackgerrit	Merged openstack-infra/elastic-recheck: Add query for cinder CLI race bug 1429151 https://review.openstack.org/162176	14:56
openstack	bug 1429151 in Cinder "SimpleReadOnlyCinderClientTest.test_cinder_encryption_type_list fails randomly" [High,Confirmed] https://launchpad.net/bugs/1429151	14:56
*** amotoki has quit IRC		14:57
dhellmann	AJaeger: +2	14:57
AJaeger	dhellmann: thanks!	14:57
AJaeger	ihrachyshka: looks like the old job definition was used. mordred, clarkb: Could you check http://logs.openstack.org/30/161930/1/check/check-neutron-dsvm-functional/7aa1c5b/console.html - it seems that patch 161569 was not applied	14:59
ihrachyshka	AJaeger, maybe puppet not applied to jenkins? I don't know much about mechanics.	15:00
*** e0ne_ has quit IRC		15:01
*** persia has quit IRC		15:01
AJaeger	ihrachyshka: that would explain it - that's why I need jeblair, mordred or clarkb to check...	15:01
ihrachyshka	AJaeger, afaik we were not sure that it will work since the author is not fluent in project-config speak	15:01
ihrachyshka	it was more a wild guess?	15:01
jd__	AJaeger: sure	15:02
AJaeger	thanks, jd__ !	15:02
ihrachyshka	ah, missed the 2nd patchset there. then I guess it's more grounded	15:02
*** yfried has quit IRC		15:04
AJaeger	jeblair, mordred, clarkb: https://review.openstack.org/161447 just failed jobs - and looking at the failures, I only get "file not found". Any problems with Jenkins/Zuul/gearman?	15:05
AJaeger	and two more failures just coming in. Something looks broken ;(	15:06
jeblair	https://jenkins02.openstack.org/job/gate-infra-puppet-apply-precise/1443/console	15:06
jeblair	2015-03-06 15:01:39.899 \| ImportError: No module named glob2	15:07
jeblair	i'm guessing a base image changed ?	15:07
mordred	yay!	15:07
*** dims is now known as dimsum__		15:07
jeblair	that was bare-precise-rax-dfw	15:08
*** deepakcs has joined #openstack-infra		15:08
*** mrmartin has quit IRC		15:08
AJaeger	jeblair: thanks for looking into this!	15:08
jeblair	found one on rax-iad too	15:09
*** achuprin_ has quit IRC		15:10
jeblair	we started using new images there about 20 mins ago	15:11
*** radez is now known as radez_g0n3		15:12
*** hdd_ has joined #openstack-infra		15:13
*** persia_ has joined #openstack-infra		15:13
*** persia_ has quit IRC		15:13
*** persia_ has joined #openstack-infra		15:13
*** jlanoux has quit IRC		15:14
jeblair	oh	15:15
jeblair	we just merged a change to the script to use glob2, but did not add it to the venv	15:15
jeblair	this may be a signal the script is getting a bit too big to be a standalone script.	15:15
*** armax has joined #openstack-infra		15:15
jeblair	perhaps we should fold it into zuul. we already install zuul to get cloner, so it would actually be simpler anywya.	15:16
*** wenlock has joined #openstack-infra		15:16
mordred	++	15:16
mordred	jeblair: I agree 100%	15:16
jroll	mordred: is this thing awful or awesome? https://wiki.jenkins-ci.org/display/JENKINS/JClouds+Plugin	15:17
mordred	jroll: it's not usable by us	15:17
jroll	mordred: not for you	15:17
jroll	for me	15:17
mordred	jroll: but we funded its creattion :)	15:17
jroll	:P	15:17
mordred	jroll: dunno	15:17
mordred	I've literally never been able to get it to work	15:17
jroll	huh	15:17
mordred	because clouds are too flaky	15:17
jroll	probably needs a real service catalog or something	15:17
mordred	that's why we wrote nodepool	15:17
jroll	I just want to use the storage bits	15:17
mordred	no - it's just that handling all of the ways clouds fail is hard	15:17
mordred	oh - well that's probably fine	15:18
jroll	which shouldn't be super flaky	15:18
jroll	yeah	15:18
jroll	ok, thanks :)	15:18
*** wenlock has quit IRC		15:19
AJaeger	jeblair: argh, the swift uploader change	15:19
*** changbl has quit IRC		15:19
jeblair	AJaeger: hrm, that change did come with a command to install glob2	15:20
jeblair	AJaeger: so i'm back to not quite understanding what's wrong...	15:20
*** esker has joined #openstack-infra		15:20
*** mrmartin has joined #openstack-infra		15:20
*** r-daneel has joined #openstack-infra		15:20
*** timcline has joined #openstack-infra		15:20
*** wenlock has joined #openstack-infra		15:21
openstackgerrit	Filip Blaha proposed openstack-infra/project-config: Congress client repository clone fix https://review.openstack.org/162188	15:21
*** timcline_ has joined #openstack-infra		15:22
jeblair	oh!	15:23
jeblair	it's because puppet is stopped on nodepool.o.o	15:23
*** marun has joined #openstack-infra		15:24
jeblair	so the image scripts are out of sync with the project_config repo that gets cloned on the image	15:25
AJaeger	jeblair: see above the discussion with ihrachyshka, where we looked into a job that looked it was an old job, so puppet did not update jenkins	15:25
AJaeger	is puppet disabled on the jenkins notes as well?	15:26
*** timcline has quit IRC		15:26
jeblair	AJaeger: it shouldn't be	15:26
jeblair	AJaeger: except something is wrong with jenkins04	15:26
jeblair	AJaeger: http://puppetboard.openstack.org/	15:26
jeblair	AJaeger: so if it ran there, it would be outdated	15:26
*** doude has quit IRC		15:27
jeblair	i think it's safe to re-enable puppet on nodepool	15:27
*** dmsimard_away is now known as dmsimard		15:28
*** jlanoux has joined #openstack-infra		15:28
jeblair	clarkb, AJaeger: i'm going to re-enable puppet on nodepool and delete all current bare- images	15:28
*** hdd_ has quit IRC		15:28
ihrachyshka	armax, ^^	15:28
*** ashleighfarnham has joined #openstack-infra		15:29
AJaeger	jeblair: seems it was run on jenkins03 - scroll back for discussion with ihrachyshka ...	15:29
AJaeger	jeblair: But let's first fix the images.	15:29
armax	ihrachyshka: but I am still confused	15:29
jeblair	mordred: you want to see why puppet isn't running on jenkins04?	15:30
armax	ihrachyshka: if I look at the expression [[ ! "$BRANCH_OVERRIDE" =~ ^stable/(icehouse\|juno)$ ]] \|\| [[ ! "$ZUUL_BRANCH" =~ ^stabl	15:30
armax	e/(icehouse\|juno)$ ]] as clarkb suggested	15:30
armax	we should still be getting in the if branch for stable branches even if we don’t want to	15:30
ihrachyshka	le sigh	15:31
armax	ihrachyshka: as a matter of fact, the aim of the fix should be ‘not to get into the if branch’ if we are stable	15:31
ihrachyshka	armax, remove '!'s?	15:31
armax	ihrachyshka: my bad for copying and pasting not paying any attention at what I was told :O	15:31
armax	I think that we should not use BRANCH_OVERRIDE at all	15:32
*** changbl has joined #openstack-infra		15:32
armax	ihrachyshka: it sounds to me that it is == default regardless of the branch, so what’s the point in using it to determine whether we are on a stable branch or not?	15:32
*** e0ne has joined #openstack-infra		15:32
armax	ihrachyshka: but I have very little clue on how the infra stuff works	15:33
armax	ihrachyshka: apparently I am not alone :)(	15:33
ihrachyshka	armax, neither me, we're both blind	15:33
*** tiswanso has quit IRC		15:33
*** ashleighfarnham has quit IRC		15:33
*** radez_g0n3 is now known as radez		15:34
*** yamamoto has quit IRC		15:34
*** zz_dimtruck is now known as dimtruck		15:34
*** yamamoto has joined #openstack-infra		15:34
*** tiswanso_ has joined #openstack-infra		15:34
*** achuprin_ has joined #openstack-infra		15:35
AJaeger	armax, ihrachyshka you confused me as well with this change ;)	15:35
openstackgerrit	Armando Migliaccio proposed openstack-infra/project-config: Remove BRANCH_OVERRIDE from bool condition https://review.openstack.org/162199	15:36
armax	AJaeger: ^^	15:36
clarkb	no dont remove it ..	15:36
ihrachyshka	AJaeger, so I guess it's both 'jenkins not updated' and 'code is wrong' issues	15:36
*** tiswanso_ has quit IRC		15:36
armax	clarkb: ok, can you elaborate why?	15:36
clarkb	should \|\| be && I may have flipped logic	15:36
clarkb	because if you are overridden you dont want to be in setting the projects...	15:36
AJaeger	might be, ihrachyshka. Note the job might fail due to the broken images.	15:37
armax	clarkb: but I still don’t understand why we’d want it to determine a condition when the value doesn’t change from one branch to another	15:37
clarkb	the idea is if icehouse or juno dont do this	15:37
clarkb	so if someone makes a job that says "this is icehouse" then you dont want to do that	15:37
armax	so if branch_override is default regarless of whether we run master or juno	15:37
clarkb	its not necessarily	15:37
armax	I beg to understand why we’d want to use it? in any condition OR or AND	15:37
clarkb	it may be icehouse or juno	15:38
clarkb	thats the whole point	15:38
armax	clarkb: when?	15:38
clarkb	when people make jobs with the override	15:38
*** dizquierdo has joined #openstack-infra		15:38
mkj_wk	I could use some guidance on normal Gerrit expectations, for my new feature changeset, now in its 20th revision... Would it be considered polite for me to +1 my own patch as an indication that I think it's ready for merge and would appreciate review/merge?	15:39
lxsli	Hmm not sure this is the right place to raise this, but has anyone suggested colour-coding badges at Vancouver?	15:39
armax	clarkb: this is utterly confusing	15:39
lxsli	Primarily for the benefit of security, and hence decreasing queuing times	15:39
armax	clarkb: so I can have BRANCH_OVERRIDE=default and ZUUL_BRANCH=juno?	15:40
clarkb	armax yes	15:40
armax	clarkb: what does that mean?	15:40
jeblair	mkj_wk: you could just leave a comment to that effect	15:40
clarkb	it means zuul has given you a change from juno so do the default behavior for juno	15:40
AJaeger	mkj_wk: what's the URL for the patch?	15:41
clarkb	armax but override juno branch master means this is a change from master but test it against juno	15:41
*** ChuckC_ has joined #openstack-infra		15:41
mkj_wk	AJaeger: https://review.openstack.org/#/c/158877/	15:41
armax	clarkb: for the sake of what it needs to be done, I still reckon that we don’t need BRANCH_OVERRIDE at all then	15:41
*** ChuckC has quit IRC		15:42
mkj_wk	jeblair: thanks :)	15:42
clarkb	armax then please completely remove it from your job.	15:42
armax	clarkb: that said, what do you mean?	15:42
clarkb	162199 is still wrong regardless	15:42
armax	like remove the export?	15:43
AJaeger	mkj_wk: ah, a patch for git-review. Let's ask clarkb and jeblair kindly for a review	15:43
clarkb	and the template parameter	15:43
*** tiswanso has joined #openstack-infra		15:43
armax	clarkb: so the options are two, if I understand you correctly, flip the \|\| to && or get rid of the BRANCH_OVERRIDE entirely?	15:44
clarkb	yes	15:44
armax	clarkb: ok then, since I am a lazy moron, I am going to go for the easiest change	15:44
mkj_wk	AJaeger: that brings up the next politeness question -- does one normally add someone else as a reviewer to ones own patch, or just ping on IRC?	15:44
openstackgerrit	Armando Migliaccio proposed openstack-infra/project-config: Flip the bool condition https://review.openstack.org/162199	15:45
armax	clarkb: assuming that ZUUL_BRANCH is a stable branch this should be good now	15:46
*** ChuckC_ has quit IRC		15:46
AJaeger	mkj_wk: I prefer pinging on IRC, people might not see the adding	15:46
superanne	anteaya: I'm trying to reach as many vendor driver writers as possible, can you think of a list that might be useful? I've been using -docs and -dev mostly.	15:46
mkj_wk	ok, well I suspect jeblair and clarkb saw your mention as a ping then :)	15:46
superanne	anteaya: wondering about 3rd party list	15:47
AJaeger	mkj_wk: hope so. But your patch is large, so reviewing needs time	15:47
jeblair	mkj_wk: we should ask mordred and/or fungi to look at it, though fungi is on vacation right now	15:48
*** achuprin_ has quit IRC		15:48
anteaya	superanne: this should get you started: https://review.openstack.org/#/admin/groups/270,members	15:48
mkj_wk	AJaeger: it was quite small when I started, but it's much better tested now	15:48
superanne	anteaya: perfect, thanks	15:48
*** arxcruz has quit IRC		15:48
anteaya	superanne: they also should all be listed here: https://wiki.openstack.org/wiki/ThirdPartySystems	15:49
AJaeger	mkj_wk: ;) Please have patience, reviewing can take some time.	15:49
mkj_wk	jeblair: yeah, I saw fungi head off on vacation just as I was really getting to work on the feature	15:49
anteaya	superanne: good luck	15:49
mkj_wk	AJaeger: oh, I'm patient. I just didn't want it to be stuck in limbo because I didn't realize I was supposed to ping someone	15:49
mkj_wk	AJaeger: I've been on the other end of this elsewhere	15:49
anteaya	mind if I ask what your project is?	15:49
superanne	anteaya: will cross posting super annoy them?	15:49
anteaya	cross posting what?	15:49
superanne	anteaya: it's informing them of the driver doc spec AJaeger completed	15:49
superanne	anteaya: basically "here's your template, expectations, and deadline for kilo"	15:50
superanne	anteaya: cross posting to 3rd party, -dev, and -docs	15:50
AJaeger	mkj_wk: you don't have to ping, people will review at their time.	15:50
openstackgerrit	Julien Danjou proposed openstack-infra/project-config: Move Gnocchi from Stackforge to OpenStack https://review.openstack.org/162146	15:50
mkj_wk	superanne: you might care to look at it, since I was attempting to implement a feature that you asked for almost two years ago. ☺	15:50
*** skolekonov has quit IRC		15:50
mkj_wk	AJaeger: thanks for the clues :)	15:51
anteaya	superanne: oh in that case, draft up and email and we can send it to third-party-announce mailing list	15:51
anteaya	since they all are expected to be subscribed to that	15:51
anteaya	jeblair: any objections to sending the driver doc spec to -announce?	15:52
superanne	anteaya: ah nice. if I send it to -docs and -dev and send you a link is that what you need?	15:52
anteaya	no don't spam dev with it	15:52
* AJaeger will be back later, time for some cooking...		15:52
anteaya	just draft it in an etherpad so I can take a peek	15:52
anteaya	then we just send it to -announce and -docs at teh same time	15:53
*** sdake__ has quit IRC		15:53
*** AJaeger is now known as AJaeger_cooking		15:53
jeblair	anteaya: no	15:54
anteaya	jeblair: okay	15:54
anteaya	jeblair: what do you suggest?	15:54
jeblair	anteaya: that you send it to announce. you asked if i had objections, and i said i had no objections.	15:54
anteaya	ah	15:55
anteaya	thank you, great	15:55
sdague	jeblair: so on this https://review.openstack.org/#/c/161911/, is that actually what you want to do? The devstack plugin architecture was really more designed with the plugin being in the project repo.	15:55
anteaya	superanne: okay give me a few minutes	15:55
jeblair	sdague: oh okay, are there projects that implement that? (why are there devstack-plugin projects then?)	15:56
sdague	the only times we've built fully new repos is for stuff like glusterfs support, which don't really have a project repo to live in, as it's not a new project but a config tweak	15:56
sdague	stackforge/ec2-api	15:56
sdague	a bunch of the neutron plugins	15:56
sdague	I think gnocchi as well	15:56
*** ayoung has joined #openstack-infra		15:57
sdague	I pushed it for tuskar recently, I don't remember if they landed it or not	15:57
