Thursday, 2014-05-08

*** manishg has quit IRC00:00
morganfainbergclarkb, not a bother at all, rather explain it if that makes the infra folks happier about the change :)00:00
*** mrmartin has joined #openstack-infra00:00
clarkbmorganfainberg: then when we are ready to gate everything on this I think some partition of the existing jobs should just use apache instead00:00
clarkbmorganfainberg: eg apache keystone with postgres or with neutron or whatever00:00
*** manishg has joined #openstack-infra00:01
morganfainbergclarkb, ++ my goal is to make keystone's default deployment in devstack apache (long term)00:01
morganfainbergclarkb, in either case, i totally agree combining deployment types at some point.00:01
clarkbnow I am waiting for gertty to catch up with the -1 I left so that it doesn't override my +2 that I want to give00:01
morganfainbergclarkb, hehe thanks :)00:01
openstackgerritA change was merged to openstack-infra/config: Clean up some variable references
clarkbshould have this approved shortly00:02
anteayaStevenK: it said it didn't work to send it to and it sent it to keys.gnupg.net00:02
anteayaStevenK: how do I confirm it is there?00:02
morganfainbergclarkb, awesome! this will make us keystone folks happy.00:02
StevenKanteaya: --keyserver argument passing to gnupg is terrible at best.00:02
*** fifieldt has joined #openstack-infra00:03
StevenKanteaya: has a web interface, but the search is broken.00:04
StevenKWhich is just lol-tastic00:04
*** zns has quit IRC00:04
clarkbmorganfainberg: and approved00:04
morganfainbergclarkb, i agree we should add those to devstack/tempest as well (perhaps ds-g)00:04
*** manishg has quit IRC00:05
morganfainbergclarkb, do you want that as a separate patch? or in the child?00:05
anteayagpg --send-keys <8 digit key id> keyserver.ubuntu.com00:05
anteayathat is the command I had used00:05
clarkbmorganfainberg: I would wait a bit and maybe we can just go straight to gating on this everywhere00:05
anteayawhat should I have done?00:05
*** ramashri has joined #openstack-infra00:05
clarkbmorganfainberg: if this works properly as is I think it would be safe to pick some subset of jobs and use apache there00:05
morganfainbergclarkb, great. works for me.00:05
StevenKanteaya: --keyserver after the key id00:05
openstackgerritA change was merged to openstack-infra/config: Upload tooz tarballs to PyPI
morganfainbergclarkb, ++00:05
anteayaStevenK: I will try that00:05
*** manishg has joined #openstack-infra00:06
morganfainbergclarkb, thanks again!00:06
StevenKanteaya: The 8 digit key id is public information, too, so you don't need to snip it out. :-)00:06
openstackgerritA change was merged to openstack-infra/config: Set Alpha Channel for better looking favicon
*** bhuvan has quit IRC00:06
clarkbmordred: arg I just approved a thing you approved00:07
*** matsuhashi has joined #openstack-infra00:07
clarkbmordred: there are definitely some rought edges to gertty00:07
morganfainbergclarkb, i keep meaning to try out gertty00:07
morganfainbergclarkb, sounds like it isn't 100% there yet.00:07
anteayamorganfainberg: worth a test drive00:08
clarkbmorganfainberg: well this most recent thing is related to it being an offline review tool00:08
clarkbmorganfainberg: which means it may get slightly out of sync. But you can review on an airplane00:08
*** yamahata has joined #openstack-infra00:08
anteayamorganfainberg: fairly easy to set up00:08
clarkbmorganfainberg: it is pretty amazing otherwise00:08
*** SumitNaiksatam has quit IRC00:08
StevenKanteaya: This is what I was saying, gnupg's option parsing is terrible.00:08
anteayait does eat cpu when you subscribe though00:08
morganfainbergclarkb, oooh on the plane!00:08
anteayaStevenK: k00:08
StevenKanteaya: gnupg --send-keys --keyserver 0x<8 digit key>00:08
anteayaStevenK: I'll try that00:09
mordredclarkb: yah. the async part, although awesome, is also not my favorite at moments00:09
morganfainbergi'll def look into it00:09
morganfainberganteaya, clarkb, thanks :)00:09
clarkbdims are you around?00:09
mordredI kinda of want a commit button that's synchronous that i can use to make sure I'm in sync00:09
clarkbmordred: jhesketh on 92016 that span class name is auto generated right?00:09
clarkbmakes me worry about breaking the css when we update gerrit00:09
anteayaStevenK: thanks, taht looks like it worked00:09
*** manishg has quit IRC00:10
anteayaStevenK: how can I verify?00:10
clarkbmordred: yeah00:10
StevenKanteaya: has a web interface, search for your key id00:10
*** james_li has quit IRC00:10
StevenKanteaya: I see it, first hex digit is A, last is 4.00:10
*** manishg has joined #openstack-infra00:10
jheskethclarkb: I'm not sure, to be honest00:11
openstackgerritA change was merged to openstack-infra/config: Add tool to parse puppet reports
jheskethclarkb: I inspected the html and the class was on the correct parts00:11
anteayaStevenK: yes, thanks00:11
*** timrc-afk is now known as timrc00:11
*** pcrews has quit IRC00:11
jheskethbut I'm not sure if it's hard coded, configured or generated00:11
StevenKanteaya: You're very welcome :-)00:12
* StevenK scampers off to raid the kitchen for some breakfast00:12
clarkbjhesketh: ya I am trying to look into that now00:13
*** crc32 has quit IRC00:13
clarkbbut not seeing that span in the source of the change I have open00:13
anteayahappy breakfast00:13
*** manishg has quit IRC00:15
*** manishg has joined #openstack-infra00:15
* clarkb has no idea what he is doing looking for that string00:15
*** oomichi has joined #openstack-infra00:16
*** Alexandra_ has quit IRC00:17
morganfainbergso i just noticed something odd. it looks like zuul popped something frmo check to gate before the check finished?00:19
*** manishg has quit IRC00:20
clarkbjhesketh: where did you find it?00:20
morganfainbergit was a dependant change.00:20
clarkbmorganfainberg: did someone push a new patchset?00:20
morganfainbergclarkb, no this was the infra ones you just approved00:20
clarkb that one?00:20
morganfainbergclarkb, the 2nd patch 91739 appeared in gate ebfore check finished00:20
morganfainbergclarkb, yeah00:20
*** manishg has joined #openstack-infra00:20
*** rwsu has joined #openstack-infra00:20
morganfainbergclarkb, i've seen that happen every now and again come to think of it00:21
clarkbit +1'd it00:21
openstackgerritA change was merged to openstack-infra/config: Add apache-configured devstack tempest test
morganfainbergclarkb, ah ok.00:21
morganfainbergclarkb, yeah it finished normally. i think i've seen a case where it still -1'd and ended up +2ing the gate00:22
openstackgerritA change was merged to openstack-infra/config: Enable apache-configured-service tempest for keystone
morganfainbergclarkb, i'll keep my eyes peeled and see if i can find one of those / let you know if it happens00:22
morganfainbergclarkb, if anything it seems infrequent / edge case at best00:22
clarkbmorganfainberg: you mean it -1'd in check then went on to gate and got +2?00:22
morganfainbergclarkb, yeah i think i've seen that before, something (child patchset) moved to gate ebfore check finished00:23
clarkbin that case it should report the -1 then next time someone bumps it it will check + gate again00:23
clarkbjhesketh: I am derpy when it comes to css but I can't find that in chromium or firefox00:23
jeblairmorganfainberg: yeah, it's possible (but fortunately we still have gate checks that will prevent that from merging)00:23
jeblairmorganfainberg: we recently merged a change to zuul that will let us plug that logic hole00:24
morganfainbergjeblair, ++ cool00:24
jheskethclarkb: expand all the comments00:24
jeblairmorganfainberg: but we're waiting on a more comprehensive test before we change the config to use it00:24
jheskethclarkb: it's the ones where they overrun the comment field00:24
jheskethbecause they are formatted as code blocks00:24
morganfainbergjeblair, makes sense to me.00:24
morganfainbergjeblair, it means i'm not taking crazy pills :)00:24
jeblairmorganfainberg: cause, well, there are a lot of possible branches :)00:24
clarkbjeblair: oh is that part of the requires change?00:25
*** manishg has quit IRC00:25
jeblairclarkb: yeah -- because a +1 verify requirement on a trigger is not checked when a change is pulled in because it's a mergable dependency00:25
jeblairclarkb: whereas putting the requirement in the pipeline enqueue check will make it enforced all the time00:25
*** manishg has joined #openstack-infra00:26
clarkbjhesketh: there it is thanks00:27
clarkbgoing to zgrep on a couple gerrit wars00:27
jeblairclarkb, mordred: if you have interesting gertty debug logs from when it behaved unexpectedly, that could be helpful00:27
clarkbjeblair: rgr so far I think most of the issues have been expected00:27
mordredjeblair: yah - not an issue as much as just the way busy things work I believe00:28
*** matsuhashi has quit IRC00:28
mordredjeblair: well, I have crashed gerrty a few times by reviewing too quickly and hitting a sqlite lock00:28
mordredjeblair: and by reviewing too quickly - I mean "marking as read" too quickly00:29
*** manishg has quit IRC00:30
dimsclarkb, pong00:30
jeblairmordred: ok.  me might need to check that review uploads get high priority (i think they do) -- but we may also need to double check that refreshes of changes that have just had their review uploaded also get high priority00:30
*** manishg has joined #openstack-infra00:31
clarkbdims: 92016 that span class name doesn't look reliable to me00:31
clarkbdims: any idea if it is consistent across gerrit builds?00:31
clarkbdims: I am looking int othat too00:31
*** matsuhas_ has joined #openstack-infra00:31
jeblairmordred: ok; i've tried to avoid the sqlite lock issue by keeping the sections that actually use the db very small, but other than that, have done basically nothing around sqlite tuning, so there's probably much more we can do there00:32
bknudsonseen this before?
