Wednesday, 2015-07-15

nibalizerpabelanger: hey wanna laugh?00:00
nibalizerwe just moved nanliu/git into
openstackgerritSpencer Krum proposed openstack-infra/system-config: Nanliu-git is now Puppet-community git
*** akshai has quit IRC00:02
camunozpleia2, any idea why Jenkins rejected my patch? logs talk about a TLS packet size error00:07
openstackgerritMerged openstack/requirements: Bump 1.9.0 to oslo.utils
pleia2camunoz: unfortunately that's just a false failure, leave a comment in the review that just says: recheck00:08
openstackgerritMerged openstack/requirements: Bump oslotest to 1.6.1 for functional tests
openstackgerritMerged openstack/requirements: Stop adding extra newline after unparseable lines.
openstackgerritMerged openstack/requirements: Place an accurate lower bound on pbr
camunozpleia2, ok, will do... thanks00:08
sdagueclarkb: thank you, I just looked at the last logs, and the new image was blowing up on parted not being installed00:10
pleia2camunoz: if you want to see something cool, go to and search for your patch number (201808) - it shows progress of tests :)00:10
sdagueand I was about to rage quit the multinode job game00:11
sdagueI'll give it another look in the morning00:11
camunozpleia2, that is cool00:11
*** FallenPegasus has quit IRC00:12
*** esmute has quit IRC00:13
openstackgerritsebastian marcet proposed openstack-infra/openstackid: OIDC - OpenId Connect Implementation
mfalaticWhat team manages the openstackstatus bot?00:13
pleia2mfalatic: that's us00:13
mfalaticpleia2 I'm curious why Closes-bug: NNNNNN triggers the bot, apparently, while Closes-bug: #NNNNN does not.00:14
openstackgerritOpenStack Proposal Bot proposed openstack/requirements: Updated from global requirements
mfalaticpleia2 scratch that00:15
mfalaticpleia2 Seems to work either way.00:16
pleia2openstackstatus does that?00:16
mfalaticpleia2 didn't seem to when I tested it earlier.00:16
mfalaticactually, "openstack" seems to do it.00:16
openstackgerritOpenStack Proposal Bot proposed openstack/diskimage-builder: Updated from global requirements
mfalaticpleia2 For future reference, does infra own that as well?00:17
pleia2that's actually not one we run00:17
mfalaticah, hmm.00:17
mfalaticWell, it seems to work for both cases so my current question is moot00:18
pleia2indeed :)00:18
mfalaticbut interesting to know its not one of yours00:18
*** esmute has joined #openstack-infra00:18
*** mtanino has quit IRC00:19
*** tsekiyam_ has joined #openstack-infra00:21
*** woodster_ has quit IRC00:21
openstackgerritOpenStack Proposal Bot proposed openstack-dev/hacking: Updated from global requirements
openstackgerritOpenStack Proposal Bot proposed openstack/diskimage-builder: Updated from global requirements
*** david-lyle has quit IRC00:25
*** hichihara has joined #openstack-infra00:25
*** tsekiyama has quit IRC00:25
*** tsekiyam_ has quit IRC00:26
mordredclarkb, greghaynes: btw - image imported successfully on retry00:26
*** sdake_ has joined #openstack-infra00:28
*** Swami has quit IRC00:28
*** Kennan2 is now known as Kennan00:29
*** achanda has quit IRC00:29
camunozpleia2, that worked... jenkins has now accepted the patch00:30
*** sdake has quit IRC00:31
*** signed8bit has joined #openstack-infra00:33
*** signed8b_ has joined #openstack-infra00:33
pleia2camunoz: looks good00:33
*** pcrews has joined #openstack-infra00:34
camunozpleia2, yay!.... so, what happens now? :-\00:35
pleia2camunoz: now you wait for reviews, once it has enough a core member can approve it for merging00:36
camunozalright... sounds good, now I wait... thanks for all the guidance!00:37
pleia2you're welcome00:38
*** esmute has quit IRC00:38
*** thorst has joined #openstack-infra00:38
*** esmute has joined #openstack-infra00:39
*** signed8b_ is now known as signed8bit_ZZZzz00:40
*** signed8bit_ZZZzz has quit IRC00:41
*** signed8bit has quit IRC00:41
*** esmute has quit IRC00:43
*** bitblt has quit IRC00:44
*** watanabe_isao has quit IRC00:45
*** baoli has joined #openstack-infra00:45
*** mhu has quit IRC00:46
*** ivar-lazzaro has quit IRC00:50
*** laughterwym has joined #openstack-infra00:50
*** ivar-lazzaro has joined #openstack-infra00:51
*** mhu has joined #openstack-infra00:52
*** dims__ has quit IRC00:54
*** fitoduarte has quit IRC00:55
*** kzaitsev_mb has quit IRC01:00
*** alexsyip has quit IRC01:03
*** hrubi has quit IRC01:07
*** spzala_ has joined #openstack-infra01:08
*** spzala has quit IRC01:08
*** spzala_ is now known as spzala01:08
*** shashank_hegde_ has quit IRC01:08
*** tqtran has quit IRC01:09
*** asettle has joined #openstack-infra01:09
*** hrubi has joined #openstack-infra01:13
*** puranamr has quit IRC01:18
*** puranamr has joined #openstack-infra01:20
openstackgerritMerged openstack/requirements: Update pylint pin to 1.4.4
*** FallenPegasus has joined #openstack-infra01:23
*** changbl has joined #openstack-infra01:24
*** rlandy has joined #openstack-infra01:25
*** bradjones has quit IRC01:26
*** dims_ has joined #openstack-infra01:28
*** bdemers has joined #openstack-infra01:29
*** bradjones has joined #openstack-infra01:30
*** bradjones has quit IRC01:30
*** bradjones has joined #openstack-infra01:30
*** signed8bit has joined #openstack-infra01:30
*** puranamr has quit IRC01:30
*** signed8b_ has joined #openstack-infra01:30
*** puranamr has joined #openstack-infra01:30
*** puranamr has quit IRC01:30
openstackgerritOpenStack Proposal Bot proposed openstack-dev/hacking: Updated from global requirements
*** signed8b_ is now known as signed8bit_ZZZzz01:31
*** puranamr has joined #openstack-infra01:31
openstackgerritOpenStack Proposal Bot proposed openstack/diskimage-builder: Updated from global requirements
*** signed8bit_ZZZzz has quit IRC01:33
*** signed8bit has quit IRC01:33
*** eantyshev has quit IRC01:35
*** dprince has quit IRC01:36
*** Ryan_Lane has left #openstack-infra01:36
*** puranamr has quit IRC01:39
*** unicell has quit IRC01:39
*** shashank_hegde_ has joined #openstack-infra01:40
*** puranamr has joined #openstack-infra01:40
*** thorst has quit IRC01:41
*** asettle has quit IRC01:46
*** geoffarnold has quit IRC01:50
*** otter768 has joined #openstack-infra01:51
*** shashank_hegde_ has quit IRC01:51
*** boris-42 has quit IRC01:52
*** rlandy has quit IRC01:52
*** rlandy has joined #openstack-infra01:53
*** signed8bit has joined #openstack-infra01:53
*** signed8b_ has joined #openstack-infra01:53
*** markvoelker has quit IRC01:54
*** geoffarnold has joined #openstack-infra01:54
*** signed8bit is now known as signed8bit_ZZZzz01:54
lifelessfungi: we should also trigger the constraints generator on merges landing in requirements01:55
*** mhu has quit IRC01:55
*** otter768 has quit IRC01:55
*** kzaitsev_mb has joined #openstack-infra01:56
*** signed8bit_ZZZzz has quit IRC01:56
*** signed8b_ has quit IRC01:56
*** mhu has joined #openstack-infra01:57
*** markvoelker has joined #openstack-infra02:00
*** kzaitsev_mb has quit IRC02:01
*** signed8bit has joined #openstack-infra02:01
*** signed8b_ has joined #openstack-infra02:01
*** ivar-lazzaro has quit IRC02:01
*** Kennan2 has joined #openstack-infra02:02
*** signed8b_ is now known as signed8bit_ZZZzz02:02
*** Kennan has quit IRC02:03
*** signed8bit has quit IRC02:04
*** signed8bit_ZZZzz has quit IRC02:04
*** shashank_hegde_ has joined #openstack-infra02:04
*** patrickeast_ has quit IRC02:10
*** coolsvap|away has quit IRC02:10
*** unicell has joined #openstack-infra02:10
*** yamahata has quit IRC02:10
*** coolsvap|away has joined #openstack-infra02:12
*** david-lyle has joined #openstack-infra02:12
*** abregman has quit IRC02:13
*** dims_ has quit IRC02:17
*** markvoelker has quit IRC02:18
*** puranamr has quit IRC02:18
*** koolhead17 has joined #openstack-infra02:18
*** koolhead17 has quit IRC02:18
*** koolhead17 has joined #openstack-infra02:18
*** yamamoto has joined #openstack-infra02:18
openstackgerritColleen Murphy proposed openstack-infra/puppet-askbot: Enable arrow alignment lint check
*** arxcruz has quit IRC02:19
*** rlandy has quit IRC02:19
*** maishsk has quit IRC02:21
openstackgerritColleen Murphy proposed openstack-infra/puppet-asterisk: Enable arrow alignment lint check
openstackgerritColleen Murphy proposed openstack-infra/puppet-bandersnatch: Enable arrow alignment lint check
openstackgerritlifeless proposed openstack/requirements: Overhaul the docs.
*** unicell1 has joined #openstack-infra02:29
*** unicell has quit IRC02:29
*** ashleighfarnham has joined #openstack-infra02:30
openstackgerritlifeless proposed openstack/requirements: Document current constraints status.
*** shashank_hegde_ has quit IRC02:31
*** zz_dimtruck is now known as dimtruck02:32
*** sdake_ is now known as sdake02:34
*** geoffarnold has quit IRC02:34
*** david-lyle has quit IRC02:35
*** ashleighfarnham has quit IRC02:35
*** unicell1 has quit IRC02:35
*** unicell has joined #openstack-infra02:36
*** larainema has quit IRC02:40
*** sdake_ has joined #openstack-infra02:43
*** asettle has joined #openstack-infra02:43
*** madong_ has joined #openstack-infra02:44
*** coolsvap has quit IRC02:45
*** sdake has quit IRC02:46
*** sdake__ has joined #openstack-infra02:46
mordredcrinkle: yay arror alignment02:47
*** madong_ has quit IRC02:49
crinklemordred: i like messing with your git blame02:49
*** baoli has quit IRC02:49
*** sdake_ has quit IRC02:50
*** cody-somerville has joined #openstack-infra02:50
*** cody-somerville has quit IRC02:50
*** cody-somerville has joined #openstack-infra02:50
*** tlian2 has joined #openstack-infra02:51
mordredcrinkle: that's what people say about you02:52
mordredthat crinkle sure does like messing with git blame02:52
*** baoli has joined #openstack-infra02:52
*** patrickeast has joined #openstack-infra02:53
*** tlian has quit IRC02:53
*** xnox has quit IRC02:54
*** sdake__ is now known as sdake02:58
openstackgerritColleen Murphy proposed openstack-infra/puppet-bugdaystats: Make git_url parameter useful
openstackgerritColleen Murphy proposed openstack-infra/puppet-bugdaystats: Add Gemfile and puppet 4 checks
openstackgerritColleen Murphy proposed openstack-infra/puppet-bup: Add Gemfile and puppet 4 checks
*** signed8bit has joined #openstack-infra03:04
*** signed8b_ has joined #openstack-infra03:04
*** larainema has joined #openstack-infra03:05
*** ParsectiX has joined #openstack-infra03:09
*** xnox has joined #openstack-infra03:10
*** achanda has joined #openstack-infra03:10
*** ddieterly has quit IRC03:10
*** dims_ has joined #openstack-infra03:19
*** xnox has quit IRC03:23
*** dims_ has quit IRC03:23
*** pal has joined #openstack-infra03:28
camunozStevenK,  (release notes contain a bunch of configuration changes that need to go to puppet)03:30
*** xnox has joined #openstack-infra03:33
StevenKcamunoz: \o/03:33
StevenKcamunoz: I shall start digging into that after lunch03:34
pleia2with translators testing now we need like, a -dev-dev03:34
camunozlet me know if you have any questions03:34
*** markvoelker has joined #openstack-infra03:35
*** sputnik13 has joined #openstack-infra03:37
StevenKpleia2: Well, it's a -dev instance, I think we can announce it and then +A the branch at the appropiate time.03:37
pleia2StevenK: yeah, I have some how-we-install-it patches to write too now that we have some new tools for handling zip files and things03:38
*** david-lyle has joined #openstack-infra03:38
pleia2so I figure we'll have a few windows for maintenance03:38
*** markvoelker_ has joined #openstack-infra03:38
StevenKpleia2: We're learning too, just like the translators are :-03:39
StevenK:-) even03:39
*** shashank_hegde_ has joined #openstack-infra03:40
*** markvoelker has quit IRC03:40
StevenKExcellent, I've imparted some wisdom, now I can go hom ... Oh, damn, I'm already here.03:40
*** arosen has quit IRC03:43
openstackgerritsebastian marcet proposed openstack-infra/openstackid: OIDC - OpenId Connect Implementation
*** otter768 has joined #openstack-infra03:52
mordredStevenK, pleia2: zomg. a -dev-dev would be teh ossum03:53
*** spzala has quit IRC03:55
*** achanda has quit IRC03:56
*** otter768 has quit IRC03:56
*** arosen has joined #openstack-infra03:56
*** achanda has joined #openstack-infra03:57
*** kushal has quit IRC04:03
*** signed8__ has joined #openstack-infra04:03
*** markvoelker has joined #openstack-infra04:04
*** signed8b_ has quit IRC04:04
*** signed8bit has quit IRC04:04
* StevenK proposes translate-dev-ops.openstack.org04:05
* nigelb recommends having dev-ops translated to a different language and using that instead.04:06
*** crc32 has quit IRC04:06
StevenKBut I don't know the French or German word for 'hipster'04:06
nigelblol, let's see what google translate says.04:07
nigelbgerman has to have a word, right?04:07
*** puranamr has joined #openstack-infra04:07
*** markvoelker_ has quit IRC04:07
*** markvoelker_ has joined #openstack-infra04:08
*** dannywilson has joined #openstack-infra04:08
*** dannywilson has quit IRC04:08
*** dannywilson has joined #openstack-infra04:09
*** dannywilson has quit IRC04:09
*** xnox has quit IRC04:09
*** markvoelker has quit IRC04:09
*** smyers has quit IRC04:11
*** ddieterly has joined #openstack-infra04:11
*** puranamr_ has joined #openstack-infra04:11
*** puranamr has quit IRC04:13
*** gildub has quit IRC04:14
*** asettle has quit IRC04:16
*** ddieterly has quit IRC04:16
*** signed8bit has joined #openstack-infra04:17
*** achanda has quit IRC04:18
*** linkedinyou has joined #openstack-infra04:18
*** signed8__ has quit IRC04:20
*** coolsvap|away is now known as coolsvap04:20
*** signed8bit has quit IRC04:21
*** hichihara has quit IRC04:22
*** dannywilson has joined #openstack-infra04:24
*** kzaitsev_mb has joined #openstack-infra04:28
*** fifieldt has joined #openstack-infra04:28
*** xnox has joined #openstack-infra04:30
*** harlowja_at_home has joined #openstack-infra04:31
*** kzaitsev_mb has quit IRC04:33
*** sputnik1_ has joined #openstack-infra04:33
*** ajo has quit IRC04:35
*** hdd has joined #openstack-infra04:36
*** sputnik13 has quit IRC04:36
*** ajo has joined #openstack-infra04:37
*** baoli has quit IRC04:39
*** oomichi has quit IRC04:40
*** baoli_ has joined #openstack-infra04:41
*** gildub has joined #openstack-infra04:41
*** boris-42 has joined #openstack-infra04:43
*** baoli_ has quit IRC04:44
*** gsagie has joined #openstack-infra04:44
*** puranamr_ has quit IRC04:47
gsagieHello everyone, i will appreciate it if anyone can look at and tells me why Jenkin fails, i am able to install it and run tox on my local enviorment and i see this but dont understand what is the problem04:47
gsagieits an initial commit i added for a new project04:47
openstackgerritSergey Vilgelm proposed openstack/requirements: Bump oslo.log to 1.6.0 version
*** koolhead17 has quit IRC04:49
*** asettle has joined #openstack-infra04:50
*** patrickeast has quit IRC04:51
*** fitoduarte has joined #openstack-infra04:51
clarkbgsagie: incompatible requirements04:51
clarkbtoggle ci and look at the comments04:51
gsagieclarkb: yes but i am wondering why, as i took the requierments from anothe project which works fine04:52
gsagieand i dont understand which requierment is incompatible04:52
*** yfried_ has quit IRC04:53
clarkbgsagie: taking them fro ma project may not give you a good list as we only enforce updates when peopl etry to make changes04:53
clarkbso it is eventually consistent04:53
clarkb(and my ipv6 is hosed agin so all the things are breaking for me)04:54
gsagieclarkb: ok thanks, i will try to narrow it down and find the problem04:54
clarkbgsagie: the job should tell you what is wrong04:54
clarkbif you read the log04:55
*** mrmartin has joined #openstack-infra04:55
*** shashank_hegde_ has quit IRC04:56
clarkblooks like a key error setup.cfg, I would have to go read the script to see what that is about04:56
*** shashank_hegde_ has joined #openstack-infra04:56
gsagieclarkb : yeah i read the log, this is the exception
*** hdd has quit IRC04:57
clarkb is the line breaking04:57
*** shashank_hegde_ has quit IRC04:58
gsagieclarkb: yeah thanks, guess i miss that header in my setup.cfg04:59
gsagielet me check04:59
clarkbI don't think you need the extras header, that code checks if it doesnt exist and returns an empty dict04:59
clarkbit is saying the project has no setup.cfg04:59
*** Adri2000 has joined #openstack-infra05:00
*** sabeen has quit IRC05:00
*** dimtruck is now known as zz_dimtruck05:01
gsagieclarkb: the initial commit has it (
gsagiemaybe something wrong since its a new project and doesnt already have it in it?05:02
clarkb is what reads the setup.cfg05:02
*** otter768 has joined #openstack-infra05:04
clarkbha I see the problem05:05
clarkbit checks out master and checks that05:05
clarkbwhich fails bceause master is current commit 000000005:05
gsagieyeah, thats why a patch i have which is based on that one actually pass05:05
gsagiethe tests05:06
clarkbwell that and if it doesn't make reqs changes they aren't checked05:06
*** markvoelker has joined #openstack-infra05:06
*** markvoelker_ has quit IRC05:06
*** kzaitsev_mb has joined #openstack-infra05:06
clarkbdid we seed this repo with something? I don't think we have run into this before because either people seed with workingish repo or we push in an initial commit with .gitreview file05:07
*** achanda has joined #openstack-infra05:07
clarkbin either case master has content05:07
*** tlian2 has quit IRC05:07
clarkboh hrm it does say the added git review file is the first commit, so maybe this is a new bug05:07
gsagieclarkb: i saw that .gitreview was added05:07
clarkbya its there05:07
*** markvoelker_ has joined #openstack-infra05:08
*** kzaitsev_mb has quit IRC05:11
openstackgerritClark Boylan proposed openstack-infra/system-config: Use config drive in hpcloud
*** markvoelker has quit IRC05:11
clarkbjogo: sdague ^ that should fix hostnames in hpcloud05:11
clarkbjhesketh: pleia2 any othercores awake at this hour?05:12
*** tjones1 has joined #openstack-infra05:12
*** ddieterly has joined #openstack-infra05:12
*** adduarte has joined #openstack-infra05:13
*** FallenPegasus has quit IRC05:14
mrmartinclarkb are you core on infra?05:15
openstackgerritVipul Sabhaya proposed openstack/requirements: Add cue and python-cueclient to projects.txt
ianwclarkb: yay, that was a pita to figure out with mod_ssl's certificate generation05:15
*** fitoduarte has quit IRC05:16
*** ddieterly has quit IRC05:17
*** smyers has joined #openstack-infra05:18
jheskethclarkb: yep, will tkae a look05:19
gsagieclarkb: what i clone the repository i dont see the .gitreview file, so that might be part of the problem05:19
*** xyang1 has quit IRC05:20
ianwjhesketh / clarkb : <- remove heat 2.6 job ... should be uncontroversial and unblock ggillies05:21
clarkbianw: it should already be removed?05:22
clarkbmrmartin: yes05:22
mrmartincan you give me an approve on that: ? :)05:22
ianwclarkb: i think it got overlooked ... it is gate-heat-templates-python2605:23
clarkbgsagie: its there. dot files are by default hidden05:23
clarkbjhesketh: did you see the os loganalyze stack is in merge conflict?05:27
clarkbmrmartin: I need to pop out now but will try to take a look later05:27
*** Longgeek has joined #openstack-infra05:28
*** puranamr has joined #openstack-infra05:30
*** harlowja_at_home has quit IRC05:30
openstackgerritVipul Sabhaya proposed openstack-infra/project-config: Move cue and cueclient docs to docs.o.o
*** tjones1 has quit IRC05:32
*** laughterwym_ has joined #openstack-infra05:32
*** puranamr_ has joined #openstack-infra05:32
*** puranamr has quit IRC05:34
*** laughterwym has quit IRC05:35
*** Sukhdev has joined #openstack-infra05:38
openstackgerritMerged openstack-infra/project-config: Make heat-templates CI passable again - remove py26 job
jogoclarkb: thanks!05:39
*** kiran-r has joined #openstack-infra05:39
jheskethclarkb: yeah I did, sorry I haven't got to it (day has been full of interruptions) will do it asap!05:42
*** jyuso1 has quit IRC05:43
StevenKcamunoz: Hmmm, so, I can't see a 3.7.1 war on sourceforge05:45
*** _nadya_ has joined #openstack-infra05:45
StevenKcamunoz: Also, seems to have its own page, but redirects to
StevenK<meta http-equiv="refresh" content="0; url=">05:47
StevenKThat would do it05:48
*** dboik has joined #openstack-infra05:53
*** fawadkhaliq has joined #openstack-infra05:54
*** dboik has quit IRC05:57
*** _nadya_ has quit IRC05:57
*** _nadya_ has joined #openstack-infra05:58
*** otter768 has quit IRC05:59
openstackgerritOpenStack Proposal Bot proposed openstack-infra/project-config: Normalize projects.yaml
*** adreznec has quit IRC06:01
*** _nadya_ has quit IRC06:02
*** adreznec has joined #openstack-infra06:03
*** k4n0 has joined #openstack-infra06:04
*** deepakcs has joined #openstack-infra06:09
*** larainema has quit IRC06:10
*** BharatK has joined #openstack-infra06:12
StevenKpleia2: Are you still around? I can't see how the ROOT.war file is generated06:13
*** ddieterly has joined #openstack-infra06:13
*** shashank_hegde_ has joined #openstack-infra06:14
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Always use gate- instead of check (devstack-gate)
*** yfried_ has joined #openstack-infra06:18
AJaegerpleia2, anteaya, jeblair : This is the final cleanup ^. Reviews would be great.06:18
*** ddieterly has quit IRC06:18
*** shashank_hegde_ has quit IRC06:21
openstackgerritGal Sagie proposed openstack/requirements: Try a fix for new projects
*** yfried__ has joined #openstack-infra06:22
*** yfried_ has quit IRC06:22
*** ParsectiX has quit IRC06:24
*** abregman has joined #openstack-infra06:25
*** shashank_hegde_ has joined #openstack-infra06:27
*** mrunge has joined #openstack-infra06:28
*** mrunge has quit IRC06:28
*** puranamr_ has quit IRC06:30
*** afazekas has joined #openstack-infra06:31
*** fawadkhaliq has quit IRC06:33
*** mrunge has joined #openstack-infra06:34
*** larainema has joined #openstack-infra06:37
pleia2StevenK: in case you haven't gotten it,
openstackgerritMerged openstack-infra/project-config: Normalize projects.yaml
StevenKDidn't work for me, patch incoming06:38
*** pal has quit IRC06:38
openstackgerritSteve Kowalik proposed openstack-infra/puppet-zanata: Make source for the ROOT.war a URI
AJaegerStevenK: Did you see the question regarding the translation memory?06:39
AJaeger(on the mailing list)06:39
StevenKAJaeger: I did, yes, but got distracted.06:39
StevenKpleia2: Source for that change is
*** mrunge has quit IRC06:40
*** pal has joined #openstack-infra06:41
gsagieAJaeger: i was trying to send an initial commit for the project (Kuryr), but it fails with strange error (  apparently 'setup.cfg' is not in the project dictionary, i was trying to add a fix-test patch as i am not sure06:41
gsagiewho calls that "extras" method06:42
*** ppai has joined #openstack-infra06:42
StevenKAJaeger: Thanks for replying, let's see what the reporter comes back with06:44
gsagiemy "correction" is here :   (it was failing on the method before, now its incompatible requierments)06:44
AJaegerStevenK: I'm not sure how to download the translation memory from transifex yet ;(06:45
*** mrunge has joined #openstack-infra06:45
*** shashank_hegde_ has quit IRC06:46
AJaegergsagie: Your requirements file content is incompatible06:46
AJaeger" Requirement set([Requirement(package=u'pbr', location='', specifiers='<2.0,>=0.11', markers=u'', comment='')]) does not match openstack/requirements value set([Requirement(package='pbr', location='', specifiers='<2.0,>=1.3', markers='', comment='')])"06:47
AJaegerSo, update the pbr line and as well:06:47
AJaeger"Requirement set([Requirement(package=u'oslotest', location='', specifiers='>=1.5.1', markers=u'', comment=u'# Apache-2.0')]) does not match openstack/requirements value set([Requirement(package='oslotest', location='', specifiers='>=1.7.0', markers='', comment='# Apache-2.0')])"06:47
pleia2StevenK: oh huh, I wonder how the old one worked at all on our server, I'll have to look in the morning06:47
AJaegeronce those two lines in the *requirements* files are updated, you should be fine06:47
pleia2since at some point it had to, else it wouldn't have installed o_o06:47
AJaeger201933 is wrong06:47
AJaegergsagie: Do you know what to do?06:48
*** HeOS has quit IRC06:48
AJaegerStevenK: Seems you can only download the translation memory via the API:
StevenKpleia2: Right06:49
gsagieAjaeger: i guess i need to check and update the requierment.txt and test-requierment.txt files according to the global openstack one, but before 201933 it wasnt giving me the same error06:50