anteaya	superanne: okay so when you are ready just cc third-party-announce@lists.openstack.org when you send it out to -docs	15:57
anteaya	superanne: keep in mind these folks are used to having deadlines and consequence	15:57
anteaya	if they don't have "this is waht will happen if you don't" they will just get confused and come in here and ask	15:58
*** nelsnelson has joined #openstack-infra		15:58
anteaya	so my take on what happens if they don't is, they don't get included in the docs	15:58
anteaya	which is fine with me	15:58
jeblair	sdague: okay, i see how that works, thanks. i'll abandon it then.	15:58
anteaya	superanne: the -announce list is moderated so I will keep an eye out for it and usher it through when I see it	15:58
*** superanne has quit IRC		15:59
jeblair	sdague: does that means 161887 is a bad idea?	15:59
*** achuprin_ has joined #openstack-infra		16:00
*** superanne has joined #openstack-infra		16:00
sdague	well... I think it won't be used a ton	16:00
openstackgerrit	gordon chung proposed openstack-infra/project-config: add experimental ceilometer elasticsearch db jobs https://review.openstack.org/160824	16:01
*** stevemar has joined #openstack-infra		16:01
jeblair	sdague: still do it, but just document that it should only be used if in-repo is not appropriate?	16:01
sdague	yes, I think that makes sense	16:01
openstackgerrit	Brian Hunter proposed openstack-infra/system-config: grammar updates for gerrit documentation https://review.openstack.org/160185	16:02
jeblair	sdague: k, thx. i'll write a patch to copy the devstack stuff into trove	16:02
ihrachyshka	clarkb, just wondering, why would I want to run master patch with juno default?	16:03
*** hashar is now known as hasharAway		16:03
jeblair	clarkb: since it's early still, i'd like to kick off bare-* image builds; how should i go about that?	16:03
sdague	the places I expect that we'll need it is where there is some 3rd party software outside the openstack space you need to configure for a thing. So filesystems like glusterfs and ceph (which we'll try to get into one of these as well).	16:04
sdague	all the neutron stuff has driver trees now, so that should be fine	16:04
clarkb	for branchless projects testing against other things	16:04
clarkb	for example neutronclient	16:04
sdague	or tempest	16:04
*** maune has joined #openstack-infra		16:04
clarkb	jeblair bare-* are not dib so just image-update	16:04
clarkb	jeblair I like to run one command per region for parallel builds in screen	16:05
jeblair	clarkb: ok, thanks	16:05
superdan	hi all, the devstack patch merged that this depends on: https://review.openstack.org/#/c/160893/	16:06
sdague	jeblair: unrelated, been playing with https://pypi.python.org/pypi/pycscope/ and emacs integration this morning. Thought you might appreciate it if you haven't seen it before.	16:06
superdan	would love to get that in soon so I can start doing testing against devstack	16:06
*** imcsk8 has quit IRC		16:07
*** imcsk8 has joined #openstack-infra		16:07
*** e0ne is now known as e0ne_		16:07
*** superanne has quit IRC		16:07
openstackgerrit	Lucas Alvares Gomes proposed openstack-infra/project-config: Ironic: Move pxe_ipa-nv to check instead of experimental https://review.openstack.org/162204	16:07
sdague	superdan: +A	16:08
superdan	sdague: thanks	16:08
*** superanne has joined #openstack-infra		16:08
*** carl_baldwin has joined #openstack-infra		16:08
*** dizquierdo has quit IRC		16:08
jeblair	sdague: oh cool, i had not seen that, thanks.	16:10
*** e0ne_ is now known as e0ne		16:10
*** mpaolino has quit IRC		16:10
*** mpaolino has joined #openstack-infra		16:11
*** ChuckC_ has joined #openstack-infra		16:12
*** deepakcs has quit IRC		16:12
*** jaypipes has joined #openstack-infra		16:14
*** mbacchi has quit IRC		16:14
*** jaypipes is now known as leakypipes		16:14
*** nelsnelson has quit IRC		16:16
*** teran_ has quit IRC		16:17
*** achuprin_ has quit IRC		16:18
clarkb	jeblair for dib images I do a single image-build the image-update per region	16:19
clarkb	er	16:19
clarkb	image-upload not image-update	16:19
jeblair	apparently we can't build centos6 images in rax	16:19
jeblair	2015-03-06 16:07:54,613 INFO nodepool.image.build.rax-dfw.bare-centos6: setuptools must be installed to install from a source distribution	16:19
jeblair	we have some that are 25 hours old	16:20
*** rfolco has quit IRC		16:20
*** dizquierdo has joined #openstack-infra		16:20
*** sabari has quit IRC		16:20
*** sabari has joined #openstack-infra		16:23
*** krtaylor has quit IRC		16:23
*** scheuran_ has quit IRC		16:25
*** dannywilson has joined #openstack-infra		16:27
*** garyk has quit IRC		16:28
*** dannywilson has quit IRC		16:28
*** ghostpl_ has quit IRC		16:30
*** dannywilson has joined #openstack-infra		16:32
*** ParsectiX has joined #openstack-infra		16:32
*** amotoki has joined #openstack-infra		16:33
*** chlong has joined #openstack-infra		16:33
*** ghostpl_ has joined #openstack-infra		16:34
*** otter768 has joined #openstack-infra		16:34
*** krtaylor has joined #openstack-infra		16:35
*** achuprin_ has joined #openstack-infra		16:35
*** nelsnelson has joined #openstack-infra		16:36
anteaya	clarkb jeblair are we having more swift log upload issues? https://review.openstack.org/#/c/161447/4	16:36
anteaya	sorry I haven't read all the backscroll yet	16:36
*** rfolco has joined #openstack-infra		16:36
anteaya	latest run has File Not Found instead of logs	16:36
anteaya	and the run prior to that as well	16:37
*** otter768 has quit IRC		16:38
jeblair	anteaya: can you link to the failed run instead?	16:38
anteaya	and can someone add kevinbenton to https://review.openstack.org/#/admin/groups/632,members when they get a moment?	16:38
jeblair	anteaya: done	16:39
anteaya	thanks	16:39
anteaya	link to the failed run? http://logs.openstack.org/47/161447/5/check/gate-infra-puppet-apply-centos6/16d57ef/	16:39
anteaya	is that what you mean?	16:39
anteaya	and another: http://logs.openstack.org/47/161447/5/check/gate-infra-puppet-apply-precise/1444292/	16:39
anteaya	I've got more	16:40
Ng	mordred: are you using a patched vhd-util in your upload-rax.sh? (hence the docker madness in the script?)	16:40
mordred	Ng: yes	16:41
jeblair	anteaya: that's from the same time we were having problems	16:41
anteaya	ah	16:41
anteaya	so the way forward is recheck? or are we still looking for a solution?	16:41
mordred	Ng: I took an initial stab at making debs here: https://launchpad.net/~mordred/+archive/ubuntu/vhd-util but haven't followed up	16:41
*** sdake_ has joined #openstack-infra		16:42
openstackgerrit	Merged openstack/requirements: Require openstack-doc-tools 0.23 https://review.openstack.org/157472	16:42
jeblair	anteaya: recheck. fixed a while ago	16:42
anteaya	thanks	16:42
openstackgerrit	Merged openstack/requirements: Update openstackdocstheme https://review.openstack.org/158379	16:42
*** unicell1 has quit IRC		16:42
mordred	jeblair: wtf on the centos6 in rax issue - aren't we using get-pip everywhere?	16:42
clarkb	what problems were we having?	16:42
openstackgerrit	Merged openstack/requirements: Adds Autobahn to global requirements https://review.openstack.org/156982	16:43
openstackgerrit	Merged openstack/requirements: Add python-openstackclient https://review.openstack.org/160887	16:43
clarkb	oh!	16:43
openstackgerrit	Merged openstack/requirements: Bump oslo.context version to 0.2.0 https://review.openstack.org/161868	16:43
*** sdake__ has joined #openstack-infra		16:43
jeblair	clarkb: i pinged you on it earlier to fyi you. puppet was stopped on nodepool, so the images prep scripts were not in sync with on-image copies of project-config, so zuul swift upload was missing glob2	16:43
clarkb	ya reading sb now	16:43
jeblair	clarkb: i deleted most recent images, re-enabled puppet, and started building new ones just cause	16:44
clarkb	cool	16:44
jeblair	but yeah, i have no idea what's up with centos6	16:44
mordred	jeblair: just cause is a great reason	16:44
clarkb	is devatack-fedora21-dib still failing to build with puppet enabled?	16:44
jeblair	clarkb, mordred: http://paste.openstack.org/show/190455/	16:44
clarkb	that is an amazing error	16:45
jeblair	clarkb: yes, f21 seems to be in a tight fail loop	16:45
jeblair	clarkb: i think we should remove it from the config	16:45
anteaya	clarkb jeblair can you post when the new images are finished building? the enable swift logs on python-jobs patch is waiting on the image builds	16:45
clarkb	jeblair +1	16:45
jeblair	anteaya: they are not building	16:46
clarkb	I wonder if setuptools released without a whl?	16:46
mordred	Ng: https://github.com/emonty/deb-vhd-util <-- there's the packaging repo for it - in case you decide to punch that in the face instead of the docker	16:46
Ng	mordred: I'm going to be honest here, I just build a separate vhd-util binary and shoved it in ~/bin/ ;)	16:46
Ng	*built	16:46
clarkb	there is a whl on pypi	16:47
*** sdake_ has quit IRC		16:47
Ng	mordred: I am curious why the convert patch has been floating around since 2012 and apparently not made it into either debian or fedora	16:47
openstackgerrit	James E. Blair proposed openstack-infra/system-config: Remove devstack-fedora21-dib https://review.openstack.org/162217	16:47
clarkb	because its terrible C	16:47
jeblair	clarkb: ^	16:47
*** devvesa has quit IRC		16:48
Ng	clarkb: well that's no excuse!	16:48
jeblair	clarkb: i guess we're not building any nodes again?	16:48
jeblair	jeez	16:48
anteaya	jeblair: oh	16:48
*** rmcall has quit IRC		16:48
*** rmcall has joined #openstack-infra		16:49
openstackgerrit	Doug Hellmann proposed openstack-infra/devstack-gate: add oslo.versionedobjects https://review.openstack.org/162218	16:49
jeblair	i am going to force-merge this patch, and then i am not going to review another nodepool code or config change again until this thing has some real testing.	16:49
*** hdd_ has joined #openstack-infra		16:49
clarkb	ok	16:49
*** gyee has joined #openstack-infra		16:49
openstackgerrit	Merged openstack-infra/system-config: Remove devstack-fedora21-dib https://review.openstack.org/162217	16:50
mordred	Ng: I have literally no idea	16:52
jeblair	nodes are being launched now	16:52
clarkb	mordred really its bad C iirc there was a review of it and no one worked to fix it	16:52
*** yamahata has joined #openstack-infra		16:52
*** boris-42 has joined #openstack-infra		16:53
mordred	clarkb: sure. but that has never stopped distros from using patches before	16:53
*** notmyname has joined #openstack-infra		16:54
*** sambetts has quit IRC		16:55
*** purp is now known as purp_away		16:55
*** sambetts has joined #openstack-infra		16:57
greghaynes	Ng: Theres also https://review.openstack.org/#/c/156433/	16:58
greghaynes	(totally untested)	16:58
*** jerryz_ has quit IRC		16:58
mordred	greghaynes: nice	16:58
jeblair	greghaynes: does that require the patched vhd-util?	16:59
greghaynes	yes	16:59
cody-somerville	What version of setuptools is openstack currently using?	17:00
mordred	yah - but it'll allow us to tell dib to output a vhd and a qcow2	17:00
mordred	cody-somerville: the latest	17:00
mordred	cody-somerville: don't ever attempt to pin setuptools, you will fail horriblyand die	17:00
mordred	cody-somerville: I mean, unless you want to :)	17:00
anteaya	2015-03-06T15:15:15 <jeblair> we just merged a change to the script to use glob2, but did not add it to the venv	17:00
*** changbl has quit IRC		17:00
anteaya	so that was me	17:00
anteaya	help me to learn what I should have done	17:01
jeblair	anteaya: that statement was wrong, keep reading	17:01
Ng	greghaynes: nice	17:01
anteaya	I'm reading	17:02
zaro	morning	17:02
anteaya	I see the problem was puppet was stopped on nodepool.o.o	17:02
anteaya	I still want to get better at doing the right thing though	17:02
mordred	greghaynes: I know you said "completely untested" ... last time I tried a multi-format output, the second format failed because a variable got stepped IIRC	17:02
mordred	greghaynes: we shoudl probably put "make sure format=qcow2,vhd works" on our TDL :)	17:03
jeblair	anteaya: okay, but you didn't do anything wrong, so good luck improving :)	17:03
bhunter71	quick question, sorry to butt in: if I was to run my devstack-trusty at say 2 vcpus versus 4 - what would be the expected impact? ie - are the tests CPU bound and concurrent/multiprocess? would I notice significantly longer runs, do you think? well, there's surely one way to find out!	17:03
anteaya	thanks, it a daily thing	17:04
openstackgerrit	Merged openstack-infra/project-config: Enable dsvm testing on oslo.versionedobjects https://review.openstack.org/160893	17:04
ihrachyshka	anteaya, please https://review.openstack.org/#/c/161303/	17:05
jeblair	clarkb: maybe there was not a wheel at the time? i will kick off builds again and see	17:05
greghaynes	mordred: wait, you want to actually use the code? Clearly we should have designed it for this.	17:05
*** jlanoux has quit IRC		17:05
clarkb	jeblair: ya either that or get-pip's pip is selecting the sdist for some reason. Maybe that wheel doesn't properly advertise support on say linux because it has to link against things?	17:06
greghaynes	but yes, that actually could be a really simpe test case - build a few combinations of super minimal images	17:06
jeblair	clarkb: i just ran it on a not entirely clean vm and it got a wheel, which is what prompted me to think something different might happen	17:06
greghaynes	since im pretty sure the different dib output formats dont care if the image is basically empty	17:07
jeblair	clarkb: and yeah, one of the builds just got a wheel and passed that point	17:07
*** Ala has quit IRC		17:07
jeblair	clarkb: so it seems temporal?	17:07
anteaya	ihrachyshka: we have a bit of a backlog in zuul and I am still reading backscroll but I will get to it	17:07
ihrachyshka	thanks	17:07
jeblair	mordred: fyi centos6 pip issue was temporary lack of wheel from pypi	17:07
clarkb	jeblair: weird	17:08
anteaya	ihrachyshka: and try to save pinging me for reviews for crisis things, because if you ping me for every patch you write, your pings will be lost to me	17:08
mordred	oh - like, "it was in the middle of being uploaded" or something?	17:08
*** superanne has quit IRC		17:08
ihrachyshka	anteaya, I don't ping for each review, only those that break branches for stable	17:08
*** vilobhmm has joined #openstack-infra		17:09
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Test that dib failures don't stop node builds https://review.openstack.org/162237	17:09
clarkb	jeblair: ^ that is the first test in response to this	17:09
*** vilobhmm1 has joined #openstack-infra		17:09
clarkb	jeblair: that covers what you fixed	17:09
anteaya	2015-03-06T15:39:31 <lxsli> Hmm not sure this is the right place to raise this, but has anyone suggested colour-coding badges at Vancouver?	17:10
*** ghostpl_ has quit IRC		17:11
anteaya	lxsli: you might better email the foundation staff with your thoughts on summit workflows	17:11
clarkb	lxsli anteaya summit@openstack.org	17:11
clarkb	forgot to use the dib-nodepool topic, will update	17:12
*** vilobhmm has quit IRC		17:13
asselin	greghaynes, from yesterday, it does try to download the ruby1.8 package. Seem's like a bug in the package generation script. http://paste.openstack.org/show/190467/	17:14