bknudson Non-zero exit code (2) from test listing. stdout='\xb00:32
mordredjeblair: I can look at it on a plane trip ;000:32
mordredwow. that was not the smiley I wanted to type00:32
dimsclarkb, is there a gerrit instance with that update?00:32
*** msabramo has quit IRC00:33
clarkbdims: right now we don't have a gerrit with different versions00:33
jeblairmordred: ++00:33
clarkbwe do have different wars at though00:33
mordredjeblair: I did use gertty a lot this afternoon and found it quite pleasant00:33
clarkbdims: I grabbed the same version we have deployed and zgrep doesn't find that JDC* string either00:33
clarkbnot sure if zgrep is at fault or if multiple levels of compression00:33
mordredjeblair: with one or two ui things that I haven't felt like I've used it long enough to suggest as bugs- so I should have real feedback by the summit00:34
dimsclarkb, looking00:34
*** manishg has quit IRC00:35
*** manishg has joined #openstack-infra00:36
*** aconrad has quit IRC00:36
clarkbdims: ok I unzipped the wars and used grep -r and it shows up in 2.8.1something and 2.8.4something00:38
clarkbso we are probably ok00:38
*** etoews has joined #openstack-infra00:38
clarkbdims: also what editor do you use? you should teach it that files end with a newline :)00:39
*** prad_ has quit IRC00:39
anteayaI just updated my gertty repo, what command to start gertty now?00:39
dimsclarkb, will do :) switching laptops now (ubuntu->osx) all my tools are in a mess/flux00:40
clarkbdims: my last comment is shouldn't you use pre-wrap instead of pre-line?00:40
clarkbdims: nnnooooo00:40
clarkbuse the linux :)00:40
anteayaI tried python gertty/ which didn't work00:40
*** manishg has quit IRC00:40
clarkbanteaya: just `gertty` after pip installing gertty00:40
*** matsuhas_ has quit IRC00:40
*** matsuhashi has joined #openstack-infra00:41
*** mrmartin has quit IRC00:41
*** manishg has joined #openstack-infra00:41
*** msabramo has joined #openstack-infra00:41
anteayaah I handn't done pip install00:41
openstackgerritK Jonathan Harker proposed a change to openstack-infra/devstack-gate: Allow for an alternate fallback branch
jeblairanteaya: if you do 'pip install -e .' in the repo, then you won't need to install it again when you git pull00:43
anteayagreat, I will do that00:44
*** manishg has quit IRC00:45
mattoliveraujeblair: awesome work on gertty btw! What an great tool :)00:46
anteayaI still can so no reviews in gertty00:46
*** manishg has joined #openstack-infra00:46
anteayaperhaps I am missing something00:46
clarkbanteaya: you have to hit l to list all projects then subscribe to some projects iwth s00:47
anteayaI have00:47
clarkbit should then clone and sync then hit enter on a project whihc will take you to the project's change list00:47
anteayaI am subscribed to 4 of them00:47
anteayaI will try hitting enter some more00:47
*** timrc is now known as timrc-afk00:48
clarkbdims: well I decided it should be pre-wrap so I -1'd00:49
*** jasondotstar has joined #openstack-infra00:50
openstackgerritA change was merged to openstack-infra/config: TripleO: use the common-venv element by default.
*** melwitt has quit IRC00:50
*** manishg has quit IRC00:50
jeblairmattoliverau: thanks -- it's early days yet, lots left to do!  (i'm trying not to over-promise it yet)00:50
dimsclarkb, cool, will resubmit00:51
*** manishg has joined #openstack-infra00:51
clarkbdims: otherwise I think anyone doing code blocks that go long will lose their formatting whitespace00:51
*** markmcclain1 has quit IRC00:52
*** jhesketh has quit IRC00:53
mgagnehas anyone installed the openstack summit android app?00:53
clarkbno, previous summits have made me use the website instead00:54
mgagneclarkb: what is the account linked to? is it a standalone account for the sole purpose of using the app or is it linked to launchpad or something?00:54
clarkbmgagne: its the account00:55
clarkbwhcih is standalone or facebook or linked in iirc00:55
clarkbI use standalone00:55
*** manishg has quit IRC00:55
mgagneclarkb: right, would have been great to mention it somewhere on the sign in screen ^^'00:55
clarkbmgagne: the mobile site works great00:56
anteayajeblair: ArxCruz likes as a suggestion for nodepool/jenkins communication00:57
anteayajeblair: he tweeted but you aren't on the twitter00:57
mgagneclarkb: doesn't look to be linked to sched.org00:57
clarkbmgagne: huh00:58
clarkbmgagne: you are going to get me to install this thing00:58
mgagneclarkb: the app looks to be written with/by guidebook.com00:59
clarkbmgagne: is it in the play store?01:00
openstackgerritDavanum Srinivas (dims) proposed a change to openstack-infra/config: Gerrit - Break up long lines to fit
mgagneclarkb: yep, 2s01:01
*** sdake_ has quit IRC01:01
mgagneclarkb: link found here:
*** mrodden has joined #openstack-infra01:02
anteayaArxCruz: it looks interesting to me01:03
clarkband suddenly I am reminded why I don't install these things01:03
clarkbit wants access to the accounts on my phone wtf01:03
ArxCruzanteaya: yes, pretty much, and much more readable01:03
anteayaArxCruz: would there be bugs it would address, do you think?01:04
mgagneclarkb: probably to store his own account01:04
*** aconrad has joined #openstack-infra01:05
clarkbmgagne: it looks like this is completely independent of sched01:07
*** Ryan_Lane has quit IRC01:08
*** mestery has quit IRC01:08
clarkbmgagne: nevermind01:10
*** jhesketh has joined #openstack-infra01:10
*** SumitNaiksatam has joined #openstack-infra01:10
mgagneclarkb: looks to be "linked" only for the event organizer, not the attendees01:11
ArxCruzanteaya: I will see if I can work this weekend with this :)01:11
ArxCruzif beers permits :D01:11
*** dims has quit IRC01:11
*** zns has joined #openstack-infra01:11
*** dims has joined #openstack-infra01:11
clarkbmgagne: confirmed I can't login with my sched account01:12
mgagneclarkb: I tried too already =)01:12
*** msabramo has quit IRC01:12
jeblairArxCruz: python-jenkins is on stackforge now01:12
*** dims has quit IRC01:13
ArxCruzjeblair: cool, I will try write some code replacing urlib with python-requests :)01:14
*** msabramo has joined #openstack-infra01:14
*** dims has joined #openstack-infra01:15
*** dims has quit IRC01:15
*** dims has joined #openstack-infra01:15
clarkbmgagne: you should twitter @openstack and get it fixed01:15
clarkbmgagne: but the mobile sched site works pretty well01:16
*** marcoemorais has quit IRC01:16
*** dims has quit IRC01:16
*** mriedem has joined #openstack-infra01:16
mgagneclarkb: I'm sure it works well. I however got curious about the app after reading about it on twitter01:17
*** dims has joined #openstack-infra01:18
*** bhuvan has joined #openstack-infra01:19
*** msabramo has quit IRC01:19
*** msabramo has joined #openstack-infra01:19
*** mrodden has quit IRC01:19
anteayaArxCruz: beer fueled patches, I hear they are good01:20
*** dims has quit IRC01:20
*** spredzy has joined #openstack-infra01:20
*** dims has joined #openstack-infra01:21
*** dims has quit IRC01:22
ArxCruzanteaya: :D01:22
*** dims has joined #openstack-infra01:22
ArxCruzanteaya: I though I was a good beer drinker until I see sdague list of beers... makes me very very sad...01:22
openstackgerritMatthew Oliver proposed a change to stackforge/gertty: Add '?' as another way show help dialog
*** mrodden has joined #openstack-infra01:23
*** mestery has joined #openstack-infra01:25
*** mbacchi has quit IRC01:25
*** aconrad has quit IRC01:25
mattoliveraujeblair: Added the ? key as another way of diesplaying the help dialog as gnome-terminal intercepts the <F1> key.01:25
*** jaypipes has quit IRC01:28
anteayaArxCruz: sdague writes some good code, I wonder if data should be collected about beer and patch quality01:28
ArxCruzanteaya: indeed01:28
anteayamattoliverau: one small grammar comment about the commit message01:30
*** spredzy has left #openstack-infra01:32
*** etoews has quit IRC01:33
*** mestery has quit IRC01:34
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Use alembic
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Add some helper methods to deal with sqlite migrations
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Increase the status field width and drop 'hidden'
*** UtahDave has joined #openstack-infra01:37
jeblairSergeyLukjanov: ^ alembic sqlite stuff based on what you did a while ago.  would be great if you have a minute to take a look01:37
openstackgerritClark Boylan proposed a change to stackforge/gertty: Remove MERGED changes from unreviewed list.
clarkbjeblair: ^ thats a quick stab at fixing a minor annoyance but as the commit message says I am not quite sure it fixes the entire problem01:38
*** msabramo has quit IRC01:38
jeblairclarkb: that's actually what i was about to start on a few days ago when i noticed i needed a schema change and went down that branch; i'll take a look real quick and see how close it is to what i was about to do01:39
clarkbjeblair: also I didn't need or_ anymore I should remove that but will wait for you to make a decision on whether or not that change is useful first01:41
openstackgerritArnaud Legendre proposed a change to openstack-infra/config: Create glance-specs git repo
*** arnaud has joined #openstack-infra01:41
jeblairclarkb: that approach wfm for now, but we should do status not in ['MERGED', 'ABANDONED']01:41
clarkboh right, I will update01:41
clarkband remove or_01:41
jeblairclarkb: maybe make the _closed_status class var a module level global and use it both places01:42
openstackgerritIan Wienand proposed a change to openstack-infra/devstack-gate: Add distro check functions
jeblairclarkb: oh, different module, that's in sync...01:42
jeblairclarkb: whatever you think best.  :)01:42
*** mrodden has quit IRC01:43
clarkboh this has a counterpart in sync? probably trying to sync abandoned and merged changes01:43
clarkbI will poke at it01:43
*** Ryan_Lane has joined #openstack-infra01:43
*** zns has quit IRC01:43
jeblairclarkb: right now it doesn't _initially_ sync abandoned and merged, but it will keep updating them once they are imported01:43
jeblairclarkb: but yeah, at some point we'll probably want to purge the db for efficiency.  but thankfully not yet.  :)01:44
*** nosnos has joined #openstack-infra01:44
openstackgerritIan Wienand proposed a change to openstack-infra/devstack-gate: Add distro check functions for Ubuntu and Fedora
*** homeless has quit IRC01:47
*** mestery has joined #openstack-infra01:48
*** unicell has joined #openstack-infra01:49
*** jasondotstar has quit IRC01:50
*** yaguang has joined #openstack-infra01:51
*** UtahDave has quit IRC01:53
*** UtahDave has joined #openstack-infra01:53
mattoliverauanteaya: thanks, I'll fix it up :)01:53
*** zehicle_at_dell has quit IRC01:55
*** zehicle_at_dell has joined #openstack-infra01:56
anteayamattoliverau: :D01:57
openstackgerritClark Boylan proposed a change to stackforge/gertty: Ignore closed changes more.