gsagieit had an exception in the "extras" method in openstack_requierments where it couldnt find 'setup.cfg' in project dir06:50
StevenKAJaeger: Huh, I saw TMX referenced on Zanata as well, I'll look into that as wekk.06:50
*** _nadya_ has joined #openstack-infra06:52
StevenKAJaeger: Right, so it's something different, Zanata supports TMX upload, I'll look at that tomorrow.06:52
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Always use gate- instead of check (devstack-gate)
*** juggler has quit IRC06:52
AJaegergsagie: your first patch really needs to  setup the repo completely so that it passes all tests...06:53
AJaegerStevenK: thanks!06:53
*** ildikov has quit IRC06:53
gsagieAJaeger: sec i will show you, i updated the requierments and re sent the patch without the dependency on the fix06:53
*** fawadkhaliq has joined #openstack-infra06:54
gsagieAJaeger:   this was the error message before this fix06:54
StevenKpleia2: So that fix gets the ROOT.war to exist, but there's no mysql server :-/06:55
*** fawadk has joined #openstack-infra06:55
AJaegerlifeless: are you around?06:56
pleia2StevenK: yeah, in production we use trove, so you'll want to use a .pp file that has the mysql stuff in it06:57
*** adduarte has quit IRC06:58
pleia2we don't use this in our prod instance, but you will:
*** jyuso1 has joined #openstack-infra06:58
*** fawadkhaliq has quit IRC06:58
StevenKpleia2: Yeah, I found it, just wondering if I can plug it into manifests/local.pp06:58
*** asettle has quit IRC06:59
*** markus_z has joined #openstack-infra06:59
AJaegergsagie: Let'S ask lifeless, he has been working on these. Seems he's currently not here. Please ping him later again and ask him to look into the requirements failure06:59
*** fawadkhaliq has joined #openstack-infra06:59
*** fhubik has joined #openstack-infra06:59
pleia2StevenK: should be able to, it uses the mysql puppet module that install_modules should give you06:59
*** dingyichen has joined #openstack-infra07:00
gsagieAJaeger: ok thanks, will try to see if i can get this to pass with this fix to pinpoint the problem07:00
*** fawadk has quit IRC07:00
*** heyongli has quit IRC07:01
pleia2aaand I should get to bed07:01
AJaegergsagie: I suggest to try contacting lifeless at another point of time. He lives in Australia, so hope you find some time together.07:01
AJaegerpleia2: Good night!07:01
StevenKAJaeger: He does not!07:01
StevenKAJaeger: New Zealand is not Australia.07:02
AJaegerStevenK: He doesn't? Oops...07:02
AJaegerStevenK: Indeed! That's a huge difference!07:02
*** heyongli has joined #openstack-infra07:02
AJaegerNew Zealand is much nicer ;)07:02
AJaeger(me has never been to Australia, so cannot compare ;)07:02
StevenKSee how much stuff I do for you now. ;-)07:02
* AJaeger apologies to all Australians for the unfair comparison07:03
*** yamamoto has quit IRC07:03
StevenKAJaeger: Haha. Dig up!07:03
*** dannywilson has quit IRC07:05
*** samuelBartel has joined #openstack-infra07:06
*** sdake has quit IRC07:07
*** yamamoto has joined #openstack-infra07:09
AJaegergsagie: what did you change so that your tests finally pass?07:09
AJaegerAh, the depends on - yeah, that explains it...07:09
*** laughterwym_ has quit IRC07:10
*** vlaza has joined #openstack-infra07:10
*** shardy has joined #openstack-infra07:10
*** ihrachyshka has joined #openstack-infra07:12
*** larainema has quit IRC07:12
openstackgerritliusheng proposed openstack/requirements: Bump gabbit to version 1.1.0
*** ddieterly has joined #openstack-infra07:14
*** mrmartin has quit IRC07:16
*** markus_z has quit IRC07:17
*** Ala has joined #openstack-infra07:18
*** shardy_ has joined #openstack-infra07:18
gsagieAJaeger: yeah thats the depends-on with my "fix" in the extras method, so i believe there is something wrong with new projects on that part but i couldnt find which project calls method or suppose to populate project['setup.cfg']07:18
*** heyongli has quit IRC07:18
*** shardy has quit IRC07:19
AJaegergsagie: yeah - please discuss with lifeless, he recently reworked that whole code07:19
ppaihi all, is there a way to invoke a hook script for every patch set posted to a stackforge project for review ? All I need is to invoke a curl request to external build server with change number as param.07:19
*** ddieterly has quit IRC07:19
*** dims_ has joined #openstack-infra07:20
*** heyongli has joined #openstack-infra07:21
*** shardy_ has quit IRC07:23
*** zeih has joined #openstack-infra07:23
openstackgerritGal Sagie proposed openstack/requirements: Add checks for setup.cfg existence in project dictionary
*** Longgeek has quit IRC07:23
*** zigo has quit IRC07:24
*** shardy has joined #openstack-infra07:24
gsagieAJaeger: will do07:24
*** zigo has joined #openstack-infra07:25
*** fhubik_afk has joined #openstack-infra07:25
*** fhubik has quit IRC07:25
*** fhubik_afk is now known as fhubik07:25
*** dims_ has quit IRC07:26
*** Kiall has quit IRC07:26
openstackgerrityolanda.robla proposed openstack-infra/puppet-gerrit: cron to clear gerrit logs
*** koolhead17 has joined #openstack-infra07:28
*** Kiall has joined #openstack-infra07:28
openstackgerritMerged openstack/requirements: Bump upper constraint of taskflow to newly released 1.15.0
openstackgerritDirk Mueller proposed openstack-infra/project-config: Start with a green field for RPM packaging
*** zeih has quit IRC07:32
*** rossella_s has joined #openstack-infra07:32
*** zeih has joined #openstack-infra07:33
*** yfried has joined #openstack-infra07:34
*** arxcruz has joined #openstack-infra07:35
*** yfried__ has quit IRC07:35
*** ParsectiX has joined #openstack-infra07:37
*** Sukhdev has quit IRC07:37
*** bexelbie has joined #openstack-infra07:39
*** e0ne has joined #openstack-infra07:39
*** jlanoux has joined #openstack-infra07:40
*** zeih has quit IRC07:42
*** sshnaidm has joined #openstack-infra07:43
*** ildikov has joined #openstack-infra07:47
*** odyssey4me has quit IRC07:51
*** odyssey4me has joined #openstack-infra07:52
*** dtantsur|afk is now known as dtantsur07:52
*** larainema has joined #openstack-infra07:53
*** zeih has joined #openstack-infra07:55
*** fhubik is now known as fhubik_afk07:55
*** fhubik_afk is now known as fhubik07:55
*** pblaho has joined #openstack-infra07:55
*** koolhead17 has quit IRC07:55
lifelessAJaeger: gsagie: whats up?07:55
*** heyongli has quit IRC07:55
AJaegerlifeless: gsagie has started a new repo from scratch - and has the requirements jobs running there. These fail for the very first patch07:56
*** sergmelikyan has joined #openstack-infra07:56
*** heyongli has joined #openstack-infra07:56
AJaegerlifeless: changes is, and look at the change set 8 (which does not have the depends-on) to see the failure for the requirements job07:57
*** bluex-pl has joined #openstack-infra07:57
*** pksingh has joined #openstack-infra07:57
lifelessAJaeger: so it looks like the initial tree was bad07:57
lifelessAJaeger: there's an initial commit which is in an invalid state07:57
*** kushal has joined #openstack-infra07:58
*** bluex-pl has quit IRC07:58
AJaegerlifeless: initial tree just contains the .gitreview file07:58
*** bluex-pl has joined #openstack-infra07:58
pksinghlifeless: Hi,07:58
AJaegerlifeless: this is the normal situation for a fresh repository from scratch07:58
AJaegerlifeless: One way is to disable the requirements check until the repo has initial content.07:59
lifelessAJaeger: but this repo *has* initial content07:59
AJaegerlifeless: only the .gitreview:
*** Longgeek has joined #openstack-infra08:00
AJaegerAnd that's how OpenStack "empty" repos are setup08:01
lifelessAJaeger: so thats initial content08:01
gsagielifeless: hi, basically the repository was created with the .gitreview, i couldnt add any more commits (tried with a simple one then in the patch) until i added this "fix" i wanted to further investigate where project dictionary is being populated but couldnt find the caller08:01
lifelesscould we fix that? I guess not, because it would be python specific.08:01
AJaegerlifeless: I guess not either08:01
*** yamamoto has quit IRC08:01
lifelessgsagie: anyhow, sure, but your patch needs a unit test.08:01
*** achanda has quit IRC08:02
*** ZZelle has quit IRC08:03
*** fhubik is now known as fhubik_afk08:04
gsagielifeless: what kind of unit test you had in mind and where do you think its best to put it?08:04
*** zeih has quit IRC08:04
gsagiethis code is a bit new to me08:04
gsagiewell, i will find the right place, nvm08:04
lifelessgsagie: one that exercises the extras() call08:04
lifelessgsagie: there are other tests of that :)08:04
gsagielifeless: ok will look around, thanks08:05
*** fawadkhaliq has quit IRC08:05
openstackgerritAndreas Scheuring proposed openstack-infra/project-config: Add networking-macvtap to Openstack
openstackgerritMerged openstack-infra/project-config: Add training-guides to populate-zanata-project
*** kzaitsev_mb has joined #openstack-infra08:06
*** yfried has quit IRC08:07
*** maishsk has joined #openstack-infra08:07
*** dmellado has left #openstack-infra08:08
openstackgerritGal Sagie proposed openstack/requirements: Add checks for setup.cfg existence in project dictionary
*** yfried_ has joined #openstack-infra08:10
gsagielifeless: ^^08:10
*** jistr has joined #openstack-infra08:12
*** boris-42 has quit IRC08:12
*** zeih has joined #openstack-infra08:14
*** bexelbie has quit IRC08:14
*** kzaitsev_mb has quit IRC08:15
*** ddieterly has joined #openstack-infra08:15
openstackgerritRui Chen proposed openstack-infra/irc-meetings: Add Congress logs and agenda url
*** ociuhandu has joined #openstack-infra08:16
*** e0ne is now known as e0ne_08:16
*** gildub has quit IRC08:18
*** fhubik_afk is now known as fhubik08:19
*** ddieterly has quit IRC08:19
dingyichenzanata-war-3.7.1.war is now in both Sourceforge and github08:19
*** yamamoto has joined #openstack-infra08:20
*** PpPyg has joined #openstack-infra08:21
*** leseb_ has quit IRC08:21
*** markvoelker_ has quit IRC08:21
*** yfried_ is now known as yfried|afk08:21
*** e0ne_ is now known as e0ne08:22
*** fawadkhaliq has joined #openstack-infra08:22
*** yfried|afk is now known as yfried_08:22
*** zeih has quit IRC08:25
*** bexelbie has joined #openstack-infra08:27
*** dizquierdo has joined #openstack-infra08:28
openstackgerritNikki Heald proposed openstack-infra/pypi-mirror: Update build tools and fix link paths
StevenKdingyichen: Thanks!08:29
*** maishsk has quit IRC08:31
*** larainema has quit IRC08:32
*** HeOS has joined #openstack-infra08:32
*** nfedotov has joined #openstack-infra08:32
*** larainema has joined #openstack-infra08:32
*** derekh has joined #openstack-infra08:35
*** markvoelker has joined #openstack-infra08:36
*** maishsk has joined #openstack-infra08:39
*** dingyichen has quit IRC08:39
openstackgerritJoshua Hesketh proposed openstack-infra/os-loganalyze: Add support for setting filters and views per file
openstackgerritJoshua Hesketh proposed openstack-infra/os-loganalyze: Allow the config to set the filter and view
openstackgerritJoshua Hesketh proposed openstack-infra/os-loganalyze: Reapply tidy up generators into contained objects
*** ociuhandu has quit IRC08:45
openstackgerritvikram.choudhary proposed openstack-infra/project-config: Adding networking-onos as a part of neutron project.
*** markvoelker has quit IRC08:45
*** _nadya_ has quit IRC08:47
*** _nadya_ has joined #openstack-infra08:47
*** dalgaaf has quit IRC08:48
*** markvoelker has joined #openstack-infra08:51
*** _nadya_ has quit IRC08:52
*** hashar has joined #openstack-infra08:53
*** ParsectiX has quit IRC08:55
*** bluex-pl has quit IRC08:56
*** zeih has joined #openstack-infra08:56
*** markvoelker has quit IRC08:56
*** hashar has quit IRC09:00
*** markus_z has joined #openstack-infra09:00
*** _nadya_ has joined #openstack-infra09:01
*** derekh has quit IRC09:02
*** kushal has quit IRC09:02
*** zeih has quit IRC09:06
*** markvoelker has joined #openstack-infra09:06
*** _nadya_ has quit IRC09:06
*** hashar has joined #openstack-infra09:06
*** btully has quit IRC09:08
*** dtantsur is now known as dtantsur|brb09:10
*** bexelbie has quit IRC09:10
*** markvoelker has quit IRC09:10
*** dalgaaf has joined #openstack-infra09:12
*** maishsk has quit IRC09:12
*** bluex-pl has joined #openstack-infra09:13
*** maishsk has joined #openstack-infra09:13
*** yamahata has joined #openstack-infra09:15
*** e0ne is now known as e0ne_09:15
*** ddieterly has joined #openstack-infra09:15
*** yfried_ is now known as yfried|afk09:16
*** e0ne_ is now known as e0ne09:16
*** ParsectiX has joined #openstack-infra09:17
*** ZZelle has joined #openstack-infra09:18
*** _nadya_ has joined #openstack-infra09:20
*** yamahata has quit IRC09:20
*** ddieterly has quit IRC09:20
*** markvoelker has joined #openstack-infra09:20
*** kiran-r has quit IRC09:21
*** bexelbie has joined #openstack-infra09:21
*** e0ne has quit IRC09:22
*** aarefiev has quit IRC09:22
*** markvoelker has quit IRC09:25
*** yfried|afk is now known as yfried_09:25
*** kzaitsev_mb has joined #openstack-infra09:27
*** doude has quit IRC09:31
*** ParsectiX has quit IRC09:34
*** dizquierdo has quit IRC09:34
*** maishsk has quit IRC09:34
*** kzaitsev_mb has quit IRC09:35
*** markvoelker has joined #openstack-infra09:35
*** dguitarbite has quit IRC09:36
*** kzaitsev_mb has joined #openstack-infra09:39
*** kzaitsev_mb has left #openstack-infra09:39
*** markvoelker has quit IRC09:40
openstackgerritlifeless proposed openstack/requirements: Move integration test dependencies into bindep
*** ParsectiX has joined #openstack-infra09:47
*** dguitarbite has joined #openstack-infra09:48
openstackgerritKairat Kushaev proposed openstack/requirements: Increase dogpile.cache versio to support NullBackend
*** markvoelker has joined #openstack-infra09:49
*** amotoki has quit IRC09:50
*** dizquierdo has joined #openstack-infra09:50
*** doude has joined #openstack-infra09:53
openstackgerritKairat Kushaev proposed openstack/requirements: Increase dogpile.cache versio to support NullBackend
*** markvoelker has quit IRC09:54
openstackgerritCyril Roelandt proposed openstack/requirements: Bump mock to at least 1.1.2
*** vponomaryov1 has joined #openstack-infra09:56
*** sergmelikyan has quit IRC09:57
*** kushal has joined #openstack-infra09:59
*** pksingh has quit IRC10:00
*** Kennan has joined #openstack-infra10:03
*** Kennan2 has quit IRC10:03
*** dims_ has joined #openstack-infra10:03
*** ParsectiX has quit IRC10:04
*** markvoelker has joined #openstack-infra10:04
*** pal has quit IRC10:05
*** dims_ has quit IRC10:05
*** sergmelikyan has joined #openstack-infra10:06
anteayaAJaeger: done10:07
*** markvoelker has quit IRC10:08
*** yfried_ is now known as yfried|afk10:09
*** vlaza has quit IRC10:11
*** gildub has joined #openstack-infra10:12
*** e0ne has joined #openstack-infra10:12
sdaguejhesketh: is an os-loganalyze change unrelated to yours, but straight forward10:13
*** zns has joined #openstack-infra10:15
*** ddieterly has joined #openstack-infra10:16
*** yamamoto has quit IRC10:18
*** markvoelker has joined #openstack-infra10:19
*** ddieterly has quit IRC10:21
*** mrmartin has joined #openstack-infra10:21
openstackgerritMerged openstack-infra/pypi-mirror: Update build tools and fix link paths
*** markvoelker has quit IRC10:23
openstackgerritValeriy Ponomaryov proposed openstack-infra/project-config: Make rally jobs for Manila non-votable
jyuso1anteaya: hi.10:24
openstackgerritValeriy Ponomaryov proposed openstack-infra/project-config: Make Rally jobs for Manila non-votable
*** fhubik is now known as fhubik_afk10:25
*** maishsk has joined #openstack-infra10:25
*** pc_m has joined #openstack-infra10:28
*** vlaza has joined #openstack-infra10:28
*** yfried|afk is now known as yfried_10:29
jheskethsdague: lgtm10:31
*** e0ne is now known as e0ne_10:32
*** markvoelker has joined #openstack-infra10:33
tchaypodhellmann: I tried %(default)s but it gives me None10:33
*** cody-somerville has quit IRC10:34
*** sputnik1_ has quit IRC10:34
*** ParsectiX has joined #openstack-infra10:34
AJaegerthanks, anteaya !10:35
*** sergmelikyan has quit IRC10:36
openstackgerritMerged openstack-infra/irc-meetings: Add Congress logs and agenda url
openstackgerritMerged openstack-infra/irc-meetings: Change Congress meeting time
openstackgerritMerged openstack-infra/irc-meetings: Remove the product team meeting
openstackgerritMerged openstack-infra/irc-meetings: updating cue team meeting date
openstackgerritMerged openstack-infra/os-loganalyze: Add Manila logs to filter regex
*** markvoelker has quit IRC10:38
*** mdenny has quit IRC10:39
*** e0ne_ is now known as e0ne10:39
*** kmartin has quit IRC10:40
*** jamielennox is now known as jamielennox|away10:41
*** coolsvap is now known as coolsvap|away10:41
*** e0ne has quit IRC10:46
*** markvoelker has joined #openstack-infra10:48
*** _nadya__ has joined #openstack-infra10:48
*** vikram has joined #openstack-infra10:49
*** _nadya_ has quit IRC10:50
*** e0ne has joined #openstack-infra10:50
vikramAJaeger: Ping10:51
AJaegerHI Vikram, what's up?10:51
vikramAJaeger: Hi10:51
vikramwhat will be next step for getting merged10:51
vikrami rebased the patch today for conflict10:51
AJaegerreview by the team ;)10:52
AJaegerLet me do my part of that now and then wait for others to review, this can take a few days...10:52
vikramcan you please add mandatory members from infra10:52
*** cdent has joined #openstack-infra10:52
AJaegervikram: the members go over it eventually...10:53
AJaegervikram: I just gave a -1, could you fix quickly so that I can give a +2?10:54
vikramAJaeger: okay10:54
*** markvoelker has quit IRC10:54
*** pcm_ has joined #openstack-infra10:54
*** zeih has joined #openstack-infra10:55
*** yfried_ is now known as yfried|afk10:56
*** pc_m has quit IRC10:57
vikramAJaeger: I need to remove "-    gate:10:57
vikram-      - gate-tempest-dsvm-networking-onos"10:57
AJaegervikram: correct, remove those two lines10:58
openstackgerritvikram.choudhary proposed openstack-infra/project-config: Adding networking-onos as a part of neutron project.
vikramAJaeger: Done :-)10:59
*** kutija has joined #openstack-infra10:59
*** hashar has quit IRC11:00
*** e0ne has quit IRC11:00
*** kutija_ has quit IRC11:01
AJaegervikram: reviewed with +211:01
vikramAJaeger: Thanks11:01
*** maishsk has quit IRC11:02
*** markvoelker has joined #openstack-infra11:02
*** bexelbie has quit IRC11:03
*** maishsk has joined #openstack-infra11:03
*** e0ne has joined #openstack-infra11:04
*** afazekas has quit IRC11:05
*** zeih has quit IRC11:06
*** sergmelikyan has joined #openstack-infra11:07
*** markvoelker has quit IRC11:07
*** bexelbie has joined #openstack-infra11:08
*** asselin has quit IRC11:09
*** ociuhandu has joined #openstack-infra11:10
openstackgerritKairat Kushaev proposed openstack/requirements: Bump oslo.cache version to 0.3.0
*** e0ne has quit IRC11:13
*** gildub has quit IRC11:13
*** pelix has joined #openstack-infra11:14
*** maishsk has quit IRC11:14
*** yfried|afk is now known as yfried_11:16
*** maishsk has joined #openstack-infra11:16
*** markvoelker has joined #openstack-infra11:17
*** kutija_ has joined #openstack-infra11:17
*** ddieterly has joined #openstack-infra11:17
*** baoli has joined #openstack-infra11:17
*** e0ne has joined #openstack-infra11:18
*** kutija__ has joined #openstack-infra11:19
*** kutija has quit IRC11:19
*** markvoelker has quit IRC11:21
*** ddieterly has quit IRC11:21
*** kutija has joined #openstack-infra11:22
*** mpaolino has joined #openstack-infra11:22
*** kutija_ has quit IRC11:22
*** fawadkhaliq has quit IRC11:23
*** kutija__ has quit IRC11:24
*** markvoelker has joined #openstack-infra11:24
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
*** openstack has joined #openstack-infra11:38
openstackgerritFatih Degirmenci proposed openstack-infra/jenkins-job-builder: Add support for forbidden file paths in gerrit trigger
*** yfried_ is now known as yfried|afk11:39
*** markvoelker has joined #openstack-infra11:39
*** e0ne_ is now known as e0ne11:39
AJaegersdague, jhesketh: If either of you has some time for review: and are two new projects, and gets rid of "change-" in job names. Any review love would be welcome (each has one +2 already)11:40
sdaguejhesketh: cool, ok, so hopefully someone else can take a look and merge the rest of that stack. I'm +2 on it.11:40
SotKHi folks, I was wondering if there is an easy way for me to test a potential change to openstack-infra/project-config, or if I just send a change for review and some CI tests it?11:41
*** erlon has joined #openstack-infra11:41
*** ociuhandu_ has joined #openstack-infra11:41
*** ociuhandu has quit IRC11:41
*** ociuhandu_ is now known as ociuhandu11:41
AJaegerSotK: most tests can be run locally via tox, e.g. tox -e zuul.11:41
sdagueSotK: run tox11:41
*** yfried|afk is now known as yfried_11:41
SotKAJaeger, sdague: OK then, thanks!11:42
*** andrey-mp has joined #openstack-infra11:43
*** markvoelker has quit IRC11:44
*** fawadkhaliq has quit IRC11:45
andrey-mpHi all. Is it possible to make periodic job in zuul for one of stackforge project?11:46
*** dims_ has joined #openstack-infra11:47
*** zeih has joined #openstack-infra11:48
*** vikram has quit IRC11:49
openstackgerritJulia Kreger proposed openstack-infra/system-config: Use bifrost for bare metal portion of infra-cloud
jheskethsdague: thanks :-)11:51
jheskethAJaeger: will shortly11:52
*** markvoelker has joined #openstack-infra11:54
*** Alexei_987 has quit IRC11:54
*** Alexei_987 has joined #openstack-infra11:56
AJaegerthanks, jhesketh !11:57
AJaegerandrey-mp: what do you want to do?11:57
*** maishsk has quit IRC11:57
*** yamamoto has joined #openstack-infra11:58
*** markvoelker has quit IRC11:58
*** puranamr has joined #openstack-infra11:58
andrey-mpAJaeger: I want to run some job with long tests one a week(for example)11:59
andrey-mpagainst project stackforge/ec2-api11:59
andrey-mpbecause some tests are very long to run in check pipeline11:59
*** dguitarbite has quit IRC12:00
andrey-mpand as I understood, periodic pipelines are not suitable for such activity as I want12:00
AJaegerandrey-mp: you can add these jobs for now to the experimental pipeline, so you can trigger them manually and figure out that they work.12:00
openstackgerritMonty Taylor proposed openstack-infra/shade: Account for Error 396 on Rackspace
andrey-mpyeah, I did it two months ago but we forget to run it once a week )12:01
mordredok. that's my best commit message to date12:01
AJaegerI'm not sure whether periodic is really the right approach, you better discuss with others in the team the proper way for this.12:01
mordredShrews: ^^ I have NO IDEA how to test that12:01
mordredSpamapS: ^^12:01
AJaegermordred: In that case: Please stop sending patches for today ;)12:02
andrey-mpAJaeger: thank you :)12:02
mordredAJaeger: :)12:02
*** maishsk has joined #openstack-infra12:02
*** radez_g0n3 is now known as radez12:02
*** yamamoto has quit IRC12:02
*** Ala has quit IRC12:05
greghayneswell done12:06
*** Ala has joined #openstack-infra12:06
sdaguemordred: zomg12:06
*** gordc has joined #openstack-infra12:08
mordredsdague: right?12:08
*** markvoelker has joined #openstack-infra12:08
mordredgreghaynes: I _think_ we have some test code somewhere which already tests a retry in one of the loops, but I haven't foudn it yet12:09
*** deepakcs has quit IRC12:10
*** puranamr has quit IRC12:11
greghayneshrm, unsure in shade12:11
*** markvoelker has quit IRC12:13
mordredgreghaynes: maybe SpamapS will have a good idea - he's good at devious testing things12:13
mordredsdague: have I mentioned how much I love that tasks never go away?12:14
sdaguemordred: oh neat12:14
mordredsdague: it's such a great feature12:14
mordredsdague: there is also no task-delete function - so if you accidentally up-arrow a task create command line action ... you will get 2 tasks and you will like it12:15
sdaguewell, on the upside, the multinode jobs seem to be doing a thing again. So now the grenade multinode errors are all of my own making.12:16
sdagueyou know what I'd kind of like, if you put 'check experimental' in a commit message, and it would automatically kick off experimental jobs on patch update12:16
openstackgerritAndrey Pavlov proposed openstack-infra/project-config: Add job for ec2-api project that run all tests.