greghaynes	asselin: nice!	17:14
*** AJaeger_cooking is now known as AJaeger		17:15
*** e0ne is now known as e0ne_		17:16
*** e0ne_ is now known as e0ne		17:16
anteaya	clarkb: thanks	17:16
jeblair	clarkb: cool, so hopefully that will fail, yeah?	17:17
clarkb	jeblair: no it passes because your fix merged I think	17:17
jeblair	oh ok	17:17
clarkb	jeblair: nodepool daemon that was running failed because we need to restart the daemon	17:17
clarkb	yup your fix is the commit that test is based on	17:18
jeblair	clarkb: do you need the sleep(3) there?	17:18
pleia2	morning	17:19
clarkb	let me check, I mostly cargo culted the test just before it (which I also wrote) so it may be wrong too	17:19
jeblair	clarkb: also, what causes the dib image to fail?	17:20
pleia2	doh, need to remove old ssh key for ci-backup-rs-ord.openstack.org	17:20
* pleia2 gets to it		17:20
*** changbl has joined #openstack-infra		17:20
anteaya	morning pleia2	17:21
clarkb	jeblair: in the yaml, I have specified an invalid base image name. The fake dib script that the tests runs exits 1 on that	17:21
asselin	greghaynes, so some bug here: https://github.com/openstack-infra/project-config/blob/master/nodepool/elements/cache-devstack/extra-data.d/55-cache-devstack-repos#L120	17:21
clarkb	jeblair: I can also add a specific assert that the dib image table is empty too and not just check nodes build	17:21
jeblair	clarkb: got it	17:21
*** lucasagomes has joined #openstack-infra		17:21
*** cdent has quit IRC		17:22
*** armax has quit IRC		17:22
lucasagomes	Hi all, I've a experimental job in Ironic and instead of building a ramdisk from scratch we download it during devstack run see: http://logs.openstack.org/44/158644/8/experimental/check-tempest-dsvm-ironic-pxe_ipa-nv/bd20459/logs/devstacklog.txt.gz#_2015-03-06_15_19_52_267	17:22
lucasagomes	is it possible to cache it somewhere?	17:22
* lucasagomes goes and look for examples		17:22
AJaeger	pleia2: do you have a few minutes to discuss https://review.openstack.org/#/c/157626/ ?	17:22
*** mpaolino has quit IRC		17:23
pleia2	AJaeger: if everyone is ok with it, that's fine, but it's still weird to me	17:23
AJaeger	pleia2: what is the weirdness? The descriptions or the check for it?	17:23
anteaya	AJaeger: me too	17:24
anteaya	I'm fine with adding descriptions	17:24
anteaya	the test seems weird to me	17:24
pleia2	AJaeger: that we're forcing projects to have descriptions when they chose not to	17:24
pleia2	I don't see the benefit	17:24
anteaya	have you looked at teh format check scripts jeblair introduced last night?	17:24
anteaya	I think checking for format is a better way to go	17:24
pleia2	and they don't know that a description has been added, so they won't know if they have to update it	17:25
jeblair	pleia2: i'm not sure that anyone has _chosen_ not to have a description	17:25
*** vilobhmm1 has quit IRC		17:25
AJaeger	anteaya: I looked briefly but don't understand what you mean checking format	17:25
AJaeger	anteaya: jeblair's script can be enhanced to check for the description entry	17:25
jeblair	pleia2: more likely, cargo-culted something that happened to not have one	17:25
pleia2	jeblair: ah, just unaware of syntax?	17:25
*** vilobhmm has joined #openstack-infra		17:25
* pleia2 nods		17:25
jeblair	pleia2: i think very early versions of that file didn't have that, so many didn't get them originally	17:26
AJaeger	pleia2, anteaya: Most recent patches have descriptions in there - the cargo cult and docu works	17:26
anteaya	I'm fine checking format of what is there	17:26
clarkb	jeblair: the test does fail without the sleep, I think it may expose a potential real problem so let me debug before approving anything	17:26
anteaya	whitespace and removing '	17:26
AJaeger	anteaya: the patch adds missing descriptions and then checks for the future...	17:26
clarkb	jeblair: though the other tests seem to ru nwithout their sleeps, so we can probably clean the sleeps up	17:26
AJaeger	anteaya: yeah, I'm fine with that kind of format cleanup	17:26
anteaya	I'm fine with a test for format	17:26
clarkb	looks like waitForNodes should do the sleep for us	17:26
*** e0ne is now known as e0ne_		17:27
AJaeger	For me it's really usefull to have some kind of descriptions for repositories. There are far too many to know what each does...	17:27
greghaynes	asselin: http://git.openstack.org/cgit/openstack-dev/devstack/tree/files/debs/trema	17:28
clarkb	jeblair: I see what the problem is, we allocate nodes to fake-provider1 even when it does not have any images ready to build that node	17:28
* asselin looking		17:28
anteaya	I have read some of the descriptions they don't help me in many cases understand any better what the project does	17:28
clarkb	jeblair: so waitForNodes() waits until it times out	17:28
anteaya	bindep for instance	17:28
greghaynes	asselin: I am still very confused why infra doesnt hit this when building images for trusty	17:28
anteaya	still have no idea what that repo does	17:28
*** unicell has joined #openstack-infra		17:28
jeblair	anteaya: er, you have an action item to adopt it into infra	17:28
AJaeger	anteaya: then let's ask for better descriptions merging it in.	17:28
greghaynes	asselin: are you building on a trusty build host or precise?	17:29
AJaeger	anteaya: and for bindep, you can surely improve it.	17:29
asselin	greghaynes, precise	17:29
anteaya	jeblair: I do and have written the patches	17:29
anteaya	I know how to move a project	17:29
greghaynes	hrm, well that kills my hypothesis then	17:29
anteaya	but I read the description it doesn't tell me what the project does	17:29
anteaya	in language I understand	17:29
*** luqas__ has quit IRC		17:29
anteaya	but that wasn't required to write the patch to move it	17:30
AJaeger	anteaya: http://git.openstack.org/cgit/stackforge/bindep/tree/README.rst	17:30
*** dtantsur is now known as dtantsur\|afk		17:30
*** dtantsur\|afk is now known as dtantsur\|pto		17:30
clarkb	jeblair: and the sleep doesn't seem to affect that, I get it to pass with and without the sleep. I am thinking a race between allocation and image build failing	17:30
AJaeger	anteaya: binary dependencies tells me already a lot - but yes, it still leaves question open	17:30
asselin	greghaynes, why?	17:30
clarkb	jeblair: eg if we allocate before the image fails it thinks it will have an image and allocates 1, if we allocate after then it knows to give both nodes to provider2	17:31
greghaynes	asselin: I was thinking infra didnt hit it because they build on precise but if youre doing that too then thats probably not the issue	17:31
openstackgerrit	Merged openstack-infra/project-config: Flip the bool condition https://review.openstack.org/162199	17:31
asselin	greghaynes, I think infra builds on trusty	17:31
*** ddieterl_ has quit IRC		17:31
*** ddieterly has joined #openstack-infra		17:31
greghaynes	ah	17:32
*** e0ne_ has quit IRC		17:32
anteaya	AJaeger: see I speak a different language	17:32
anteaya	so forcing someone to write a description that amounts to binary dependencies doens't help me	17:32
AJaeger	anteaya: I guess I did a bit more packaging than you did in my past...	17:32
anteaya	no notice I didn't vote on teh patch	17:32
anteaya	I have done zero packaging	17:33
*** jcoufal_ has quit IRC		17:33
anteaya	so if others want a test to force pepole to write stuff that amounts to binary dependanceies who am I to say no	17:33
*** rwsu-afk is now known as rwsu		17:33
anteaya	but if you ask me does that add value for me? no, no it doesn't	17:33
anteaya	formating does	17:33
*** claudiub has joined #openstack-infra		17:34
AJaeger	anteaya: Reading the README, what about as description for bindep: "Tool to check that binary packages need by applications and libraries exist"	17:34
asselin	greghaynes, for now I will work around by taking away the set -e.	17:34
anteaya	if you want	17:35
AJaeger	or "are available". It's basically like requirements.txt file but not for python packages but for binary (distribution) packages	17:35
anteaya	my point is a test for the presence of a description doesn't do anything for me	17:35
*** mpaolino has joined #openstack-infra		17:35
anteaya	formating any description that is there does	17:35
AJaeger	anteaya: I agree that only the presence is not enough. We need to review better to get good descriptions in.	17:35
anteaya	okay	17:36
greghaynes	asselin: so my reading of devstack indicates that the trema dependencies file is never actually used	17:36
AJaeger	And I'm happy to +2 a formatting patch, I just need some more time to review it.	17:36
anteaya	I haven't bee reviewing for presence, just reasonableness if they put one there	17:36
greghaynes	asselin: which would explain why everyone isnt hitting this	17:36
anteaya	AJaeger: of course	17:36
AJaeger	anteaya: but I liked where it was going...	17:36
anteaya	AJaeger: your description of bindep makes sense to me	17:36
anteaya	thanks	17:36
*** dizquierdo has quit IRC		17:36
greghaynes	clarkb: is it possible that youall are pointing your nodepool-dib builds for ubuntu at a precise apt mirror?	17:36
anteaya	AJaeger: of course because you like to automate everything	17:37
anteaya	and i can see that	17:37
clarkb	greghaynes: I don't think so	17:37
AJaeger	anteaya: ;)	17:37
anteaya	but somethings, like language choices aren't a good fit	17:37
*** jistr has quit IRC		17:37
clarkb	greghaynes: we use the ubuntu base images	17:37
clarkb	greghaynes: no funny business	17:37
anteaya	but language format is	17:37
greghaynes	heh, ok	17:37
*** pblaho has quit IRC		17:38
AJaeger	anteaya, pleia2: The patch has now 2 +2s on it. Since I didn't hear a veto, would you be ok if I approve it now - or do we need to discuss this in the infra-meeting some more?	17:39
openstackgerrit	Mauro Rodrigues proposed openstack-infra/subunit2sql: Convert tags metadata to string before persist it https://review.openstack.org/147205	17:40
*** ivar-lazzaro has joined #openstack-infra		17:40
*** armax has joined #openstack-infra		17:40
pleia2	AJaeger: you can go ahead	17:40
anteaya	AJaeger: like I said if others feel there is value there, they can have it	17:40
anteaya	I see no value in it	17:40
anteaya	I'm not blocking it	17:41
*** rkukura has joined #openstack-infra		17:41
asselin	greghaynes, ok looking here: https://github.com/openstack-infra/project-config/blame/master/nodepool/elements/cache-devstack/extra-data.d/55-cache-devstack-repos#L137	17:41
*** ivar-lazzaro has quit IRC		17:41
asselin	should pick either debs or apts? I see both...?	17:41
clarkb	jeblair: the allocation granting system does not seem to take into account image availability	17:42
AJaeger	anteaya, pleia2: understood - thanks	17:42
*** ivar-lazzaro has joined #openstack-infra		17:42
clarkb	jeblair: probably something to try updating in a subsequent change?	17:42
*** ivar-lazzaro has quit IRC		17:42
AJaeger	anteaya: speaking about checks, I added one that every channel in gerritbot is in accessbot as well: Do you see value in that? https://review.openstack.org/162070	17:43
*** ivar-lazzaro has joined #openstack-infra		17:43
*** david-lyle_afk is now known as david-lyle		17:43
clarkb	actually I think there may be a simple fix	17:43
greghaynes	asselin: I think its correct - I think apts is just a symlink to debs	17:43
AJaeger	clarkb: with 162070 in, all channels are cleaned up ;)	17:44
greghaynes	so worst case it would be double installing everything	17:44
asselin	greghaynes, ok I see	17:44
*** hodos has joined #openstack-infra		17:45
anteaya	AJaeger: wasn't there some discussion you had with jeblair to create one bot file instead of many?	17:45
asselin	greghaynes, sorry going to a meeting	17:45
*** armax has quit IRC		17:45
anteaya	so that getting channel services was one config entry rather than many?	17:45
*** ashleighfarnham has joined #openstack-infra		17:46
anteaya	am I remembering correctly?	17:46
*** AJaeger_ has joined #openstack-infra		17:46
jeblair	anteaya: i'm working on a plan, but don't have it ready yet	17:47
*** Guest94194 has quit IRC		17:47
anteaya	okay thanks, just was wondering if I understood the direction correctly	17:48
*** claudiub has quit IRC		17:48
*** AJaeger has quit IRC		17:48
*** afazekas has quit IRC		17:49
anteaya	AJaeger: and yes in the meantime ensuring that channels that recieve gerrit patch status events also have infra access is a good thing to have	17:49
anteaya	oh he left	17:49
*** stevemar has quit IRC		17:50
*** stevemar has joined #openstack-infra		17:50
AJaeger_	anteaya: I'm back	17:52
*** bnemec is now known as beekneemech		17:52
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Test that dib failures don't stop node builds https://review.openstack.org/162237	17:52
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Don't allocate nodes to providers without images https://review.openstack.org/162255	17:52
clarkb	jeblair: ^ I don't fully understand how that wasn't a 100% failure before, but this makes my test pass after 10 consecutive runs when before the fail rate was about 50%	17:53
anteaya	AJaeger_: oh okay great	17:53
*** purp_away is now known as purp		17:53
clarkb	jeblair: not sure that that is the correct place to handle this case in the allocator either	17:53
AJaeger_	anteaya: having it in one file would be great - together with the eavesdrop, and then still we would need some checking. This is the best I came up with for now	17:53
anteaya	AJaeger_: understood	17:53
anteaya	and yes it has value for me	17:54
anteaya	just keep in mind that even when we disagree I admire your hard work	17:54
anteaya	and appreciate your perspective	17:54
anteaya	I just don't share it sometimes	17:54
anteaya	:)	17:54
*** jcoufal has joined #openstack-infra		17:54
clarkb	jeblair: though this exposes a bunch of fails in the other tests :(	17:55
AJaeger_	anteaya: that's the spirit. Disagreement is fine if we handle it good as we do	17:55
clarkb	because they depend on a positive allocation leading to waiting for a node to be made	17:55
clarkb	hrm	17:55
*** AJaeger_ is now known as AJaeger		17:57
anteaya	AJaeger: can you look at https://review.openstack.org/#/c/161447/5/zuul/layout.yaml for me	17:58
anteaya	perhaps I am confused	17:58
anteaya	I see tarballs and publish to pypi as separate things	17:58
anteaya	does publish to pypi need tarballs?	17:58
anteaya	I didn't think it did	17:58
*** ssam2 has quit IRC		17:59
openstackgerrit	Merged openstack-infra/project-config: Add a lot of missing descriptions https://review.openstack.org/157626	18:02
jeblair	clarkb: i feel like your code change is correct, but that means we need a different strategy for testing :/	18:03
*** lucasagomes is now known as lucas-beer		18:04
clarkb	jeblair: ya I am working on that	18:04
*** superanne has joined #openstack-infra		18:04
jeblair	pushed gertty 1.1.0	18:04
clarkb	jeblair: I think the key is each test should explicitly state what resources it is waiting on early in a test	18:05
jeblair	clarkb: that sounds like a good solution	18:05
clarkb	jeblair: so waitForImage(X) waitForImage(Y) waitForNodes() then asserts	18:05
jeblair	clarkb: ++	18:05