clarkbjeblair: ^ I want to say I perceived that chewing more cpu time but that may just be initial startup where it goes and syncs all the things01:57
mattoliverauanteaya: I don't see the extra d you mention?... maybe my spelling is just that bad :P01:58
*** UtahDave has quit IRC01:58
*** thomasem has joined #openstack-infra01:59
*** UtahDave has joined #openstack-infra01:59
*** thomasem has quit IRC01:59
openstackgerritMatthew Oliver proposed a change to stackforge/gertty: Add '?' as another way show help dialog
anteayashould that not be an02:00
anteayayes, you got it02:00
jeblairclarkb: you went too far :)02:01
jeblair(inline review)02:01
fungiokay, has now been updated too. calling it a night02:01
anteayafungi: great, the md5sum works for me02:01
anteayanight all02:01
jeblairfungi: good night, and thanks!02:01
*** harlowja has quit IRC02:03
clarkbjeblair gah. I know how to fix that though02:03
clarkbmy original pass at it was correct then I optimised02:03
*** harlowja has joined #openstack-infra02:04
*** mengxd has joined #openstack-infra02:04
jeblairclarkb: ok -- i mean, i'm happy with the logic as-is there... :)02:04
*** LarsN has left #openstack-infra02:04
openstackgerritArnaud Legendre proposed a change to openstack-infra/config: Create glance-specs git repo
mattoliverauanteaya: lol, I see, there extra d was in and.. to make it an.. of course! sorry, maybe I need another coffee or something :)02:04
*** arnaud has quit IRC02:05
mattoliveraunight anteaya and fungi02:06
*** thomasem has joined #openstack-infra02:06
*** thomasem has quit IRC02:06
*** thomasem has joined #openstack-infra02:07
*** signed8bit has quit IRC02:09
clarkbjeblair no I had an intermediate step02:11
clarkbI didnt push it02:11
*** gokrokve has quit IRC02:11
clarkbwalking home then dinner then I will fix02:12
*** mriedem has quit IRC02:12
*** zz_gondoi is now known as gondoi02:12
openstackgerritJames E. Blair proposed a change to stackforge/gertty: Sync parent changes
jeblairclarkb: ok.  btw, ^ that change probably wants to be based on yours because it's another place that needs to know the list of closed states.02:14
jeblair(i just marked that a todo for now)02:14
*** dims has quit IRC02:16
*** thomasem has quit IRC02:16
jheskethfungi: so I missed the keysigning sign up because I'm lame... are others allowed to participate somehow? For example, I could bring a printout of my key to hand to people and they can verify my ID as I give it to them02:16
*** dims has joined #openstack-infra02:16
*** zhiyan_ is now known as zhiyan02:17
*** UtahDave has quit IRC02:19
jeblairjhesketh: well, i for one will be happy to find time to verify your id regardless.  :)02:20
jeblairmattoliverau: is there a pass-through modifier you can use with gnome-terminal for f1?02:20
* jhesketh will bring some printed fingerprints anyway and see what happens - even if just in the hallway02:21
jeblairmattoliverau: i tried to provide helpful suggestions in the welcome message about dealing with that problem02:21
jeblairmattoliverau: but i don't actually have it, so i'm writing blind.  :)02:21
greghaynes"...bring a printout of my key": I hope youre not actually going to print out your key and hand those pages out02:22
clarkbI disable f102:22
jeblairmattoliverau: i think i'm ok with adding ?, but i also expect that we might add more fkeys02:22
clarkbfor help02:22
jeblairclarkb: ++02:22
clarkbworks great02:22
jeblairjhesketh: sounds like a plan02:23
*** Alexandra_ has joined #openstack-infra02:27
openstackgerritNikhil Manchanda proposed a change to openstack-infra/config: Add new gate-trove-buildimage-mysql job
*** amcrn has joined #openstack-infra02:31
Alex_GaynorCan someone who understands how testr works take a look at Alastair's comment ?02:34
*** oomichi has quit IRC02:35
*** maxbit has joined #openstack-infra02:36
*** etoews has joined #openstack-infra02:36
*** dims has quit IRC02:37
*** dims has joined #openstack-infra02:38
clarkbAlex_Gaynor: failed to catch this02:39
clarkbAlex_Gaynor: which is annoying02:39
Alex_Gaynorclarkb: yeah, trying to understand why it's running 0 on py3k though, tests run fine on py202:39
clarkbAlex_Gaynor: what does `.tox/py33/bin/testr last` spit out?02:39
clarkbafter running tox -epy2702:39
*** UtahDave has joined #openstack-infra02:40
clarkber tox -epy3302:40
Alex_Gaynorchecking locally02:40
*** dims has quit IRC02:40
Alex_Gaynorclarkb: PASSED (id=42)02:40
clarkbAlex_Gaynor: but not Ran X tests?02:41
clarkbAlex_Gaynor: can you also paste `.tox/py33/bin/testr run --parallel` and `.tox/py33/bin/python -m discover`02:42
clarkbAlex_Gaynor: I think no tests are being matched and thus it says +102:42
*** gokrokve has joined #openstack-infra02:42
clarkbAlex_Gaynor: we have a check in place to try and catch that but its failing according to that console log02:42
Alex_Gaynorclarkb: discover prints... tons of binary stuff out, do you want me to pastebin it?02:43
clarkbAlex_Gaynor: no thats fine02:43
*** UtahDave has quit IRC02:43
*** harlowja has quit IRC02:44
clarkbAlex_Gaynor: `.tox/py33/bin/testr run --parallel tests.unit` would be the last intereting thing I can come up with02:44
*** gokrokve_ has joined #openstack-infra02:44
clarkboh you know I know02:44
*** harlowja has joined #openstack-infra02:44
clarkbthe test filter happens after discovery02:44
clarkbis discovery failing and testr with a fitler does something different than testr without it?02:44
clarkblifeless: ^02:44
clarkbhrm no its fails in both cases correctly02:45
*** zns has joined #openstack-infra02:45
lifelessAlex_Gaynor: that looks like a sane error :)02:45
*** zns has quit IRC02:45
lifelessclarkb: whats up ?02:45
clarkbya running testr directly works fine02:45
clarkbrunning tox -epy33 runs no tests02:46
Alex_Gaynor(and passes!)02:46
clarkbpython swiftclient tox -epy33 runs no tests and apsses02:46
*** zns has joined #openstack-infra02:46
clarkbour check for that isn't working but also curious to why it runs zero tests02:46
Alex_GaynorThere also doesn't appear to be an "all" import anywhere in swiftclient :/02:46
clarkbI don't have python33 handy on this machine (still precise)02:46
*** gokrokve has quit IRC02:47
Alex_Gaynorclarkb: if you feel like doing stuff, the deadsnakes PPA may be useful02:47
*** gokrokve_ has quit IRC02:49
*** mestery has quit IRC02:49
mattoliveraujeblair: fair enough, using function keys is nice, but I think help should be a special case, as users will need to be able to easily use it so they can figure out how to use each view. We could make it 'h' instead, as you use 'l' and other letters for using most the views.02:50
mattoliverauAnyway lunch time :)02:50
*** gondoi is now known as zz_gondoi02:51
*** otherwiseguy has joined #openstack-infra02:52
*** bhuvan has quit IRC02:53
*** dims has joined #openstack-infra02:54
*** moted has quit IRC02:55
*** zz_gondoi is now known as gondoi02:56
*** moted has joined #openstack-infra02:57
*** moted_ has joined #openstack-infra03:00
*** msabramo has joined #openstack-infra03:00
openstackgerritClark Boylan proposed a change to openstack-infra/config: Fix zero tests ran unittest check.