mordredyeah - I see that12:16
*** bswartz has quit IRC12:17
mordredgreghaynes: so - I guess what we need to do now is add another multi-node experimental to the current multi-node experimental12:17
mordredgreghaynes: that runs on the ubuntu-trusy nodes12:17
mordredgreghaynes: so that we can verify we have all the glean/minimal issues sorted before migrating again12:18
sdaguethe multinode smoke job has a good pass rate12:18
*** ddieterly has joined #openstack-infra12:18
sdagueso that's the one you should use for stuff like that12:18
ihrachyshkasdague, can you +A?
sdagueihrachyshka: look...12:19
sdagueihrachyshka: done12:19
sdaguemordred: hey, so my daily bugging about the wsgi_scripts pbr patch :)12:20
mordredsdague: looks great to me12:22
Shrewsmordred: umm, wow. But next time, try to have a protagonist in the commit message to counteract the antagonist. It reads better.12:22
*** markvoelker has joined #openstack-infra12:23
openstackgerritMerged openstack-infra/tripleo-ci: Drop Fedora python-ipaddress package cherry pick
*** ddieterly has quit IRC12:23
mordredShrews: that's good feedback12:23
mordredShrews: isn't the implied protagonist "your poor soul" though? and it just loses mightily?12:24
Shrewsmordred: oh, i may have to re-read. this is as complex as Dante's Inferno12:25
* Shrews re-reads for subtext12:26
AJaegermordred: isn't it Cthulhu? Let me send you an extra "h"12:26
*** fbo- has quit IRC12:27
sdaguereview: -1 for misspelling the dark lord name12:27
mordredAJaeger: oh, I think you're right12:27
*** markvoelker has quit IRC12:28
openstackgerritMonty Taylor proposed openstack-infra/shade: Account for Error 396 on Rackspace
openstackgerritMonty Taylor proposed openstack-infra/shade: Account for Error 396 on Rackspace
mordredspelling errors in commit message fixed12:28
openstackgerritMerged openstack-infra/project-config: disable python34 job for lbaas stable branches
*** fbo has joined #openstack-infra12:30
*** woodster_ has joined #openstack-infra12:30
*** ppai has quit IRC12:33
openstackgerritMonty Taylor proposed openstack-infra/system-config: Trim the elements list from nodepool.yaml
*** wznoinsk has quit IRC12:33
sdagueok, so my current failure is because jenkins owns /opt/stack/new/devstack during runs, but not /opt/stack/new/grenade12:33
sdagueshould I just make jenkins own /opt/stack/new/grenade as well?12:34
openstackgerritMonty Taylor proposed openstack-infra/project-config: Express nodepool base's dependency on puppet
*** dprince has joined #openstack-infra12:34
mordredsdague: seems consistent?12:34
gsagieHello all, if anyone from requierments can look   that would be great, its a fix so new projects can commit their initial commit (without setup.cfg)12:35
mordredgsagie: I'm confused - your initial commit _has_ a setup.cfg12:36
sdaguemordred: though I'm trying to figure out where that happens12:36
gsagiemordred : yes, but jenkins failed on it12:36
gsagiei could get it to pass the tests until i put it dependent on that fix patch12:37
gsagiewith Depends-On12:37
*** dtantsur|brb is now known as dtantsur12:37
mordredlet me read slightly furter - the patch seems sane enough12:37
*** markvoelker has joined #openstack-infra12:37
gsagiemordred: the log =>
openstackgerritliusheng proposed openstack/requirements: Bump gabbi to version 1.1.0
AJaegermordred: it'S the first commit after repo creation - and the requirements scripts compare setup.cfg from old HEAD with current version12:39
*** aysyd has joined #openstack-infra12:39
mordredAJaeger: AH yes12:39
mordredso old head is empty12:40
*** dims_ has quit IRC12:41
*** bnemec has joined #openstack-infra12:41
*** markvoelker has quit IRC12:42
mordredgsagie: ok. +2 from me12:42
gsagiemordred: thanks12:43
mordredAJaeger, sdague: if you have a sec -
mordredSergeyLukjanov, jhesketh, fungi, clarkb: (if any of you happen to be around)12:43
*** Alexei_987 has quit IRC12:43
*** andrey-mp has quit IRC12:44
openstackgerritAndrey Pavlov proposed openstack-infra/project-config: Add job for ec2-api project that run all tests.
*** afazekas has joined #openstack-infra12:44
*** Alexei_987 has joined #openstack-infra12:46
AJaegermordred: +212:46
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
*** ppai has joined #openstack-infra12:46
sdaguemordred: base patch seems good12:47
openstackgerritFatih Degirmenci proposed openstack-infra/jenkins-job-builder: Add support for forbidden file paths in gerrit trigger
sdagueso... it turns out ordering of operations here in the grenade path is important, and I was doing things in an order that was fail-o-potamus. Will be interesting to see if this new patch gets further.12:48
*** signed8bit has joined #openstack-infra12:48
electricalzaro: finally got some time again to work on JJB.
electricalreviewed your comment and the line break is required :-)12:50
*** pal has joined #openstack-infra12:50
*** tlian has joined #openstack-infra12:51
*** markvoelker has joined #openstack-infra12:52
jheskethmordred: lgtm12:53
openstackgerritMonty Taylor proposed openstack-infra/project-config: Add experimental multi-node job to use dib images
mordredjhesketh: thanks!12:53
mordredsdague: hahaha12:53
*** zns has quit IRC12:53
*** gordc has quit IRC12:54
mordredsdague, greghaynes: ^^ added experimental multi-node job on the new images12:54
*** gordc has joined #openstack-infra12:54
electricalmordred: AJaeger hiya guys. been a while12:54
AJaegerhi electrical !12:54
mordredhi electrical !12:54
*** hdd has joined #openstack-infra12:54
electricalbeen a bit to busy lately to spend time on JJB. maintaining puppet modules is a lot of work :-)12:55
mordredelectrical: bah. we know nothing about about the work required for maintaining puppet modules around here12:55
*** eharney has joined #openstack-infra12:55
*** bhunter71 has joined #openstack-infra12:56
dhellmannfungi: apparently I forgot to ask to have the releases-core team populated in gerrit. Can you add me, so I can add the rest of the folks, please?,members12:56
*** markvoelker has quit IRC12:56
mordreddhellmann: done12:57
*** zeih has quit IRC12:58
*** timrc_ has joined #openstack-infra12:58
*** bswartz has joined #openstack-infra12:58
*** zns has joined #openstack-infra12:58
*** yamamoto has joined #openstack-infra12:59
openstackgerritOleksii Zamiatin proposed openstack-infra/project-config: Make zeromq functional gate check non-voting (should pass in oslo.messaging/feature/zmq)
*** jaypipes has joined #openstack-infra12:59
*** zeih has joined #openstack-infra12:59
*** julim has joined #openstack-infra13:00
*** liam_ has joined #openstack-infra13:01
*** markvoelker has joined #openstack-infra13:01
*** yamamoto has quit IRC13:03
*** rlandy has joined #openstack-infra13:04
*** radez is now known as radez_g0n313:04
*** sergmelikyan has joined #openstack-infra13:05
*** dims_ has joined #openstack-infra13:06
openstackgerritRichard Pijnenburg proposed openstack-infra/jenkins-job-builder: Adding support for Hipchat plugin 0.1.9 as publisher
*** markvoelker has quit IRC13:09
*** markvoelker has joined #openstack-infra13:09
openstackgerritMerged openstack-infra/project-config: Adding networking-onos as a part of neutron project.
*** signed8bit is now known as signed8bit_ZZZzz13:10
*** markvoelker_ has joined #openstack-infra13:10
*** kiran-r has joined #openstack-infra13:10
*** signed8bit_ZZZzz has quit IRC13:10
openstackgerritMerged openstack-infra/project-config: Express nodepool base's dependency on puppet
openstackgerritMerged openstack-infra/project-config: Add missing projects to specs list
openstackgerritMerged openstack-infra/project-config: Always use gate- instead of check (devstack-gate)
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
sdaguemordred: what's up with experimental-tempest-dsvm-full-dibtest13:13
openstackgerritVictor Stinner proposed openstack-infra/project-config: Make gate-heat-python34 check voting and add gate
sdaguethat seems pretty face plant, dies really early13:13
*** markvoelker has quit IRC13:14
*** akshai has joined #openstack-infra13:15
*** krtaylor has joined #openstack-infra13:16
*** amitgandhinz has joined #openstack-infra13:16
mordredsdague: yeah - that's sad panda13:17
mordreddunno - that's a new one13:17
sdagueit looks like basically the issues I was running into, the minimal images are so minimal they don't work13:17
*** kiran-r has quit IRC13:18
mordredwell - so we HAVE successfully run devstack jobs on them before13:18
mordredso I'm keen to find out what changed and what we missed expressing somewhere13:19
mordredhowever ... while we're marvelling at that ...13:19
mordredthis is what I'm getting from the rackspace errors in nodepool13:20
mordredbasically, we're getting nodes with empty addresses dicts13:20
*** baoli has quit IRC13:20
mordredjhesketh: ^^ in addition to just finding it interesting, it might also be interesting to your employer13:20
*** ppai has quit IRC13:21
*** peristeri has joined #openstack-infra13:22
mordredI've verified that I do still get addresses dicts on existing nodes I have in rax13:22
*** hdd has quit IRC13:24
mordredsdague: is there any chance in nova that a server could go into ACTIVE state and then some amount of time later get its networking information and fill in the addresses fields?13:25
*** dustins has joined #openstack-infra13:27
*** peristeri has quit IRC13:27
*** haypo has joined #openstack-infra13:27
*** peristeri has joined #openstack-infra13:28
haypohi. a colleague proposed : to get a bugfix, should we modify upper-constraints.txt & global-requirements.txt? or only upper-constraints.txt?13:28
*** esmute has joined #openstack-infra13:29
*** shardy_ has joined #openstack-infra13:29
*** Steap has joined #openstack-infra13:29
*** derekh has joined #openstack-infra13:30
*** dkranz has joined #openstack-infra13:30
*** shardy has quit IRC13:31
*** maishsk has quit IRC13:31
*** samuel-dmq has quit IRC13:31
*** samueldmq has joined #openstack-infra13:32
*** maishsk has joined #openstack-infra13:33
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for python-qpid-proton build bug 1474852
*** mriedem_away is now known as mriedem13:34
openstackbug 1474852 in oslo.messaging "gate-oslo.messaging-python34 fails in the gate with "Failed building wheel for python-qpid-proton"" [Undecided,Confirmed]
*** dguitarbite has joined #openstack-infra13:34
*** dguitarbite is now known as Guest428613:34
dhellmannmordred: thanks!13:34
*** Guest4286 has quit IRC13:35
*** dbite has joined #openstack-infra13:35
*** shardy_ has quit IRC13:35
fungimordred: care to weigh in on so mrmartin can work on getting a test copy of our ask database imported and validate the new version? (he's trying to get google auth working for people again and that means an upgrade, so staging first)13:35
*** dbite has quit IRC13:35
sdaguemordred: yes13:35
*** shardy has joined #openstack-infra13:35
sdagueat least I think so, I know there is no guaruntee that by ACTIVE there is working networking, and I don't think that there is a guaruntee there is network info13:36
pabelangernibalizer, ack13:37
fungilifeless: on constraints generator triggering, i agree that we have basically the same touchpoints that we had with the old pypi mirror update jobs (on requirements change merged, on pypi release upload)13:37
sdaguefungi / mordred can we get this -
mordredsdague: AWESOME13:39
sdaguebecause our horrible stack traces are now mostly hidden in logstash13:39
sdagueand that will fix it13:40
*** hdd has joined #openstack-infra13:40
openstackgerritJens Rosenboom proposed openstack-infra/puppet-iptables: Use service-name netfilter-persistent for Vivid
mordredsdague: I don't even want to ask how that active state might relate to network info in config-drive13:40
openstackgerritPaul Belanger proposed openstack-infra/system-config: Remove color output from puppet list command
*** vikram has joined #openstack-infra13:41
*** signed8bit has joined #openstack-infra13:41
pabelangerpleia2, ^ nice catch with above, did think of checking that.13:41
vikramAJaeger: Ping13:41
openstackgerritMonty Taylor proposed openstack-infra/nodepool: Wait for servers to have addresses
pabelangereither way, real issue was color output that mucked up jenkins logs13:41
*** BharatK_ has quit IRC13:41
mordredfungi: ^^ (also, sdague check me on that - also, Shrews we should be aware of this in shade and probably port it in)13:41
vikramAJaeger: I can find the changes for "" is merged..13:41
openstackgerritMerged openstack-infra/elastic-recheck: Add query for python-qpid-proton build bug 1474852
openstackbug 1474852 in oslo.messaging "gate-oslo.messaging-python34 fails in the gate with "Failed building wheel for python-qpid-proton"" [Undecided,Confirmed]
greghaynesmordred: 2015-07-14 19:40:53.835 | /home/jenkins/workspace/experimental-tempest-dsvm-full-dibtest/devstack-gate/ line 82: ping: command not found13:42
greghaynesmordred: lol13:42
openstackgerritAndrey Pavlov proposed openstack-infra/project-config: Add job for ec2-api project that run all tests.
mordredgreghaynes: yeah - working up a patch right now13:42
*** gyee has joined #openstack-infra13:43
Shrewsmordred: ah, yes. I just ran into this yesterday writing a playbook for a devstack server using our new modules. After creating a server, there needs to be a delay in the playbook waiting for the addresses to be filled before I can do more actions on the new server.13:43
*** signed8b_ has joined #openstack-infra13:43
*** btully has joined #openstack-infra13:44
*** bknudson has joined #openstack-infra13:44
vikramAJaeger: My intention was not to trouble you.. I just want to know what will be the next step towards the new repo creation. Am I suppose to do anything now?13:44
vikramAJaeger: I apologies if I did13:44
AJaegervikram: you need to ask for an infra root to setup the groups - as explained in the infra manual13:44
AJaegervikram: naked pings are not really helpfull - that's all I wanted to say...13:45
sdaguemordred: ... so I'm looking at the code, actually the network info seems like it should be there before the active transition13:45
sdagueit doesn't mean the network works yet though13:45
*** signed8bit has quit IRC13:45
vikramAJaeger:  I understand.. will take care.. :-)13:45
openstackgerritMonty Taylor proposed openstack-infra/system-config: Add ping to our machines
gsagievikram: i have recently gone throught the same process, notice that needs to merge before you will be able to add your first commit13:46
mordredsdague: networking not working is fine- we already do a second poll wait to wait for the network to come up13:46
sdagueyeh, so I'm pretty sure, at least with master, network info is there already13:46
mordredsdague: so it's possible something is just going horribly wrong occasionally at rax and we should not necessarily wait for it to come later13:46
sdagueby ACTIVE13:46
*** prad has joined #openstack-infra13:47
mordredjohnthetubaguy: ^^ fyi - we're seeing this at rax from time to time that we're booting nodes and they're hitting ACTIVE with an empty addresses dict13:47
sdaguealthough... if there is a call out to neutron... we're not blocking on that13:47
*** sergmelikyan has quit IRC13:47
vikramgsagie: thanks for the update13:47
sdagueso... I don't know13:47
sdaguethat interaction pattern is something that's not fully in my head13:48
mordredsdague: you know, specifically at rax where there is no dhcp and one needs to depend on the network info in config-drive - it seems like if  it hits ACTIVE without networking, the config drive is also going to be blank, which means you can predict successfully that the node will not work13:48
mordredand go ahead and treat it as a launch error rather than as a thing to wait for13:49
*** jistr is now known as jistr|mtg13:49
mordredjroll: ^^ seems like the type of thing you might also want to know things about maybe13:49
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Do not run ironic-inspector-dsvm job on stable/1.1 branch
sdaguehey, all my multinode jobs went NOT_REGISTERED13:49
greghaynesmordred: curious what the motivation for was?13:49
mordredsdague: yay!13:49
greghayneswe kind of actively avoided that sort of thing in tripleo13:49
mordredgreghaynes: because the one element literally would not work without the other13:50
sdaguemordred: any idea what happened there?13:50
greghaynesmordred: the node-devstack one?13:50
mordredsdague: yikes, no13:50
pcm_anteaya: SergeyLukjanov: Can you folks re-review (experimental job for neutron), so I can start testing it?13:50
AJaegersdague: what exactly is the problem?13:50
mordredgreghaynes: yah. there are no cases in which node-devstack will work without puppet13:51
greghaynesmordred: ah, ok13:51
jrollmordred: is that actually a thing that happens?13:51
mordredgreghaynes: also - I was starting to work on splitting the puppet element up so that long-lived servers could use an "install puppet" element without also getting a "run puppet" action13:51
AJaegersdague: Seeing those now - strange.13:51
mordredjroll: yes. we are seeing it in our logs reasonably frequently right now13:51
sdaguealso, - file upload issue13:51
AJaegersdague: I guess a fallout of my recent change13:51
sdagueAJaeger: which change?13:51
openstackgerritMerged openstack-infra/system-config: Handle 'ERROR' for multiline oslofmt
jrollmordred: argh. how new is this? all regions / one in particular? is it still happening?13:52
mordredjroll: 81 times in the current active log which was last rotated about 4 or 5 hours ago13:52
sdagueAJaeger: yeh, so that's kind of ruining my day. Do you mind a fast revert on that one?13:52
*** bexelbie has quit IRC13:53
*** ddieterly has joined #openstack-infra13:53
jrollmordred: instance and/or request IDs welcome :)13:53
AJaegersdague: I would first like to understand why it fails - and then I'm fine with a quick revert13:53
mordredjroll: started yesterday13:53
* AJaeger prepares the revert...13:53
sdagueAJaeger: honestly, it's probably a naming clash13:53
AJaegersdague: I hate those ;(13:54
mordredjroll: 56be97b8-0ca9-4bd9-b672-46eee022b64e f55c7cea-ed24-4df4-b92f-4919c07d8600 7a988e79-cd6a-450a-b3f5-63ccca7a9929 15d33115-a81e-49ae-b0e7-c752e1aaae9c e82c6221-54ea-44fe-9aa8-a29e9a23506b13:54
openstackgerritAndreas Jaeger proposed openstack-infra/project-config: Revert "Always use gate- instead of check (devstack-gate)"
jrollmordred: thanks13:54
AJaegersdague: I'll self-approve and split up the change in smaller chunks if possible13:54
mordredjroll: the first 4 are all in iad - the last one is DFW (all of those were just the first five in my debug log)13:54
openstackgerritMerged openstack-infra/system-config: Trim the elements list from nodepool.yaml
mordredjroll: and those are instance ids13:55
sdagueAJaeger: yeh, I just fast approved13:55
AJaegersdague: you're faster than my +2 ;)13:55
vikramPlease help to add "" to the Gerrit Group Members for "" git repo approved via ""13:55
jrollmordred: ok, ty13:55
AJaegersdague: If you later find time to look into the patch and figure out why it's broken, it would be great - I'll split a few bits out now...13:56
*** zeih has quit IRC13:56
*** radez_g0n3 is now known as radez13:56
fungivikram: you have two gerrit accounts with that e-mail address13:58
fungivikram: give me a moment to see if i can figure out the situation with that and clean it up for you13:58
*** tvanderwiel has joined #openstack-infra13:58
vikramfungi: thanks13:58
*** fhubik_afk is now known as fhubik13:58
*** alivigni has joined #openstack-infra13:58
AJaegerfungi, do you have any idea why let to all the NOT_REGISTERED jobs?13:59
AJaegerfungi, we're currently reverting but I'd like to know what I did wrong ;)13:59
jrollmordred: one more request, an instance ID of a good one?14:00
mordredjroll: weirdly harder ... one sec14:00
fungiAJaeger: yes, and reverting may not help a lot14:00
fungiAJaeger: massive job name changes mean that all new jobs are created in the jenkins masters14:01
mordredjroll: 350f2fcb-23d7-4280-a86a-ef916460092314:01
mordredjroll: also in IAD14:01
fungiAJaeger: in this case my guess is around 3000 new jobs created by that patch14:01
jrollthanks mordred14:01
fungiAJaeger: it takes jenkins-job-builder a while to update those14:01
AJaegerfungi, ah!14:01
fungiAJaeger: often longer than the puppet exec timeout which runs it14:01
fungiAJaeger: and almost certainly longer than it takes puppet to update the zuul configuration with the same new names14:02
AJaegersdague, fungi: In that case, we don't need the revert - and just wait for puppet to finish - correct.14:02
AJaegerI meant: Correct?14:02
* AJaeger just WIP the revert while we talk here...14:02
fungiAJaeger: well, i will probably need to log into each jenkins master and check that jjb is still in progress14:02
*** otter768 has joined #openstack-infra14:03
fungiAJaeger: and i should also spot-check that the new job names are registering in zuul14:03
AJaegersorry, fungi!14:03
AJaegerHow do you suggest to move forward?14:03
*** dboik has joined #openstack-infra14:03
openstackgerritThomas Morin proposed openstack-infra/project-config: Add check-requirements job to 2 projects
fungiAJaeger: looking at jenkins01, `jenkins-jobs update --delete-old /etc/jenkins_jobs/config` has been running since 13:45 utc14:04
AJaegerso, for 20 mins ;(14:04
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Add flavor admin support
*** kutija_ has joined #openstack-infra14:04
mordredShrews: also - update to the above thing about polling for addresses14:04
mordredShrews: we probably do not want to port in that patch - but we probably _do_ want to check that addresses has content when it hits active and else return error14:05
mordredShrews: oh - wait - I just saw your response - on devstack you got addresses later after the thing went into ACTIVE?14:06
*** spzala has joined #openstack-infra14:06
*** annegentle has joined #openstack-infra14:06
mordredgreghaynes, Shrews: I think we should try to figure out if we can reproduce (even intermittently) so we can see what winds up in config drive on a server that hits active without addresses14:07
Shrewsmordred: i'm not sure what shade gets back, but it took a while before 'nova list' showed the requested floating IP as attached. I haven't dug into it yet, but my theory was that was the reason SSH timed out14:07
sdaguefungi: oh, so we need to block that patch?14:07
*** kutija has quit IRC14:07
mordredShrews: oh - that's a whole other thing14:07
AJaegersdague, fungi: Looking at newer jobs, they don't have NOT_REGISTERED anymore.14:07
*** otter768 has quit IRC14:07
AJaegersdague: Yeah, I think we can abandon it.14:08
Shrewsmordred: then ignore me14:08
fungiAJaeger: sdague: yeah, i think it's slowly fixing itself, but not sure how long to expect14:08
mordredShrews: yeah - floating ips can take a whiel to attach, and you definitely need to wait for ssh to go active for a while after server active14:08
sdaguefungi: ok14:08
*** ParsectiX has quit IRC14:08
sdagueAJaeger: you want to kill it?14:08
fungiAJaeger: sdague: we did merge some recent fixes for jjb which should make it a _lot_ faster than it used to be14:08
fungiso maybe rsn14:08
*** peristeri has quit IRC14:08
mordredShrews: wait_for: port=22 (*syntax hand wave*) is a great thing to put in after os_server14:08
AJaegersdague: yes, will kill it14:08
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for broken ironic unit tests due to mock 1.1
openstackgerritMatthias Runge proposed openstack-infra/project-config: Add a gate job for Django-1.8
electricalfungi: I'm hitting more and more things that break my wishes with JJB :p14:10
AJaegerfungi, thanks a lot for your hand-holding ;)14:10
*** tonytan4ever has joined #openstack-infra14:10
*** amitgandhinz has quit IRC14:10
*** zz_dimtruck is now known as dimtruck14:11
*** sigmavirus24_awa is now known as sigmavirus2414:12
anteayajyuso1: sorry I missed you, I was up early to catch a flight14:12
AJaegerfungi: Besides changing a lot of jobs, we now have 170 jobs less ;)14:12
*** jsavak has joined #openstack-infra14:13
sdagueit's a bummer we can't do a two phase commit there14:14
anteayapcm_: you have a comment from AJaeger and also the patch is in merge conflict can you take a look?