*** lucas-beer has left #openstack-infra		18:06
*** yamahata has quit IRC		18:07
clarkb	jeblair: I think the sleeps were there to approximate this behavior in the past	18:07
jeblair	clarkb: the negative is still tricky there... you are working on a test where you don't expect the image to appear, and the dib image construction is asynchronous...	18:07
clarkb	jeblair: yup I will need to assert failed buidls some other way	18:08
jeblair	clarkb: perhaps it will generally work, but it doesn't seem foolproof	18:08
jeblair	clarkb: maybe record failed builds in something the test can inspect?	18:08
clarkb	jeblair: we do :) the log	18:08
*** yamamoto has quit IRC		18:09
*** SumitNaiksatam has quit IRC		18:09
clarkb	though thats a bit fragile	18:09
jeblair	https://pypi.python.org/pypi/gertty/1.1.0 exists	18:09
*** marcusvrn has quit IRC		18:09
jlvillal	jeblair: I got an exception about the database in gertty after I did a fresh git pull. Is this a known issue?	18:10
superanne	anteaya: did you get a chance to read through the email, does it make sense for the 3rd party audience?	18:11
jeblair	jlvillal: can you paste the error?	18:12
jlvillal	jeblair: http://paste.openstack.org/show/190472/	18:12
*** marcusvrn has joined #openstack-infra		18:13
*** bdpayne has joined #openstack-infra		18:13
jlvillal	jeblair: It did immediately on startup. So I deleted the ~/.gertty.db file But then it died again.	18:13
jlvillal	s/did/died/	18:13
jeblair	jlvillal: did it die with the same error the second time?	18:14
jlvillal	jeblair: The one I pasted is the 2nd time. Let me look at scrollback for first error	18:15
*** edmondk has left #openstack-infra		18:15
jlvillal	jeblair: Seems to be the same error. I'll paste	18:15
jlvillal	jeblair: 1st run error: http://paste.openstack.org/show/190473/	18:15
jeblair	jlvillal: i believe it's an instance of this bug: https://storyboard.openstack.org/#!/story/2000073	18:16
jeblair	mordred: what's the latest thinking on that?	18:16
jeblair	jlvillal: it seems to be somehowe related to pbr and the way gertty is installed from git	18:17
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Test that dib failures don't stop node builds https://review.openstack.org/162237	18:17
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Don't allocate nodes to providers without images https://review.openstack.org/162255	18:17
jeblair	jlvillal: if you create a new venv, and run "pip install -d ." it should work. or if you create a new venv and install from pypi with "pip install gertty"	18:17
mordred	jeblair: still unable to reproduce	18:17
clarkb	jeblair: ^ ok that should work for the first change, now I just need to sort out positively asserting that image builds failed in the new test	18:17
jeblair	jlvillal: do you have any idea how you originally installed it?	18:18
jlvillal	jeblair: I did the virtualenv setup	18:18
jlvillal	jeblair: Installed it	18:18
jlvillal	jeblair: Then I just installed it again after doing a git pull	18:18
jeblair	jlvillal: do you know what command you used to install it?	18:18
*** purp is now known as purp_away		18:18
clarkb	jeblair: mordred the fix for that is git clean -xfd then reinstall	18:18
jlvillal	jeblair: pip install .	18:19
clarkb	the stale egg tells pbr/setuptools it doesn't need to update the alembic files	18:19
*** morganfainberg is now known as needscoffeebadly		18:19
jeblair	clarkb: so theoretically wou should be able to checkout an old version, pip install ., checkout a new version, pip install . and reproduce?	18:19
jeblair	s/wou/we/	18:20
*** dolphm is now known as dolphin		18:20
*** mpaolino has quit IRC		18:20
*** dolphin is now known as Guest82734		18:20
jeblair	clarkb: so theoretically we should be able to checkout an old version, pip install ., start gertty, stop gertty, checkout a new version, pip install ., start gertty and reproduce?	18:20
jeblair	fixed ^ :)	18:20
jlvillal	jeblair: Did a brand new virtualenv. activate. 'pip install .' And crashed :(	18:21
clarkb	jeblair: yes I think so, just be sure you start with a clean slate otherwise your old version may also be affected by a stale egg	18:21
jeblair	jlvillal: what's the error this time?	18:21
jlvillal	jeblair: Same one	18:21
jlvillal	jeblair: sqlalchemy.exc.OperationalError: (OperationalError) no such column: ...	18:22
*** esker has quit IRC		18:22
*** needscoffeebadly is now known as CaptainMorgan		18:22
*** e0ne has joined #openstack-infra		18:22
pc_m	clarkb: pping	18:23
clarkb	pc_m: pong	18:23
*** achanda has joined #openstack-infra		18:23
*** armax has joined #openstack-infra		18:23
pc_m	clarkb: Can you review (and hopefully approve) https://review.openstack.org/#/c/162135/?	18:23
jlvillal	jeblair: It happens when I hit the key 'L'	18:23
pc_m	clarkb: I goofed with the file filtering thing.	18:23
*** e0ne has quit IRC		18:24
clarkb	pc_m: sure done	18:24
pc_m	clarkb: Thank you!	18:24
jeblair	jlvillal: i just did that and it worked for me. are you sure you were in a completely new venv?	18:24
*** tonytan4ever has quit IRC		18:25
jlvillal	jeblair: I'm sure. I'll do it again :)	18:25
jeblair	jlvillal: best also remove the .gertty.db just to be sure	18:25
*** jp_at_hp has quit IRC		18:25
jlvillal	jeblair: I did that. Testing again.	18:26
AJaeger	anteaya: what's the problem with pypi?	18:27
*** superanne has quit IRC		18:27
AJaeger	anteaya: the template generates the tarballs and uploads them	18:27
*** jogo is now known as flashgordon		18:27
jeblair	clarkb: those steps don't reproduce :(	18:28
*** salv-orl_ has quit IRC		18:28
jeblair	2015-03-06 10:32:44,963 Running upgrade 312cd5a9f878 -> 4a802b741d2f, add starred	18:28
anteaya	AJaeger: right so with publish to pypi we don't need any additional information for tarballs do we?	18:28
jeblair	clarkb: https://etherpad.openstack.org/p/LeqN5lpQ4d are the steps	18:29
anteaya	AJaeger: see my comments: https://review.openstack.org/#/c/161447/4/jenkins/jobs/projects.yaml	18:29
*** coolsvap has joined #openstack-infra		18:29
anteaya	superanne: yes, go ahead do send	18:29
AJaeger	ah, dig a bit...	18:30
anteaya	AJaeger: am I wrong?	18:30
AJaeger	anteaya: jenkins/jobs/pypi-jobs.yaml uses tarball-site	18:31
*** yamahata has joined #openstack-infra		18:31
AJaeger	so, the variable is needed	18:31
AJaeger	and the jjb template python-jobs includes the {name}-tarball job-group	18:32
AJaeger	I meant:	18:32
AJaeger	and the jjb python-jobs job-group includes the {name}-tarball job	18:32
jeblair	clarkb: i don't see egg dirs in gertty when i do that; could an older version of pbr have created them?	18:32
AJaeger	And the Zuul publish-to-pypi template uses {name}-tarball job	18:32
openstackgerrit	Michael Krotscheck proposed openstack-infra/storyboard-webclient: Added sane result set indicators. https://review.openstack.org/162265	18:33
AJaeger	So, if you remove the two liines from projects.yaml, our tests should fail because tarball-site is not set and the tarball job does not exist	18:33
anteaya	ah okay	18:34
anteaya	so doug was right the first time and i"m wrong	18:34
anteaya	thanks	18:34
jeblair	clarkb, mordred: i found that if i run python setup.py install with the old checkout, it creates an egg dir and triggers the problem	18:34
mordred	jeblair: AH. LOVELY	18:34
*** otter768 has joined #openstack-infra		18:34
jeblair	mordred: see the steps in https://etherpad.openstack.org/p/LeqN5lpQ4d	18:35
AJaeger	anteaya: dhellmann used just ''{name}-tarball' in his most recent patch - so just the one job...	18:35
anteaya	but in the patch I commented on	18:35
mordred	jeblair: /me grumps	18:35
openstackgerrit	Merged openstack-infra/project-config: Partial revert of func job setup for VPNaaS repo https://review.openstack.org/162135	18:36
anteaya	AJaeger: did doug have it right in patchset 4?	18:36
jeblair	jlvillal: so clark's part of the solution is important. you have to clean out the old egg dir from your git checkout (recreating the venv is not enough)	18:36
*** Guest82734 is now known as dolphinator		18:36
mordred	dstufft: there isn't any way to detect if you're being run by pip vs. being run directy by python setup.py is there?	18:36
jlvillal	jeblair: Oh. let me try that :)	18:36
jeblair	jlvillal: "the fix for that is git clean -xfd then reinstall"	18:36
dstufft	mordred: do you want the real answer or the ideal answer	18:37
mordred	dstufft: the real answer	18:37
mordred	dstufft: I've long since given up on idealism	18:37
jlvillal	jeblair: Woo hoo!!! and woo hoo to clarkb :) It works now. Thanks. Hopefully there can be a way to detect that for upgraders.	18:38
mordred	jeblair: (I'm thinking that if we can detect it - we can at least print a warning like "dude, don't run setup.py directly")	18:39
AJaeger	anteaya: setting tarball-site was right, using python-jobs included the tarball job but many other jobs as well - like many other include.	18:39
*** markvoelker has quit IRC		18:39
jeblair	jlvillal: yeah, i think that's what mordred is working on. but from now on, as long as you don't run "python setup.py install" and just do "pip install ." you shouldn't hit the error again, even on the next upgrade	18:39
AJaeger	anteaya: so it was good - but using only tarballs job is perfect like its now ;)	18:39
*** otter768 has quit IRC		18:40
*** markvoelker has joined #openstack-infra		18:40
dstufft	mordred: so pip calls setup.py in a specific way that's highly unlikely to be replicated by someone whose not pip, also I'm pretty sure that pip sets some environment variables too that you could check for. But none of those are promised things and may change in the future (and may not hold in the past)	18:40
jlvillal	jeblair: I only did 'pip install .' as a note	18:40
anteaya	AJaeger: ah so the latest patchset is better, cool	18:40
anteaya	AJaeger: thanks for looking at it	18:41
mordred	dstufft: could I request that we add a thing that always sets a "RUNNING_IN_PIP" env or something? I'll even write the patch ...	18:41
jeblair	jlvillal: do you have a "gertty.egg-info" in your gertty directory now?	18:41
*** BobBall is now known as BobBall_AWOL		18:41
*** sarob_ has joined #openstack-infra		18:41
anteaya	dhellmann: AJaeger likes your latest aeromancer patchset, and shortly I will too	18:41
jlvillal	jeblair: I don't see one	18:41
*** thingee has joined #openstack-infra		18:41
jeblair	jlvillal: ok good. i wonder if maybe an older version of pbr caused it to be created or something...	18:42
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Test that dib failures don't stop node builds https://review.openstack.org/162237	18:42
jeblair	jlvillal: do you have your old virtualenv -- the one that had the initial problem?	18:42
thingee	jeblair: I noticed the #openstack-cinder channel topic changed. Not sure if I upset someone or if there was some global reset in topics.	18:42
dstufft	mordred: I'd need to think about that a little bit, I understand the desire to check for that, but I worry about the precidence it would set - We're trying to move towards a world where what installer you use doesn't matter	18:42
jeblair	thingee: global reset of topics with outdated info due to technical error. sorry :( feel free to change it back	18:42
jlvillal	jeblair: Sorry. The git -xfd wiped it out.	18:42
pleia2	need to heads down for a bit to work on some slides+practice elasticon talk for next week, bbiab	18:43
thingee	jeblair: Ok just double checking I didn't do something wrong :)	18:43
dstufft	mordred: open a ticket for discussion?	18:43
clarkb	jeblair: I also see http://paste.openstack.org/show/190485/ occasionally when running these tests with --until-failure	18:43
jeblair	jlvillal: np. we're a bit closer to understanding now, and i'm at least pretty sure that you should be all set going forward.	18:43
jeblair	jlvillal: thanks and sorry	18:43
*** wenlock has quit IRC		18:43
jlvillal	jeblair: Thanks for the help :)	18:43
mordred	dstufft: fair concern... and yes, I shall do that	18:43
*** markvoelker has quit IRC		18:44
*** spzala has joined #openstack-infra		18:44
openstackgerrit	Rafael Folco proposed openstack-infra/system-config: Updates to running-your-own CI docs: Changes required https://review.openstack.org/162268	18:45
*** amitgandhinz has quit IRC		18:45
anteaya	can we get third party account 9751 IBM DB2 CI disabled please? http://paste.openstack.org/show/190486/ https://review.openstack.org/#/c/157158/	18:45
jlvillal	jeblair: On another note. How hard would it be to have j/k scroll down/up ? Or are those keys already used? vim user here :)	18:45
anteaya	keystone has tried for 4 days to get them to fix it and they won't	18:45
anteaya	so CaptainMorgan has requested they be disabled	18:45
CaptainMorgan	i also looked, they are not properly posting logs for success. and haven't been for a while.	18:46
CaptainMorgan	404 when you click the link.	18:46
openstackgerrit	Rafael Folco proposed openstack-infra/system-config: Updates to running-your-own CI docs: Changes required https://review.openstack.org/162268	18:47
CaptainMorgan	since i haven't seen failures outside of merge conflict with no logs, i don't know if failures would have logs either.	18:47
jeblair	jlvillal: you can override keys in your local config file. and if you create a "vim-style" map that others might like, we can add that as a built-in keymap too, so people can just say "keymap: vim" in their config files and get the same thing.	18:47
jlvillal	jeblair: Thanks. I will check out the config file	18:48
* mordred keeps wanting "u" to do the same thing as esc - but hasn't bothered doing it ...		18:48
mordred	also, I have no idea why I want that ... my fingers just want to do it all the time	18:48
mordred	was that an elm keystoke I wonder?	18:48
jeblair	jlvillal: i am not (much of) a vim user, so if you come up with the mapping, i'll be happy to help out with the patch	18:48
jlvillal	jeblair: Okay. j/k are the things I naturally go for when I want to scroll up/down. Not sure if any other keys come to mind as useful.	18:49
*** stevemar is now known as stevedore		18:50
jeblair	jlvillal: it currently doesn't have support for multi-key commands, but probably could with a little work (so :q isn't an option yet). i hope to add that at some point if someone doesn't beat me to it.	18:51
jeblair	jlvillal: (i would use it for c-x c-q of course ;)	18:52
*** achanda has quit IRC		18:52
jlvillal	jeblair: :D	18:52
jlvillal	jeblair: I will get back to you and ask how to set it up in the config file. Time for lunch now...	18:52
*** ivar-lazzaro has quit IRC		18:52
*** rkukura has quit IRC		18:53
sabari	I have a glance bp for k3 and it needs https://review.openstack.org/#/c/161878/.	18:56
sabari	It would be great if someone can review.	18:56
*** ddieterly has quit IRC		18:56
anteaya	sabari: these are the folks you need to find: https://review.openstack.org/#/admin/groups/131,members	18:57
*** achanda has joined #openstack-infra		18:57
*** ddieterly has joined #openstack-infra		18:57
*** sdake__ has quit IRC		18:57
sabari	anteaya: Thanks :). Will do.	18:58
openstackgerrit	yolanda.robla proposed openstack-infra/system-config: Expose jenkins_gitfullname and jenkins_gitemail https://review.openstack.org/161695	18:58