clarkbAlex_Gaynor: ^ that should fix the test and make it fail03:01
*** moted has quit IRC03:01
*** moted_ is now known as moted03:01
clarkbbut its still a defensive thing not an actual make py33 work03:02
*** david-lyle has joined #openstack-infra03:02
clarkbAlex_Gaynor: after you run those tox commands that say non zero exit code from test listing what is $? ?03:02
clarkbis it not 0?03:02
Alex_Gaynorclarkb: 303:02
Alex_Gaynor3 is a cool status03:02
*** renlt has joined #openstack-infra03:04
*** dims has quit IRC03:05
* clarkb tries a thing03:06
Alex_Gaynoryay for things :-)03:07
*** signed8bit has joined #openstack-infra03:07
clarkbAlex_Gaynor: lets see how that does03:07
Alex_Gaynorclarkb: it's going to fail the same way default does (import errors inside the tests), you need to rebase it on to get useful knowledge03:08
clarkbAlex_Gaynor: right but if it fails then we figured it out03:09
clarkbpass or fail if it does the correct thing thats fine03:09
clarkb is the job03:09
clarkbI think the testr command comes from testr itself and mordred made it broken03:10
clarkbthe test command from pbr however should work03:10
*** otherwiseguy has quit IRC03:10
clarkbmordred shoudl buy a round at the summit if I am correct ;)03:11
Alex_GaynorI'll buy anyone who wants anything they want if this solves it :-)03:11
clarkband by success I mean the job failed but it did so properly03:11
Alex_Gaynorclarkb: FWIW a casual grep shows that " testr" is used allover03:11
clarkbAlex_Gaynor: I will rebase it now and see if things are good03:12
clarkbAlex_Gaynor: ya03:12
*** dkehn_ has joined #openstack-infra03:12
clarkbok rebased03:12
clarkbAlex_Gaynor: long story short mordred made a small mistake in the testr testr command03:13
clarkbnever fixed it and updated pbr instead with a test command03:13
clarkbhwoever the testr command seems to have stuck around03:13
Alex_Gaynor~~computers~~ :-)03:14
clarkbI can propose a ton of changes tomorrow to s/testr/test/ if mordred doesn't beat me to it03:14
clarkbor maybe he did fix it but then upstram never cut a release03:14
*** dkehn has quit IRC03:15
*** dkehnx has quit IRC03:15
Alex_Gaynorclarkb: "2014-05-08 03:16:00.132 | PASSED (id=0)" womp womp :-(03:16
*** dkehn has joined #openstack-infra03:17
clarkbwhy did the behavior change after the rebase?03:17
clarkb clearly did the right thing03:18
Alex_Gaynorclarkb: I think the errors we saw before (import errors inside swiftclient/tests) happen before it gets to whatever makes this fail03:18
clarkboh I see, so this change gets us past that03:19
*** maxbit has quit IRC03:19
*** otherwiseguy has joined #openstack-infra03:19
*** aconrad has joined #openstack-infra03:20
*** zns has quit IRC03:21
clarkb is the code running03:21
*** changbl has quit IRC03:21
*** oomichi has joined #openstack-infra03:21
*** lcheng_ has joined #openstack-infra03:21
*** etoews_ has joined #openstack-infra03:22
renltclarkb: I have queries about manage stackforge project. yesterday we got our stackforge/powervc-driver created, then we open a bug, to request initial group member, could you please take a look at it if time permits?03:22
uvirtbotLaunchpad bug 1316893 in openstack-ci "Initial group member request for stackforge/powervc-driver" [Undecided,New]03:22
renltclarkb: and another query is do we need create a launchpad project to trace bugs and blueprints03:23
*** david-lyle has quit IRC03:23
*** etoews has quit IRC03:23
*** matsuhashi has quit IRC03:24
renltclarkb: and if we need to create a launchpad project, I still don't quite understand the relationship between this prject and the 2 teams of project-bugs & project-drivers team mentioned in
*** etoews_ has quit IRC03:27
renltclarkb: thanks in advance!03:27
clarkbrenlt: launchpad projects are all on you03:27
clarkbwe don't manage them for you03:27
renltclarb: ok03:28
clarkbrenlt: bug teams manage bugs, drivers manage blueprints03:28
openstackgerritClark Boylan proposed a change to stackforge/gertty: Ignore closed changes more.
clarkbjeblair: ^ I think that should do it03:29
*** james_li has joined #openstack-infra03:29
clarkbrenlt: if you look at the bug the email addresses are hidden03:30
renltclarkb: so we need create a project first, and then create 2 teams if we need03:31
clarkbtypically we just add a single member and they can manage the rest. Is there one person I should add to the core group and the ptl group in gerrit?03:31
clarkbrenlt: yes03:31
lifelessclarkb: so discover is finding no tests03:32
lifelessclarkb: and not indicating any error03:32
renltclarkb: I'm sorry? if you look at the bug the email addresses are hidden?03:32
clarkbrenlt: yes, open the bug all of the email addresses you provided are <email address hidden>03:33
clarkbrenlt: if you give me a gerrit account here in irc I can go ahead and add that person and they can add the rest03:33
*** krtaylor has joined #openstack-infra03:33
renltclarkb: oh..why I can see them03:34
renltgerrit initial group members:03:34
renlt powervc-driver-core:,,,
renlt powervc-driver-milestone:,,,
renlt powervc-driver-ptl:
*** blamar has quit IRC03:35
clarkbrenlt: there is no milestone group03:35
clarkbrenlt: if I add megxd to -core and -ptl they can add the remaining members03:36
renltclarkb: thanks, then we can do the remain management work by ourselves.03:37
clarkbooohhhh I had to log in now I can see the email addresses03:37
renltclarkb: oh..that must be some launchpad acl..03:39
renltclarkb: thanks again03:39
*** nosnos has quit IRC03:41
*** lcheng_ has quit IRC03:45
*** gokrokve has joined #openstack-infra03:45
*** lcheng_ has joined #openstack-infra03:46
*** gokrokve has quit IRC03:50
*** arnaud has joined #openstack-infra03:50
*** unicell1 has joined #openstack-infra03:53
*** unicell has quit IRC03:54
*** eharney has quit IRC03:56
*** unicell1 has quit IRC03:58
*** unicell has joined #openstack-infra03:58
*** etoews has joined #openstack-infra04:00
*** otherwiseguy has quit IRC04:01
*** UtahDave has joined #openstack-infra04:04
*** etoews has quit IRC04:04
openstackgerritClark Boylan proposed a change to stackforge/gertty: Ignore closed changes more.
clarkbjeblair: ^ is tested to actually do stuff04:06
clarkband seems to work04:06
*** esker has quit IRC04:07
clarkbI couldn't figure out how to make the sqlalchemy relationship use the list of statuses04:07
*** ArxCruz has quit IRC04:08
*** blamar has joined #openstack-infra04:08
*** msabramo has quit IRC04:12
*** msabramo has joined #openstack-infra04:13
openstackgerritClark Boylan proposed a change to stackforge/gertty: Delete closed changes from local DB
*** blamar has quit IRC04:16
*** blamar has joined #openstack-infra04:16
jgriffithmordred: hey-yo... is there an example of what I need for abandon/unbandon on that patch?04:20
jgriffithmordred: suppose I could check what's in the queue04:20
clarkbjgriffith: we just merged a ton of edits04:20
clarkbjgriffith: just open an acl file from another project on tip of master04:21
clarkbjgriffith: or if you give me a link to your change I can commetn on it04:21
jgriffithclarkb: I did , but maybe my pull from master was no bueno04:21
jgriffithI didn't see it04:21
*** yfried has quit IRC04:23
*** MarkAtwood has joined #openstack-infra04:23
clarkbjgriffith: done04:23
jgriffithclarkb: thank you sir04:23
*** mestery has joined #openstack-infra04:24
jgriffithclarkb: perfect04:24
openstackgerritJohn Griffith proposed a change to openstack-infra/config: Create cinder-specs git repo
*** aconrad has quit IRC04:26
clarkbjgriffith: ETABS04:27
clarkbjgriffith: iirc gerrit actually complains about that04:27
*** che-arne has quit IRC04:27
*** che-arne has joined #openstack-infra04:27
*** mengxd has quit IRC04:28
clarkbit does weird stuff with whitespace in its ini conf files04:28
openstackgerritJohn Griffith proposed a change to openstack-infra/config: Create cinder-specs git repo
jgriffithclarkb: :)04:28
jgriffithclarkb: I need to set my vimrc for yaml and .config04:28
jgriffithI've slipped on that one far too many times04:29
*** nosnos has joined #openstack-infra04:29
clarkbI do so much python I just default to no tabs and set noexpandtab if necessary04:29
jgriffithclarkb: indeed...I set that by file type, I should just make it global04:30
jgriffithour C++ standard sadly enough is tabs :(04:31
*** wenlock has joined #openstack-infra04:32
*** arnaud has quit IRC04:33
*** gondoi is now known as zz_gondoi04:34
*** wenlock has quit IRC04:34
*** wenlock has joined #openstack-infra04:35
*** matsuhashi has joined #openstack-infra04:37
*** rcarrill` has quit IRC04:38
*** lcheng_ has quit IRC04:39
*** rcarrillocruz has joined #openstack-infra04:44
*** gokrokve has joined #openstack-infra04:44
*** zhiyan is now known as zhiyan_04:46
*** gokrokve has quit IRC04:49
*** Longgeek has joined #openstack-infra04:55
*** lcheng_ has joined #openstack-infra04:56
*** MarkAtwood has quit IRC05:02
*** FallenPegasus has joined #openstack-infra05:02
*** FallenPegasus has quit IRC05:02
*** salv-orlando has quit IRC05:18
*** amotoki has joined #openstack-infra05:19
*** matsuhashi has quit IRC05:19
*** matsuhashi has joined #openstack-infra05:20
*** UtahDave has quit IRC05:23
openstackgerritKhai Do proposed a change to openstack-infra/config: add a build-timeout macro
*** lcheng_ has quit IRC05:32
*** etoews has joined #openstack-infra05:32
openstackgerritKhai Do proposed a change to openstack-infra/config: add a build-timeout macro
*** chandan_kumar has joined #openstack-infra05:37
*** james_li has quit IRC05:38
*** marcoemorais has joined #openstack-infra05:38
openstackgerritAndreas Jaeger proposed a change to openstack-infra/config: Add database-api publishing jobs to trove
*** etoews has quit IRC05:39
*** krotscheck has quit IRC05:40
*** marcoemorais1 has joined #openstack-infra05:41
openstackgerritKhai Do proposed a change to openstack-infra/config: update the jjb timeout global defaults