openstackgerritMerged openstack-infra/project-config: Fix Cinder configuration for tempest-dsvm-full-lio
pcm_anteaya: WIll check the merge conflict - missed that. AJaeger commented to check the naming, but it is already correct.14:15
*** shardy_ has joined #openstack-infra14:16
anteayapcm_: k14:16
*** amitgandhinz has joined #openstack-infra14:16
AJaegerpcm_: It is correct now, yes. You said that *check* is used and I corrected you on that14:16
AJaegerpcm_: No action needed from my comment, just rebase and you're good ;)14:17
*** shardy has quit IRC14:17
pcm_AJaeger: Yeah, apparently it just changed, as previously check- was used in neutron-vpnaas (separate from this commit which is in neutron project).14:18
AJaegerpcm_: yes, we changed it14:18
pcm_AJaeger: I'll quickly fix conflict and repost.14:18
*** yamahata has joined #openstack-infra14:18
AJaegerpcm_: and I'll quickly +2 afterwards if my meeting here allows ;)14:18
pcm_AJaeger: Thanks!14:19
*** mtanino has joined #openstack-infra14:20
openstackgerritMerged openstack-infra/elastic-recheck: Add query for broken ironic unit tests due to mock 1.1
sdaguefungi: so, is there a current status on job definition builds?14:20
sdagueso I can figure out if trigging the devstack-gate experimental bits is useful or not14:20
openstackgerritNikki Heald proposed openstack-infra/puppet-bandersnatch: Sync with gozer's puppet-bandersnatch
AJaegersdague: looking at zuul, I think we're fine again14:20
fungisdague: still running, but now puppet has started a second one because we don't have any locking. i'll kill the interlopers so that they don't tread on the earlier run14:21
openstackgerritPaul Michali proposed openstack-infra/project-config: Run neutron-vpnaas functional tests for Neutron
*** haypo has left #openstack-infra14:21
*** shardy_ has quit IRC14:21
AJaegerpcm_: wrong merge conflict resolution, I commented ^14:22
openstackgerritSzymon Wróblewski proposed openstack-infra/project-config: [WIP] Add non-voting experimental dsvm job for nova
fungisdague: we can actually do a two-phase commit if we want. it's just a matter of making sure we don't delete any existing jobs in the first change and only create additional ones14:22
*** shardy has joined #openstack-infra14:22
pcm_AJaeger: yeah, I was just fixing.14:22
fungihowever that means it's not a straight rename14:22
fungiit's a create and then later a delete14:22
openstackgerritPaul Michali proposed openstack-infra/project-config: Run neutron-vpnaas functional tests for Neutron
fungisdague: if we'd thought ahead, that probably would have been a better idea, however more complex14:22
*** peristeri has joined #openstack-infra14:23
*** sdake has joined #openstack-infra14:23
openstackgerritMerged openstack-dev/pbr: Add wsgi_scripts support to PBR
pcm_AJaeger: Thanks for reviewing.14:23
* AJaeger doesn't plan to do any massive job renaming in the near future ;)14:24
* AJaeger already tried to split this changeset in smaller ones but this final one was obviously too large - thanks to all the templates ;(14:24
fungisdague: AJaeger: it looks like it's probably done everywhere except jenkins01. i expect that one may have thread leak issues and so is making the api calls take longer14:24
fungisdague: were there any specific job names you were interested in? i can confirm manually whether they're registered14:25
openstackgerritNikki Heald proposed openstack-infra/puppet-bandersnatch: Sync with gozer's puppet-bandersnatch
openstackgerritKairat Kushaev proposed openstack/requirements: Bump oslo.cache version to 0.3.0
sdaguefungi: yes - gate-grenade-dsvm-multinode14:27
pcm_anteaya: Commit is all fixed up (rebased, conflicts resolved, mistake on conflict resolution fixed). Can you take a look at it?14:28
*** xyang0 has joined #openstack-infra14:28
*** sergmelikyan has joined #openstack-infra14:28
*** markvoelker has joined #openstack-infra14:28
fungisdague: we have it registered on devstack-trusty-2-node and there are 9 available workers for that14:29
sdaguefungi: woot, thanks14:29
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
*** nfedotov has quit IRC14:30
*** doug-fish has joined #openstack-infra14:30
*** markvoel_ has joined #openstack-infra14:31
*** markvoelker has quit IRC14:31
*** rbrooker has joined #openstack-infra14:31
*** markvoelker_ has quit IRC14:32
openstackgerritDirk Mueller proposed openstack-infra/project-config: Start with a green field for RPM packaging
*** jistr|mtg is now known as jistr14:33
*** markvoel_ has quit IRC14:33
*** wenlock has joined #openstack-infra14:35
*** haypo has joined #openstack-infra14:35
haypothe merge of my nova patch failed because of "gate-tempest-dsvm-nova-v21-full: UNSTABLE". the link to the job logs is a dead link (File Not Found). is someone aware of the issue?14:36
AJaegerhaypo: add a simple "recheck" comment. We just renamed some jobs.14:36
AJaegerhaypo: this is fixed now, sorry for this14:37
openstackgerritMatt Riedemann proposed openstack-infra/devstack-gate: Add a loop to git_prune
*** pal has quit IRC14:37
fungialso jenkins01 seems to have eventually completed14:37
sdagueso... as we've been at full node count for.... a day14:38
sdaguehow is the mythical 3rd cloud into infra coming along?14:38
haypoAJaeger: no problem. i just try to avoid dummy "reverify"14:38
sdaguebecause, come L3... things are going to be slamming14:38
fungiwow, we really have been too. i'm probably out of touch with our developer base... what's the particular uptick in the past week that has us running full-out?14:39
fungijust people scrambling for l2 milestone?14:40
mordredsdague: I just got an email from our 3rd cloud people about starting to ramp up our quota there14:41
mordredsdague, fungi: so, we should probably start doing the things that will be required for that to be a thing14:41
sdaguefungi: no idea, I've just noticed the load14:41
sdagueas the tent embiggens, that's going to happen as well14:41
anteayapcm_: I think I did14:41
electricalfungi: weird JJB use case perhaps. I have 2 lists in a defaults part. Now i would like to have a 3rd list to be compiled based on the first 2 lists that grabs the first and last value of those 2 lists. Any idea's or reaching to far? :-)14:41
*** signed8b_ is now known as signed8bit_ZZZzz14:42
*** signed8bit_ZZZzz is now known as signed8b_14:42
pcm_anteaya: Thank you very much!14:42
haypoto schecule a recheck, is it possible to write something after "recheck" in a comment? for example on a new paragraph?14:42
openstackgerritMatt Riedemann proposed openstack-infra/devstack-gate: Add a loop to git_prune
anteayapcm_: welcome14:42
jeblairmordred: what do we need to do to start using them?14:42
fungisdague: now i so want to rename it the embiggened tent14:42
mordredjeblair: 2 things, I believe14:42
mordredjeblair: a) set up pypi mirror14:42
mordredjeblair: b) add their info to our nodepool.yaml and clouds.yaml14:43
AJaegersigmavirus24: I'd like to explain my -2 to - are you around?14:43
*** maishsk has quit IRC14:43
*** jsavak has quit IRC14:43
jeblairmordred: is their info on puppetmaster?14:43
sigmavirus24AJaeger: is that prevent it from merging before the next round?14:43
fungielectrical: honestly, no clue. we use jjb for some fairly basic templating and try not to get too magical with it since we expect casual contributors to our projects to actually configure their own jobs14:44
* sigmavirus24 wasn't sure if that's what you meant or if you were just trying to prevent an accidental merge on Friday14:44
AJaegersigmavirus24: exactly14:44
sigmavirus24AJaeger: makes sense14:44
sigmavirus24Sorry for the confusion :D14:44
openstackgerritNikki Heald proposed openstack-infra/puppet-bandersnatch: Sync with gozer's puppet-bandersnatch
AJaeger"Let me mark this with -2 to avoid accidential merge. The -2 is a procedural one, it should be a +2 ;)"14:44
pcm_sdague: Do  you think we're ready for approvals of (disables adv svcs from grenade jobs)?14:44
electricalfungi: makes sense yeah14:44
AJaegersigmavirus24: no problem14:44
*** Zara_ has left #openstack-infra14:44
mordredjeblair: nope - lemme do that real quick14:44
fungipcm_: yes, as long as the comment starts with the word "recheck" you can add anything else you want after it14:44
jeblairmordred: cool, yeah, that's probably step... er... what comes before a?14:44
fungipcm_: i usually follow recheck with an explanation of why i've decided to recheck14:45
sdaguepcm_: yes, I do, that needs fungi / mordred / jeblair / clarkb / pleia2 / jhesketh  to vote on14:45
openstackgerritNikki Heald proposed openstack-infra/puppet-bandersnatch: Sync with gozer's puppet-bandersnatch
pcm_fungi: Cool. I didn't know I could add more to the recheck line.14:45
fungijeblair: care to weigh in on so mrmartin can work on getting a test copy of our ask database imported and validate the new version? (he's trying to get google auth working for people again and that means an upgrade, so staging first)14:46
pcm_sdague: Great. Looking forward to getting this in.14:46
openstackgerritAdam Coldrick proposed openstack-infra/irc-meetings: Add StoryBoard Meeting
sdaguefungi: fwiw on I did recheck, but they were on jobs not touched by it14:46
jeblairfungi: +2; want more or aprv?14:47
fungijeblair: it's just a staging server, so approve now is fine14:47
jeblairfungi: +3 then :)14:47
mordredjeblair: ok - added to hiera and also to ci-launch in the file openstack-ovh-sbg1.sh14:48
*** sergmelikyan has quit IRC14:48
sdagueok, grenade mulitnode got as far as running the base - - lets see if my call out is going to work at all14:48
fungimrmartin: i'm working on copying the production backup onto the staging server now14:48
*** ociuhandu_ has joined #openstack-infra14:48
*** vlaza has quit IRC14:49
*** jasondotstar has joined #openstack-infra14:49
fungias soon as i clear up a duplicate gerrit account that's preventing me from adding an initial user to a core group for a new repo14:49
*** yfried_ is now known as yfried|afk14:49
*** jsavak has joined #openstack-infra14:50
*** gyee has quit IRC14:50
*** tsekiyama has joined #openstack-infra14:50
openstackgerritMerged openstack-infra/project-config: Run neutron-vpnaas functional tests for Neutron
*** markvoelker has joined #openstack-infra14:53
jaypipesanteaya, mordred, pabelanger: OK, I have just now received confirmation from on high that stackalytics should move ASAP to Infra.14:53
*** FallenPegasus has joined #openstack-infra14:53
*** sergmelikyan has joined #openstack-infra14:54
*** yfried|afk is now known as yfried_14:54
*** sergmelikyan has quit IRC14:54
*** gyee has joined #openstack-infra14:54
*** ociuhandu_ has quit IRC14:54
*** markvoelker_ has joined #openstack-infra14:54
*** pal has joined #openstack-infra14:55
fungijaypipes: the marketing team have waved their royal sceptre?14:55
pabelangerfungi, trolo14:55
pabelangerjaypipes, excellent!14:56
anteayajaypipes: yay14:56
anteayajaypipes: \o/14:56
openstackgerritMerged openstack-infra/system-config: Give ssh access to ask-staging.o.o
anteayajaypipes: thank you thank you thank you14:56
docaedojaypipes: that's excellent!14:56
jaypipesfungi: yup.14:56
openstackgerritMerged openstack-infra/project-config: Do not run ironic-inspector-dsvm job on stable/1.1 branch
*** sergmelikyan has joined #openstack-infra14:57
fungivikram: okay, i've added you to networking-onos-core now. you had an older account that hadn't submitted contact info and didn't have a username, so i blanked its conflicting e-mail address and deactivated it for you14:57
jaypipespabelanger: let me know how I can help. Looks like SergeyLukjanov will help from the technical side. SergeyLukjanov, if you run into any issues, just ping me.14:57
*** markvoelker has quit IRC14:58
vikramfungi: thanks..14:58
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
SergeyLukjanovjaypipes, yay! cool, did you kill somebody to get the confirmation? :)14:59
pabelangerjaypipes, SergeyLukjanov thanks again for the help, I'll focus on the spec again today14:59
jaypipesSergeyLukjanov: yes. body was dropped off in a swampy area outside Milpitas, CA.14:59
jeblairthat's what that smell is14:59
openstackgerritgaryk proposed openstack/requirements: Bump oslo.vmware to 1.16.0
jaypipespabelanger: no worries.14:59
openstackgerritMerged openstack-infra/tripleo-ci: Revert "Switch to using pypi"
*** _nadya__ has quit IRC15:00
*** sergmelikyan has quit IRC15:00
*** _nadya_ has joined #openstack-infra15:01
*** fhubik is now known as fhubik_afk15:01
*** yfried__ has joined #openstack-infra15:01
*** sergmelikyan has joined #openstack-infra15:02
*** yamahata has quit IRC15:02
*** yfried_ has quit IRC15:02
*** yamahata has joined #openstack-infra15:03
*** abregman has quit IRC15:03
*** shardy_ has joined #openstack-infra15:03
*** sergmelikyan has quit IRC15:03
*** laughterwym has joined #openstack-infra15:04
*** shardy has quit IRC15:04
*** _nadya_ has quit IRC15:05
*** kashyap has joined #openstack-infra15:06
*** fhubik_afk is now known as fhubik15:07
*** sergmelikyan has joined #openstack-infra15:07
*** shardy_ has quit IRC15:08
*** shardy has joined #openstack-infra15:09
*** amitgandhinz has quit IRC15:10
*** dkranz has quit IRC15:11
*** amitgandhinz has joined #openstack-infra15:11
*** yamahata has quit IRC15:12
*** yamahata has joined #openstack-infra15:12
*** markus_z has quit IRC15:13
*** amitgandhinz has quit IRC15:15
*** rm_work|away is now known as rm_work15:16
*** amitgandhinz has joined #openstack-infra15:16
*** crc32 has joined #openstack-infra15:17
*** kmartin has joined #openstack-infra15:17
*** ParsectiX has joined #openstack-infra15:18
*** hashar has joined #openstack-infra15:20
*** dannywilson has joined #openstack-infra15:22
*** dannywilson has quit IRC15:24
*** dannywilson has joined #openstack-infra15:24
Shrewsmordred: do you have any recollection as to why we have the python-cinderclient<1.2 requirement in shade?15:24
*** dtantsur is now known as dtantsur|brb15:25
Shrewsmordred: ah, nm. commit 29790315:25
Shrewswonder if we can allow >1.215:26
*** BharatK has joined #openstack-infra15:27
*** puranamr has joined #openstack-infra15:30
*** dkranz has joined #openstack-infra15:30
SotKI'm trying to run `tox -e jjb` to test a change to a job in project-config, but I'm getting a traceback (, even without the change. Does anyone know what would cause this?15:30
openstackgerritMerged openstack-infra/project-config: Switch to devstack-{ostype} for beaker infra jobs
openstackgerritLucas Alvares Gomes proposed openstack-infra/project-config: Add non-voting python34 job for ironic-python-agent
*** peristeri has quit IRC15:31
*** sshnaidm has quit IRC15:31
*** sdake has quit IRC15:32
fungiSotK: you might try clearing the .test directory and rerunning. i've noticed git clean doesn't remove it because it contains clones of jjb and zuul15:32
*** sdake has joined #openstack-infra15:32
fungiand then ends up testing with old versions15:32
*** yfried__ is now known as yfried|afk15:33
SotKfungi: Same issue still :(15:33
fungiSotK: delete the .tox directory too (or use tox -r -e jjb)?15:34
fungicould be a stale tox virtualenv, and if so that'll take care of it15:34
SotKfungi: Same again15:35
fungii'm running from the tip of the project-config master branch now to see if i can reproduce the same error15:35
*** e0ne is now known as e0ne_15:36
*** yfried|afk has quit IRC15:37
fungiSotK: "jjb: commands succeeded"15:37
fungican't reproduce here15:38
openstackgerritMerged openstack-infra/release-tools: clean up repodir handling
fungiSotK: i did a git remote update, checked out origin/master and ran `rm -rf .test;git clean -dfx;tox -e jjb`15:38
openstackgerritMerged openstack-infra/release-tools: don't assume all projects are in the openstack namespace
*** ddieterly has quit IRC15:38
*** crc32 is now known as crc32_soda_break15:39
openstackgerritMerged openstack-infra/release-tools: use more efficient query to find milestone
*** ig0r__ has quit IRC15:39
fungiSotK: i'm testing on fairly updated debian/stable but should be fairly similar to ubuntu15:39
SotKfungi: hmm, I'll try on a different machine15:39
*** ig0r_ has joined #openstack-infra15:40
*** ajmiller has joined #openstack-infra15:40
openstackgerritPaul Belanger proposed openstack-infra/infra-specs: Host Stackalytics Service
SotKheh, its worked on my debian machine too, thanks15:42
*** sergmelikyan has quit IRC15:44
fungifwiw we test it on ubuntu 14.04 machines in our ci system too and haven't been seeing that failure as far as i'm aware15:44
*** rm_work is now known as rm_work|away15:45
*** e0ne_ is now known as e0ne15:45
mrmartinfungi: a quick question, do we need to open the ssh port of ask-staging.o.o somewhere?15:45
*** koolhead17 has joined #openstack-infra15:46
*** peristeri has joined #openstack-infra15:46
*** vikram has quit IRC15:46
fungimrmartin: no, we have ssh open on all our servers by default15:46
openstackgerritPaul Belanger proposed openstack-infra/system-config: Partial migrate to puppet-httpd module
openstackgerritPaul Belanger proposed openstack-infra/puppet-openstackci: Migrate to puppet-httpd module
fungimrmartin: remind me where askbot dumps its backups? it uses some sort of internal cron-like mechanism and not the actual cron subsystem right?15:47
mrmartinfungi: I got it, but we have a cloudflare in front of the instance15:47
mrmartinfungi: it is using the standard backup as I know15:48
fungimrmartin: no, i mean askbot has some sort of mechanism whereby it creates tarballs of its static content15:48
fungii know where the database dumps are, just looking for the static tarballs15:49
mrmartinI guess we don't have static tarballs, as bup is backing up the entire server15:49
*** dustins has quit IRC15:50
fungioh, maybe that never got turned on in the new server then. i remember copying one off the old ask server15:50
*** efried has joined #openstack-infra15:50
mrmartinI can add it to the new one15:51
fungimrmartin: so you'll want the latest pgsql dump and a tarball of the current contents of the static directory?15:51
mrmartinwhat is the real public ip of the instance?15:52
*** k4n0 has quit IRC15:52
*** ajmiller has quit IRC15:52
*** ajmiller has joined #openstack-infra15:52
fungimrmartin: oh! i see why i'm able to ssh into it. i'm using ipv6 and that never got repointed at cloudflare. also we probably need to rethink the way cloudflare is implemented since this will likely prevent us from being able to puppet the servers if we use the same name for the server as we use for the site being hosted by the server15:53
mrmartinor create a vpn for admin access15:54
fungimrmartin: vpn to what?15:54
*** kiran-r has joined #openstack-infra15:54
mrmartinto get behind cloudflare15:54
fungimrmartin: there is no "behind cloudflare"15:54
fungimrmartin: it's just dns changes15:54
fungithe network topology is the same as it ever was15:55
mrmartinso I can bypass cloudflare if I know the real ip address15:55
*** laughterwym has quit IRC15:55
fungiwhat tipit did was remove the correct dns entry for and replace it with a cname to a cloudflare round-robin15:55
mrmartinI got it15:56
*** patrickeast has joined #openstack-infra15:56
*** armax has joined #openstack-infra15:56
mrmartinbut if someone now the real ip address of the instance, he can bypass the cloudflare checking15:56
*** kiran-r has quit IRC15:56
mrmartinnow / know15:56
*** ashleighfarnham has joined #openstack-infra15:57
*** scheuran has joined #openstack-infra15:57
*** jaypipes has quit IRC15:58
*** Sherif has joined #openstack-infra15:59
fungii'm still pretty unconvinced by cloudflare for anything besides a caching layer15:59
fungii mean, it would be possible to adjust the firewall rules on the server so that it only accepts https connections from the cloudflare proxies to prevent them being bypassed16:00
*** rm_work|away is now known as rm_work16:00
mrmartinyeap, we need to modify the fw rules too16:00
*** shashank_hegde_ has joined #openstack-infra16:01
fungithough i have doubts that cloudflare is actually smart enough to know that people are posting spam and block them automatically, especially since it's not the termination point of the encrypted session, so all it sees is garbage/noise16:01
mrmartinand where is the SSL terminated, at CF?16:01
*** salv-orl_ has joined #openstack-infra16:01
*** _nadya_ has joined #openstack-infra16:01
*** amotoki has joined #openstack-infra16:02
fungimrmartin: apache16:02
mrmartinand CF is bypassing SSL?16:02
mrmartinor how are we serving the certs?16:02
*** jsavak has quit IRC16:03
fungimrmartin: i assume it's just a socket proxy to the server, though honestly since it's only ipv4 i'll need to do something a little convoluted to find out16:03
mrmartincheck the certs of ask-staging.o.o in a browser hehe16:03
*** Sukhdev has joined #openstack-infra16:03
fungimrmartin: my browser is hitting the ipv6 address of the ask-staging server16:03
mrmartinneed to properly configure that for prod ask.o.o if we want to use CF16:03
*** otter768 has joined #openstack-infra16:03
*** ajmiller has quit IRC16:03
mrmartincert's CN is ssl2583.cloudflare.com16:04
*** jsavak has joined #openstack-infra16:04
fungimrmartin: okay, so i'm guessing tipit has access to be able to generate certs with sni for any hostname and add them to cloudflare. that would make sense16:05
*** unicell has quit IRC16:05
*** salv-orlando has quit IRC16:05
*** earlephilhower_ has joined #openstack-infra16:05
*** earlephilhower has quit IRC16:06
mrmartinit's a great fun16:06
*** amitgandhinz has quit IRC16:07
mrmartinthe alt names in the cert16:07
fungiyep, here's what i'm seeing from openssl
anteayaEmilienM: the puppet sprint is a virtual sprint, it should be removed from this page:
anteayaand added to the virtual sprint page which is linked in the first line of the sprints page16:08
*** otter768 has quit IRC16:08
mrmartinX509v3 Subject Alternative Name:16:08
*** sdake has quit IRC16:08
EmilienManteaya: ack16:08
anteayaand can be found here:
EmilienManteaya: cool, we have 2 pages, I didn't know :)16:09
anteayathe virtual sprint page is its own page, yes16:09
fungimrmartin: ahh, i see what the deal is with ipv6... they renamed the aaaa record for it from to
*** yfried|afk has joined #openstack-infra16:09
anteayaglad we clarified :)16:09
EmilienManteaya: makes sense16:09
fungimrmartin: but it's still just the ipv6 address of the server16:09
*** _nadya_ has quit IRC16:09
*** kiran-r has joined #openstack-infra16:09
fungimrmartin: i sort of wonder if cloudflare lacks ipv6 support16:09
mrmartinfungi: LOLz16:09
mrmartinwe can turn of the ipv6 support of the instance16:10
fungimrmartin: noting that embarrassingly does not have any ipv6 addresses too, probably for the same reason16:10
mrmartinI don't like to know everything :)16:11
openstackgerritNikki Heald proposed openstack-infra/puppet-bandersnatch: Sync with
*** jlanoux has quit IRC16:11
*** sdake has joined #openstack-infra16:11
*** rm_work is now known as rm_work|away16:11
mrmartinmaybe the spammers still don't have ipv6 support16:12
EmilienManteaya: updated!16:12
anteayaEmilienM: thank you16:12
*** unicell has joined #openstack-infra16:13
*** fhubik has quit IRC16:13
*** rm_work|away is now known as rm_work16:13
*** yfried has joined #openstack-infra16:13
*** yfried|afk has quit IRC16:14
*** signed8b_ is now known as signed8bit_ZZZzz16:14
*** ildikov has quit IRC16:14
*** unicell has quit IRC16:15
*** krtaylor has quit IRC16:15
*** arxcruz has quit IRC16:16
*** jsavak has quit IRC16:16
openstackgerritMerged openstack-infra/system-config: Use config drive in hpcloud
*** jsavak has joined #openstack-infra16:17
clarkbsdague ^ that shluld be the last piece of making mulyinode work16:17
*** hdd has quit IRC16:17
*** bluex-pl has quit IRC16:18
*** asselin has joined #openstack-infra16:18
*** gyee has quit IRC16:19
*** spzala has quit IRC16:19
*** bswartz has quit IRC16:20
*** bradjones has quit IRC16:20
*** ZZelle has quit IRC16:20
*** akscram has quit IRC16:20
*** mrda-away has quit IRC16:20
*** crc32_soda_break is now known as crc3216:20
*** bradjones has joined #openstack-infra16:20
*** bradjones has quit IRC16:20
*** bradjones has joined #openstack-infra16:20
*** ZZelle has joined #openstack-infra16:20
*** mrda-away has joined #openstack-infra16:20
*** akscram has joined #openstack-infra16:20
*** bswartz has joined #openstack-infra16:20
dgonzalezhi clarkb: if you have some time could you have a look at and ?16:21
*** puranamr has quit IRC16:21
*** notnownikki has quit IRC16:21
*** hashar has quit IRC16:22
fungimrmartin: i've got the backup copying to ask-staging now, estimated time to completion seems to be about 10 minutes. i'll move it into your homedir once it's finished16:22