*** rmcall has quit IRC		18:58
*** sarob_ has quit IRC		18:58
clarkb	jeblair: "# granted but not requested? shouldn't happen" comment around the raise for the KeyError I pasted above	19:00
*** ivar-lazzaro has joined #openstack-infra		19:00
clarkb	jeblair: any thoughts on how that is happening would be awesome. I am going to see if it happens without my changes now	19:00
*** dangers_away is now known as dangers		19:00
*** amitgandhinz has joined #openstack-infra		19:01
SpamapS	clarkb: funny story, on your question "why did you remove the base class?" ... I actually didn't even look at the base class and figured that was just useless boiler plate. ;)	19:01
* SpamapS will push patch up for that later		19:01
*** markvoelker has joined #openstack-infra		19:03
*** MaxV has quit IRC		19:03
*** superanne has joined #openstack-infra		19:04
clarkb	jeblair: so far it hasn't happened without my change, going to let that run as much as it wants though while I step out for some food	19:05
clarkb	but its likely my change has a bug	19:05
*** rkukura has joined #openstack-infra		19:06
clarkb	CaptainMorgan: anteaya did that account get inactivated?	19:06
clarkb	I am reading scrollback to catch up	19:06
cody-somerville	Have you ever seen a case where all the jobs have ran for a change but they just sit there in the queue?	19:06
anteaya	clarkb: I have not seen that it has	19:07
clarkb	anteaya: ok will do that now	19:07
anteaya	clarkb: thank you	19:07
clarkb	anteaya: CaptainMorgan all done	19:07
*** melwitt has joined #openstack-infra		19:07
clarkb	cody-somerville: which queue?	19:08
cody-somerville	clarkb: gate	19:08
anteaya	clarkb: thank you	19:08
* anteaya composes email to -announce		19:08
openstackgerrit	jose-idar proposed openstack-infra/project-config: Adding pep8 gate check to Opencafe https://review.openstack.org/159642	19:09
*** yamamoto has joined #openstack-infra		19:09
clarkb	cody-somerville: check your zuul event and results queue	19:10
clarkb	cody-somerville: if the results queue value is not 0 that means zuul is processing the results and finished jobs will hang around until zuul gets to them	19:10
cody-somerville	clarkb: Queue lengths: 0 events, 0 results.	19:10
clarkb	cody-somerville: could it be that gerrit is taking a long time to merge the change?	19:11
clarkb	cody-somerville: zuul will sit and wait for either the successful or fail response	19:12
clarkb	jeblair: it almost looks like getNeededNodes is not properly reentrant	19:12
clarkb	jeblair: and both _run and waitForNodes call it	19:13
openstackgerrit	Merged openstack-infra/subunit2sql: Convert tags metadata to string before persist it https://review.openstack.org/147205	19:13
clarkb	and reentrancy is broken by my change somehow, I will need to think this thorugh	19:13
*** yamamoto has quit IRC		19:14
cody-somerville	clarkb: Unlikely. It's been an hour and a half.	19:15
cody-somerville	clarkb: its noop jobs fwiw	19:15
cody-somerville	clarkb: is it because there are other changes in the change queue that aren't finished?	19:15
clarkb	cody-somerville: I don't think the jobs are at fault, it would be something zuul is doing	19:16
clarkb	cody-somerville: if it is a proper gating queue zuul always pops off the head	19:16
clarkb	cody-somerville: no changes can be removed elsewhere	19:16
clarkb	jeblair: I think I see it, this isn't an issue with the old code because it always requests consistently	19:16
*** thingee has quit IRC		19:17
clarkb	jeblair: but now when we check the labels first there is a race between the label having an image ready and multiple getNeededNodes calls	19:17
clarkb	jeblair: except hrm, all of these objects appear to be on the stack	19:18
jeblair	clarkb: yeah, things are going to be more dynamic... things that might have done one node from each of 2 providers before could end up doing 2 nodes from 1 and none from another...	19:19
cody-somerville	clarkb: but doesn't it do the grouping by job name? Why would noop jobs be in a change queue with a change that has no noop jobs?	19:19
clarkb	cody-somerville: yes its based on job but its transitive	19:20
clarkb	so if you have noop job in a project along side another gate job then that can happen	19:20
nikhil_k	hey, we're looking to possibly retag glance_store version(s)	19:20
*** dkranz has quit IRC		19:20
nikhil_k	was hoping to find out the best way to do that in our setup	19:21
cody-somerville	clarkb: they only have noop jobs then there is a cinder change with the standard cinder jobs	19:21
jeblair	nikhil_k: what do you mean by retag?	19:21
nikhil_k	jeblair: we've this issue where the release consisted of a feature addition but we released it with a bump on X in 0.0.X	19:21
clarkb	cody-somerville: and no other project has both noop and cinder jobs?	19:21
jeblair	doesn't have to just be cinder	19:22
nikhil_k	jeblair: so, we are wondering if it is possible to retag 0.1.12 to 0.2.0	19:22
nikhil_k	or have parallel tags, maybe?	19:22
jeblair	clarkb, cody-somerville: it could be, but it could also be noop + foo and foo + bar and bar + devstack-gate	19:22
cody-somerville	and weird, clarkb jeblair: the noop jobs have finally disappeared but the change at the top of the queue is still running.	19:22
jeblair	cody-somerville: the status page tells you whether changes are in the same change queue or not	19:23
cody-somerville	jeblair: they were in the same change queue, yes.	19:23
jeblair	cody-somerville: if they are, then something is very, very broken. if they aren't, it was probably just waiting for gerrit.	19:23
cody-somerville	the list of projects for the change queue seemed to list almost all the projects we have	19:23
*** superanne has quit IRC		19:24
jeblair	cody-somerville: then it seems likely that you have at least one project with noop and another job.	19:24
*** ivar-lazzaro has quit IRC		19:24
jeblair	cody-somerville: but if they were in the same queue, it should not have been removed if it was not at the top.	19:24
jeblair	nikhil_k: that sounds terribly confusing :(	19:24
nikhil_k	umm	19:25
*** coolsvap has quit IRC		19:25
nikhil_k	jeblair: one sec	19:25
clarkb	ok I really need to step out for a bit, jeblair your thoughts on the nodepool change and possible bug much appreciated if you have time	19:25
jeblair	clarkb: yeah, i've been thinking about it, but i don't have a good answer yet	19:26
nikhil_k	jeblair: http://eavesdrop.openstack.org/irclogs/%23openstack-glance/%23openstack-glance.2015-03-06.log you would need to search for "2015-03-06T15:05:50"	19:26
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Allow JJB to set connection timeout to jenkins server https://review.openstack.org/151472	19:27
nikhil_k	jeblair: the basic issue is that a condition is asking for the release number to be 0.2.0 but we have a release of 0.1.11	19:27
nikhil_k	jeblair: we were hoping to see if that can be corrected and the release tag for 0.1.12 can be equivalent to 0.2.0 by any chance?	19:28
jeblair	nikhil_k: i think that's only going to make things worse. i think sigmavirus24 is correct -- it's done, and we just need to deal with it. releasing 0.2.0 will not make 0.1.12 go away, and 0.1.12 existing and adding new features is the part that causes problems.	19:29
*** _shaps_ has joined #openstack-infra		19:29
jeblair	nikhil_k: so i think the best choice is just to deal with it (we may need to adjust stable branch version caps)	19:29
*** thingee has joined #openstack-infra		19:29
cody-somerville	jesusaurus: ^^ can you check on that noop thing?	19:30
openstackgerrit	Ian Cordasco proposed openstack-infra/system-config: Add logging to #openstack-ansible https://review.openstack.org/161412	19:30
jeblair	nikhil_k: if, for some reason, we really, really, really want to cause more pain -- you could release 0.1.13 which is equivalent to 0.1.11, and then release 0.2.0 as equivalent to 0.1.12.	19:31
jeblair	nikhil_k: i would recommend against that though, unless there's a really good reason	19:31
nikhil_k	jeblair: heh, that seems worse	19:31
nikhil_k	jeblair: yeah, ok. I will try to convince doug otherwise	19:31
jeblair	dhellmann: oh, you want a release of 0.2.0 that is == to 0.1.12?	19:32
nikhil_k	jeblair: he wanted for us to do it as we'r trying to bump up the cap https://review.openstack.org/#/c/161079/	19:32
*** ayoung is now known as SugarAddy		19:32
*** SugarAddy is now known as ayoung		19:32
*** SumitNaiksatam has joined #openstack-infra		19:32
*** ja is now known as zz_ja		19:33
jeblair	dhellmann: my thought is, 0.1.12 is already out there, and breaks any caps that people may have had. releasing 0.2.0 doesn't fix that, but only makes things confusing for users (why are these versions the same?)	19:33
jeblair	dhellmann: releasing 0.1.13 as a revert of 0.1.12 (and then releasing 0.2.0) would "fix" people with version caps, but is complex.	19:34
jeblair	dhellmann: (and still slightly confusing to users, but only if they wonder what 0.1.12 was all about)	19:34
jeblair	dhellmann: whereas the state we are in now is a more comprehensible "oops -- semver fail" situation :)	19:34
jeblair	nikhil_k: ^	19:34
jeblair	so anyway, my recommendation would be (in order of preference): a) do nothing (except fix version caps where necessary); or b) release 0.1.13 and 0.2.0 as described above	19:36
nikhil_k	jeblair: I personally prefer to let things be as is and promise dhellmann to release 0.2.0 instead of 0.1.13	19:36
nikhil_k	jeblair: however, we really need that review merged https://review.openstack.org/#/c/161079/	19:37
jeblair	nikhil_k: yeah, i just don't see what releasing 0.2.0 without 0.1.13 solves	19:37
jeblair	it only creates confusion and solves no technical problems	19:37
nikhil_k	jeblair: :)	19:38
nikhil_k	jeblair: I was trying to resolve conflicts. What can we do to get that merged?	19:38
*** tnovacik has joined #openstack-infra		19:39
jeblair	nikhil_k: i'm not going to review that change, so you should probably wait to see if i have convinced dhellman to change his mind and then do what he suggests.	19:40
*** ivar-lazzaro has joined #openstack-infra		19:41
*** ghostpl_ has joined #openstack-infra		19:42
Ng	hmm, I am apparently being 17 kinds of stupid today	19:43
nikhil_k	jeblair: heh, ok.. appreciate it, nevertheless	19:43
nikhil_k	zhiyan: ^^	19:43
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Make manual checklang target voting https://review.openstack.org/162033	19:44
Ng	mordred: other than using rackspace-glanceclient, did you have to do some magic to be able to talk to rax glance? I seem to be talking fine to their nova, but glance is insisting on 401s	19:45
*** thingee has quit IRC		19:46
jeblair	Ng: at some point, their url wasn't what was actually in the catalog... i wonder if that got fixed?	19:46
*** ddieterly has quit IRC		19:48
*** dkranz has joined #openstack-infra		19:49
*** ddieterl_ has joined #openstack-infra		19:49
Ng	jeblair: huh	19:49
*** mrunge has quit IRC		19:50
Ng	well this is just bizarre, keystone 401s too, but nova is very clearly working	19:50
*** CaptainMorgan is now known as morganfainberg		19:50
*** superanne has joined #openstack-infra		19:50
morganfainberg	Ng, how is keystone 401ing	19:50
morganfainberg	or..	19:50
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Normalize Gerrit projects.yaml https://review.openstack.org/161954	19:50
morganfainberg	nothing i need to worry about ;)	19:50
mordred	Ng: yes. please dont use rackspace-glanceclient	19:50
mordred	Ng: use python-glanceclient	19:51
mordred	Ng: you need to set an env var	19:51
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Wrap projects.yaml at 80 characters https://review.openstack.org/161955	19:51
mordred	Ng: export OS_IMAGE_API_VERSION=2	19:51
Ng	mordred: OS_IMAGE_API_VERSION?	19:51
Ng	k	19:51
mordred	yup	19:51
mordred	Ng: it's ... bannanas	19:51
*** superanne has quit IRC		19:51
*** carl_baldwin has quit IRC		19:52
morganfainberg	oh	19:52
*** salv-orlando has joined #openstack-infra		19:52
morganfainberg	glanceclient	19:52
* morganfainberg goes back to identity-land		19:52
*** superanne has joined #openstack-infra		19:52
Ng	hmm, still 401	19:53
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Remove upstreams from projects.yaml https://review.openstack.org/161956	19:53
jesusaurus	clarkb: jeblair: oh the test name check for dependent pipelines is transitive?	19:54
openstackgerrit	Merged openstack-infra/system-config: Add Wildfly Puppet module https://review.openstack.org/143512	19:54
jeblair	jesusaurus: yes -- if project A affects B and project B affects C, then it follows that A affects C	19:55
jesusaurus	clarkb: jeblair: is there an easy way to get the list of projects grouped into a dependent pipeline? hovering over the list on the zuul status page gives a box too large to be useful	19:56
*** superanne has quit IRC		19:56
*** superanne has joined #openstack-infra		19:57
jeblair	jesusaurus: it should log all of the decision making about it at startup. or you can grab the status.json directly.	19:57
mrmartin	AJaeger: are you here?	19:58
AJaeger	mrmartin: yes, I am	19:58
*** _et has joined #openstack-infra		19:58
clarkb	jeblair: I think I figured it out, there is a pool.allocation_history that is used	19:59
clarkb	jeblair: so we need to lock around access to that	19:59
jeblair	clarkb: just for testing?	20:00
mtreinish	jeblair: qq about depends on. Would it work if I had a tempest patch and linked it as depending on a stable/icehouse and stable/juno devstack patch?	20:01
clarkb	jeblair: yes, the only place getNeededNodes() is called in prod is the run loop	20:01
sigmavirus24	mtreinish: is there any way I can help with https://review.openstack.org/#/c/152224/ ?	20:02
clarkb	it might also be possible toget the info we need with calling the side effecting getNeededNodes	20:02
jeblair	mtreinish: tell me if this answers your question: http://docs.openstack.org/infra/manual/developers.html#multiple-changes	20:02
mtreinish	sigmavirus24: hmm, yeah if you want to pick that up and run with it that would be great	20:03
mtreinish	sigmavirus24: my next step was going to be tracking down which bugs are still there (which is causing the failures) and reopen them	20:03
sigmavirus24	mtreinish: heh okay	20:03
*** che-arne has quit IRC		20:03
*** bcourt has joined #openstack-infra		20:03
sigmavirus24	Elastic recheck caught a bunch of bugs it apparently recognized but let me rebase to see what's going on	20:04
mtreinish	jeblair: I think it does, so it should work as I would expect. Thanks	20:05
jeblair	mtreinish: woo!	20:05
openstackgerrit	Merged openstack-infra/system-config: Add Zuul Gerrit Trigger Way & JJB https://review.openstack.org/149032	20:05
*** mriedem has quit IRC		20:05
*** Longgeek has quit IRC		20:06
mtreinish	jeblair: the pending change is to set the tempest.conf on stable to disable an option which is being added (default enabled) to the config file	20:06
mtreinish	sigmavirus24: sure, I just assumed the bugs were still open, but there's also a good chance the tempest tests have bit rotted while skipped too	20:07
jeblair	mtreinish: perfect	20:07