*** marcoemorais has quit IRC05:42
openstackgerritAndreas Jaeger proposed a change to openstack-infra/config: Add database-api publishing jobs to trove
openstackgerritNikhil Manchanda proposed a change to openstack-infra/config: Add new gate-trove-buildimage-mysql job
*** gokrokve has joined #openstack-infra05:44
*** Alexandra_ has quit IRC05:45
*** krotscheck has joined #openstack-infra05:45
*** gokrokve_ has joined #openstack-infra05:46
*** gokrokve has quit IRC05:49
*** gokrokve_ has quit IRC05:50
zaroclarkb: check out my comment on 8843205:55
*** wenlock has quit IRC05:55
clarkbzaro: so I am really confused05:55
clarkbzaro: we explicitly stopped logging at debug level on those loggers05:55
clarkbwhich is why that file says WARNING today05:56
*** rcarrill` has joined #openstack-infra05:56
clarkbwe updated gearman plugin to log less verbosely and now we are going to go more verbose on our installs?05:56
*** yfried has joined #openstack-infra05:56
clarkbI get logging the plugin more verbosely because it reduced its default verbosity05:56
clarkbbut I don't think we want the gearman lib to be more verbose05:56
*** rcarrillocruz has quit IRC05:57
*** dstanek_zzz is now known as dstanek05:59
*** zhiyan_ is now known as zhiyan05:59
*** oomichi_ has joined #openstack-infra06:00
*** oomichi has quit IRC06:02
*** zhiyan is now known as zhiyan_06:05
*** morganfainberg is now known as morganfainberg_Z06:05
openstackgerritMaria Nita proposed a change to openstack/requirements: Add xvfbwrapper package for running Selenium tests headless
*** gyee has quit IRC06:06
*** zhiyan_ is now known as zhiyan06:08
*** zhiyan has left #openstack-infra06:08
*** signed8bit is now known as signed8bit_ZZZzz06:08
*** harlowja is now known as harlowja_away06:08
*** zhiyan has joined #openstack-infra06:09
*** lcheng_ has joined #openstack-infra06:10
openstackgerritIan Wienand proposed a change to openstack-infra/config: Ensure jenkins user has useful secure_path
*** zhiyan is now known as zhiyan_06:12
*** zhiyan_ is now known as zhiyan06:12
*** WaltNik has joined #openstack-infra06:15
*** wenlock has joined #openstack-infra06:16
*** lcheng_ has quit IRC06:17
*** ildikov has quit IRC06:29
*** zhiyan is now known as zhiyan_06:30
*** zhiyan_ is now known as zhiyan06:34
*** zhiyan is now known as zhiyan_06:37
*** gokrokve has joined #openstack-infra06:44
*** salv-orlando has joined #openstack-infra06:47
*** gokrokve has quit IRC06:49
nibalizerkrotscheck: working with your module now06:52
*** praneshp has quit IRC06:58
*** etoews has joined #openstack-infra07:00
*** ramashri has quit IRC07:00
*** praneshp has joined #openstack-infra07:01
*** mrmartin has joined #openstack-infra07:02
*** praneshp has quit IRC07:02
*** etoews has quit IRC07:05
*** amcrn has quit IRC07:06
*** sandywalsh has quit IRC07:06
*** _nadya_ has joined #openstack-infra07:06
*** sandywalsh has joined #openstack-infra07:07
*** flaper87|afk is now known as flaper8707:08
*** ildikov has joined #openstack-infra07:08
*** Ryan_Lane has quit IRC07:10
*** SumitNaiksatam_ has joined #openstack-infra07:11
*** SumitNaiksatam has quit IRC07:13
*** SumitNaiksatam_ has quit IRC07:17
*** dizquierdo has joined #openstack-infra07:21
*** _nadya_ has quit IRC07:25
*** _nadya_ has joined #openstack-infra07:29
mattoliverauI'm stopping for the day, night all07:30
*** ominakov has joined #openstack-infra07:37
*** gokrokve has joined #openstack-infra07:44
*** gokrokve has quit IRC07:49
*** matsuhashi has quit IRC07:54
*** dstanek is now known as dstanek_zzz07:55
*** e0ne has joined #openstack-infra07:55
*** matsuhashi has joined #openstack-infra07:55
*** ildikov has quit IRC07:57
*** ildikov has joined #openstack-infra07:57
*** wenlock has quit IRC07:58
*** geertj has joined #openstack-infra07:58
*** jpich has joined #openstack-infra08:00
*** mrmartin has quit IRC08:03
*** etoews has joined #openstack-infra08:03
*** andreaf has joined #openstack-infra08:07
*** etoews has quit IRC08:08
openstackgerritMaria Nita proposed a change to openstack/requirements: Add xvfbwrapper package for running Selenium tests headless
*** lcheng_ has joined #openstack-infra08:10
*** derekh_ has joined #openstack-infra08:11
*** lcheng_ has quit IRC08:12
*** renlt has quit IRC08:22
*** dkehn__ has joined #openstack-infra08:22
*** jcoufal has joined #openstack-infra08:24
*** dkehn__ has quit IRC08:24
*** dkehn_ has quit IRC08:25
*** dkehn has quit IRC08:25
*** dkehn__ has joined #openstack-infra08:25
*** dkehn has joined #openstack-infra08:27
*** jcoufal has quit IRC08:27
*** salv-orlando_ has joined #openstack-infra08:27
*** darraghb has joined #openstack-infra08:27
*** salv-orlando has quit IRC08:29
*** salv-orlando_ is now known as salv-orlando08:29
*** dkehn_ has joined #openstack-infra08:38
*** jcoufal has joined #openstack-infra08:40
*** dkehn_ has quit IRC08:40
*** dkehn has quit IRC08:40
*** dkehn__ has quit IRC08:41
*** dkehn_ has joined #openstack-infra08:41
*** ociuhandu has quit IRC08:42
*** dkehn has joined #openstack-infra08:42
*** gokrokve has joined #openstack-infra08:44
*** jcoufal has quit IRC08:45
*** jcoufal has joined #openstack-infra08:45
*** markmc has joined #openstack-infra08:48
*** gokrokve has quit IRC08:49
*** salv-orlando has quit IRC08:56
*** andreykurilin_ has joined #openstack-infra08:57
*** e0ne has quit IRC09:03
*** katyafervent is now known as katyafervent_awa09:03
*** e0ne has joined #openstack-infra09:04
*** katyafervent_awa is now known as katyafervent09:05
*** rlandy has joined #openstack-infra09:08
*** jcoufal has quit IRC09:09
*** jcoufal has joined #openstack-infra09:09
*** rcarrillocruz has joined #openstack-infra09:09
*** hashar has joined #openstack-infra09:10
*** rcarrill` has quit IRC09:11
*** skolekonov has joined #openstack-infra09:11
*** _nadya_ has quit IRC09:15
*** hashar has quit IRC09:20
*** dangers is now known as dangers_away09:28
*** jasondotstar has joined #openstack-infra09:28
*** marun is now known as marun_afk09:35
*** danpb has joined #openstack-infra09:39
danpbcan anyone here see by Jenkins has gone insane on this
danpbit is starting the test jobs over & over every minute, without any re-check request that I see09:40
*** dkehn__ has joined #openstack-infra09:42
*** gokrokve has joined #openstack-infra09:44
*** dkehn_ has quit IRC09:45
*** jasondotstar has quit IRC09:46
*** marcoemorais1 has quit IRC09:46
*** ociuhandu has joined #openstack-infra09:49
*** gokrokve has quit IRC09:49
*** jasondotstar has joined #openstack-infra09:49
*** Alexei_987 has joined #openstack-infra09:50
*** e0ne has quit IRC09:59
*** e0ne has joined #openstack-infra10:05
*** etoews has joined #openstack-infra10:07
*** e0ne has quit IRC10:08
*** e0ne has joined #openstack-infra10:08
*** e0ne has quit IRC10:12
*** etoews has quit IRC10:13
*** Alexandra_ has joined #openstack-infra10:15
*** _nadya_ has joined #openstack-infra10:23
*** matsuhashi has quit IRC10:27
*** Alexandra_ has quit IRC10:28
*** oomichi_ has quit IRC10:29
*** matsuhashi has joined #openstack-infra10:40
*** jasondotstar has quit IRC10:41
*** maxbit has joined #openstack-infra10:43
*** e0ne has joined #openstack-infra10:43
*** gokrokve has joined #openstack-infra10:44
*** e0ne has quit IRC10:44
*** e0ne has joined #openstack-infra10:45
*** e0ne_ has joined #openstack-infra10:47
*** e0ne has quit IRC10:47
*** _nadya_ has quit IRC10:48
*** KurtMartin has joined #openstack-infra10:48
*** _nadya_ has joined #openstack-infra10:48
*** miqui_ has joined #openstack-infra10:49
*** gokrokve has quit IRC10:49
*** thingee_ has joined #openstack-infra10:50
*** YorikSar_ has joined #openstack-infra10:51
*** e0ne_ has quit IRC10:51
*** maurosr- has joined #openstack-infra10:54
*** geertj has quit IRC10:54
*** dkliban1 has joined #openstack-infra10:55
*** spy1 has joined #openstack-infra10:55
*** thingee has quit IRC10:55
*** Clabbe has quit IRC10:55
*** maurosr has quit IRC10:55
*** antonym has quit IRC10:55
*** dkliban has quit IRC10:55
*** kmartin has quit IRC10:55
*** miqui has quit IRC10:55
*** YorikSar has quit IRC10:55
*** maurosr- is now known as maurosr10:55
*** Clabbe has joined #openstack-infra10:55
*** ildikov is now known as ildikov_fan_repa10:57
*** ildikov_fan_repa has quit IRC10:57
openstackgerritDerek Higgins proposed a change to openstack-infra/config: Scale back on tripleo testing
*** danpb has left #openstack-infra10:59
*** _nadya_ has quit IRC11:00
*** davidlenwell_ has joined #openstack-infra11:02
*** YorikSar has joined #openstack-infra11:04
*** dizquierdo is now known as dizquierdo_afk11:05
*** che-arne has quit IRC11:06
*** maxbit has quit IRC11:08
*** dims has joined #openstack-infra11:12
*** rcarrill` has joined #openstack-infra11:15
anteayadanpb the comment on patchset 3 seems to have set it off11:17
*** rcarrillocruz has quit IRC11:17
anteayait seems to have stopped 21/2 hours ago11:17
sdagueanteaya: I was assume the positive results are actually what stopped it11:20
*** mbacchi has joined #openstack-infra11:21
sdagueI think there is some logic challenges with clean check and the gerrit11:21
anteayaI agree11:21
anteayaand yes, I assume the same about the positive results being the cause the exited the loop11:22
anteayaI look forward to fungi's assessment of the logs11:22
*** matsuhas_ has joined #openstack-infra11:22
*** YorikSar_ has quit IRC11:23
*** matsuhashi has quit IRC11:23
*** Alexei_987 has quit IRC11:23
*** ociuhandu has quit IRC11:23
*** davidlenwell has quit IRC11:23
*** isviridov has quit IRC11:23
*** jamielennox is now known as jamielennox|away11:26
*** che-arne has joined #openstack-infra11:29
skolekonovhello. could you please take a look on this patch ?11:36
anteayaskolekonov: what are you seeing?11:38
anteayaoh you are looking for reviews11:39