*** hdd has joined #openstack-infra16:23
*** pal has quit IRC16:23
*** puranamr has joined #openstack-infra16:24
openstackgerritMadhusudhan Kandadai proposed openstack-infra/project-config: New jobs for neutron-lbaasv2 scenario tests
*** BharatK has quit IRC16:26
*** amitgandhinz has joined #openstack-infra16:26
*** jsavak has quit IRC16:27
*** Steap has left #openstack-infra16:27
*** yfried has quit IRC16:27
*** jsavak has joined #openstack-infra16:27
*** afazekas has quit IRC16:27
*** krtaylor has joined #openstack-infra16:28
openstackgerritFatih Degirmenci proposed openstack-infra/jenkins-job-builder: Validate and set compare types in gerrit trigger
*** scheuran has quit IRC16:32
*** crc32 has quit IRC16:32
*** shashank_hegde_ has quit IRC16:33
fungiyolanda: if you have a moment, i'm trying to implement your request from 200698 in 200826 but it doesn't appear that variable corresponds to an existing project_config file resource
fungiyolanda: which solution would you prefer for that?16:33
*** scheuran has joined #openstack-infra16:34
*** Ala has quit IRC16:34
*** shashank_hegde_ has joined #openstack-infra16:34
*** scheuran has quit IRC16:35
fungiyolanda: looking at the project_config module, we achieve that by cloning the project-config repo, so can't really set puppet dependencies on subtrees of that clone16:35
fungiyolanda: which explains why we previously just required $::project_config::config_dir16:35
*** crc32 has joined #openstack-infra16:36
fungi(e.g. the vcsrepo resource for it)16:36
SpamapSmordred: first, zomg thank you for writing about the sad life of error 396. Second... I was wondering about the API itself.. and I realized it doesn't actually exist here: ???16:36
*** haypo has left #openstack-infra16:36
*** heyongli has quit IRC16:37
openstackgerritIvan Kolodyazhny proposed openstack-infra/project-config: Move cinderclient-dsvm-functional job to non-voting queue
*** hdd has quit IRC16:37
*** gsagie_ has joined #openstack-infra16:38
*** heyongli has joined #openstack-infra16:38
rhe00_Is there a way to tell nodepool to only allow x number of concurrent instances of a label? there is a min-ready, but I don't see a way to specify a max? Currently I have a separate provider for FC and iSCSI and therefore I have an image for each provider even though the images are identical and the FC passthrough is done during run time on one provider but not the other. Would be nice to have only one provider and image and limit at the label level16:40
gsagie_Hello all, if any core from requirements can view this patch    it fixes a problem which blocks new projects from committing there first commit after the repository was created16:40
openstackgerritMerged openstack-infra/irc-meetings: Add StoryBoard Meeting
*** markvoelker_ has quit IRC16:40
openstackgerritAdam Coldrick proposed openstack-infra/project-config: Fix failing StoryBoard job
fricklermordred: with you might want to clean up nodepool.yaml even further16:41
*** markvoelker has joined #openstack-infra16:42
*** fdegir has quit IRC16:43
openstackgerritPaul Belanger proposed openstack-infra/project-config: Add voting jobs for puppet-openstack-integration
*** heyongli has quit IRC16:43
*** Sherif has quit IRC16:43
*** heyongli has joined #openstack-infra16:44
clarkbrhe00_ demand determines it based in the min ready ratios16:44
clarkbso it should be fairly automagic16:44
fungimrmartin: there are now ask-static.tar.gz and askbotdb.sql.gz files in your homedir on ask-staging16:44
clarkbjust set min ready to a reasonable ratio and you are done16:44
*** spzala has joined #openstack-infra16:45
clarkbSpamapS its unproposed un merged completely not upstresmed rax special righf now16:45
clarkbso we cant test with devstack16:45
SpamapSclarkb: Oh I tried testing it in devstack. it _does not work_16:45
SpamapSthe code is there16:46
SpamapSand the endpoint is available to any glance user16:46
SpamapSbut it _does not work_16:46
clarkbno the cod eis not there...16:46
clarkbthe tasj api is but not the import stuff16:46
rhe00_clarkb: unfortunately when the demand for FC is higher than iSCSI nodepool spins up more FC nodes and failure occurs during the passthrough set up16:46
clarkbso it doesnt work16:46
* SpamapS tosses an ikea flůgren chair16:46
SpamapSclarkb: right, the API is there and actually if you leave it turned on, a user can fill up the tasks table with a neverending infinite supply of tasks16:47
clarkbrhe00_ I dont understand the issue then. if under quota and demand is higher for another lable you should boot more of that label16:47
*** fdegir has joined #openstack-infra16:47
SpamapSnow that I say that.. I wonder if OSSG would like to fix that..16:47
*** mjturek1 has joined #openstack-infra16:47
*** dtantsur|brb is now known as dtantsur16:48
rhe00_clarkb: but I can only have 1 running at a time because I can only have one active FC passthrough at a time on the provider.16:48
clarkbrhe00_ then quota should be 116:49
rhe00_ideally I would have a 4 port HBA so I could run 2 at a time, but then I would still be limited to 2.16:49
*** melwitt_ is now known as melwitt16:49
mordredSpamapS: yes. that is right. tasks do not work in devstack16:49
mordredSpamapS: and that error message and code are also not in openstack16:49
*** tvanderwiel has quit IRC16:49
*** unicell has joined #openstack-infra16:49
*** tvanderwiel has joined #openstack-infra16:50
*** jistr has quit IRC16:50
*** amotoki has quit IRC16:50
*** hichihara has joined #openstack-infra16:52
mordredSpamapS: because, you see, the tasks are all pluggable16:52
*** sputnik13 has joined #openstack-infra16:52
rhe00_clarkb: so min-ready is currently set to 1 for FC, 2 for iSCSI. max-servers on the provider is set to 3. if FC demand is higher then nodepool will spin up 1-3 FC nodes and starve out iSCSI. Two of the FC nodes will quickly fail during passthrough set up.16:52
mordredSpamapS: so, the name of teh task you create with task-create may or may not be a thing that the cloud in question has deployed a plugin for16:52
clarkbrhe00_ max servers should be 1 in nodepoola model16:52
mordredSpamapS: there is also no way to get a list of the tasks your cloud knows how to perform16:53
*** hdd has joined #openstack-infra16:53
clarkbit doesnt understand non quota quota and relies purely on per provider instance quota to determine max nodes16:53
*** ajmiller has joined #openstack-infra16:53
openstackgerritPaul Belanger proposed openstack-infra/project-config: Add puppet-jobs for puppet-openstack-integration master
*** ajmiller_ has joined #openstack-infra16:54
Shrewsmordred: that's for sad making16:54
*** ajmiller__ has joined #openstack-infra16:55
mordredShrews: it's so sad making16:56
*** Sherif has joined #openstack-infra16:56
ShrewsGheRivero: congrats! your README file change has failed all functional testing16:56
*** HeOS has quit IRC16:56
SpamapSmordred: I'm pretty sure we need to kick tasks out of glance, and make that a mistral thing.16:56
*** akshai has quit IRC16:56
SpamapSmordred: or a "simpler workflow service"16:57
*** pabelanger has quit IRC16:57
*** pabelanger has joined #openstack-infra16:57
*** yamahata has quit IRC16:57
*** ptoohill is now known as pothole16:57
mordredSpamapS: it's not a workflow though - it's just simply exposing the internals of the internal task processing queue to the end user16:57
*** ajmiller__ has quit IRC16:57
mordredSpamapS: it needs to be removed with prejudice and replaced with literally nothing16:57
*** ajmiller__ has joined #openstack-infra16:57
mordredfor concerns about different api purposes and api vs. streaming - that's what apache is for16:58
*** pcm_ is now known as pc_m16:58
*** yamahata has joined #openstack-infra16:58
*** akshai has joined #openstack-infra16:58
mordredfrickler: agree. once we've done that, I have a hunch that we should probably go combine nodepool-base and node-devstack16:58
GheRiveroShrews: yeah! master level achievement!16:58
clarkbmordred did you catch Ibhad to enable config drive in hpcloud for dib built images?16:58
*** pal has joined #openstack-infra16:58
clarkbit not beinf there is whybwe had ubuntu as a hostname16:59
*** patrickeast has quit IRC16:59
*** ajmiller_ has quit IRC16:59
*** ajmiller has quit IRC16:59
mordredclarkb: no. on the old images?16:59
mordredthat's awesome16:59
*** bswartz has quit IRC16:59
clarkbfix merged si all should be well soon16:59
mordredclarkb: sweet. I'm so excited about these things17:00
mordredclarkb: they increase my love of clouds17:00
*** pblaho has quit IRC17:00
clarkbfor multinode on ubuntu-trusty sdague mentioned parted was missing as well17:01
*** derekh has quit IRC17:01
mordredclarkb: k. good to know17:01
*** esmute has quit IRC17:01
mordredclarkb: speaking of missing things: 20211617:02
*** kiran-r has quit IRC17:02
*** jsavak has quit IRC17:02
*** jsavak has joined #openstack-infra17:03
clarkbshould rsync et al move into that list?17:03
*** maishsk has joined #openstack-infra17:03
mordredclarkb: possibly - I left them out because they were the same name on each17:03
mordredclarkb: but I think you're right - how about I make a follow up that does that and adds parted17:03
clarkbsounds good17:04
mordredSpamapS, clarkb: got word from internal that they're about to apply our new network range in HP117:04
mordred15.184.52.0/22 for anyone who cares17:05
SpamapSmordred: sweet!17:05
clarkbso can haz VMs?17:05
SpamapSmordred: I'm poking at getting a new box deployed w/ glean instead of cloud-init to work around the resizing issue.17:05
mordredclarkb: still need to get hardware remediation completed17:05
mordredSpamapS: cool!17:05
SpamapSmordred: because zomg reconfiguring cloud-init to not have a single module in its list is o_O17:05
mordredSpamapS: I need to write you a glean patch - unless you want to write it17:06
SpamapSmordred: also I thought about it. Lets not add resizing.17:06
mordredSpamapS: to support the vlan structure in config-drive17:06
mordredSpamapS: agree. it's an easy enough to thing to do otherwise if needed17:06
SpamapSmordred: oh yes I very much want the vlan thing. :)17:06
SpamapSmordred: well and harlowja_ had a good point. Resizing is _slow_17:06
mordredyah  man17:06
SpamapSmordred: so rather than do it automatically for people.. they _always_ disable it at Yahoo!17:07
harlowja_SpamapS yes, avoid blocking users access to VM(s) imho to auto-resize crap for them17:07
harlowja_let user resize it, take there own time if they want17:07
SpamapSharlowja_: or let user just make a new partition and start putting data on it immediately.17:07
mordredharlowja_: turns out they may also not care17:07
harlowja_we are moving to provide fixed size stuff in our images, rest of disk, they can expand as they want17:07
fungimrmartin: okay, tipit had a test setting on for ask-staging in cloudflare which was bypassing ipv6, but that's corrected now and looking more like what i would expect17:07
harlowja_because people were like it takes forever to boot, and blah blah17:07
*** hichihara has quit IRC17:07
*** ihrachyshka has quit IRC17:07
harlowja_and most of the time is in resizing (especially with qcow2...)17:07
*** bswartz has joined #openstack-infra17:08
sc68calbig spender! a /22!17:08
fungimrmartin: this does also mean that we won't be able to puppet machines whose server names are the same as site names being proxied by cloudflare, so we probably need to rethink the design (rebuild the servers with different names than the site names for example) if there's interest in moving forward there17:08
mordredsc68cal: there are some fun things about HP and Digital being the same company ... having 2x class A is one of them17:08
SpamapSharlowja_: before I 'splode everything and use glean instead of cloud-init.. do you have a quick trick to just remove one module from the list of modules? Or do I have to rewrite the whole list minus the one I want removed?17:09
openstackgerritIvan Kolodyazhny proposed openstack-infra/project-config: Move cinderclient-dsvm-functional job to non-voting queue
*** yamahata has quit IRC17:09
clarkbthey are adjacebt too but not a /717:09
harlowja_SpamapS hmmm, would take me a few to get u a trick, probably just easier to replace the whole list17:10
clarkber maths fail they are two adjacent /7s17:10
*** ajmiller__ is now known as ajmiller17:10
clarkbgah /817:10
*** sigmavirus24 is now known as sigmavirus24_awa17:10
clarkbI give up on typing17:10
openstackgerritKhai Do proposed stackforge/python-jenkins: Add support for getting information for the last executed build
mordredSpamapS: 'splode everything and use glean17:11
mordredSpamapS: because
mordredSpamapS: I intend to use it for infra base images in the vm clouds - so this'll be a great start17:12
SpamapSharlowja_: ty17:12
SpamapSmordred: I know. I just like decoupling workarounds from progress. :)17:12
clarkbSpamapS sed in a #17:13
mordredSpamapS: good point17:13
SpamapSclarkb: <sigh> right? :-P17:13
clarkbsed -i 's/- resize/# no resize/' /etc/cloudsomething17:13
SpamapSclarkb: its a yaml file, I can actually just delete the item exactly and dump it again.17:14
SpamapSbut.. glean ftw17:14
clarkbya but thats >1 line :)17:14
*** koolhead17 has quit IRC17:15
mordredwe shoudl write ysed - that is a yaml aware sed17:15
clarkbdib golf17:15
SpamapSYou have awakend my oneliner dragon and it is battling with my TDD dragon.17:15
SpamapSmordred: I'm sure thats how jq came to be.17:15
* mordred sits back to watch the dragon battle17:15
mordredSpamapS: yah. I am also sure of that17:15
SpamapSwhich is like.. zomg o_O wtf in a bottle.17:15
*** Sherif has quit IRC17:16
SpamapSbut I'm sure version 0.1 was like.. "hey look we can remove a single key easily in a one liner"17:16
nibalizeryear later...17:16
openstackgerritPaul Belanger proposed openstack-infra/project-config: Add puppet-jobs for puppet-openstack-integration master
*** Sherif has joined #openstack-infra17:17
*** heyongli has quit IRC17:17
*** ddieterly has joined #openstack-infra17:18
*** patrickeast has joined #openstack-infra17:18
*** heyongli has joined #openstack-infra17:19
lifelessfungi: yup17:20
lifelessfungi: since its doing the job we were abusing the mirror for, its not surprising ;)17:21
fungilifeless: indeed. i thought the same ;)17:21
*** jsavak has quit IRC17:21
*** jsavak has joined #openstack-infra17:21
fungilifeless: just a sec and i'll get the ref-driven variant of the periodic constraints proposal job whipped up17:21
lifelessclarkb: touching base on requirements clone17:22
clarkblifeless I havent had a chance to update the change yet.need to incorporate some improvements fungi is amking to our use of zuul cloner17:24
*** heyongli has quit IRC17:24
clarkband probably pick a single job to start with. python27 best for that?17:24
fungiclarkb: speaking of that, since you have that problem in your head anyway a review of would be appreciated when you have a moment17:24
fungibecause it's also delaying my bindep jobs poc17:25
*** ajmiller_ has joined #openstack-infra17:25
*** heyongli has joined #openstack-infra17:25
lifelessclarkb: python27 sounds dandy17:26
*** ajmiller__ has joined #openstack-infra17:26
fungii mean, i could just switch my jobs over to gerrit-git-prep instead, but this is a good opportunity to nail down the ggp replacement with zuul-cloner17:26
clarkbfungi +117:26
*** ajmiller has quit IRC17:26
fungiand i'm definitely open to cleaner suggestions on 201692 if you have any17:27
openstackgerritKhai Do proposed stackforge/python-jenkins: Check a job's build status
*** mjturek1 has quit IRC17:28
*** esmute has joined #openstack-infra17:28
*** shashank_hegde_ has quit IRC17:29
*** achanda has joined #openstack-infra17:29
*** ajmiller_ has quit IRC17:29
*** tonytan4ever has quit IRC17:30
mkodererhi folks :) I need some help with devstack-plugin for Murano and the gate jobs17:31
clarkbfungi: I am not sure we absolutely need the cleanup trap but its hygenic so meh17:31
harlowja_mordred hey,,,, u could like help out with the cloud-init2.0 stuff and make sure its good enough for u (and then not need glean) ;-)17:31
harlowja_a brilliant idea, i know17:31
clarkbharlowja_: my major request: stop using distro specific users, have a "cloud" user instead and use it everywhere17:32
fungiclarkb: yeah, i hate using mktemp without a trap17:32
harlowja_clarkb thats really a setting the distros are putting in17:32
fungii don't want to be the one who littered your tmpdir with trash from failed invocations of my scripts ;)17:32
harlowja_not something cloud-init is forcing afaik17:32
*** gsagie_ has quit IRC17:32
clarkbharlowja_: then make it unsettable17:32
harlowja_should be i think17:32
clarkbno I mean have cloud init fail if any other user is configured17:33
clarkbor just always only create the one user17:33
*** yamahata has joined #openstack-infra17:33
clarkbdon't let the distros be jerks to your users17:33
harlowja_ (basically this is doing what u are saying)17:33
harlowja_u can send in17:33
harlowja_users: []17:33
harlowja_annnd profit17:33
clarkbI am saying I don't want to do that17:33
clarkbyou do it17:33
*** dustins has joined #openstack-infra17:33
*** pelix has quit IRC17:33
clarkb(and this is why glean exists /me goes back to reviewing fungi's change)17:33
harlowja_confused, what do u want me to do , lol17:34
clarkbharlowja_: make it impossible for the distros to screw me over17:34
*** dustins has quit IRC17:34
fungiharlowja_: not _you_ but the software17:34
clarkbharlowja_: don't force me to make config changes17:34
mordredremove the config setting17:34
greghaynesSpamapS: mordred what is this about glean and resizing?17:34
*** dkehn has quit IRC17:34
mordredclarkb: btw ... the current beta ibm public cloud configures cloud-init to make a "ibmcloud" user :)17:34
harlowja_thats a user-data entry u can send in, not a config tweak17:34
*** dustins has joined #openstack-infra17:34
* greghaynes goes hunting for the "no" hammer17:35
*** jaypipes has joined #openstack-infra17:35
mordredgreghaynes: we're not going to do it17:35
harlowja_like in userdata u'd send in "users: []"17:35
mordredharlowja_: and yes - I've been meaning to engage on cloud-init 2.017:35
clarkbharlowja_: right stop allowing that at all17:35
clarkbharlowja_: just be opinionated and fix it17:35
fungiharlowja_: yeah, i think the main concern here is that distros preconfigure cloud-init. i'm not sure that's something cloud-init can solve other than maybe having a user-data override for "please ignore your configuration"17:35
mordredharlowja_: I have never in my life sent in userdata17:35
jogoclarkb: hmm the failure rate for multinode looks like it is very high
harlowja_fungi ya, don't use distro stuff, lol17:35
mriedemhrm, i haven't seen this before with latest pbr17:35
jogosdague: ^17:36
greghaynesclarkb: I saw a blurb in backscroll about cloud-init on hp getting enabled?17:36
greghaynescurious what was up with that17:36
jogoclarkb: even after one fix17:36
clarkbjogo: well the config drive fix only just went in17:36
fungiclarkb: well, if cloud-init doesn't do something the distros want, they'll fork/patch it or just add some plugin/separate program17:36
clarkband it takes time for nodes to cycle out17:36
harlowja_mordred u don't know what u've been missing, lol17:36
mordredfungi: yup17:36
clarkbgreghaynes: config drive getting enabled17:36
greghayneser, right17:36
clarkbgreghaynes: because cloud init is configured to only use config drive so without it we weren't cloud initing very well17:36
*** dkehn has joined #openstack-infra17:36
greghaynesclarkb: oh, and it was for some reason turned off?17:37
clarkbgreghaynes: no cloud turns it on by default17:37
greghaynesin hp I thought we didnt config drive, just dhcp17:37
*** _nadya_ has joined #openstack-infra17:37
clarkbgreghaynes: that was the problem17:37
*** maishsk has quit IRC17:37
*** shashank_hegde_ has joined #openstack-infra17:37
clarkbwith just dhcp our hostname was always "ubuntu"17:37
mordredclarkb: why was the hostname being ubuntu a problem?17:38
greghaynesdoesnt work for multinode17:38
clarkbmordred: bceause nova requires unique hostnames17:38
*** jerryz has joined #openstack-infra17:38
greghaynesyou need different hostnames17:38
greghaynessuch fun17:38
clarkbmordred: because bonghits17:38
mordredyah. that thing17:38
mordredwith you17:38
clarkbmordred: but they are 5 year old bonghits so we live with them17:38
clarkbfungi: approved17:38
fungiclarkb: thanks17:38
*** mtanino has quit IRC17:39
mordredclarkb: yup17:39
harlowja_mordred u can like pass in shell scripts as user data, cloud-init will then later execute them, its pretty cool, u should use it :-P17:39
*** achanda has quit IRC17:39
* harlowja_ will now sell u a used car17:39
mordredgreghaynes: back to resize - we are in fact going to use glean in the infra-cloud nodes so that we can _avoid_ resizing17:39
mordredharlowja_: hehe. so ... funny story17:39
*** heyongli has quit IRC17:39
mordredharlowja_: we tried that ages ago - but ran in to 2 problems17:39
openstackgerritArmando Migliaccio proposed openstack-infra/project-config: Drop unstable configuration for tempest-dsvm-neutron job
greghaynesmordred: oo fun17:40
mrmartinfungi: thanks for the .tgz, and yes we can rethink the naming there, but as a workaround I'm using an /etc/hosts entry for external instance name resolution.17:40
*** maishsk has joined #openstack-infra17:40
mordredharlowja_: a) no cloud-init on rackspace b) too hard to debug failures ... so we moved to the model of just using cloud-init for minimal things on hp and using in-bound ssh for script execution17:40
mordredharlowja_: I _do_ think it's a neat feature - it's just one that does not work well for me17:40
greghaynesmordred: just because we dont need to waste the i/o?17:40
*** ivar-lazzaro has joined #openstack-infra17:40
mordredgreghaynes: well, also the disks are 2TB17:40
mordredgreghaynes: so grub has a sad17:40
harlowja_mordred i used horses back in the day to, they are reliable, they never need gas17:40
openstackgerritJeremy Stanley proposed openstack-infra/project-config: Update constraints after releases and reqs changes
fungilifeless: ^17:41
openstackgerritArmando Migliaccio proposed openstack-infra/project-config: Drop unstable configuration for tempest-dsvm-neutron job
mordredharlowja_: I actually think that now that ansible exists, there is no real need fora  proto-config-management system to bootstrap your config management system17:41
fungilifeless: turns out i think we can get by reusing the same job template for both periodic and ref-driven events17:41
harlowja_mordred possibly17:41
mordredget an ssh key on teh machine- all the rest can be done normally17:41
*** esmute has quit IRC17:41
mordredharlowja_: I mean - don't get me wrong - I get the appeal, and if it works for people, AWESOME17:41
mordredI'm super happy that cloud-init exists17:41
harlowja_mordred understood, thats also the usage for cloud-init mainly at yahoo, configure chef, get the heck out of the way17:42
mordredit just doesn't solve my problem in the way I need - but luckily that's ok17:42
mordredI thnik cloud-init is AMAZEBALLS at bootstrapping chef and puppet stuffs17:42
mordredif that's your environment and stuff17:42
lifelessgreghaynes: / SpamapS: check your HP mail - network stuff for the infra cloud17:42
*** dtantsur is now known as dtantsur|afk17:42
*** dangers_away is now known as dangers17:43
*** heyongli has joined #openstack-infra17:43
*** ivar-laz_ has joined #openstack-infra17:44
mordredlifeless: thanks! I replied and copied them and removed you so as not to bother you ... but come to think of it, keeping you on the cc list so that you knew I'd done that would have been more efficient :)17:44
*** jsavak has quit IRC17:44
lifelessmordred: yes, yes it would17:44
greghaynesah, there is the confustion, I was wondering "is there another person with that name?"17:44
*** dizquierdo has quit IRC17:45
clarkbbwhahahahahahah I think config drive may have broken d-g17:45
clarkbis it too early to rage quit clouds?17:45
*** jsavak has joined #openstack-infra17:45
nibalizerclarkb: containers *are* the future...17:45
clarkbnibalizer: containers hvae the same problems17:45
greghaynesclarkb: oh, weve run into that before17:45
*** esmute has joined #openstack-infra17:45
clarkbnibalizer: specifically the "configure this templated thing after it boots so that it appears unique to its peers"17:46