sigmavirus24	mtreinish: yeah, also this isn't going to rebase cleanly (oh joy ;))	20:07
mtreinish	sigmavirus24: it might just make sense to abandon the single patch and split it into a bunch of separate patches for each skip being removed	20:08
clarkb	jeblair: commented on my change with the info	20:08
clarkb	jeblair: we might be able to get away with using two different history objects	20:08
clarkb	jeblair: though really we just need to record how many nodes we need to launch somewhere	20:09
sigmavirus24	mtreinish: makes sense	20:09
jeblair	clarkb: 162255?	20:09
clarkb	jeblair: yes	20:09
sigmavirus24	mtreinish: I'll start working on splitting it up. Thanks for the pointers. =)	20:10
jeblair	clarkb: how is that getting called twice?	20:10
clarkb	jeblair: _run() calls it and waitForNodes() in the tests call it	20:11
jeblair	clarkb: thx	20:11
*** salv-orlando has quit IRC		20:11
mtreinish	sigmavirus24: sure, np. Thanks for taking it over. I haven't had a chance to pick it back up	20:11
asselin	greghaynes, another possibility of why -infra isn't running into the issue is that they already have the packages cached?	20:11
*** ctlaugh_ has joined #openstack-infra		20:11
*** esker has joined #openstack-infra		20:11
sigmavirus24	mtreinish: no worries. trying to turn on testing for a heat thing and noticed that review.	20:11
mtreinish	sigmavirus24: heh, have you seen: https://review.openstack.org/#/c/158536/	20:13
Ng	greghaynes: on a scale from zero to legal injunction, how offended would you be if I pushed some tweaks to your changeset for adding VHD support to DIB?	20:13
*** mriedem has joined #openstack-infra		20:13
sigmavirus24	No but it's relevant to my interests heh	20:13
*** rkukura has quit IRC		20:14
*** bswartz has joined #openstack-infra		20:14
*** bcourt has quit IRC		20:14
anteaya	superanne: i see the post to -dev but not -announce yet	20:16
superanne	anteaya: ah yes, meant to ask, will teh same topics in the subject be useful on that list?	20:16
anteaya	no, just the subject	20:17
superanne	ok	20:17
superanne	thanks anteaya	20:17
anteaya	thank you!	20:17
jeblair	clarkb: if we made allocation_history a param to getNeededNodes, would it work?	20:18
*** _shaps_ has quit IRC		20:18
openstackgerrit	Merged openstack-infra/project-config: Normalize Gerrit projects.yaml https://review.openstack.org/161954	20:18
jeblair	clarkb: nodepool could use its own allocation history, and the tests could use theirs? the tests don't actually care about the history, they mostly just want to know whether demand is satisfied.	20:18
clarkb	jeblair ya I think that would work	20:18
jeblair	clarkb: or....	20:19
jlvillal	jeblair: In keymaps. If I want to do: cursor-down: ['j'] how would I also specify the arrow-up key? Or do I need to?	20:19
jeblair	clarkb: actually i don't like my other idea, nevermind the "or...". :)	20:19
openstackgerrit	Merged openstack-infra/project-config: Wrap projects.yaml at 80 characters https://review.openstack.org/161955	20:20
jeblair	clarkb: i did have one more idea which was to change waitfornodes from a general wait to a specific one (eg, wait for X nodes of this typo and Y nodes of that type -- like waitforimage)	20:20
*** aysyd has quit IRC		20:20
clarkb	ok I can update to do the first thing	20:21
jlvillal	jeblair: I didn't seem to need to have to specify arrow-up. So I think I am good :)	20:22
cinerama	hey umm pleia2	20:22
jeblair	jlvillal: i don't think you need to	20:22
*** rkukura has joined #openstack-infra		20:22
*** aysyd has joined #openstack-infra		20:22
jeblair	jlvillal: yeah :) but i think if you want to, you can say ['j', 'down'].	20:23
jlvillal	jeblair: Okay. Thanks!	20:23
jeblair	jlvillal: but you should only need to do that if you want to change, say, down to up. :)	20:23
jlvillal	jeblair: Is it just me or are the times being displayed in UTC?	20:23
openstackgerrit	nikhil komawar proposed openstack/requirements: Bump glance_store to 0.2.0 https://review.openstack.org/161079	20:23
jeblair	jlvillal: they are in utc	20:23
jlvillal	jeblair: :(	20:23
jlvillal	jeblair: Is there a way to be in local time?	20:23
jeblair	jlvillal: see 127260	20:23
nikhil_k	jeblair: new release ^	20:23
jeblair	nikhil_k: did dhellmann continue to disagree?	20:24
jlvillal	jeblair: So I assume all the clocks in your house are set to UTC? :P	20:24
nikhil_k	jeblair: I've passed along the info on ML too.	20:24
*** mrmartin has quit IRC		20:25
nikhil_k	jeblair: we just figured this was the fastest way to resolve conflicts ...	20:25
nikhil_k	and hopefully it works with everyone!	20:25
jeblair	nikhil_k: not sure i follow -- dhellmann still isn't around to weigh in on this	20:25
* dhellmann hears his name		20:25
jeblair	jlvillal: the one i use for work is, and all of fungi's clocks are :) but anyway, i think it's a good idea, i just think it should be an option	20:26
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Remove upstreams from projects.yaml https://review.openstack.org/161956	20:26
nikhil_k	jeblair: ah, we just got a +2! \o\ \o/ /o/	20:26
jeblair	jlvillal: "20:26 < jeblair> jlvillal: the one i use for work is," :)	20:26
nikhil_k	please do the honors if possible :)	20:26
dhellmann	jeblair, nikhil_k : sorry, I was at lunch for most of that earlier discussion.	20:26
jlvillal	jeblair: Okay and thanks. Not sure I want to tackle that patch.... Hopefully the author will update with your suggestions.	20:27
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Propose projects.yaml normalization daily https://review.openstack.org/161957	20:27
nikhil_k	dhellmann: no issues..	20:27
nikhil_k	glad we got it resolved!	20:27
dhellmann	jeblair, nikhil_k: Did the new release end up including things that were not in the earlier patch release?	20:27
openstackgerrit	Matthew Treinish proposed openstack-infra/puppet-subunit2sql: Add option to disable uneccesary data migration https://review.openstack.org/162289	20:27
mtreinish	clarkb, jeblair: ^^^ if you guys get a sec that branch is needed before I push the next release	20:28
* mtreinish hopes to keep the proud tradition of Fri. releases		20:29
*** tonytan4ever has joined #openstack-infra		20:29
AJaeger	jeblair: I loved your patch series and rebased and +2ed - thanks!	20:30
openstackgerrit	Antoine Musso proposed openstack-infra/zuul: Fix testtools requiring unittest2 https://review.openstack.org/162291	20:30
jeblair	dhellmann: it looks like it has a few more commits in it, so at least it's not exactly the same	20:30
dhellmann	jeblair: ok. we've re-tagged libs in the past without anyone getting confused, so I really thought that was a simple solution.	20:31
*** radez is now known as radez_g0n3		20:32
nikhil_k	dhellmann: yes	20:32
nikhil_k	ah, see that is figured out earlier..	20:33
nikhil_k	no issues with newer commits, they needed to go in soon anyways :)	20:33
*** otter768 has joined #openstack-infra		20:35
*** dmsimard has quit IRC		20:36
*** achanda has quit IRC		20:37
*** SumitNaiksatam has quit IRC		20:37
marcusvrn	asselin: ping	20:38
*** ddieterl_ has quit IRC		20:39
*** ddieterly has joined #openstack-infra		20:39
*** ivar-lazzaro has quit IRC		20:39
*** otter768 has quit IRC		20:40
asselin	hi	20:41
*** rkukura has quit IRC		20:41
asselin	marcusvrn, pong	20:41
*** crc32 has joined #openstack-infra		20:43
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Log info about updated and deleted jobs https://review.openstack.org/162293	20:46
marcusvrn	asselin: hi! I was looking at my CI...and I saw that zuul is checking out the patches into its git repo (configured in zuul.conf)...but how can slaves use this repo to run the tempest against the code with that patch? Because I think my CI is running tempest against the devstack master, with no patch applied...	20:46
*** sc68cal_ is now known as sc68cal		20:46
lxsli	anteaya: clarkb: thanks	20:47
asselin	marcusvrn, which os is zull running on?	20:47
marcusvrn	asselin: ubuntu	20:47
asselin	marcusvrn, which version?	20:47
marcusvrn	asselin: ubuntu 12.04.5	20:48
asselin	marcusvrn, ok ...and are you using devstack-gate scripts?	20:49
marcusvrn	asselin: yes	20:49
asselin	marcusvrn, do you have a link to one of the jobs in question?	20:50
*** ivar-lazzaro has joined #openstack-infra		20:51
*** ivar-lazzaro has quit IRC		20:51
*** ivar-laz_ has joined #openstack-infra		20:52
*** ivar-laz_ has quit IRC		20:52
marcusvrn	asselin: http://177.84.241.119:10000/28/159528/2/check/hds-hnas-iscsi/e332854/	20:53
*** tnovacik has quit IRC		20:54
asselin	marcusvrn, http://177.84.241.119:10000/28/159528/2/check/hds-hnas-iscsi/e332854/logs/devstack-gate-setup-workspace-new.txt.gz#_2015-02-28_08_28_03_195	20:54
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Test that dib failures don't stop node builds https://review.openstack.org/162237	20:55
openstackgerrit	Clark Boylan proposed openstack-infra/nodepool: Don't allocate nodes to providers without images https://review.openstack.org/162255	20:55
clarkb	jeblair: ^ local testing seems to indicate that is working as expected	20:55
clarkb	an I forgot to set the topic again :/	20:55
clarkb	but consider that ready for review now	20:55
erlon	asselin: see the devstack.log for https://review.openstack.org/#/c/161365/	20:57
marcusvrn	asselin: yep, this remote repo is the one that we have configured, but it's not the one that zuul is checking out and merging patches.	20:58
erlon	asselin: if you seek for 'git show --oneline', it says: "e34e6ae Add oslo.policy to global requirements", which is note the change in test	20:58
asselin	marcusvrn, not sue this is right: http://177.84.241.119:10000/28/159528/2/check/hds-hnas-iscsi/e332854/console.html	20:58
asselin	ZUUL_BRANCH=master	20:58
asselin	marcusvrn, I'm looking at ours for that patch and we don't set it: http://15.126.198.151/65/161365/3/check/3par-iscsi-driver-master-client-pip-eos10-dsvm/7c788c8/console.html	20:59
*** dprince has quit IRC		20:59
marcusvrn	asselin: do you have a local openstack git repo for your CI?	21:00
*** achanda has joined #openstack-infra		21:00
asselin	marcusvrn, we're using a mirror	21:01
*** dimsum__ has quit IRC		21:01
asselin	marcusvrn, you should see something like this which pulls it from zuul: http://15.126.198.151/65/161365/3/check/3par-iscsi-driver-master-client-pip-eos10-dsvm/7c788c8/logs/devstack-gate-setup-workspace-new.txt.gz#_2015-03-06_13_11_48_770	21:01
marcusvrn	asselin: how did you configure the repo to make zuul and slave use the same patch?	21:03
asselin	marcusvrn, does your job look something like this? https://github.com/rasselin/os-ext-testing-data/blob/master/etc/jenkins_jobs/config/dsvm-cinder-driver.yaml.sample#L67	21:03
asselin	marcusvrn, zuul sets up the change, passes the ref to the jenkins job which gets consumed by devstack-gate script	21:04
asselin	marcusvrn, in jenkins, you can see all the environment variables passed in for completed jobs	21:04
marcusvrn	asselin: yes, my job is something like that you sent me	21:05
clarkb	jeblair: any word on the leaked ports? wondering if maybe we should just go ahead and dlete them	21:06
asselin	marcusvrn, then you need to look at the env variales in jenkins. There should be a ZUUL_BRANCH set there	21:06
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Log info about updated and deleted jobs https://review.openstack.org/162293	21:08
asselin	marcusvrn, e.g. this is what I see in mine. please paste yours: http://paste.openstack.org/show/190514/	21:08
*** bcourt has joined #openstack-infra		21:08
*** arosen has left #openstack-infra		21:09
marcusvrn	asselin: http://paste.openstack.org/show/190515/	21:10
AJaeger	clarkb: could you review https://review.openstack.org/162093 today, please? We want to translate RST files in openstack-manuals...	21:10
clarkb	AJaeger: uh oh bash arrays, this will be fun :)	21:11
AJaeger	clarkb: we use them already in openstack-manuals	21:11
AJaeger	clarkb: but I had to write a test script to check that everything works ;)	21:11
clarkb	ya its fine I just have to lookup what the symbols mean each time	21:11
*** yamamoto has joined #openstack-infra		21:12
asselin	marcusvrn, strange. these lines seem to indicate nothing is set...? http://paste.openstack.org/show/190516/	21:12
asselin	marcusvrn, are the variables getting unset anywehre?	21:13
clarkb	AJaeger: doc-tools-check-languages.conf does that exist in the repo(s) yet?	21:13
marcusvrn	asselin: I think I know what's going on...I have 2 jobs...one that handles the slave node VM and another one (triggered after the VM is ready) that runs the devstack-gate scripts..... the ZUUL env variables are on the master node, but it's not being passed to the slave node	21:13
clarkb	AJaeger: its set -e so will error if not	21:13
AJaeger	clarkb: it does	21:14
asselin	marcusvrn, so the job doesn't get sent directly to the jenkins slave? I guess you have a different setup	21:15
AJaeger	clarkb: http://git.openstack.org/cgit/openstack/openstack-manuals/tree/doc-tools-check-languages.conf	21:15
marcusvrn	asselin: I think I have to pass those variables to the slave which will run the devstack-gate scripts	21:15
*** ldnunes has quit IRC		21:15
*** kgiusti has left #openstack-infra		21:15
clarkb	AJaeger: thank you	21:15
AJaeger	the other repos don't use SPECIAL_BOOKS yet but have the file	21:15
*** yamamoto has quit IRC		21:16
marcusvrn	asselin: yes, the job get sent to the slave that will handle the VM and then, this slave will trigger another slave to runs devstack-gate on the VM	21:17
clarkb	AJaeger: lgtm	21:17
marcusvrn	asselin: so, I think all that I have to do is to pass the env variables to that slave which will run the devstack-gate..	21:18
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Change log level for job deletion and add logging for job creation https://review.openstack.org/162302	21:18
asselin	marcusvrn, yes	21:18
AJaeger	clarkb: thansk for the review	21:18
*** rkukura has joined #openstack-infra		21:18
*** thingee has joined #openstack-infra		21:18
marcusvrn	asselin: do you know which env variables the devstack-gate uses? Does it use only ZUUL variables?	21:19
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Allow JJB to set connection timeout to jenkins server https://review.openstack.org/151472	21:20
asselin	marcusvrn, yes, definitely the ZUUL*. Not sure what else.	21:21
marcusvrn	asselin: and if I pass the variables to the slave, I will not need this (http://paste.openstack.org/show/190520/) anymore, will I?	21:22
marcusvrn	asselin: perfect, I'm going to fix it now!	21:22
asselin	marcusvrn, actually, I removed those from my jobs already....	21:22
marcusvrn	asselin: nice	21:23
openstackgerrit	Andreas Jaeger proposed openstack-infra/project-config: Support translation of RST files https://review.openstack.org/162093	21:24
AJaeger	clarkb: yes, we can use the same way - just different directories ^	21:24
*** rkukura has quit IRC		21:24
marcusvrn	asselin: Thanks!! You helped me a lot!	21:25
asselin	marcusvrn, you're welcome	21:25