*** Alexei_987 has joined #openstack-infra11:41
skolekonovyes, if it's possible :) I've made a mistake in my previous patch when I didn't enable Tempest11:41
*** e0ne has joined #openstack-infra11:42
*** gokrokve has joined #openstack-infra11:44
*** e0ne_ has joined #openstack-infra11:44
*** e0ne has quit IRC11:44
anteayaskolekonov: +111:46
*** ArxCruz has joined #openstack-infra11:47
*** isviridov has joined #openstack-infra11:48
*** ociuhandu has joined #openstack-infra11:48
*** e0ne_ has quit IRC11:49
*** gokrokve has quit IRC11:49
skolekonovanteaya, thank you11:52
*** weshay has joined #openstack-infra12:02
*** pdmars has joined #openstack-infra12:03
*** tteggel has quit IRC12:04
*** cyeoh has quit IRC12:04
*** mkoderer has quit IRC12:04
*** exz has quit IRC12:04
*** mancdaz has quit IRC12:04
*** johnthetubaguy has quit IRC12:04
*** blamar has quit IRC12:04
*** blamar has joined #openstack-infra12:04
*** e0ne has joined #openstack-infra12:07
*** e0ne has joined #openstack-infra12:07
*** yaguang has quit IRC12:08
*** blamar has quit IRC12:09
*** blamar has joined #openstack-infra12:09
*** che-arne has quit IRC12:09
*** e0ne has quit IRC12:09
*** etoews has joined #openstack-infra12:10
*** etoews has quit IRC12:17
*** tteggel has joined #openstack-infra12:18
*** cyeoh has joined #openstack-infra12:18
*** mkoderer has joined #openstack-infra12:18
*** exz has joined #openstack-infra12:18
*** mancdaz has joined #openstack-infra12:18
*** johnthetubaguy has joined #openstack-infra12:18
*** adalbas has joined #openstack-infra12:19
*** geertj has joined #openstack-infra12:32
*** julim has joined #openstack-infra12:33
*** sandywalsh has quit IRC12:35
phschwartzMorning infra12:35
*** sandywalsh has joined #openstack-infra12:36
*** etoews has joined #openstack-infra12:38
anteayamorning phschwartz12:39
*** aysyd has joined #openstack-infra12:41
*** gokrokve has joined #openstack-infra12:44
*** rfolco has joined #openstack-infra12:47
*** bknudson has quit IRC12:48
*** gokrokve has quit IRC12:49
*** signed8bit_ZZZzz is now known as signed8bit12:50
*** zehicle_at_dell has quit IRC12:51
openstackgerritJoão Vale proposed a change to openstack-infra/jenkins-job-builder: Support variable-specific substitutions in templates.
*** smarcet has joined #openstack-infra12:51
*** dkranz has joined #openstack-infra12:52
*** geertj has quit IRC12:59
*** dprince has joined #openstack-infra13:00
*** jcoufal has quit IRC13:00
fungii'll take a look13:00
*** e0ne has joined #openstack-infra13:04
*** zehicle_at_dell has joined #openstack-infra13:05
*** geertj has joined #openstack-infra13:05
*** e0ne has quit IRC13:05
*** andreykurilin_ has quit IRC13:05
*** e0ne has joined #openstack-infra13:06
*** fifieldt has quit IRC13:06
openstackgerritA change was merged to openstack-infra/config: Fix zero tests ran unittest check.
*** bknudson has joined #openstack-infra13:09
*** e0ne has quit IRC13:10
*** azneita has joined #openstack-infra13:11
*** david-lyle has joined #openstack-infra13:11
fungidanpb seems to have vaporied13:12
openstackgerritArx Cruz proposed a change to openstack-infra/devstack-gate: Add qpid config
*** dkranz has quit IRC13:12
*** dkranz has joined #openstack-infra13:13
fungilooks like jenkins was looping trying to report on patchset #3, which was closed13:14
fungithe current patchset was #613:14
fungis/jenkins/zuul/ really13:14
anteayawhy would it pick patchset #3 to chew on?13:15
anteayabecause that was the one with the comment13:15
fungioho, because the comment which set it off was on patchset #313:15
anteayabut patchset 6 was the latest13:16
fungii think we have a logic issue with the patchset matching13:16
*** skolekonov has quit IRC13:16
anteayalooks like it13:16
fungiyes, latest patchset was #6 but you can still leave comments on older patchsets13:16
fungizuul then tried to retest *that* patchset and vote on it, which gerrit won't let you do13:16
fungibut it still left a comment, saw its comment and no recent vote... loop13:16
anteayawhy did zuul behave as expected with gerrit 2.4?13:17
fungiso another flavor of what we were fixing with the "don't test closed changes in some pipelines" fix13:17
fungito accommodate a difference in the behavior of 2.8 we made it so that zuul would leave a comment even if gerrit refused to accept a vote with it13:18
*** changbl has joined #openstack-infra13:18
anteayaoh yes now I remember13:18
*** mriedem has joined #openstack-infra13:19
*** james_li has joined #openstack-infra13:21
openstackgerritJoão Vale proposed a change to openstack-infra/jenkins-job-builder: Support variable-specific substitutions in templates.
*** mflobo has joined #openstack-infra13:22
*** zehicle_at_dell has quit IRC13:23
*** chandan_kumar has quit IRC13:28
*** james_li has quit IRC13:28
anteayaif I star a gerrit patchset, does a new patchset remove the star?13:28
BobBall_Awayno anteaya - at least it didn't used to :)13:29
*** dstanek_zzz is now known as dstanek13:29
BobBall_AwayI had a starred changeset for several patches before I'm sure13:29
*** mwagner_ has quit IRC13:29
anteayaBobBall_Away: okay thanks, not asking because I noticed a change in behaviour, asking to learn what the behaviour is13:29
anteayaI've never known13:30
BobBall_AwayI was just qualifying because I don't know for sure the new behaviour is the same as the old :)13:30
* anteaya nods13:30
BobBall_AwayLooks like it should be because the star is associated with the change rather than patchset (i.e. where you select the star and how it appears in your list)13:30
anteayataking advantage of the lull to improve my gerrit search foo13:31
anteayaI'm looking to see if there is anything that toggles triggered on a new patchset13:31
*** azneita has quit IRC13:31
anteayaas in, have _I_ reviewed this patchset already?13:32
*** lcostantino has joined #openstack-infra13:33
anteayareviewer:self appears to be any changeset I have reviewed or subscribed to13:33
openstackgerritA change was merged to openstack-infra/config: Enable Tempest in Mistral devstack job
*** james_li has joined #openstack-infra13:34
*** zns has joined #openstack-infra13:34
*** david-lyle has quit IRC13:34
*** dkehn__ is now known as dkehnx13:35
anteayaif is:reviewed had a self option, that would fit what I am looking for13:35
*** lcostantino has quit IRC13:36
*** maxbit has joined #openstack-infra13:36
anteayaalas I do not see it13:36
*** mriedem has quit IRC13:37
*** dizquierdo_afk is now known as dizquierdo13:37
*** nosnos has quit IRC13:38
*** bknudson has quit IRC13:41
anteayaoh, this looks nice and cumbersome: reviewer:self label:Code-Review=-1,user=anteaya13:42
*** changbl has quit IRC13:42
*** changbl has joined #openstack-infra13:43
fungiyeah, i think i used to query on changes where i had no vote or a vote of exactly 0 (which would catch new patchsets to changes on which i was still a reviewer)13:43
anteayaso this gets me changesets I have reviewed in the past, with a current patchset I have not yet reviewed => reviewer:self label:Code-Review=0,user=anteaya13:44
anteayayup, just found that13:44
*** gokrokve has joined #openstack-infra13:44
BobBall_AwayThat would be a useful link to add to everyone's board if that was easy...13:44
anteayawe can create dashboards13:44
fungiBobBall_Away: i believe the custom dashboards are already in progress13:44
anteayafungi: is there a dashboard with that as an option?13:45
fungii think right now we only have the important-changes dashboard live13:46
*** chandan_kumar has joined #openstack-infra13:46
*** doug-fish has joined #openstack-infra13:46
openstackgerritA change was merged to openstack-infra/storyboard: Tasks can now be filtered by assignee
*** gokrokve has quit IRC13:48
*** zul has joined #openstack-infra13:49
anteayasee this query label:Code-Review>=0,self returns patches I have already +1'd13:49
* clarkb really isnt awake yet but question over in -dev13:49
anteayathis query label:Code-Review=0,self returns patches I have not reviewed the latest patchset13:50
* anteaya goes to -dev13:50
*** yamahata has quit IRC13:50
*** james_li has quit IRC13:51
*** yamahata has joined #openstack-infra13:51
anteayathe second query is what is useful for me13:51
openstackgerritA change was merged to stackforge/gertty: Add '?' as another way show help dialog
*** markwash has joined #openstack-infra13:52
*** miqui_ is now known as miqui13:53
*** weshay has quit IRC13:53
*** changbl has quit IRC13:54
*** maxbit has quit IRC13:54
*** bknudson has joined #openstack-infra13:54
*** maxbit has joined #openstack-infra13:55
*** weshay has joined #openstack-infra13:55
sdagueanteaya: I think you want13:55
openstackgerritA change was merged to openstack-infra/storyboard: Introducing the event timeline
sdagueNOT label:Code-Review>=0,self13:55
*** zhiyan_ is now known as zhiyan13:56
Alex_Gaynorclarkb: Should the 0 tests == failure thing be live at this point?13:56
anteayasdague: I will try that13:56
sdaguethat will also filter out comments you've provided without a vote13:56
clarkbAlex_Gaynor: no, my change is still in review to fix it I think13:57
sdaguein current patcheset13:57
anteayafungi: can you look at the gerrit db for account 8932, there is a key permissions error for that user13:57
clarkbAlex_Gaynor: after that we need to build new images13:57
fungianteaya: yeah, was just catching up in -dev... checking into it13:57
Alex_Gaynorclarkb: for some reason I thoguht it had landed, what I get for reviewing email first thing in the morning I guess13:57
anteayafungi: thanks13:57
*** salv-orlando has joined #openstack-infra13:57
*** arnaud has joined #openstack-infra13:59
clarkbAlex_Gaynor: oh maybe it merged over night13:59
*** thomasem has joined #openstack-infra14:00
Alex_Gaynorclarkb: but it still needs the new image to take affect -- I wanted to make sure the fact that I hadn't seen the effects was normal14:00