greghaynesclarkb: I think in the nodepool ready script we sed that device out of mount because hp started advertising it as a device we should mount17:46
lifelessmmm nuts
*** bauzas has quit IRC17:46
lifelessfungi: ^ :/17:46
*** alivigni has quit IRC17:46
clarkbideas on fixing that? otherwise I think I likely need to revert17:47
mordredclarkb: maybe we shoudl account for blkid -t LABEL="config-2"17:47
*** ivar-lazzaro has quit IRC17:47
clarkbthis is so ridiculous17:47
mordredclarkb: let me look at the function real quick17:47
clarkbmordred: but is it config-2 on rax too?17:47
mordredit's always config-217:47
mordredthat's part of the interface17:47
greghaynes is the previous ordeal FWIW17:47
SpamapSso far the new range does not yet work on the servers17:48
clarkbmordred: we basically assume on hpcloud our ephemeral drive is /dev/vdb17:48
*** Alexei_987 has left #openstack-infra17:48
* SpamapS goes to run errands17:48
nibalizerso if pbr handles the actual 'version bumb', how are changelogs updated, is that a gerrit review that lands right before someone tags and pushes the tag?17:48
mordredclarkb: yah - we should probably not do that17:48
*** heyongli has quit IRC17:48
clarkbmordred: but I think what we want is check /dev/disk/by-label for whatever rax/hpcloud call their ephemeral drives17:48
clarkbthen use that device17:48
*** achanda has joined #openstack-infra17:48
lifelessfungi: suggestions on how to figure out where the script was run from - $0 hackery?17:48
greghaynesclarkb: yea, that is the issue17:48
*** dalgaaf has quit IRC17:48
greghaynesclarkb: ++17:48
mordredsdague, jogo: is the epehemeral drive guaranteed to be ephemeral0 ?17:49
mordredclarkb: hp calls it ephemeral017:49
mordredclarkb: well, and on rax we just get the one disk and don't rework it17:49
mordredclarkb: /dev/xvda1: UUID="c1aa92c8-2961-429c-b04f-8f6c7e9b4cf4" TYPE="ext4"17:49
*** ajmiller__ is now known as ajmiller17:49
mordredis on a rax node17:49
clarkbmordred: the vdxe check in d-g is for rax17:50
openstackgerritMerged openstack-infra/shade: Allow use of admin tokens in keystone
clarkband we get two with performance nodes17:50
mordredyah. gotcha17:50
mordred/dev/vda1: LABEL="cloudimg-rootfs" UUID="3c32313b-8459-400d-84fb-e75554d8add1" TYPE="ext4"17:50
mordred/dev/vdb: LABEL="ephemeral0" UUID="199007d8-1bc3-43d9-8a71-0a96a8017ed8" TYPE="ext3"17:50
*** heyongli has joined #openstack-infra17:50
mordredI get that for an hp node17:50
*** sabeen has joined #openstack-infra17:50
*** bauzas has joined #openstack-infra17:50
*** salv-orl_ has quit IRC17:50
mordredclarkb: you doing a patch or want me to?17:50
clarkbmordred: why don't you as you have a better handle of the options here17:51
mordredk. I'll take stab #117:51
mordredclarkb: review.o.o is a performance, right?17:51
clarkbmordred: yes with an ssd volume iirc17:52
mordredI'm so excited ... there are no labels on the disks in rax17:52
*** mtanino has joined #openstack-infra17:52
mordred/dev/xvda1: UUID="bee42aed-9d8e-4377-b273-28db31edee74" TYPE="ext4"17:52
*** hdd has quit IRC17:53
greghaynesthe config drive must have one?17:54
clarkbok so we probably want to have if blkid | grep ephemeral0 then use whatever that is else if /dev/vdxe use that17:54
greghaynesbecause I think thats all glean looks for17:54
clarkbgreghaynes: well we don't care about the config drive really, we want to know which is the ephemeral disk17:54
*** heyongli has quit IRC17:54
greghaynesim just hopings its not actually no labels ;)17:55
fungilifeless: see comments on 19978217:55
clarkbI am going to disable puppet on nodepool.o.o and just trim out the config drive = true from that for a short period until we ge tthe d-g change in17:55
openstackgerritlifeless proposed openstack/requirements: Move integration test dependencies into bindep
openstackgerritMerged openstack-infra/project-config: Create clone map outside workspace in zuul-.*-prep
*** puranamr has quit IRC17:56
*** akshai has quit IRC17:56
openstackgerritMonty Taylor proposed openstack-infra/devstack-gate: Use blkid to find ephemeral disk on HP
mordredclarkb: ^^17:56
asselinfungi, how are things in infra?17:56
openstackgerritlifeless proposed openstack/requirements: Move integration test dependencies into bindep
*** heyongli has joined #openstack-infra17:57
fungiasselin: i think we're at our baseline/typical levels of insanity17:57
clarkbmordred: you replaced the vdxe handling not the vdb handling17:57
mordredI did?17:57
*** gyee has joined #openstack-infra17:57
clarkbmordred: I think you need to move vdxe into the if case and make else the ephemeral label stuff17:58
mordredone sec17:58
fungiasselin: nothing massively broken, so if you want to break something now is probably the time ;_17:58
jogoclarkb: what about RAX?17:58
asselinfungi, ok :) i'll try not too however17:58
jogomordred: not sure17:58
openstackgerritMonty Taylor proposed openstack-infra/devstack-gate: Use blkid to find ephemeral disk on HP
*** Swami has joined #openstack-infra17:58
clarkbjogo: rax is fine because its a separate cse17:58
mordredjogo: it seems rax does not do disk labels17:58
clarkband they dont do disk labels for whatever reason17:58
mordredalthough they _do_ do them for config-drive17:59
fungi(at the moment anyway. that's always subject to change!)17:59
sdaguemordred: yeh, only libvirt backend labels disks17:59
sdagueconfig drive is common code17:59
*** akshai has joined #openstack-infra17:59
mordredsdague: I propose that the disk labeling become common code - but I understand it may be harder than that :)17:59
sdaguemordred: at least in what I'm looking at17:59
*** _nadya_ has quit IRC18:00
mordredsdague: awesome. well - our anecdotal experience is matching with that18:00
*** hdd has joined #openstack-infra18:00
sdaguemordred: yeh, well xen isn't even running on the same host as where the disks are, so it's even more complicated18:00
clarkbI bet this is why we don't check labels already18:00
greghaynesmordred: commented18:00
lifelessso is dsvm down right now? should we not doa  notify ?18:00
*** _nadya_ has joined #openstack-infra18:00
mordredlifeless: eyeballs from you on too?18:00
clarkblifeless: no one reads them anyways (/me should stop being cranky)18:00
lifelessclarkb: moar sleep. In about 15 years.18:01
openstackgerritMonty Taylor proposed openstack-infra/devstack-gate: Use blkid to find ephemeral disk on HP
mordredgreghaynes: thanks!18:01
* cdent reads everything he can get his hands on18:01
fungioh, have we broken more than the multi-node job with switching to configdrive?18:01
openstackgerritRamy Asselin proposed openstack-infra/puppet-openstackci: Move jenkins job builder into jenkins_master
mordredfungi: yah. possibly18:01
mordredfungi: eyes on ?18:01
fungii guess i wasn't paying close enough attention after all!18:01
asselinfungi, nibalizer yolanda zaro  ^^ Let's get JJB landed18:02
fungiproving clarkb's cynical point18:02
clarkbfungi: yes config drive is mounted at /dev/vdb18:02
*** ayoung has joined #openstack-infra18:02
clarkbfungi: which is where ephemeral is normally located18:02
fungiand that's where... yeah18:02
clarkbit WOULD be nice if nova was at least stable in that sorting18:02
clarkbeven if we can't have labels everywhere18:02
clarkbbut having labels everywhere would also be nice18:02
*** puranamr has joined #openstack-infra18:02
mordredclarkb: well - to be fair - nova _Does_ give us a label on config-2 always18:02
clarkbsdague: is it worth filing bugs for ^18:02
asselinnibalizer, updated commit message since zaro doesn't seem to be around. ^^18:02
*** Kennan has quit IRC18:02
clarkbmordred: right but thats one of N block devices18:02
clarkbmordred: in rax one of 16 potential devs unsure of hpcloud18:03
mordredclarkb: yah. but I guess I'm saying it seems that nova itself is doing what it can on this ... and some of the drivers do not implement it18:03
lifelessmordred: this code is running without -e ?18:03
*** Sukhdev has quit IRC18:03
clarkbmordred: imo drivers not implementing it are novas problem when those drivers are in tree18:03
*** Kennan has joined #openstack-infra18:03
clarkbmordred: and both libvirt and xen are in tree18:03
mordredclarkb: so I'd blame the drivers18:03
mordredclarkb: fair18:03
clarkblifeless: good question18:04
*** signed8bit_ZZZzz is now known as signed8b_18:04
*** signed8b_ is now known as signed8bit_ZZZzz18:04
lifelessmordred: if its running with -e blkid will exit 2 on missing labels18:04
clarkblifeless: I think it is set -e based on the logs where this fails (it short circuits immediately)18:04
*** otter768 has joined #openstack-infra18:04
*** _nadya_ has quit IRC18:05
fungimordred: clarkb if 202217 is a likely solution, we can enqueue/promote it and get it confirmed that way while speeding up resolution of the issue18:05
clarkbalso I have ninja'd config drive out of the nodepool config so its not self testing at this point18:05
lifelessmordred: commented on rev 218:05
fungioh, nevermind18:05
mordredlifeless: how about EPHEMERAL_DEV=$(blkid -L ephemeral0 || true)18:05
clarkbwasn't sure how quickly we would turn around the d-g fix when the gate was thrashing so figured not thrashing was best18:05
mordredclarkb: yah18:06
*** sigmavirus24_awa is now known as sigmavirus2418:06
greghaynesthat works18:06
lifelessmordred: yes, also see my other point in the review :)18:06
*** annegentle has quit IRC18:06
clarkbthough it will be self testing just without config drive so I think we cover the same code jus tas well18:06
mordredlifeless: on it18:06
lifelessso - *are* we mid-incident?18:07
lifelessI am confused18:07
clarkblifeless: we were, it should no longer be a problem18:07
lifelessack, thanks18:07
mordredlifeless: we have manual fix in place18:07
openstackgerritMonty Taylor proposed openstack-infra/devstack-gate: Use blkid to find ephemeral disk on HP
clarkbso we can review 202217 without rushing18:08
mordredok - that should take care of greghaynes and lifeless concerns I think18:08
sdagueso, xen probably could be made to do this, but I'd bug the rax folks, as they are pretty much the only people working on that code18:08
*** slagle_ is now known as slagle18:08
clarkbsdague: what about stable sorting the available devices?18:08
*** otter768 has quit IRC18:09
sdagueclarkb: in what context?18:09
clarkbso that adding in config drive doesn't change the location of ephemeral. This may be hard in linux as iirc linux doesn't garuntee it is stable18:09
clarkbbut there may be some way to do that with uuids18:09
clarkband udev?18:09
mordredbut then you'd need special udev in-guests18:09
sdagueyeh, I think you are at the mercy of the base OS18:09
mordredand I yell every time someone suggests in-guest config special18:09
lifelessVM enumeration is muuuuch more stable than PCI bus walking18:10
*** alexsyip has joined #openstack-infra18:10
clarkbsdague: basically /dev/vdb was always our ephemeral drive then we added config drive and now its the config drive and everything fails18:10
clarkblifeless: oh cool so maybe it can be made t owork18:10
lifelessyes, its possible for the guest to do wtf it wants to18:10
sdagueclarkb: so that's about plug order right18:10
lifelesssdague: yeah18:10
sdagueit's kind of out of the control with a fresh guest18:11
lifelessclarkb: I'm not sure its doable18:11
lifelessclarkb: but not because of the in-guest thing, rather what sdague is onto18:11
fungia terrible thought, but we could just test mount every available virtual block device and inspect the contents to decide what it is18:11
clarkbok, in that case I think we should really push for labels everywhere18:11
lifelessfungi: welcome to the internet of the 90's18:11
lifelessfungi: YOU BAD BAD PERSON18:11
mordredclarkb: yah - I think pushing the xen backend to adopt stable labels is more win18:11
clarkbfungi: except config drive can come in a half dozen varieties iirc18:12
greghaynesyeaa, seems like labels are a known fix for this that is arleady out there18:12
* mordred revokes fungi's talking18:12
fungiyep. i needed a shower as soon as i thought about it18:12
lifelessfungi: acid ?18:12
clarkbfungi: some of which overlap what your ephemeral disk may presetn as iirc18:12
fungilifeless: it takes a bullet to scrub out the memory18:12
sdagueclarkb: sure, but a config drive is going to be < 1G18:12
mordredand iso9660 right?18:12
clarkbmordred: iirc it doesn't have to be iso966018:12
sdagueit can be vfat under some circumstances18:13
clarkbit could be fat or ext or etc18:13
*** heyongli has quit IRC18:13
sdagueI don't think it can be ext atm18:13
mordredin any case ... disk labels18:13
sdaguethere was some objection to parallels adding that support for that reason18:13
clarkbsdague: ah ok18:13
*** hdd has quit IRC18:13
openstackgerritDoug Hellmann proposed openstack/requirements: Use oslotest 1.9.0
mordreddisk labels are super useful here ... in fact, one could even imagine attaching cinder volumes with a disk label matching the name given ...18:13
*** HeOS has joined #openstack-infra18:14
mordredbut I don't want to get ahead of myself18:14
fungiwell, regardless the kernel can guess the filesystem reasonably well, and in the short term we can attempt to determine from the contents what it's being used for, if we're stuck waiting on providers to implement labels18:14
sdagueI think cinder volumes get labels under libvirt, I thought I saw that code18:14
*** puranamr_ has joined #openstack-infra18:14
mordredsee - that's pretty awesome18:14
*** tvanderwiel has quit IRC18:14
*** puranamr has quit IRC18:14
*** puranamr has joined #openstack-infra18:15
sdagueI might be wrong, I was trying to figure out a lot of code quickly, but at least the mkfs call wrapper for libvirt is setup to do label calls pretty easily18:15
clarkbmordred: as long as 202217 runs at least one devstack job on hpcloud I think we can be convinced its working and approve18:15
sdagueand I can see that being useful thing to get in there18:15
clarkbmordred: so I iwll be watching for that18:15
fungisdague: cinder adds filesystems for you? i guess i always just assumed you got a blank raw block device and still needed to format it yourself18:16
*** puranamr_ has quit IRC18:16
*** puranamr has quit IRC18:16
*** heyongli has joined #openstack-infra18:16
*** alivigni has joined #openstack-infra18:16
clarkbfungi: yes but I think you can apply a label wihtout a filesystem right?18:16
clarkblabels are in partitions? I am probably wrong18:16
*** esmute has quit IRC18:16
*** puranamr has joined #openstack-infra18:16
fungipretty sure labels are in the filesystem metadata18:17
*** ociuhandu has quit IRC18:17
fungiwhich is why mkfs is what does it18:17
sdaguelabels are in the filesystem18:18
sdagueso not all fs types support it18:18
*** tqtran has joined #openstack-infra18:18
*** FallenPegasus has quit IRC18:18
sdaguefungi: yeh, I thought I saw libvirt code around that, maybe I'm wrong18:18
persiaSome partition layouts support adding labels there.  Some filesystems support adding labels there.  There exist exceptions to both models.  For ext*, labels are easiest to get from filesystems.18:18
sdagueanyway, I'd poke some of the rax folks about xenserver add here. I think it's a reasonable thing to do. It's not really a contract we had though.18:19
*** dalgaaf has joined #openstack-infra18:19
mordredsdague: totally18:19
fungipersia: is there support in traditional dos partition tables? or do you need something like guid?18:19
mordredsdague: seems like if we can get them to add it, proposing that it becomes a contract is a conversation that could start18:19
fungier, gpt18:19
clarkbsdague: the problem with that is it is entirely unclear to the end user what is a contract and wat isnt18:20
clarkbsdague: so we end up doing the broken thing of assuming /dev/vdb is safe18:20
* dims_ peeks at backlog to see if someone is seeing "timeout: failed to run command ‘/opt/stack/new/devstack-gate/’" problem18:20
clarkbbecause /dev/vdxe is safe18:20
sdaguemordred: yeh, I agree, if we had 2 backends doing the same thing, we could start saying "we expect you's all to do this"18:20
persiafungi: I think they do for post PC-DOS 2.0 "traditional dos", but I'm not successfully parsing the spec right now.18:21
anteayadims_ we are currently post incident something to do with config drive18:21
clarkbdims_: anteaya I do not think this would result in any timeouts, it sa fail fast issue18:21
clarkbso timeouts would be different unless we find more evidence to link them18:22
*** tvanderwiel has joined #openstack-infra18:22
*** puranamr has quit IRC18:22
*** tvanderwiel has left #openstack-infra18:22
sdagueclarkb: sure, I understand the frustration. It's just not an area where this was sorted yet. The fact that infra repartitions the devices is a little different than I think the original expectation of how they'd get used.18:22
dims_thanks clarkb anteaya18:22
*** sabeen3 has joined #openstack-infra18:22
anteayaclarkb: ah I hadn't fully understood the incident, thanks18:22
mordredsdague: we'd prefer user-defined flavors I think :)18:23
fungipersia: yeah, it also doesn't help that disk label already has a much different meaning in other contexts (especially for those versed in *bsd)18:23
mordredsdague: it turns out that cloud providers are terribly at defining useful flavors18:23
*** heyongli has quit IRC18:23
persiafungi: BSD labels are entirely different :)18:23
mordredsdague: because they do things like say "zomg. nobody needs swap"18:23
*** boris-42 has joined #openstack-infra18:23
nibalizerasselin: ok18:23
clarkbmordred: an dmake you consume 30GB of memory to get useable cpu time18:24
mordredclarkb: ++18:24
persiaMy main point was that it would help the conversation to *decide* that we were talking about filesystem labels, rather than "volume labels"18:24
mordredsdague: that said - I do understand why that is the case18:24
*** sabeen has quit IRC18:24
mordredpersia: fwiw, I'm talking about whatever it is that blkid looks at18:24
sdaguemordred: yep, understand. Anyway, I think the label approach has merrits18:24
persiamordred: Excellent.  Be specific when you do :)18:24
mordredsdague: you think steps here are "talk to rax about xen disk labels" then "submit nova spec suggesting it as a contract"?18:25
*** ajmiller_ has joined #openstack-infra18:25
*** esmute has joined #openstack-infra18:25
*** tonytan4ever has joined #openstack-infra18:25
*** ajmiller_ has quit IRC18:25
sdaguemordred: yes18:26
sdagueI'm using "rax" as proxy for "people who work on xenserver"18:26
mordredsdague: sure18:26
fungipersia: but also in the more general sense, "can openstack provide meaningful labels for virtual block devices which can be inspected from within the guest to determine their intended use even when they have not been formatted with a filesystem?"18:26
sdaguewhich I think is nearly a perfect vend diagram at this point18:26
*** heyongli has joined #openstack-infra18:26
mordredsdague: it's also possible that s/rax/(rax|citrix)/18:26
mordredBobBall: ^^ you can fix everyting, right?18:27
lifelessfungi: yes it can18:27
clarkbmordred: that job18:27
* mordred hands BobBall a fluffy bunny18:27
clarkbmordred: its been running long enoug I think it did the right thing but we can inspect logs t oconfirm when it is done18:27
mordredclarkb: it seems as if it has not died18:27
persiafungi: I think the answer to that is "yes", but that we'll get there faster if we pick one class of information (that from libblkid(3) is a good example) that we're attempting to provide.18:27
mordredclarkb: yah18:27
fungilifeless: write a preamble to the beginning of the device before handing it to the guest?18:27
lifelessfungi: yes18:28
openstackgerritPaul Belanger proposed openstack-infra/elastic-recheck: Remove gate-infra-puppet-apply-centos6 check
*** hashar has joined #openstack-infra18:28
sdaguemordred: oh right, citrix is back in it18:28
sdagueso yes18:28
fungipersia: right, i'm thinking there are also cases where there is no partition table (most of my cinder volumes start out this way)18:28
* mordred hands BobBall another fluffy bunny18:28
*** notnownikki has joined #openstack-infra18:28
*** otter768 has joined #openstack-infra18:29
sdague seems to be where the inner change to add an fs_label to the mkfs call would go, and then the callers need to send the right label based on circumstances.18:29
*** hdd has joined #openstack-infra18:29
*** ivar-laz_ has quit IRC18:29
fungiso anyway, i guess getting cinder metadata written as a signature within the beginning of raw volumes would be a nifty solution18:29
*** ivar-lazzaro has joined #openstack-infra18:29
sdaguefungi: I'll say, I highly support the idea of getting more metadata directly in devices. I'd bring that up with some of the cinder folks as well to see what's viable, as I think it's a neat idea18:30
*** tjones1 has joined #openstack-infra18:30
*** tjones1 has left #openstack-infra18:30
fungithough that may be tricky with some solutions where i suppose it's possible that the volume is presented directly by the storage backend and only orchestrated by cinder, so may not have the right access to write to the volume itself18:31
*** heyongli has quit IRC18:31
*** ivar-lazzaro has quit IRC18:31
sdaguefungi: sure, that's why you'd need an expert over there18:31
*** maishsk_ has joined #openstack-infra18:31
fungisomeone with more familiarity with cinder internals, right18:31
*** zns has quit IRC18:31
*** ivar-lazzaro has joined #openstack-infra18:32
*** FallenPegasus has joined #openstack-infra18:32
clarkbsdague: so on the multinode front, you should expect rax nodes t owork and hpcloud nodes to work once d-g is fixed to handle the config drive ordering18:32
sdagueclarkb: ok, neat18:33
sdaguewell they were working this morning18:33
*** heyongli has joined #openstack-infra18:33
*** maishsk has quit IRC18:33
*** maishsk_ is now known as maishsk18:33
*** ddieterly has quit IRC18:33
clarkbthey were working on rax, the problem with hpcoud is without config drive cloud init doesn't set our hostsnmaes so both nodes are "ubuntu" which confuses nova and things break18:34
clarkbonce we have config drive working without exploding d-g we can get hostnames set to unique values and all should be happy18:34
*** shashank_hegde_ has quit IRC18:34
*** jsavak has quit IRC18:34
sdagueoh, well, I hadn't gotten quite that far yet18:34
pabelangerpleia2, when you have some time:
pabelangerand any other +2s would be super :D18:35
*** FallenPegasus has quit IRC18:35
*** kushal has quit IRC18:35
pabelangernibalizer, crinkle ^ you too18:35
*** koolhead17 has joined #openstack-infra18:35
*** hashar has quit IRC18:35
*** annegentle has joined #openstack-infra18:35
openstackgerritSean Dague proposed openstack-infra/devstack-gate: POC: grenade partial multinode
*** ddieterly has joined #openstack-infra18:36
*** ddieterly has quit IRC18:36
*** kushal has joined #openstack-infra18:36
*** juggler has joined #openstack-infra18:37
*** jsavak has joined #openstack-infra18:37
*** charlesztan has joined #openstack-infra18:38
sdagueclarkb: something is dying in the subnode ansible call, I'm trying to make it more verbose to figure out18:38
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel LDAP plugin integration
charlesztanDoes anybody know if Summit ATC passes are given for Devstack commits?18:39
clarkbsdague: it should log everything18:39
clarkboh this is the new command maybe it doesn't18:39
sdague   $ANSIBLE subnodes -f 5 -i "$WORKSPACE/inventory" -m shell \18:40
sdague        -a "cd '$BASE/new/devstack' && sudo -H -u stack stdbuf -oL -eL ./ executable=/bin/bash" \18:40
sdague        > /dev/null18:40
sdaguewas the command18:40
openstackgerritMateusz Matuszkowiak proposed openstack-infra/project-config: Added new SF repository for Fuel LDAP plugin integration
sdagueso I got rid of the dev/null18:40
anteayacrinkle: do you feel like using your -1 on 201584?18:40
*** koolhead17 has quit IRC18:40
clarkbsdague: it should likely redirect to the devstack log for that run18:41
anteayacharlesztan: summit ATC passes are based on openstack commits, yes18:42
anteayacharlesztan: have you a commit and are you wondering if you should be expecting a pass?18:42
crinkleanteaya: I think I just overlooked the conversation where these things were decided, presumably there is good reason for it18:42
*** puranamr has joined #openstack-infra18:42
charlesztananteaya: ty. No commit yet, just wondering as I was told only commits to "core" projects are applicable18:43
clarkbsdague: that way its not nested in the grenade log? maybe that is preferable18:43
*** puranamr has quit IRC18:43
anteayacrinkle: fair enough, I will stand by for answers to your questions prior to reviewing, you are asking good questions18:43