*** ihrachyshka has quit IRC		21:26
*** teran has joined #openstack-infra		21:27
*** julim has quit IRC		21:27
*** AJaeger has quit IRC		21:28
*** jp_at_hp has joined #openstack-infra		21:28
*** persia_ is now known as persia		21:32
*** pc_m has quit IRC		21:32
*** jp_at_hp has quit IRC		21:33
*** harlowja_away has quit IRC		21:34
*** salv-orlando has joined #openstack-infra		21:36
bhunter71	If one's gerrit account is Administrator group, and Administrator is granted 'Start Replication' - doesn't this mean I should be able to "gerrit replicate" over SSH? [could have sworn this worked in the past -any ideas?]	21:37
*** dboik_ has quit IRC		21:37
bhunter71	"gerrit: replicate: not found"	21:37
*** dustins has quit IRC		21:38
*** rfolco has quit IRC		21:38
clarkb	they changed the command its `replication start` now	21:38
*** jcoufal has quit IRC		21:39
*** hodos\|2 has joined #openstack-infra		21:39
bhunter71	hrm. gerrit: replication: not found	21:39
openstackgerrit	Khai Do proposed openstack-infra/jenkins-job-builder: Change log level for job deletion and add logging for job creation https://review.openstack.org/162302	21:40
clarkb	it is a plugin maybe it isnt installed?	21:40
bhunter71	further, if I just say 'gerrit', I get a list of commands with nothing resembling replication	21:40
zaro	bhunter71: yeah, depend on which version of Gerrit. but it's a core plugin in gerrit 2.9	21:41
*** hodos has quit IRC		21:41
*** dimsum__ has joined #openstack-infra		21:41
zaro	clarkb: was wondering if you could help with this? https://review.openstack.org/#/c/158387/	21:42
bhunter71	interesting. I still have 2.8. maybe I downloaded a plugin in the past.	21:42
*** koolhead17 has quit IRC		21:42
bhunter71	thanks zaro/clarkb	21:42
clarkb	zaro ya sorry keep getting distracted by fires	21:42
*** superanne has quit IRC		21:44
*** _shaps_ has joined #openstack-infra		21:45
clarkb	zaro wow that picture of the drawing I made	21:45
marcusvrn	asselin: I tested and now it's working! thanks again! :)	21:46
asselin	marcusvrn, glad to hear! :)	21:46
*** signed8bit has quit IRC		21:46
*** marun has quit IRC		21:47
*** bradm has quit IRC		21:47
zaro	clarkb: you must be looking at pleia2 slides.	21:47
*** amotoki has quit IRC		21:48
*** signed8bit has joined #openstack-infra		21:48
clarkb	zaro https://review.openstack.org/#/c/158387/5/images/ci_team_infrastructure_draft.png	21:48
clarkb	I approved	21:48
mtreinish	heh, I remember that	21:48
*** baoli_ has quit IRC		21:49
zaro	clarkb: thanks, that's one of the openstack works of arts. maybe should enter it into next tshirt contest :)	21:51
clarkb	ha	21:51
*** mattymo has quit IRC		21:51
*** mattymo has joined #openstack-infra		21:52
*** superanne has joined #openstack-infra		21:52
flashgordon	still looking for reviews for https://review.openstack.org/#/c/160548/ and https://review.openstack.org/#/c/160933/	21:52
flashgordon	expected to shave 5 minutes of all devstack jobs	21:52
jesusaurus	do zuul pipelines get starved by other pipelines with a higher precedence? will nodes be allocated to a normal queue over a low queue without regard to the length of time a job is waiting in a low queue?	21:53
flashgordon	and over 1k dsvm jobs in in zuulat this very moment that is a lot of CPU hours	21:53
flashgordon	clarkb: ^	21:54
flashgordon	mordred: ^	21:54
*** nelsnelson has quit IRC		21:55
mordred	flashgordon: looking	21:55
zaro	jesusaurus: not sure, but that seems reasonable. did you expect something else?	21:55
clarkb	flashgordon do all three devstack branches have that script?	21:56
mordred	flashgordon: I have a question you may not like	21:56
jesusaurus	zaro: it does seem reasonable, but I was wondering if it might be a bit smarter for cases when zuul is under prolonged resource contention	21:56
*** prad has joined #openstack-infra		21:56
flashgordon	clarkb: I don't think so but not 100% sure	21:56
mordred	flashgordon: how do you have the c library dependencies installed such that you can build the c extensions	21:57
clarkb	mordred hahaha good point	21:57
mordred	flashgordon: because normally those are installed by devstack	21:57
flashgordon	mordred: oh yeah ......	21:57
flashgordon	fuuuuuuuuu	21:57
* mordred apologizes		21:57
flashgordon	ck	21:57
flashgordon	mtreinish: ^	21:57
clarkb	mordred please leave review comment :)	21:57
mordred	clarkb: I will	21:57
mordred	I FULLY support the goal	21:57
mordred	btw	21:57
mtreinish	mordred: because nodepool installs the apt list before we build the wheel	21:58
mordred	no it doesn't	21:58
flashgordon	it installs some things for sure not sure if its enough	21:58
mordred	it shoudl install none of the things	21:58
mordred	and if it does, it'sa bug	21:58
mtreinish	then I have no idea, let me check when devstack calls the wheel build script	21:59
mtreinish	because I though it did that before we run apt for anything	21:59
jlvillal	jeblair: gertty --print-keymap prints the keymaps that can be set. But it doesn't print what they are set to. Is there a way to get a list of the keymap settings?	21:59
flashgordon	mtreinish: I am running it inside an empty env now to test	21:59
flashgordon	mtreinish mordred: http://paste.openstack.org/show/190527/	22:00
zaro	jesusaurus: so under prolonged resource contention you want low pri jobs to run before stuff in normal queue?	22:00
mordred	flashgordon: what caused that?	22:00
flashgordon	./tools/build_wheels.sh	22:00
*** marcusvrn has quit IRC		22:00
jesusaurus	zaro: under heavy resource contention i want low pri jobs to run more than 0% of the time, but less often than normal pri jobs	22:01
flashgordon	using a base docker ubunut:14:04 image with just a few things isntalled	22:01
mtreinish	d'oh it's kinda obvious it's called right in the script	22:01
mtreinish	http://git.openstack.org/cgit/openstack-dev/devstack/tree/tools/build_wheels.sh#n50	22:01
flashgordon	mtreinish: doh	22:01
mordred	so - we coudl run this in a docker container, copy out the results and delete the docker container	22:02
openstackgerrit	Ramy Asselin proposed openstack-infra/nodepool: Fix Nodepool DIB Uploads to Glance w/o region_name https://review.openstack.org/162313	22:02
mordred	just as a way to get the wheelhouse pre-done but not pollute the machine	22:02
clarkb	or use our mirrors again	22:02
mordred	clarkb: or that	22:02
clarkb	I am now convinced this is what we should do	22:02
flashgordon	I still think the long term answer is: stack.sh ; unstack.sh	22:03
mordred	I'm fairly convinced of that - it would also lessen the frequency of rebuilds of things like lxml even more	22:03
flashgordon	and have a new job for devstack that runs on a clean env that didn't have that	22:03
flashgordon	but that doesn't work with chroot ...	22:04
*** tiswanso has quit IRC		22:05
openstackgerrit	Gene Snider proposed openstack-infra/zuul: Migrate zuul timer to apscheduler version 3.0 https://review.openstack.org/162315	22:06
*** _shaps_ has quit IRC		22:06
flashgordon	but install_prereqs.sh ; build_wheels.sh would get us pretty far	22:06
*** jgrimm is now known as zz_jgrimm		22:06
mordred	flashgordon: yah - install_prereqs.sh ; build_wheels.sh in a chroot or a docker would be a very clean way to do it	22:07
*** ddieterly has quit IRC		22:07
*** ddieterly has joined #openstack-infra		22:07
flashgordon	right now there are some issues with dib though	22:07
flashgordon	greghaynes: spotted it	22:07
mordred	oh - right.	22:08
mordred	sorry - I forgot the dib part of this	22:08
flashgordon	it cleans out the apt-cache and links to the host by default	22:08
flashgordon	but turns out the apt-get install bits aren't that slow	22:08
flashgordon	maybe save a minute all together tops	22:08
mordred	and we can't build-host cache it because it's wheels	22:08
*** radez_g0n3 is now known as radez		22:08
*** mattfarina has quit IRC		22:09
*** timcline_ has quit IRC		22:09
jeblair	clarkb, mordred: so...	22:09
jeblair	clarkb, mordred: the case i opened with hp public cloud says "Awaiting Customer Input"	22:09
jeblair	clarkb, mordred: but i can't actually see anywhere they have said or asked anything	22:09
mordred	jeblair: that's awesome	22:09
*** baoli has joined #openstack-infra		22:10
jeblair	the description field seems the same as i left it, and there is nothing in the comments section	22:10
*** woodster_ has quit IRC		22:10
mordred	jeblair: have you tried the noc channel? or would you like me to?	22:10
*** hdd_ has quit IRC		22:10
jeblair	mordred: i'll go there now	22:10
openstackgerrit	James Polley proposed stackforge/gertty: Expand sample keymaps to ameliorate OSX features https://review.openstack.org/157488	22:13
zaro	jesusaurus: from looking available zuul params it's unlikely that it does that.	22:13
*** pblaho has joined #openstack-infra		22:13
*** pblaho has quit IRC		22:13
*** bradm has joined #openstack-infra		22:14
*** dimsum__ has quit IRC		22:16
*** dimsum__ has joined #openstack-infra		22:17
jeblair	mordred, clarkb: are we using any neutron api calls with hpcloud?	22:17
mordred	jeblair: not in nodepool	22:17
mordred	jeblair: we currently use no neutron calls at all	22:18
mordred	(although I want to change that wrt floating ips when we move to shade, because it's a more solid approach)	22:18
jeblair	right	22:18
mordred	but that's future	22:18
openstackgerrit	Gene Snider proposed openstack-infra/zuul: Migrate zuul timer to apscheduler version 3.0 https://review.openstack.org/162315	22:19
*** tnovacik has joined #openstack-infra		22:20
*** hdd_ has joined #openstack-infra		22:21
*** dimsum__ has quit IRC		22:21
*** e0ne has joined #openstack-infra		22:22
*** dimsum__ has joined #openstack-infra		22:23
*** r-daneel has quit IRC		22:23
jlvillal	jeblair: I feel like this is a dumb question. But when I'm making a comment on a line in a patch. How do I get out of the comment mode? Hitting <ESC> takes me out of diff mode. I was hoping for a better method.	22:23
jeblair	clarkb: short version: they agree none of the instances are active, and they also are not in the deleted instances list which goes back to july 2014, and don't know what they are. they had some basic followup questions which i have answered.	22:23
jeblair	jlvillal: hit up or down to move the cursor out of the box	22:24
*** chlong has quit IRC		22:24
*** andreykurilin_ has joined #openstack-infra		22:24
jeblair	jlvillal: i'm guessing the regular arrow keys are going to work better than j/k for this one? ;) also, i'm hoping you can still type j/k in the comment box, is that correct?	22:25
jlvillal	jeblair: Oh. Sounds simple! :)	22:25
jlvillal	jeblair: Yep j/k work in the comment box. I have to admit I would think escape should exit the comment box	22:25
jlvillal	and not the whole diff mode	22:25
jlvillal	jeblair: Thanks!	22:26
clarkb	jeblair so we should delete them?	22:27
jeblair	clarkb: gimme a second, they pointed me at a jira ticket... so i'm vpning	22:27
*** aysyd has quit IRC		22:28
*** radez is now known as radez_g0n3		22:31
*** ddieterly has quit IRC		22:31
tchaypo	jlvillal: i found that confusing at first as well	22:32
tchaypo	also I’d love to hack in keybindings for deleting all the comment as well	22:32
tchaypo	having to hold down backspace for half an hour is so tedious	22:32
tchaypo	then again maybe i should just write shorter comments	22:32
*** marcusvrn has joined #openstack-infra		22:33
*** dangers is now known as dangers_away		22:36
*** otter768 has joined #openstack-infra		22:36
jeblair	clarkb: i don't think they have a good answer on the deletion question, so perhaps we try to stay on the safe side and maybe delete 20-50 of them?	22:37
jeblair	clarkb: also jesusaurus reports this internally two, so extra data points	22:37
jeblair	too even	22:37
jeblair	jlvillal, tchaypo: that's good to know (i love user feedback like that).... do you expect that to happen after you've typed a comment in, or only when you've selected a line to start a comment, but then changed your mind and want to leave no comment there after all?	22:39
jeblair	jlvillal, tchaypo: (referring to the esc thing)	22:40
jlvillal	jlvillal: I was hoping <ESC> would just exit the comment line, but not erase it.	22:41
*** otter768 has quit IRC		22:41
*** julim has joined #openstack-infra		22:41
jeblair	jlvillal: where should the cursor go in that case?	22:41
tchaypo	Hrm.	22:42
jlvillal	jeblair: Hmmm. Good question! :)	22:42
tchaypo	Thinking back on the times when I’ve hit esc by mistake	22:42
tchaypo	I think it’s usually when I’m in the middle of writing something and decide I should comment on a different line, or that I don’t want to comment at all	22:43
*** dimsum__ has quit IRC		22:43
jlvillal	jeblair: I was thinking that you had to press <enter> to start editing the comment. But now I see that if there is a comment there that it automatically puts you into editing mode if you land on the line.	22:43
*** dimsum__ has joined #openstack-infra		22:43
jeblair	jlvillal, tchaypo: the paradigm is that the cursor is always on a line, either ready to create a comment or edit an existing comment, so that makes it a bit hard to figure out what to do in that case....	22:44
jlvillal	jeblair: I was thinking that you wouldn't be in editing mode, and that maybe could do 'd' to delete the comment at that point. And if you hit 'esc' while in editing mode it would leave it.	22:44
*** tonytan4ever has quit IRC		22:44
jeblair	jlvillal, tchaypo: i think our options are to rethink the paradigm, so that it behaves more like what jlvillal just described; or perhaps disable esc while in comment editing, so at least you have to leave comment editing some other way before you can leave the screen (so it won't happen by accident)	22:45
*** mriedem has quit IRC		22:47
*** dimsum__ has quit IRC		22:48
tchaypo	jlvillal: let me guess	22:49
tchaypo	jlvillal: you use vim?	22:49
tchaypo	;)	22:49
jlvillal	tchaypo: Yep :)	22:50
tchaypo	I don’t think I like either of those options	22:52
tchaypo	often when I’m editing a comment I want to pop out to check another change or something	22:52
jlvillal	Is there an IRC channel for Stackalytics? Engineer stats don't show up it seems: http://stackalytics.com/?release=kilo&module=ironic-group&metric=commits	22:53
tchaypo	if esc was disabled thinks I’d have to arrow up, then hit esc	22:53
tchaypo	i guess that’s not too bad	22:53
*** salv-orlando has quit IRC		22:53
jeblair	jlvillal: i'm not certain, but NikitaKonovalov is in here and may be able to help	22:53
tchaypo	but I think I’d be confused about why esc didn’t work straight away	22:53
*** peristeri has quit IRC		22:54
jeblair	tchaypo: yeah. :\|	22:54
jlvillal	tchaypo: jeblair: I think <ESC> should exit the comment editing, but that would require a paradigm change. Or have it move up to the line being commented on.	22:54
tchaypo	one thing we could do to make this more clear	22:55
tchaypo	would be to edit the comment in a popup rather than inline	22:55
tchaypo	but i think that would be even more disruptive	22:56
tchaypo	can’t see the code you’re commenting on, or comments you’re responding to	22:56
jlvillal	jeblair: Also it would be nice when starting a comment if it defaulted to the right hand side, as I think that is the current patch revision. It seems odd to default to the previous revision to start commenting on.	22:56
jlvillal	Unless I'm doing something wrong.	22:57