*** james_li has joined #openstack-infra14:00
*** otherwiseguy has joined #openstack-infra14:00
clarkbyup new images needed. I can try and kick those off when I am actually awake14:00
clarkbthough nodepool builds them right about now iirc14:00
anteayasdague: NOT label:Code-Review>=0,self gives me a patch I have -1'd14:01
jeblairclarkb: i left some comments on your gertty changes14:01
fungiclarkb: it merged a few minutes ago, so whatever nodepool images fail to update will need manual intervention14:01
*** lcostantino has joined #openstack-infra14:01
Alex_GaynorNo worries, that's a much lower priority for me then figuring out why no tests run.14:01
anteayawhich isn't exactly what I am looking for14:01
*** gokrokve has joined #openstack-infra14:01
*** prad_ has joined #openstack-infra14:01
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard-webclient: TimeLine events support
jeblairclarkb: i think we want to sync closed changes that we have in the db; i was wondering if maybe you didn't notice that in all cases we only try to sync changes that have been updated since the last sync though14:02
*** zz_gondoi is now known as gondoi14:02
*** terryw has joined #openstack-infra14:03
jeblairclarkb: that should keep the traffic down. of course, it might not be working quite right.14:03
*** otherwiseguy has quit IRC14:03
anteayaso far, is:watched status:open verified=1 reviewer:self label:Code-Review=0,self is giving me what I am looking for14:03
anteayachanges I have already reviewed, on watched repos, which have a new patchest I need to review14:04
clarkbjeblair: responded14:04
clarkbI agree with the -1 but think we still want a bit more checking in that loop14:05
*** blamar has quit IRC14:05
jeblairclarkb: if what you describe is true then there is a bug and we should find that14:05
sdagueanteaya: sure >=-2 then :)14:05
jeblairclarkb: because it's _only_ supposed to sync changes that have been updated14:05
jeblairclarkb: that's not an always-increasing number14:06
jeblairclarkb: it should tend toward 014:06
clarkbok, next time I run it I will have it log debug messages14:06
*** Hal_ has joined #openstack-infra14:06
*** julim has quit IRC14:06
clarkband I can update my change to address your comment until I have hard data on what is happening14:06
*** changbl has joined #openstack-infra14:06
*** azneita has joined #openstack-infra14:07
jeblairclarkb: ok, thx.  similarly, i don't think we should automatically delete changes from the local db immediately -- if we ever delete them, it should probably be something like changes that are closed and haven't been updated in 1 month or something.  but i don't want to do that for a while yet.14:07
openstackgerritVerónica Musso proposed a change to openstack/requirements: Add django-angular as requirement
clarkbjeblair: ok14:08
*** aysyd has quit IRC14:08
*** etoews_ has joined #openstack-infra14:08
anteayasdague: yes, NOT >=-2 yields the same results as =014:08
sdagueanteaya: NOT >=0,self works for me because I also typically hide reviews with negative feedback on them14:09
*** mriedem has joined #openstack-infra14:10
* jeblair decamps14:10
anteayasdague: negative feedback from any reviewer?14:10
*** blamar has joined #openstack-infra14:10
*** etoews has quit IRC14:11
sdagueanteaya: much of the time14:11
anteayaat a certain point it is the toothpaste lid on or off type conversation, everyone has a way they like best14:12
*** e0ne has joined #openstack-infra14:12
*** jgrimm has joined #openstack-infra14:12
thervesdague, Hey, got five to talk about ?14:14
sdaguetherve: sure14:15
thervesdague, Can you develop on not using devstack and custom gate hook?14:15
therveI already use a gate hook, I thought devstack was the way to go when you want db+rabbit+x services14:16
openstackgerritMonty Taylor proposed a change to openstack-infra/publications: Updated projects list to match today
*** james_li has quit IRC14:16
*** e0ne has quit IRC14:17
*** e0ne has joined #openstack-infra14:18
fungimordred: on 92837 it's worth keeping in mind that openstackclient is i guess not actually official. i've been torn on whether we should keep it marked as such on the slide14:18
fungithough it balanced our horizon (which doesn't have a cli) rather well, which is why i had originally arranged them that way ;)14:18
sdaguetherve: there aren't hard and fast rules here, but the feature matrix was largely about "I want an openstack cloud that supports X", and I really don't think that heat_templates definition fits well into that.14:19
*** prad__ has joined #openstack-infra14:19
thervesdague, OK I understand. I was just trying to find a way to keep using devstack but make it lighter14:20
therves/devstack/devstack-gate I suppose14:20
*** e0ne has quit IRC14:20
sdaguetherve: the overhead of starting those other services isn't that much14:20
sdagueor are you seeing perf issues for some reason doing it in a general environment14:20
therveNot it's really about setup14:21
therveIt's taking around 13 mins to run a build, I think it would shave about 514:21
*** prad_ has quit IRC14:22
sdaguewe could propose a change to make the ENABLED_SERVICES not additive but exclusive14:22
sdaguethat would let you do the change at the job level14:22
*** dstanek is now known as dstanek_zzz14:23
therveI don't know if it's used a lot though14:23
*** aysyd has joined #openstack-infra14:23
sdagueit's used some, but I actually think a lot of people think it's exclusive in the way they've set things up14:23
therve8 builders to fix14:23
openstackgerritA change was merged to openstack-infra/storyboard-webclient: TimeLine events support
therveThat would be okay for me, I just don't know the jobs using it to tell if it's diruptive or not14:25
therveIt looks hard to verify beforehand too :)14:25
openstackgerritA change was merged to openstack-infra/nodepool: Create snapshots when min-ready is >= 0
*** kgriffs|afk is now known as kgriffs14:26
*** zns has quit IRC14:27
sdaguetherve: well we probably would want to do ENABLED_SERVICES and ADDITIONAL_SERVICES14:27
sdagueI agree, that it's unclear what some of those things intend14:28
*** BobBall_Away is now known as BobBall14:28
therveThat's one way too14:28
therveSo hum14:28
therve1) Add ADDITIONAL_SERVICES which does the same as ENABLED_SERVICES currently14:28
therve2) Change the builds to use that14:28
therve3) Change the semantic of ENABLED_SERVICES14:28
therveDoes that sound fine?14:29
*** azneita has quit IRC14:29
therveAlright let me try that then14:29
sdaguethe heat templates are only going to syntax check on heat?14:30
sdagueI guess part of it is not really understanding what the job ends up intending to be over time14:30
*** vhoward has joined #openstack-infra14:31
therveIt's not just syntax check it validates them14:31
sdaguegotcha. I'm surprised there isn't a way to do that without heat spun up14:32
openstackgerritA change was merged to openstack-infra/storyboard-webclient: Last location improvement
*** homeless has joined #openstack-infra14:33
therveBecause it uses configuration, resources defined on the server side, etc.14:33
*** adalbas has quit IRC14:33
*** dstanek_zzz is now known as dstanek14:33
*** chandan_kumar has quit IRC14:34
*** chandan_kumar has joined #openstack-infra14:34
fungialso, part of the argument i heard in favor of it is that this is an integration test, making sure that changes to the heat service don't cause it to stop being able to validate the templates, as well as making sure that changes to templates still validate properly14:34
thervefungi, That's correct14:35
*** terryw has quit IRC14:36
*** otherwiseguy has joined #openstack-infra14:36
fungithough i agree that in a perfect world the validation mechanism would be extracted from heat and usable as a stand-alone module, it sounds like (at least for now) it's too intertwined with the running service to be a viable option14:36
*** e0ne has joined #openstack-infra14:37
*** markmcclain has joined #openstack-infra14:37
fungiand would probably involve a lot of mocking and fake backends if "validation" means making sure that the template results in a certain number of systems spun up with the right configuration and interconnectivity14:38
*** thedodd has joined #openstack-infra14:38
*** _nadya_ has joined #openstack-infra14:38
sdaguebut if you need to verify that, then you need all the services14:39
*** Hal_ has quit IRC14:40
sdagueI guess that's where my questioning really is. If it's a syntax check, then it doesn't seem anything should be running. If it's a semantic check that includes handing off the properties from one resource to another then it seems like you need to be able to spin up those resources.14:40
*** mengxd has joined #openstack-infra14:41
therveIt doesn't go as far as the last part14:41
therveBut it's more than a syntax check14:41
clarkbso this came up the first time around14:42
anteayawhat is the openstack-ptl gerrit group for?,members14:42
*** markmcclain has quit IRC14:42
clarkband I specifically questioned the need for devstack because I knew it would take a long time14:42
clarkband requires a lot of resources14:42
clarkbanteaya: git grep says nothing14:43
therveI feel we talked about it already, and I gave the arguments14:43
clarkbmight be in all-projects though14:43
clarkbtherve: but this makes me requestion it if we are suddenly trying to make the thing we knew would suck not suck14:43
clarkband are poitning at d-g instead of heat14:43
anteayaclarkb: k14:43
therveclarkb, It doesn't suck14:44
therveI'm trying to optimize it a bit, that's all14:44
*** adalbas has joined #openstack-infra14:44
clarkbanteaya: doesnt' look like an all projects thing. It might be related to something older14:45
fungianteaya: openstack-ptl was originally going to have voting rights on something... i think maybe it was intended to get included in the requirements-core group14:45