*** puranamr has joined #openstack-infra18:43
*** achanda has quit IRC18:43
sdagueclarkb: oh, right, let me see if there is anything useful18:44
sdagueit died after 9s though, so that's doubtful18:44
anteayacharlesztan: this is the list of projects:
anteayacharlesztan: do commit18:44
fungicharlesztan: commits to any repos maintained by the project-teams listed at get you a 100% discount code for summit registration18:44
*** radez is now known as radez_g0n318:44
anteayacharlesztan: here is a good starting point:
sdagueclarkb: yeh, there is nothing notable on the subnode at this stage18:45
sdagueis there a way to get ansible to be verbose for this?18:45
fungicharlesztan: or in that yaml file anteaya linked (which is what the governance project teams page is generated from)18:45
charlesztananteaya, fungi: ty for helping out a newb18:45
anteayafungi: yours is shinier18:45
anteayacharlesztan: thanks for asking18:45
anteayacharlesztan: starting off on the right foot18:45
clarkbsdague: yes its verbose by default18:46
*** _nadya_ has joined #openstack-infra18:46
clarkbsdague: we redirect to files to keep that sane iirc18:46
fungicharlesztan: though it's only for the initial registration fee. when the price goes up on september 1 the discount code will no longer cover all of the registration cost for a summit pass18:46
sdagueok, we'll see what this run looks like18:46
clarkboh hrm unless we do the redirect within the ansible call and not of the ansible call18:46
clarkbbut I think that may be a hack for our fancy timestamps18:47
openstackgerritMadhusudhan Kandadai proposed openstack-infra/project-config: New jobs for neutron-lbaasv2 scenario tests
*** puranamr has quit IRC18:49
sdaguewell, is the output somewhere I'd be able to find now?18:49
pleia2pabelanger: sure, I'll have a look18:50
asselinnibalizer, thanks. the other change I'd like to merge soon is
clarkbsdague: by default it should go to stdout out after the rc => 0 thing18:50
clarkbsdague: it will just be without fancy timestamps iirc18:50
sdagueok, cool, so hopefully the remove of /dev/null will get me something18:50
nibalizerasselin: looking18:50
*** ParsectiX has quit IRC18:50
sdagueright now it's just failing18:50
*** puranamr has joined #openstack-infra18:50
sdaguewith no output18:51
asselinnibalizer, added a comment18:51
*** pal has quit IRC18:53
openstackgerritJeremy Stanley proposed openstack-infra/bindep: Add an --exit-code option
fungilifeless: ^18:55
*** puranamr has quit IRC18:55
*** jsavak has quit IRC18:57
*** markvoelker has quit IRC18:58
*** annasort has quit IRC18:58
*** Hal has joined #openstack-infra18:58
*** ZZelle_ has joined #openstack-infra18:58
*** Hal is now known as Guest6243318:58
*** FallenPegasus has joined #openstack-infra19:00
jogoclarkb: but it doesn't look like rax job are passing multinode  (or maybe rax isn't being used for multinode?)19:00
*** jsavak has joined #openstack-infra19:00
clarkblikely rax just wasnt used?19:02
*** sabeen3 has quit IRC19:03
*** e0ne has quit IRC19:05
*** koolhead17 has joined #openstack-infra19:05
*** radez_g0n3 is now known as radez19:05
*** shashank_hegde_ has joined #openstack-infra19:05
*** sabeen1 has joined #openstack-infra19:06
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Prototype script to release from the new YAML files
*** Longgeek has quit IRC19:08
clarkbjogo but also possible there are other fails19:08
lifelessfungi: replied, assuming gertty managed to upload19:10
*** fifieldt_ has joined #openstack-infra19:10
*** achanda has joined #openstack-infra19:11
*** fitoduarte has joined #openstack-infra19:11
openstackgerritfumihiko kakuma proposed openstack-infra/devstack-gate: Allow for git base override
*** scheuran has joined #openstack-infra19:13
*** fifieldt has quit IRC19:13
mordredgreghaynes: <-- here's the chunk that defines what a dhcp network looks like in the config-drive19:13
openstackgerritfumihiko kakuma proposed openstack-infra/devstack-gate: post_test_hook is not executed when the gate test returns an error
*** yfried has joined #openstack-infra19:15
*** yfried is now known as yfried|afk19:15
*** Sherif has quit IRC19:16
*** asadoughi has joined #openstack-infra19:16
lifelessquick poll. mordred jeblair clarkb fungi : are we safe to turn off the dependency-links-processing thing in requirements integration checks now ?19:17
mordredlifeless: probably - but can you say the same thing with different words?19:18
*** prad has quit IRC19:19
*** Sherif has joined #openstack-infra19:20
*** hdd has quit IRC19:20
*** esmute has quit IRC19:21
*** prad has joined #openstack-infra19:22
*** Sherif has quit IRC19:23
lifelesshave a look at this failure19:23
lifelessignore the xattr failure19:24
*** pcrews__ has joined #openstack-infra19:24
lifelesswe've got spew about github because dependency links processing has been turned on19:24
lifelessit used to be that we had to have that to install everything19:24
*** gyee has quit IRC19:25
*** bswartz has quit IRC19:25
*** koolhead17 has quit IRC19:26
*** Sherif has joined #openstack-infra19:26
*** bswartz has joined #openstack-infra19:26
*** pcrews_ has quit IRC19:27
*** jsavak has quit IRC19:27
*** hdd has joined #openstack-infra19:28
*** prad has quit IRC19:28
openstackgerritlifeless proposed openstack/requirements: Move integration test dependencies into bindep
fungii think that's a vestige. i'm trying to remember where, but at one point we were explicitly whitelisting package names for --allow-external/--allow-unverified19:29
*** spzala has quit IRC19:30
lifelesswe had some packages that could only be discovered by traversing dependency links19:30
*** jsavak has joined #openstack-infra19:31
*** pvaneck has joined #openstack-infra19:31
*** charlesztan has quit IRC19:31
fungihowever i'm not finding it in the git histories of the repos where i would have thought i remembered adding that19:32
*** gyee has joined #openstack-infra19:34
*** scheuran has quit IRC19:35
*** jsavak has quit IRC19:37
*** jsavak has joined #openstack-infra19:37
lifelessI'll get this use of bindep into trunk19:39
lifelessthen see if it passes with the deplink thing off19:39
*** puranamr has joined #openstack-infra19:39
*** prad has joined #openstack-infra19:40
fungilifeless: do you have a suggestion for changing the "failed to execute" exit code on bindep -b? it seems like it's not straightforward to override unless there's an easy answer i'm missing19:40
fungiwondering whether sys.exit(2) would make more sense for the "some packages not installed" under -b instead19:41
lifelessfungi: I don't have the code paged in, but - generally speaking: raise an exception // return a value that is rich. Then at the CLI handling layer inspect that to decide whether to exit 0/1/219:41
*** sdake_ has joined #openstack-infra19:41
lifelessfungi: some packages not installed is exactly where we want -b to not error19:41
lifelessfungi: isn't it?19:41
fungilifeless: unless --exit-code is passed19:42
lifelessfungi: apt-get install $(bindep -b)19:42
openstackgerritMerged openstack/requirements: Bump oslo.cache version to 0.3.0
lifelessfungi: oh sure, but perhaps if we invert it its a better default19:42
lifelessfungi: I'm not worried about backward compat yet, still early days19:42
lifelessanyone using -b in a script these days is most likely either || true, or not -e19:43
fungilifeless: just seems like a lot less code for a relatively cosmetic difference (i wanted to get this nailed down before we tag 1.0.0 since it's _not_ backwards compatible)19:43
*** markvoelker has joined #openstack-infra19:43
lifelessfungi: if you want me to poke around it, I can probably do so19:43
fungii'll give this a shot with exit 2 getting used as the "executed but some packages not installed" exit code19:45
*** sdake has quit IRC19:45
*** markvoelker has quit IRC19:45
*** markvoelker has joined #openstack-infra19:46
openstackgerritMonty Taylor proposed openstack-infra/glean: Support explicitly configured DHCP interfaces
mordredgreghaynes: when you awaken ^^19:48
mordredclarkb: ^^ you may also be interested19:48
mordredSpamapS, TheJulia: also - since bifrost is making config drives ^^ this is patch one of two you may care about19:48
mordredI'll get the vlan one done next19:48
mordredSpamapS: how much do we care about nic bonding?19:49
jogoclarkb: yeah that is my guess19:50
clarkbin that env its one or two 10Gbe nics right? probably dobt really need bonding19:50
*** ihrachyshka has joined #openstack-infra19:50
mordredclarkb: agree. although I'm tempted to add support for it for completeness as its one of the things possible for the incoming data to express19:50
mordredclarkb: if you haven't read it19:51
clarkbin general I would expect a smaller nic for mgmt that glean spins up then you use somethibg else to configure your 4 bonded 10Gbe nics19:51
mordredclarkb: (and the same data goes into config-drive)19:51
*** ParsectiX has joined #openstack-infra19:51
mordredclarkb: unless you're us, where a management network is a waste19:51
*** salv-orlando has joined #openstack-infra19:51
clarkbso is bonding19:51
mordredfair enough19:52
*** heyongli has quit IRC19:52
mordredI'll leave it alone until someone needs it then19:52
mordred_we_ need vlan suppot19:52
mordredso I'll be doing that19:52
mordredbut luckily that's easy19:52
*** sdake_ has quit IRC19:52
*** annegentle has quit IRC19:53
*** annegentle has joined #openstack-infra19:53
*** heyongli has joined #openstack-infra19:53
*** asadoughi has quit IRC19:54
mordredI say that ... phone call in 5 minutes - will probably wait until after to implement vlan support :)19:54
clarkbbonding is hard for the same reasons getting disk order is hard. we dont have stable device order19:54
clarkbso now you need to do udev too19:54
mordredclarkb: I disagree with you based on the data structure passed in19:55
mordredclarkb: all of it has mac addresses associated with it - which the current code uses to match to requested logical names19:55
mordredso the data is saying "please associate a bond with the two interfaces that have mac X and mac Y"19:56
clarkbmac addrs are not necessarily static either (its crazy)19:56
clarkbthough if bonding hopefully your hardware has real addresses19:56
mordredlets say "I do not care to support you if yuor hardware is super stupid"19:57
*** salv-orlando has quit IRC19:57
*** spzala has joined #openstack-infra19:57
*** heyongli has quit IRC19:58
*** jsavak has quit IRC19:59
mordredin either case - clarkb, pleia2: if you get bored:
mordredcheck my brain logic hole19:59
*** jsavak has joined #openstack-infra19:59
mordredoh! I see bug in test output20:00
*** sdake has joined #openstack-infra20:00
mordred"iface bc:76:4e:05:7b:16 inet dhcp"20:00
*** heyongli has joined #openstack-infra20:00
openstackgerritMonty Taylor proposed openstack-infra/glean: Support explicitly configured DHCP interfaces
mordredthat'll fix that20:01
tqtranhello, when someone is free, can we get a few eyes on ? thanks!20:01
clarkbslight style preference to if elif istead of all the continues but first glance seems reasonable20:02
*** hdd has quit IRC20:05
*** tjones1 has joined #openstack-infra20:05
pleia2the logic seems fine, but I'm missing some background as to why this is eth2 (and what eth1 is here)20:05
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Move into releasetools package
*** jsavak has quit IRC20:06
mordredpleia2: arbitrary interfaces20:06
*** jsavak has joined #openstack-infra20:06
mordredpleia2: we fake them out with those files in sys/*20:06
pleia2mordred: ok, so one we set up as static, and the other dhcp?20:06
pleia2oh ok20:06
mordredpleia2: yah - I mean - the ACTUAL setup is crazypants20:06
pleia2right, I see20:06
pleia2go us20:06
mordredpleia2: next patch I'll be adding a vlan to the mix20:07
mordredor, an interface _on_ a vlan rather20:07
mordredjust what we've always wanted, right?20:07
*** hdd has joined #openstack-infra20:08
*** Longgeek has joined #openstack-infra20:08
clarkbmordred: that lgtm I say we approve the d-g change20:09
clarkband done20:09
*** abregman has joined #openstack-infra20:11
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Fix debug logging lines
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Move into releasetools package
*** Longgeek has quit IRC20:13
*** puranamr has quit IRC20:15
*** puranamr has joined #openstack-infra20:15
*** shashank_hegde_ has quit IRC20:15
openstackgerritSteve Martinelli proposed openstack/requirements: Bump oslo.cache to 0.3.0
*** annegentle has quit IRC20:18
mriedemclarkb: mordred: was there a bug for this?
mriedemwanted to do an e-r query for that20:18
clarkbmriedem: not yet20:18
clarkbit was fixed right away by ninja config change the d-g change juts allows us to move forwrad with config drive again20:19
clarkbmriedem: if you want me to file one for e-r I an do that real quick20:19
*** heyongli has quit IRC20:19
mriedemclarkb: that would be good thanks20:19
*** cdent has quit IRC20:20
*** heyongli has joined #openstack-infra20:20
*** alexsyip has quit IRC20:21
mriedemclarkb: nvm
openstackLaunchpad bug 1475012 in OpenStack-Gate "hpcloud: node setup fails due to: *** /dev/vdb appears to already be mounted" [Undecided,New]20:22
clarkbmriedem: gah derp20:23
openstackLaunchpad bug 1475013 in OpenStack-Gate "/dev/vdb is already mounted" [Critical,In progress] - Assigned to Clark Boylan (cboylan)20:23
mriedemduped it :)20:23
*** tonytan4ever has quit IRC20:25
*** shashank_hegde_ has joined #openstack-infra20:25
openstackgerritMatt Riedemann proposed openstack-infra/elastic-recheck: Add query for hpcloud config-drive bug 1475012
openstackbug 1475012 in OpenStack-Gate "hpcloud: node setup fails due to: *** /dev/vdb appears to already be mounted" [Undecided,In progress] - Assigned to Monty Taylor (mordred)20:26
*** alexsyip has joined #openstack-infra20:26
*** gordc has quit IRC20:27
*** cody-somerville has joined #openstack-infra20:27
*** cody-somerville has quit IRC20:27
*** cody-somerville has joined #openstack-infra20:27
*** dustins has quit IRC20:28
*** dkranz has quit IRC20:28
*** dustins has joined #openstack-infra20:28
*** gordc has joined #openstack-infra20:28
*** pblaho has joined #openstack-infra20:30
*** tjones1 has left #openstack-infra20:30
*** FallenPegasus has quit IRC20:32
*** alivigni has quit IRC20:33
*** FallenPegasus has joined #openstack-infra20:33
*** tonytan4ever has joined #openstack-infra20:34
*** jsavak has quit IRC20:34
*** jsavak has joined #openstack-infra20:35
jeblairmordred: what's this?20:35
jeblair| 91a2504b-dfab-41df-beaf-c4a5696853af | gw-out | ACTIVE | -          | Running     | Ext-Net= |20:35
*** esmute has joined #openstack-infra20:36
*** annegentle has joined #openstack-infra20:36
*** Guest62433 has quit IRC20:36
*** jasondotstar has quit IRC20:37
*** cbader has joined #openstack-infra20:38
openstackgerritMerged openstack/requirements: Bump oslo.log to 1.6.0 version
openstackgerritDavid Shrewsbury proposed openstack-infra/shade: Add flavor admin support
*** krtaylor has quit IRC20:40
openstackgerritDoug Wiegley proposed openstack-infra/project-config: Add repo for Horizon panels for Neutron LBaaS
*** FallenPegasus has quit IRC20:42
lifelessjeblair: I'm seeing 5-7 second pauses in gertty when exiting comment fields  in reviews20:42
lifelessjeblair: have you seen that ?20:42
jeblairlifeless: no, that certainly would annoy me enough to debug; any idea what's going on during that time? (debug logs, etc?)20:43
lifelessjeblair: no idea20:44
*** jsavak has quit IRC20:44
lifelessjeblair: I do know that gertty is going into error a lot, and I see long pauses elsewhere20:44
lifelessjeblair: but i haven't had time to debug; I just switch between trying it again and gerrit20:44
jeblairlifeless: _generally_ pauses should be due to db locks, which should have minimal (or no) overlap with network ops20:44
*** jsavak has joined #openstack-infra20:45
*** ihrachyshka has quit IRC20:46
*** jasondotstar has joined #openstack-infra20:46
jeblairmordred: and does ovh have cinder?20:47
*** dboik has quit IRC20:47
*** notnownikki has quit IRC20:48
*** bswartz has quit IRC20:49
asselinproject-config changes are backing up in zuul20:50
asselingate-openstackci-beaker-trustyqueued (non-voting)20:50
asselinthese jobs are queued fr quite a while20:50
jeblairasselin: there's a 1.2k job backlog right now so that's not surprising.20:51
jeblairasselin: or do you think these jobs are staying in the queue longer than others?20:52
asselinjeblair, where do you see that?20:52
asselinwell they're all at the top of the list20:52
jeblairasselin: red line on 'zuul job queue' graph20:52
*** krtaylor has joined #openstack-infra20:52
asselintop 5-6 oldest patches are all waiting for that job20:52
Shrewsjeblair: time to start rate-limiting developers!20:53
*** dboik has joined #openstack-infra20:53
jeblairasselin: ah, and have been for 2+ hours, so that is suspicous20:53
jeblairShrews: nah, this is not a big backlog, and we're going to get 2 new clouds :)20:54
* Shrews offers to gallantly do his part and stop working20:54
anteayaShrews: taking one for the team20:55
Shrewsanteaya: ever the modest hero20:55
anteayatweet your modesty20:55
anteayasome may have missed it20:55
*** FallenPegasus has joined #openstack-infra20:56
*** pc_m is now known as pcm_m_AWAY20:56
asselinjeblair, which link do you use to see 'zuul job queue' graph?20:57
asselinok I was always going here:
*** ParsectiX has quit IRC20:59
*** ildikov has joined #openstack-infra21:00
*** jpeeler has quit IRC21:02
jeblair    if 'trusty' in
jeblair        params['ZUUL_NODE'] = 'bare-trusty'21:02
jeblairthe job has been reconfigured to only be able to run on devstack-trusty, but zuul wants to run it on bare-trusty because of the job name21:03
*** puranamr_ has joined #openstack-infra21:03
*** thorst has quit IRC21:03
clarkbyou need to dsvm21:04
jeblairi need to zuulv321:05
*** FallenPegasus has quit IRC21:05
*** puranamr has quit IRC21:05
*** FallenPegasus has joined #openstack-infra21:05
*** unicell has quit IRC21:06
clarkbthat too21:06
*** bknudson has quit IRC21:06
*** dims_ has quit IRC21:07
*** dprince has quit IRC21:07
*** jasondotstar has quit IRC21:09
*** gyee has quit IRC21:09
openstackgerritJames E. Blair proposed openstack-infra/project-config: Rename openstackci beaker jobs to include dsvm
jeblairasselin, clarkb: we'll need to force-merge that21:09
*** julim has quit IRC21:10
*** shashank_hegde_ has quit IRC21:10
openstackgerritMerged openstack/requirements: Bump version of pyOpenSSL to >= 0.14
clarkbianw ^ did they fix the memory leaks?21:11
openstackgerritAssaf Muller proposed openstack/os-testr: Make regex path a positional argument in addition to --regex
openstackgerritMerged openstack-infra/elastic-recheck: Add query for hpcloud config-drive bug 1475012
openstackbug 1475012 in OpenStack-Gate "hpcloud: node setup fails due to: *** /dev/vdb appears to already be mounted" [Undecided,In progress] - Assigned to Monty Taylor (mordred)21:14
*** rm_work is now known as rm_work|away21:14
*** ldnunes has quit IRC21:15
*** fawadkhaliq has joined #openstack-infra21:15
*** ayoung has quit IRC21:15
*** rossella_s has quit IRC21:15
*** dustins has quit IRC21:17
*** vponomaryov1 has left #openstack-infra21:17
*** rm_work|away is now known as rm_work21:20
jeblairclarkb: local tests of that patch pass, so i'm going to go ahead and merge it21:20
openstackgerritMerged openstack-infra/project-config: Rename openstackci beaker jobs to include dsvm
*** pcm_m_AWAY is now known as pc_m21:21
davideagnellowe have a very small change in our Rally CI job that we would like merged, can someone please take a look:   thank you21:22
*** mrmartin has quit IRC21:22
openstackgerritwill soula proposed openstack-infra/jenkins-job-builder: Add Ability to use the AWS Cloudformation Plugin
*** peristeri has quit IRC21:23
clarkbjeblair sounds good21:24
*** FallenPegasus has quit IRC21:26
*** FallenPegasus has joined #openstack-infra21:26
*** dboik_ has joined #openstack-infra21:26
*** mrunge has quit IRC21:26
*** xyang0 has quit IRC21:27
ianwclarkb: it was really pycparser, who merged some memory reduction21:27
openstackgerritMerged openstack-infra/puppet-openafs: Add support for CentOS6
ianwclarkb: i don't suppose there is much to be done, but i'm pretty sure all our nodes are still making efforts to install packaged versions21:29
clarkbwell the lower end isnt met by packages now21:29
clarkbso it should upgrade21:29
*** dboik has quit IRC21:30
*** sdake_ has joined #openstack-infra21:30
*** sdake_ has quit IRC21:30
ianwyeah, just something that could be optimised away21:30
*** Sukhdev has joined #openstack-infra21:30
*** sdake_ has joined #openstack-infra21:30
fungii expect our minimal images to take care of that21:31
*** dboik_ has quit IRC21:31
*** _nadya_ has quit IRC21:31
fungii mean, they don't have ping installed, so the chances of having pycparser are slim ;)21:31
*** FallenPegasus has quit IRC21:31
*** _nadya_ has joined #openstack-infra21:32
*** FallenPegasus has joined #openstack-infra21:32
ianwjust cross your fingers and hope nothing goes wrong inside that cffi -> pycparser chain.  debugging that is not my idea of a good time :)21:32
*** FallenPegasus has quit IRC21:32
*** sdake has quit IRC21:33
EmilienMhey infra, I'm trying to create a jenkins job macro similart to zuul-git-branch-prep, but instead of checkouting master, I want to checkout the latest stable branch before the patchset21:34
openstackgerritMerged openstack-infra/puppet-askbot: Enable arrow alignment lint check
*** FallenPegasus has joined #openstack-infra21:34
EmilienMjeblair: it's about we talked yesterday ^21:34
*** _nadya_ has quit IRC21:36
*** akshai has quit IRC21:36
*** salv-orlando has joined #openstack-infra21:36
*** alexsyip has quit IRC21:36
*** Swami has quit IRC21:37
jeblair(i think we need docstrings for those builder macros)21:37
EmilienMthat means it's already existing :)21:37
jeblairEmilienM: i think the main things you'd need to change are the 'dest' in the clonemap (so that you can call it twice with two different destinations, it should probably be a substitution var)21:38
EmilienMjeblair: I guess I can use I guess I can use zuul-git-branch-prep21:38
*** shashank_hegde_ has joined #openstack-infra21:38
*** mriedem is now known as mriedem_away21:38
EmilienMjeblair: well, not only21:38
crinklepabelanger: nibalizer!topic/puppet-dev/4gZmQ_28gYU21:38
*** sabeen1 has quit IRC21:38
EmilienMjeblair: I need to define the name of the latest stable branch21:39
fungiin theory (this might be venturing into zuul v3 territory!) we could pass in a clonemap as a parameter ;)21:39
jeblairEmilienM: yep; did you want to do that automatically?21:39
EmilienMjeblair: when the commit is on master, it's easy21:39
EmilienMjeblair: yes21:39
EmilienMjeblair: but when you are somewhere in the tree, it's not easy21:39
openstackgerritMerged openstack-infra/puppet-asterisk: Enable arrow alignment lint check
jeblairEmilienM: yeah, then i think you want to add a file to project-config/jenkins/scripts to do that21:39
nibalizercrinkle: we are unpinned right?21:40
crinklenibalizer: yep21:40
jeblairEmilienM: so something vaguely like "zuul-cloner master; run that script to get the old branch, then zuul-cloner the old branch"21:40
EmilienMjeblair: I thought I could do a macro but even for a script I need to figure out how to do21:40
*** Sherif has quit IRC21:40
openstackgerritMerged openstack-infra/puppet-bandersnatch: Enable arrow alignment lint check
*** FallenPe_ has joined #openstack-infra21:40
*** FallenPegasus has quit IRC21:40
EmilienMjeblair: the #2 is what I'm looking for21:40
*** Sherif has joined #openstack-infra21:41
openstackgerritMerged openstack-infra/puppet-bugdaystats: Make git_url parameter useful
jeblairEmilienM: yeah, that's going to be a few git commands and some code to figure that out from the available branches21:41
openstackgerritMerged openstack-infra/puppet-bugdaystats: Add Gemfile and puppet 4 checks
EmilienMjeblair: ok. So rather a script than a macro, right?21:42
*** Sherif has quit IRC21:42
jeblairEmilienM: yeah, i think it's going to be enough code to warrant a script.  if it ends up being 10 lines, we can make it a macro :)21:43
*** sdake has joined #openstack-infra21:43
EmilienMjeblair: ack21:43
*** gordc has quit IRC21:43
*** radez is now known as radez_g0n321:43
*** bnemec has quit IRC21:43
*** alexsyip has joined #openstack-infra21:43
fungispeaking of zuul-git-prep, it's working now!