jeblair	jlvillal: oh interesting, that might work... it looks like you might currently be able to create multiple comments by selecting the same line multiple times, that seems buggy. at any rate, i think your suggestion of having esc move the cursor back to the line being commented on would work....	22:57
jeblair	jlvillal: would you then want to have to hit enter on the line again to edit your comment, or is the 'move cursor into edit box' paradigm okay for that?	22:58
*** hodos\|2 has quit IRC		22:58
tchaypo	on a different note	22:59
jlvillal	jeblair: Either way is fine for me. The only benefit of the <enter> paradigm. Is then could add a way to delete a whole comment or other actions on the comment.	22:59
tchaypo	as lifeless mentioned yesterday	22:59
*** hasharAway has quit IRC		22:59
tchaypo	I’m finding that gertty takes a really long time to sync, and seems to spend more time with the counter increasing than decreasing	22:59
*** superanne has quit IRC		22:59
openstackgerrit	Kyle Mestery proposed openstack-infra/project-config: Add functional test job for python-neutronclient https://review.openstack.org/162329	22:59
jeblair	jlvillal: and yes, i think it should default on the right. i note that once you leave a comment on the right, it defaults there. i believe i was fighting urwid to get the initial focus on the right, but lost. may be time for a rematch. :)	23:00
jlvillal	jeblair: LOL :D	23:00
*** ParsectiX has quit IRC		23:00
*** ParsectiX has joined #openstack-infra		23:00
openstackgerrit	Kyle Mestery proposed openstack-infra/project-config: Add functional test job for python-neutronclient https://review.openstack.org/162329	23:00
jeblair	tchaypo: if you run with verbose or debug, do log messages give you insight? is it working but really slow? is it piling up duplicate sync requests? are some requests hanging and not timing out properly?	23:01
tchaypo	i seem to be subscribed to 69 projects	23:01
*** r-daneel has joined #openstack-infra		23:02
jeblair	tchaypo: (also note that many sync actions are effectively noop -- i find that when my counter gets really high, it's really only about half as bad as it looks)	23:02
tchaypo	I’m running with verbose at the moment	23:02
tchaypo	https://www.irccloud.com/pastebin/umEjgE8h	23:02
*** achanda_ has joined #openstack-infra		23:03
*** achanda has quit IRC		23:03
tchaypo	yeah, it does seem to eventually hit a point where all the remaining changes are noops and then it drops quickly, but it can take hours to get there	23:03
*** Ryan_Lane has joined #openstack-infra		23:04
tchaypo	let me switch to debug	23:04
*** andreykurilin_ has quit IRC		23:04
*** cdent has joined #openstack-infra		23:04
tchaypo	the database lock messages usually have quite short timestamps - 0.03<x<0.1s	23:04
tchaypo	so I think it’s mostly network latency	23:05
*** claudiub has joined #openstack-infra		23:05
tchaypo	https://www.irccloud.com/pastebin/ALjS1XPZ	23:06
*** r-daneel_ has joined #openstack-infra		23:06
jlvillal	jeblair: Probably another dumb question. Is it possible to set in the config the default sort method? I searched for 'sort' in the reference example doc, but did not have success. I would like it to default to reverse-sorted by change date.	23:06
*** r-daneel has quit IRC		23:07
jeblair	tchaypo: i synced the same changes in your paste and the times in the "Synced change .. in .." messages are roughly similar for me	23:07
*** achanda_ has quit IRC		23:07
*** amitgandhinz has quit IRC		23:07
*** esker has quit IRC		23:08
tchaypo	right after i say the database locks tend to be fairly short, I find one that holds a lock for 14 seconds..	23:08
tchaypo	I’m interpreting the “revision x needs parent commit”, “…. synced” messages as meaning that it’s doing a git fetch of individual patchsets?	23:09
*** adalbas has quit IRC		23:09
*** achanda has joined #openstack-infra		23:10
*** r-daneel_ has quit IRC		23:10
*** r-daneel_ has joined #openstack-infra		23:10
*** gordc has quit IRC		23:10
*** tnovacik has quit IRC		23:12
tchaypo	Reasons I know I’m a bad person: seeing that https://review.openstack.org/#/c/161047/2/specs/stable-requirements.rst has 37 comments makes me chuckle	23:12
*** cdent has quit IRC		23:12
*** sandywalsh has quit IRC		23:12
*** crc32 has quit IRC		23:13
jeblair	tchaypo: actually, no network traffic is happening during those comments...	23:13
*** sandywalsh has joined #openstack-infra		23:13
jeblair	er log entries	23:13
*** yamamoto has joined #openstack-infra		23:14
jeblair	tchaypo: basically the result of all those "needs parent commit" lines is that later, after this sync is finished, it will go do some git fetches for those...	23:14
tchaypo	ah	23:14
tchaypo	in that case	23:14
jeblair	tchaypo: most of what's going on for those 14 seconds is actually local db queries... so there might be something really unoptimized there	23:15
tchaypo	why does it take so long? there’s like a second in between each comment	23:15
tchaypo	ah	23:15
tchaypo	also	23:15
tchaypo	I get the feeling that something is firing off a fresh sync before the previous sync is finished	23:15
tchaypo	when i first fire up gertty it will quickly get up to about 256 or so	23:15
tchaypo	if it’s been synced recently that drops back to 0 almost immediately	23:15
tchaypo	but if it hasn’t that number just grows and grows - and it seems like every now and then another ~200 or so get added	23:16
jeblair	jlvillal: i believe the defaults are hard coded, but i think a config option for them would be a great idea. if you don't want to write a patch, would you mind filing a story?	23:16
*** prad has quit IRC		23:16
tchaypo	once it hits 800 or so my usual remedy is to kill gertty and re-open it	23:16
tchaypo	in this case it was somewhere north of 800 when i killed and restarted to get debug logs, and now it’s fully synced	23:17
openstackgerrit	Joe Gordon proposed openstack-infra/devstack-gate: force_config_drive always=>True https://review.openstack.org/162339	23:17
flashgordon	mtreinish: have a chance to re-spin your patch?	23:17
tchaypo	in about 10 minutes; where before I’d had it open for >20 minutes and the count was still increasing	23:17
jlvillal	jeblair: I can investigate a patch. If I did make it a config option. What should the config option be called and where should it be placed?	23:17
*** yamamoto has quit IRC		23:18
jlvillal	jeblair: As in top_level_name -> second_level_name -> etc...	23:19
jeblair	jlvillal: there's a top level "change-list-query" option, so probably "change-list-sort: <keywords: number, updated>" and "change-list-reverse: <bool, default False>" also at the top level	23:19
tchaypo	while we’re talking feature requests	23:19
jlvillal	jeblair: Okay	23:19
tchaypo	I’d love to be able to switch between side-by-side and unified diff on the fly	23:20
jeblair	tchaypo: the periodic sync should wait until it's completely finished before it adds anything else to it... so the += 200 tasks should only be coming from things it's discovering it has to do while syncinc...	23:20
mtreinish	flashgordon: which patch? I have many	23:21
mtreinish	oh, probably the wheels one. What did I have to do, I missed the trail end of that discussion	23:21
*** e0ne has quit IRC		23:22
tchaypo	flashgordon: so this might just be me being overly relaxed because it’s saturday	23:22
jeblair	tchaypo: interesting; i thought about that but didn't expect anyone to want to do that. it should be possible, though would be a little bit of work. out of curiousity, what would you get from that? also -- literally toggle while you're already on a diff, or would going back to the change, switching, then hitting diff again be okay?	23:22
tchaypo	flashgordon: but I’m wondering if we should be thinking about adding features to pip	23:22
tchaypo	flashgordon: the ability to specify “here are the allowed packages and their allowed versions” seems like it would be useful in many contexts other than ours.	23:23
jlvillal	jeblair: And what do you think about unittests?	23:23
tchaypo	jeblair: the latter would be fine	23:24
tchaypo	when I’m looking at a modification of existing code, side-by-side is often helpful	23:24
tchaypo	but just now I was looking at a brand new file, so the left half of the window is entirely blank - unified would have used the space better	23:24
*** ChuckC_ has quit IRC		23:25
tchaypo	also, I often find myself splitting the pane in tmux and having gertty on the left with docs or some other code on the right, or even running the code. In that circumstance side-by-side often makes the columns too narrow to really see the change; unified would work better	23:25
flashgordon	mtreinish: a sudo or two, clarkb commented inline	23:25
jeblair	jlvillal: i think they might be cool for some things. i really think the search could benefit from testing, for instance, keymaps, palettes, and maybe the sync stuff too. i'm not sure what some of the ui behavioral stuff would look like unit tested. understand that i'm skeptical of mocks and generally only want to test real behaviors.	23:25
mtreinish	flashgordon: oh, just that. I thought there was more to do. That's simple enough, one sec	23:26
jlvillal	jeblair: Okay.	23:26
flashgordon	tchaypo: its not a bad idea, adam_g mentioned something similar	23:26
jeblair	tchaypo: got it, thanks :)	23:26
flashgordon	tchaypo: I agree others have the same issues but on a smaller scale	23:26
tchaypo	flashgordon: maybe it’s a sign I’ve spent too much time in Enterprise environments though	23:27
flashgordon	tchaypo: hehe	23:27
tchaypo	I think I’m thinking of an ITIL software librarby style thing	23:27
*** armax has quit IRC		23:27
flashgordon	if you look at the bottom of the spec there is some related work there	23:27
tchaypo	A different way to manage this might be to run a devpi server and control which versions are available on it	23:27
openstackgerrit	Matthew Treinish proposed openstack-infra/project-config: Build wheel house during nodepool devstack caching https://review.openstack.org/160548	23:28
mtreinish	flashgordon: ^^^	23:28
*** doug-fish has left #openstack-infra		23:28
flashgordon	clarkb: ^^	23:29
jeblair	jlvillal: not sure how i missed this, but check out change 161032	23:29
flashgordon	mtreinish: so the best part of this, is we built numpy but don't even use it	23:29
flashgordon	we still have the apt-get installed version in there	23:29
mtreinish	flashgordon: it's a 2nd level dep for something in nova	23:30
mtreinish	oh, heh I thought we removed the package, or is something else pulling it in	23:30
jlvillal	jeblair: :) Just as I was doing it....	23:30
* jlvillal goes to stash changes and cherry-pick patch		23:30
flashgordon	mtreinish: sigh	23:31
*** stevedore has quit IRC		23:31
*** signed8bit has quit IRC		23:32
mtreinish	flashgordon: nope, I'm wrong it looks like python-numpy is still in the apts lists for nova	23:35
mtreinish	if we want to use the wheel that's probably fair game to remove	23:35
flashgordon	mtreinish: so have a short term answer https://review.openstack.org/162343	23:36
flashgordon	which is the opposite	23:36
flashgordon	heh building numpy is single threaded	23:37
flashgordon	fail	23:37
mtreinish	flashgordon: well that only works until dtroyer's patches to use a venv for services lands	23:37
flashgordon	mtreinish: yup, as I said short term	23:37
flashgordon	timing a numpy wheel build on my box to get a rough estimate on how bad it is	23:38
flashgordon	mtreinish: how many dsvm jobs do we run a day?	23:38
mtreinish	flashgordon: I'm sure graphite can give you the complete answer. I can check subunit2sql for how many gate tempest/grenade dsvms we ran yesterady, one sec	23:39
*** woodster_ has joined #openstack-infra		23:39
*** mattfarina has joined #openstack-infra		23:40
flashgordon	wow 4 minutes	23:41
flashgordon	so yeah	23:41
flashgordon	IMHO worth temporary hacks just to get rid of that	23:41
jlvillal	jeblair: That patch worked for me :)	23:41
flashgordon	mtreinish: that is 4 out of the 20 minutes or so devstack takes ...	23:41
jeblair	jlvillal: cool, thanks	23:42
flashgordon	mtreinish: on zuul I see 675 dsvm jobs that are running or recently finished so this would save about 45 instance hours off of that	23:42
*** ChuckC_ has joined #openstack-infra		23:43
mtreinish	flashgordon: accoring to the db we ran 836 gating jobs between '2015-03-05 00:00:00' and '2015-03-06 00:00:00'	23:44
mtreinish	gating tempest or grenade	23:44
*** ChuckC_ is now known as ChuckC		23:44
jeblair	mtreinish: not counting check?	23:44
*** dmsimard has joined #openstack-infra		23:45
mtreinish	jeblair: nope the subunit2sql db only collects the streams from gate jobs	23:45
flashgordon	mtreinish: wow that is a lot of gate jobs	23:45
jeblair	check is much more	23:45
flashgordon	I assume that is only completed jobs too	23:45
mtreinish	http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/files/logstash/jenkins-log-client.yaml#n14	23:45
jeblair	flashgordon, mtreinish: yesterday zuul launche 8766 jobs with 'dsvm' in their names	23:47
flashgordon	jeblair: wow!	23:48
*** ghostpl_ has quit IRC		23:48
mtreinish	jeblair: heh, nice	23:48
flashgordon	mtreinish: we build numpy wheels 4,782 in the past 24 hours per logstash	23:48
flashgordon	mtreinish: or 318 hours of wall time	23:49
flashgordon	roughly	23:49
jeblair	flashgordon, mtreinish: two days ago 10671 dsvm jobs, but that may have been an abnormally heavy day due to the queue from the rax maint work.	23:49
flashgordon	out of that we are getting about 150-200 merges to openstack, openstack-infra, openstack-dev	23:50
flashgordon	https://github.com/openstack/openstack/graphs/commit-activity	23:50
flashgordon	that is really impressive	23:50
* flashgordon wonders how big travis CI is		23:50
jeblair	flashgordon, mtreinish: yesterday we launched 34841 jobs total	23:51
mtreinish	flashgordon: heh, I'm trying to get the run count for yesterday as meaninglessly large number to brag about. But the query is really slow...	23:51
jeblair	flashgordon: we are significantly larger than travisci	23:52
jeblair	mordred: ^ what was it last time you checked?	23:52
flashgordon	mtreinish: haha fail	23:52
* mordred forgets		23:52
mordred	but we were decently bigger	23:52
flashgordon	at 34k jobs I would be surprised if we are just 'decently' bigger	23:53
jeblair	flashgordon, mtreinish: 35300 total jobs launched 2 days ago.	23:53
jeblair	istr it was an integer factor larger	23:53
mtreinish	flashgordon: 627394 tempest tests as part of the gate queue yesterday	23:54
mtreinish	(that took 6 min 42.55 sec I probably need to make the schema better)	23:54
mtreinish	jeblair: hmm, 35300 over 2 days but 34841 yesterday	23:56
mtreinish	oh nm misread	23:56
jeblair	mtreinish: status:open (topic:enable_swift OR topic:dib-nodepool OR topic:zanata OR topic:downstream-puppet OR topic:askbot-site OR topic: gerrit-upgrade)	23:56
jeblair	gah	23:56
jeblair	mtreinish: 35300 on tuesday	23:57
*** mattfarina has quit IRC		23:57
jeblair	mtreinish: 34841 on thursday	23:57
jeblair	for UTC days	23:57
mtreinish	jeblair: yeah, I just misread what you put earlier	23:58
*** dimsum__ has joined #openstack-infra		23:59

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!