*** annegentle has quit IRC21:44
*** crc32 has quit IRC21:44
*** crc32 has joined #openstack-infra21:44
*** Sherif has joined #openstack-infra21:44
*** annegentle has joined #openstack-infra21:44
mordredjeblair: unsure what that is ... I'm guessing perhaps a test vm from one of the cloud admins - their names are all jd21:45
openstackgerritSukhdev Kapur proposed openstack-infra/irc-meetings: Fixing Chair person infromation for this meeting
*** sdake_ has quit IRC21:46
jeblairyolanda, nibalizer: i have a query on based on usage in
*** shardy has quit IRC21:47
*** Sherif has quit IRC21:48
*** jsavak has quit IRC21:48
mordredjeblair: also - no cinder yet - let me check with the guy on if we can spin up one of the instances they have with a lot of disk on it (I know some of them are actual full machines and I can't remember which right now)21:48
jeblairmordred: ok21:48
mordredgah. sorry21:49
mordredjeblair: looks like we're using 176G on pypi.dfw.openstack.org21:49
openstackgerritMerged openstack-infra/puppet-bup: Add Gemfile and puppet 4 checks
openstackgerritTerry Wilson proposed openstack/requirements: Avoid envirnoment marker parsing error
*** dkranz has joined #openstack-infra21:49
*** jamielennox|away is now known as jamielennox21:49
openstackgerritMerged openstack-infra/puppet-iptables: Use service-name netfilter-persistent for Vivid
*** annegentle has quit IRC21:49
*** rm_work is now known as rm_work|away21:50
nibalizerjeblair: i agree that ensurability on the logrotate::file resource is the correct approach21:50
jeblairnibalizer: cool21:50
nibalizerlogrotate::fileremoval is there more to handle a transition case because we were going to change the configuration that logrotate::file was going to use21:51
nibalizerinstead of /etc/logrotate.d/`basename $file` it would make a /etc/logrotate.d/_path_to_file21:51
jeblairnibalizer: but that's no longer the case?21:51
*** Sherif has joined #openstack-infra21:51
*** mjturek1 has joined #openstack-infra21:51
nibalizeri just dont have it in my head any more21:51
nibalizerI think I wrote the patch to do full paths thinking the bug in the logrotate::file resource was more severe21:52
*** hdd has quit IRC21:52
*** Swami has joined #openstack-infra21:52
nibalizerand then we discovered it wasn't severe, but a nice to have anyways21:52
*** rm_work|away is now known as rm_work21:52
nibalizeroh it landed21:53
jeblairnibalizer: oh, are you saying that currently logrotate::file will use /_path_to_file, but at some point it did not, so we need a way to clean old versions that used just /file ?21:53
* jeblair looks21:53
SukhdevHi there Infra folks - I made a change to the one of the IRC meetings that I chair - please have a review of this -
nibalizerjeblair: thats the reasoning behind yolandas change to add logrotate::fileremoval yes21:54
*** puranamr_ has quit IRC21:54
jeblairSukhdev: there is not a general need to request reviews on a change21:54
Sukhdevjeblair: Oh I see - so, what is the procedure then?21:55
jeblairSukhdev: wait for someone to review it.  if you like, you can review changes yourself while you are waiting :)21:55
*** fawadkhaliq has quit IRC21:55
*** earlephilhower_ has quit IRC21:56
*** puranamr has joined #openstack-infra21:57
*** cody-somerville has quit IRC21:57
*** jaypipes has quit IRC21:57
*** crc32 has quit IRC21:57
*** Longgeek has joined #openstack-infra21:57
nibalizerjeblair: so yea it actually takes an ensure21:58
jeblairnibalizer: well, now i'm stuck on your change (189378); why was that necessary? :)21:59
Sukhdevjeblair: I thought it would require a vote from an Infra core, no?21:59
fungijeblair: mordred: clarkb: lifeless: milestone! nova unit tests on our minimal ubuntu-trusty, checked out via zuul-cloner, databases configured and system package deps fulfilled at job run-time
nibalizerjeblair: if you had /var/log/thing/debug.log and /var/log/otherthing/debug.log22:00
*** crc32 has joined #openstack-infra22:00
jeblairSukhdev: it requires votes from the people who review that repository.  generally, reviewers use gerrit or gertty to find changes to review.  most of us have about a 2-4 week backlog of about 200 changes.  if your change is etremely urgent, you are welcome to make a case for us to look at it out of order.  otherwise, it would be most helpful for you to review changes so that you can join the core team and address the backlog.  :)22:00
nibalizeralso, out of curiosity, would you mind hopping on a server and seeing if we've grown extra logrotate configuration in /etc/logrotate.d ?22:01
lifelessfungi: bindep ftw?!22:01
nibalizerIm expecting to see one now unmanaged file and one new managed file22:02
jeblairnibalizer: wouldn't you just sa "logrotate::file {'thing':}" and "logrotate::file {'otherthing':}"?22:02
Sukhdevjeblair: I would be happy to do it - presently I do not have such privilege to do so.22:02
jeblairnibalizer: what server should i check?22:02
*** pblaho has quit IRC22:02
jeblairSukhdev: anyone is welcome to review any repository.  see
*** patrickeast has quit IRC22:03
*** pblaho has joined #openstack-infra22:03
*** krtaylor has quit IRC22:03
nibalizerjeblair: the gerrit server22:03
*** earlephilhower has joined #openstack-infra22:03
nibalizerwell maybe not22:04
jeblairnibalizer: we have manage_projects.log, _var_log_jeepyb_gerritfetchremotes.log, gerrit-rotate22:04
jeblair(among other more standardish things)22:04
Sukhdevjeblair: I am aware of the review process - however, I do not have authority to give a +2 and my understanding is that a review requires +2 for the merge22:04
nibalizerokay so it looks like it has managed to not duplicate things22:05
jeblairSukhdev: indeed, but +/-1 reviws are helpful in that they can find problems before a core reviewer gets to the change, and that is how people are nominated for core teams22:05
*** dizquierdo has joined #openstack-infra22:05
Sukhdevjeblair: got it - understood22:05
*** achanda has quit IRC22:06
nibalizerjeblair: so what prompted me to make that change was I expected  logrotate::file { '/var/log/thing/thing.log': } to do the right thing22:06
*** achanda has joined #openstack-infra22:06
fungijeblair: mordred: clarkb: lifeless: job runtime was just under 17 minutes, time spent installing our everything in our bindep fallback list was under 2 minutes, though i'm not convinced our package caching is working correctly. i need to dig deeper on that front22:06
jeblairnibalizer: there's to parameters though, the name and the log, right?22:06
nibalizerbut but because it tries to create /etc/logrotate.d/${name}22:06
*** signed8bit has joined #openstack-infra22:06
lifelessfungi: my requirements change is +2 ready I think22:06
nibalizerthat didn't work22:06
lifelessfungi: it failed dsvm neutron, but thats known flaky22:07
jeblairnibalizer: yeah, i actually expect _that_ to be the right thing -- i'm telling it to create a logrotate config file, and part of that config file is the log specification22:07
nibalizerand thats reasonable, I think22:08
jeblaircould have better docs for sure (which you added!)22:08
mordredfungi: WOOT22:09
clarkbdid bindep install ping ;)22:09
*** tonytan4ever has quit IRC22:09
*** mjturek1 has quit IRC22:09
*** crc32 has quit IRC22:09
*** pblaho has quit IRC22:09
*** signed8bit_ZZZzz has quit IRC22:10
* mordred pings clarkb22:11
*** puranamr has quit IRC22:12
openstackgerritJames E. Blair proposed openstack-infra/puppet-logrotate: Sketch an alternate API
jeblairnibalizer: what do you think about that ^?22:12
* jeblair hopes mordred does not notice that i just uploaded a change with a comment instead of implementing code22:14
*** Sherif has quit IRC22:14
mordredjeblair: BRILLIANT22:14
nibalizerjeblair: ya im oaky with it22:15
nibalizerI've made the snippet you need too22:15
*** patrickeast has joined #openstack-infra22:15
*** krtaylor has joined #openstack-infra22:15
jeblairoh thanks, i was still searching stackoverflow ;)22:16
*** Longgeek has quit IRC22:16
jeblairand indeed, i had just found the 'in' operator, so cool :)22:16
*** tjones1 has joined #openstack-infra22:17
*** zul has quit IRC22:17
mordredjeblair: something tells me you've done more stackoverflowprogramming related to puppet ...22:20
openstackgerritJames E. Blair proposed openstack-infra/puppet-logrotate: Require logrotate::file names not be paths
jeblairnibalizer: ^22:20
fungiclarkb: i think nova unit tests don't care whether ping exists22:21
jeblairClint: i'm not sure i understand your comment22:21
openstackgerritJames E. Blair proposed openstack-infra/puppet-logrotate: Require logrotate::file names not be paths
jeblairnibalizer: now with co-authored-by ;)22:21
nibalizerjeblair: ooh goodie22:21
*** rickbra has joined #openstack-infra22:21
nibalizerjeblair: okay dont shoot me but22:21
nibalizerwhat if we called it logrotate::config22:22
Clintjeblair: i think i utterly confused myself just now22:22
fungilifeless: cool, i'll try to take a look at the reqs change later tonight. i'm transitioning into take-more-carloads-of-boxes-to-the-new-house mode for the next several hours of my evening but will be on later i think22:22
nibalizerlike I know its a config /file/ but I think the reason I expected logrotate::file { '/path/to/file': } to work is related to that22:22
*** EmilienM is now known as EmilienM|off22:22
mtreinishjeblair: btw, if you were serious about helping with the effor to stop using oslo.db in subunit2sql I wrote up some brief notes here:
jeblairnibalizer: wfm?  it doesn't seem to set of the alarm in my bikeshed22:22
*** doude has quit IRC22:23
*** amitgandhinz has quit IRC22:23
mordredjeblair, nibalizer: I WANT IT PAINTED PURPLE!!!22:23
*** Sherif has joined #openstack-infra22:23
*** crc32 has joined #openstack-infra22:23
nibalizerya I wouldn't say I feel deeply about this issue but if the resource title is 'logrotate configuraton' then i know im passing some parameters into it22:24
*** FallenPe_ has quit IRC22:24
nibalizerwhen its logrotate::file I want to just tell it my file and have it get it done22:24
*** pblaho has joined #openstack-infra22:24
nibalizerbut really its not a big deal22:24
nibalizerand we'd have to change everywhere in our infra that uses ::file22:24
nibalizerso maybe just leave it ::file22:24
mtreinishmordred: why not paint it black?22:24
jeblairmtreinish: yeah, i might be able to help with some of the sqla stuff, but i'll need to understand what oslo.db is doing in order to undo that, which will be the hard/time-consuming part22:24
mtreinishjeblair: so what I tried to outline there is the actual sqla parts of the migration are the easy parts22:25
jeblairnibalizer: your call22:25
openstackgerritMerged openstack-infra/puppet-gerrit: cron to clear gerrit logs
mtreinishbecause what subunit2sql is doing is relatively simple22:25
mtreinishit's just get a session and do a query22:25
jeblairmtreinish: i agree, though nothing either of them strikes me as easy22:26
mtreinishwhich luckily the oslo.db parts are pretty thin22:26
*** doude has joined #openstack-infra22:26
jeblairmtreinish: yeah, it's just unwraping the facadefactoryconstructors or whatever22:26
mordredmtreinish: because it's not currently red22:26
*** tonytan4ever has joined #openstack-infra22:26
mordredjeblair: you know? nobody talks about aspects anymore22:26
nibalizerjeblair: ps3 looks fine to me22:26
jeblairmordred: aop.  time machine.22:27
*** patrickeast has quit IRC22:27
nibalizerleaving a +1 only because Im kind of caught up in the moment here and would benefit from letting some time pass22:27
mtreinishjeblair: hah, I was just thinking we could replace it with a straight sessionmaker() call and pretend that facade stuff didn't ever exist :)22:28
mtreinishand cross our fingers that it doesn't break anything22:28
pleia2clarkb: when you have a moment, I think I caught the only issue in the first patchset, but it would be nice to have you look at this one to make sure it won't cause problems for our kibana instance
nibalizersometimes i get all caught up in the energy of these conversations and the critical thinking part of my brain doesn't fire loud enough22:28
pabelangercrinkle, \o/22:28
*** crc32 has quit IRC22:29
lifelessfungi: cool22:29
*** ivar-laz_ has joined #openstack-infra22:30
*** doug-fish has left #openstack-infra22:30
*** crc32 has joined #openstack-infra22:31
*** jsavak has joined #openstack-infra22:31
*** crc32 has quit IRC22:31
*** aysyd has quit IRC22:31
SpamapSzigo: dunno if you saw the announcement, but pysaml2 is getting a major version bump and official py3.4 support! :)22:31
*** patrickeast has joined #openstack-infra22:32
SpamapSzigo: or rather, _GOT_ a bump
zigoSpamapS: Yeah ! \o/22:33
mtreinishmordred: heh, fair enough22:33
*** crc32 has joined #openstack-infra22:33
zigoSpamapS: I knew py34 support was in, and I was waiting for the release.22:33
zigoSpamapS: Thanks for all the hard work.22:33
*** ivar-lazzaro has quit IRC22:33
SpamapSzigo: it's even in Travis CI so in theory it won't backslide. :-D22:33
SpamapSzigo: it was fun work to be honest. :)22:33
* zigo is going to package 3.0.0 in Debian Experimental, together with the rest of Liberty.22:33
pleia2jeblair: thanks22:35
mordredSpamapS: yay!22:35
*** bswartz has joined #openstack-infra22:36
SpamapSI believe that means keystone's deps are all py3422:36
*** signed8bit is now known as signed8bit_ZZZzz22:37
*** EmilienM|off is now known as EmilienM22:37
asselinjeblair, fungi I know it's getting late in the day, but can we merge these? &
openstackgerritMerged openstack-infra/puppet-jenkins: Add dpkg architeture facts
asselinI can merge the first, but the 2nd needs some more cores to look at22:38
jeblairasselin: yeah, give me a few mins22:39
jeblairEmilienM: okay i could see that as a macro :)22:39
jeblairEmilienM: oh, but we actually don't have a way of passing that information between builders22:39
jeblairEmilienM: unless you write it to a file or something22:39
* mordred is so happy that is still not landed22:40
jeblairEmilienM: so maybe what we need is one nice long macro or script that does that as well as both of the checkouts22:40
mordredjeblair: (or zuulv3)22:40
EmilienMjeblair: hum. I'll go for a script22:40
jeblair2015-07-15 20:50:15.717 |   Could not resolve ''22:40
mordredjeblair: of course22:41
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: Move into releasetools package
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move list-repos into the releasetools project
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move commands into their own package
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move into the releasetools package
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move sanity-check-version into releasetools package
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move to milestone-rename
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move to milestone-ensure
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move to milestone-close
openstackgerritDoug Hellmann proposed openstack-infra/release-tools: move to milestones-create
*** dangers is now known as dangers_away22:42
*** yamahata has quit IRC22:42
*** shashank_hegde_ has quit IRC22:43
jeblairasselin: +3 +222:44
*** pc_m has quit IRC22:44
mordredlifeless: re: your email22:45
*** prad has quit IRC22:45
mordredlifeless: when I was tracking down something the other day - I noticed another tox-ism that we may want to think about once "pip install .[test]" is possible22:45
mordredlifeless: which is that tox does two passes - one to install requirements.txt and test-requirements.txt - and then one to install .22:46
openstackgerritMerged openstack-infra/puppet-jenkins: Change installation of the JDK to use the architecture of the host
mordredwhich of course pulls in requirements.txt22:46
*** dkranz has quit IRC22:46
lifelessmordred: well its either install -e ., or its sdist + install-the-sdist22:46
lifelessmordred: install . by itself never happens :)22:46
mordredlifeless: but it certainly does two phases22:47
lifelessone complication with .[test] is that in use_develop mode we'll have develop installed and normal installed22:47
lifelessgo on22:47
mordredjust saying - we might want think about how we can express what we want "please just install .[test]" to tox22:47
openstackgerritMerged openstack-infra/puppet-kibana: Add some missing dependencies
openstackgerritMerged openstack-infra/system-config: Fixes in git_backend class
jeblairclarkb: your presence is requested on at your convenience.22:48
mordredlifeless: I don't have good suggestions yet22:48
lifelessmordred: ah yeah. SO I think holger might not want that at all22:49
mordredlifeless: but I think what we're curently expressing to tox is not 100% what we _want_ to express22:49
lifelessmordred: OTOH I think it doesn't matter22:49
mordredlifeless: that is always a possibility - as is that22:49
lifelessmordred: so - what defects are you worried about with the 2 stages22:49
*** heyongli has quit IRC22:49
*** shashank_hegde_ has joined #openstack-infra22:50
openstackgerritMerged openstack-infra/puppet-log_processor: Start processes after network and named
openstackgerritMerged openstack-infra/puppet-log_processor: Retry on EAI_AGAIN name resolution failures
openstackgerritMerged openstack-infra/puppet-log_processor: Lazily connect to logstash
*** dims_ has joined #openstack-infra22:51
*** sigmavirus24 is now known as sigmavirus24_awa22:51
*** heyongli has joined #openstack-infra22:51
mordredlifeless: dunno. just noticed a second processing of the requirements.txt list which was pretty unnecessary22:51
*** xyang1 has joined #openstack-infra22:52
openstackgerritMonty Taylor proposed openstack-infra/glean: Support explicitly configured DHCP interfaces
mordredif helps ^^ when you actually git add new files22:53
mordredSpamapS: I'm assuming if I make a vlan support patch for glean, you should be able to test it ish, yeah? OR - we need to get TheJulia to add vlan output support in bifrost don't we?22:54
lifelessmordred: ok, so yeah - the possibility of flip-flop resolution is there, but USE_CONSTRAINTS will save us22:54
mordredlifeless: sweet22:54
mordredall hail USE_CONSTRAINTS22:54
openstackgerritPaul Belanger proposed openstack-infra/project-config: Add puppet-check-jobs for puppet-openstack-integration master
pabelangercrinkle, EmilienM ^22:55
*** efried has quit IRC22:55
pleia2pabelanger: feel free to ping me again once they chime in, I just wanted to see agreement from our puppet friends :)22:56
pabelangerpleia2, roger, looks like we are still discussing22:56
* pleia2 nods22:56
mordredpleia2: "puppet friends" makes me think of the mordred and jeblair puppets which makes me smile22:57
openstackgerritMerged openstack-infra/puppet-openstackci: Move jenkins job builder into jenkins_master
*** dizquierdo has quit IRC23:00
*** mestery_ has joined #openstack-infra23:00
*** dizquierdo has joined #openstack-infra23:00
*** achanda_ has joined #openstack-infra23:02
pleia2morning jhesketh23:02
*** dimtruck is now known as zz_dimtruck23:02
*** achanda has quit IRC23:02
pabelangerpleia2, okay, I think we have +1 from puppet folks23:03
*** FallenPegasus has joined #openstack-infra23:03
*** EmilienM is now known as EmilienM|off23:03
*** mestery has quit IRC23:03
pleia2cool, having a look now23:03
lifelessNakato: tchaypo: do you folk need anything from me?23:03
lifelessNakato: tchaypo: I have today and tomorrow to be tz compatible, then I'm all over the place for ~ 3 weeks23:04
*** shashank_hegde_ has quit IRC23:04
openstackgerritMerged openstack-infra/puppet-mailman: Fix variable access warnings
*** pblaho has quit IRC23:06
*** jasondotstar has joined #openstack-infra23:07
*** shashank_hegde_ has joined #openstack-infra23:07
*** tjones1 has quit IRC23:07
*** puranamr has joined #openstack-infra23:09
*** jasondotstar has quit IRC23:10
*** annegentle has joined #openstack-infra23:10
*** puranamr has quit IRC23:11
*** annegentle has quit IRC23:15
*** fitoduarte has quit IRC23:18
*** ZZelle_ has quit IRC23:24
*** annegentle has joined #openstack-infra23:26
*** rbrooker has quit IRC23:27
*** dizquierdo has quit IRC23:28
*** pblaho has joined #openstack-infra23:30
*** Sukhdev has quit IRC23:30
*** tonytan4ever has quit IRC23:31
*** gildub has joined #openstack-infra23:34
openstackgerritK Jonathan Harker proposed openstack-infra/puppet-kibana: Add kibana3 support
openstackgerritK Jonathan Harker proposed openstack-infra/puppet-kibana: Refactor kibana-ruby specific bits to kibana::ruby
*** tonytan4ever has joined #openstack-infra23:39
*** doude has quit IRC23:40
*** doude has joined #openstack-infra23:40
*** doude has quit IRC23:40
*** doude has joined #openstack-infra23:40
*** oomichi has joined #openstack-infra23:40
*** thorst has joined #openstack-infra23:41
lifelessmordred: so hey23:41
mordredlifeless: sup?23:41
lifelessmordred: python install23:41
lifelessmordred: doesn't create ChangeLog23:41
lifelessmordred: this breaks building docs in readthedocs (if you want the changelog in  your docs)23:41
mordredlifeless: interesting23:42
lifelessmordred: ChangeLog is triggered by sdist, and by build_sphinx23:42
lifelessmordred: I'd like to suggest that egg_info should probably create it, because egg_info has a manifest23:42
lifelessmordred: and we're actually making different manifests depending on call sequence, today.23:42
lifelessmordred: thoughts?23:42
mordredwell... except egg_info gets called a lot in day to day operations people do on things with tox, no?23:45
lifelessmordred: tox calls sdist23:45
mordredno it does not23:45
mordredwe do use_Develop in all openstack projects23:45
lifelessmordred: yeah, it does - it makes a .zip and installs that23:45
lifelesswe don't23:45
lifelesswe did23:45
lifelessits now mixed23:45
lifelessall the namespaced projects disabled use_develop23:46
lifelessbecause it was utterly broken23:46
mordredright. but then namespaces got removed23:46
mordredbecuase it's actually namespaces that are broken23:46
lifelessnot arguing that23:46
mordrednova uses it23:46
mordredwhich is actually my test case in mind23:46
mordredthe nova devs are touchy about making operations in tree take longer23:47
mordredso if tox -epep8 has to generate a changelog, they're going to complain23:47
mordred(beacuse they have before)23:47
lifelessso the changelog was under a second23:47
lifelessI just timed it23:47
mordredthis is not saying I think it's a bad idea23:47
lifelessauthors was a little longer23:47
mordredI'm just giving data points of where people have complained before23:47
lifelesssdist as a whole was 11 seconds23:47
openstackgerritDarragh Bailey proposed openstack-infra/jenkins-job-builder: Support copyartifact build selector param
lifelessI'm trying to quantify that23:47
mordred11 seconds will draw howls of complains23:47
mordred11 seconds will draw howls of complaints23:47
lifelesswight, thats sdist23:48
lifelessauthors and changelog are a lot faster. let me instrument more23:48
mordredso - success is somewhere in the middle :)23:48
lifeless[pbr] Writing ChangeLog23:51
lifeless[pbr] Generating ChangeLog23:51
lifeless[pbr] ChangeLog complete (0.9s)23:51
lifelesstiming AUTHORS now23:51
mordredlifeless: (tl;dr - I'm perfectly fine doing both generations where we need to for happiness ... just want to make sure we don't thousand-cuts ourselves to unahppy nova devs)23:52
*** thorst has quit IRC23:52
lifelesswe could export SKIP_GENERATE_* for tox too perhaps23:52
mordredlifeless: also - we should check if that's going to get added to "tox -evenv -- nova.tests.TestFoo.foo_bar"23:52
*** esmute has quit IRC23:52
*** esmute has joined #openstack-infra23:53
lifeless[pbr] Generating AUTHORS23:53
lifeless[pbr] AUTHORS complete (1.7s)23:53
lifelessmordred: the output ?23:53
mordredno - the time23:54
mordredbecause 2s isn't much for a full run - it's a lot for a single test23:54
lifelesstrying now23:54
tchaypolifeless: nothing from me, I don’t think.23:54
tchayposee you in brisneyland23:54
lifelessmordred: amusingly that single test case is doing a full set of requirements installs23:55
lifelessmordred: currently takes 18 seconds to run 0 tests.23:56
*** dingyichen has joined #openstack-infra23:56
lifelessmordred: so, if we build them on install23:57
lifelessmordred: (vs develop)23:58
lifelessmordred: that won't affect nova at all23:58
openstackgerritlifeless proposed openstack-dev/pbr: Show how long the git querying takes
mordredlifeless: hahaha23:58
*** Sukhdev has joined #openstack-infra23:59

Generated by 2.14.0 by Marius Gedminas - find it at!