fungiand was supposed to itself include all the individual -ptl groups for official projects14:45
clarkbtherve: sdague: I guess I am confused why a job explicitly requesting openstack via devstack wouldn't want that14:46
therveTo me openstack via devstack is a matrix of services14:46
anteayaclarkb: k14:46
therveYou may want a subset of it14:46
clarkbtherve: sort of. devstack has always been opinionated14:47
anteayafungi: ah, which it doesn't but since it doesn't do anything as a group, no sense wasting energy to make it sync up with governance/programs.yaml14:47
therveclarkb, And always configurable?14:47
fungianteaya: right, and it far predates programs.yaml14:47
clarkbtherve: not like it is today and I feel like that is a bug (we are going to talk about it at the summit)14:47
*** e0ne_ has joined #openstack-infra14:48
clarkbthere is far too much variance today to sanely test the resulting matrix14:48
clarkbwe need to rein it in14:48
clarkband be properly opinionated14:48
*** gokrokve has quit IRC14:48
fungianteaya: yeah, i just confirmed from one of my earlier queries that it's not actually included in any of the official core groups anywhere, so we could probably just delete the group (well, empty it and hide it unless gerrit recently added a group delete feature)14:49
clarkb(that is my opinion :) the summit session should hash out what others think of the situation)14:49
therveI'm fine abandoning my patches and move on, I just want to save some time and resources here and there14:49
clarkbtherve: well I think we end up trading resources14:49
anteayafungi: no objection from to clearing it and hiding it14:49
*** zehicle_at_dell has joined #openstack-infra14:50
anteayaI haven't seen a group delete feature so far in my travels14:50
openstackgerritA change was merged to openstack-dev/hacking: Speed up hacking_has_correct_license
clarkbtherve: automated resources for human resources. Every time we add a new item to the list of things tested is another things that may break without explanation. If we reduce that number of variables humans spend less time debugging (I hope and if not we have problems elsewhere)14:50
fungianteaya: yeah, doesn't look like i have group deletion privs, so i assume it's still not implemented14:50
*** e0ne has quit IRC14:51
*** andreykurilin_ has joined #openstack-infra14:51
therveclarkb, At the abstract level I agree with you. On the concrete problem of devstack and openstack, we want to support people using various sets of components14:51
fungiin a similar vein, anybody know what the Salt Formulas group in gerrit was intended for? it's visible and self-owned but empty14:52
anteayayeah, I saw that but didn't click14:52
clarkbyup so we will always haev to support some variance, but my position is we need to be explicit with that and not ad hoc14:52
anteayaI have no idea14:52
*** gokrokve has joined #openstack-infra14:52
clarkbso far it has all be ad hoc14:52
clarkbwhich has led to spaghetti14:52
*** annegentle_ has joined #openstack-infra14:52
fungii've made salt formulas owned by administrators and hidden it14:53
therveFair enough. I think devstack has grown way outside of its original goals, and that's likely to be the source of the issues you're thinking about.14:53
clarkbI don't know what Salt Formulas is for14:53
*** aysyd has quit IRC14:54
openstackgerritA change was merged to openstack-dev/hacking: Speed up hacking_import_alphabetical
fungilooks like we stopped using openstack-common-core and openstack-common-milestone at some point14:55
clarkbtherve: if  you will be at the summit14:55
clarkbtherve: there are a few other sessions that should touch on it too14:55
*** mengxd has quit IRC14:55
*** mengxd has joined #openstack-infra14:55
*** mengxd has quit IRC14:55
sdagueclarkb: so here's an interesting thing - - if you look at that view, you don't see any negative votes14:55
anteayafungi clarkb so if we ever need the Salt Formulas group we can un-hide it14:55
*** mengxd has joined #openstack-infra14:56
sdaguebut if yuo look at that change in a list of changes, robert has a -1 on it14:56
therveclarkb, I will, thanks for the pointer14:56
anteayafungi: you are doing internal gerrit queries, I am assuming, queries I can't replicate myself, yeah?14:56
sdaguewhich I assume is because he has a -1, however after he voted he was removed from the voting capable group14:56
*** maxbit has quit IRC14:56
clarkbsdague: huh14:56
clarkbsdague: we have seen other weirdness like that with drafts14:57
clarkbthe list view leaks draft info but change view doesn't14:57
fungianteaya: well, you can git grep for those group names in the acls in openstack-infra/config14:57
*** smarcet has quit IRC14:57
sdagueclarkb: fun :)14:57
fungianteaya: and it's possible group inclusion can now be mapped from the new rest api14:57
fungianteaya: though i've been doing db queries for that14:57
clarkbsdague: I'm going to guess that list view code for determining max vote is different than code that iterates and displays in the change view14:58
anteayak, I will add rest api to my list of things to grok, right now I am digging into acls better14:58
anteayaand thanks for the confirmation14:58
clarkbsdague: that may be a worth an upstream bug /me looks for one that exists already14:58
anteayaclarkb sdague the governance repo also has different acls on it than other projects14:59
anteayafor instances registered users can't vote14:59
sdagueanteaya: yep14:59
clarkbanteaya: yup thats why lifeless' -1 was removed when he was removed from the group14:59
*** smarcet has joined #openstack-infra15:00
clarkbI am not finding anything upstream15:00
anteayafungi had voted -1 workflow to get wip changes migrated to new 2.8 and then lost the -1 votes when he removed himself from the group15:01
anteayaI think it was runAS15:01
fungiclarkb: yeah, i've assumed this was intentional behavior in gerrit. for example, put yourself in project bootstrappers, approve a change for a random project which you wouldn't normally have access to, then remove yourself from project bootstrappers and your approval disappears from the change15:01
*** andreaf has quit IRC15:02
fungianteaya: it was actually the workflow +1 votes. for workflow -1 i used impersonation in the ssh api to leave votes as the change owner15:02
sdagueyeh, I was more interested in the fact that the -1 is there enough to leak through on the list view15:02
sdagueso it wasn't actually removed15:03
clarkbfungi: hmm good point. I think the UI should be consistent though15:03
sdagueeven if it's now ignored15:03
clarkbsdague: the comment won't be removed from the DB15:03
clarkbit is the interpretation of the comment that matters15:03
clarkband it seems to have at least 2 ways of doing that15:03
anteayasdague: ah now I see what you are seeing, on the list view15:03
*** yfried has quit IRC15:03
fungiclarkb: yeah, the changes that i approved in that manner failed to merge since i didn't leave myself in project bootstrappers long enough. something to keep in mind for the future15:04
*** dims has quit IRC15:04
clarkbfungi: wow15:05
clarkbthankfully cases where we need to do that are very few and far between15:05
clarkbin fact I think I have only ever done it on projects I am core on15:05
*** dims has joined #openstack-infra15:05
fungiclarkb: well, i didn't submit them, just gave myself elevated privs to re-add missing approval votes15:06
clarkbI see becaus a submit is independent of votes15:06
fungiso they started getting enqueued into the gate, but when it came time for them to merge they bounced out due to lack of approval15:06
clarkb(ish, happens immediately so no problem removing yourself form the group)15:07
anteayawhere is this page published?
clarkbanteaya: http://ci.openstack.org15:08
anteayaI thought it was here:
anteayabut they don't match15:08
anteayawrong link15:09
anteayathere it is, thanks15:09
*** julim has joined #openstack-infra15:09
*** aysyd has joined #openstack-infra15:10
*** e0ne_ has quit IRC15:10
dkranzfungi: I can't seem to figure out how to look at a tempest jobs in jenkins. I heard there was some problem associated with the recent log streaming change. Any clues?15:10
*** freyes has joined #openstack-infra15:11
clarkbdkranz: which log streaming change?15:11
dkranzclarkb: Sorry, I meant streaming to the console.15:12
clarkbdkranz: the tempest one?15:12
clarkbdkranz: anyways if you click on a job in zuul status it takes you to the jenkins job15:12
dkranzclarkb: Yes, the tempest gate jobs now show errors in line15:12
clarkbyou can view the console there15:12
dkranzclarkb: ok, thanks15:13
clarkbor click the link in gerrit when the jobs report back15:13
*** zehicle_at_dell has quit IRC15:14
*** mengxd has quit IRC15:14
*** mengxd has joined #openstack-infra15:15
*** arnaud has quit IRC15:16
*** vhoward has left #openstack-infra15:18
*** azneita has joined #openstack-infra15:18
*** WaltNik has quit IRC15:18
anteayain all project access there is no section for refs/tags/*15:18
*** marcoemorais has joined #openstack-infra15:19
anteayadoes refs/tags/* inherit anything from refs/* ?15:19
*** marcoemorais has quit IRC15:19
anteayayes, I guess it would15:19
clarkbanteaya: it does sort of. Technicaly it does but gerrit knows to treat tags as special15:19
clarkbanteaya: so even if it says registered users can push to refs/* (or whatever it says) you need different permissions for refs/tags15:19
anteayaI'm trying to see where the create line for refs/tags/* is necessary for some repos if indeed it is15:20
anteayaclarkb: k15:20
clarkbanteaya: it is necesssary because tags are special15:20
anteayait is?15:20
clarkbanteaya: in individual project ACLs it is15:21
clarkbanteaya: because we give project specific groups access to push tags15:21
anteayamordred said it isn't, but that fungi said it is for some15:21
clarkbanteaya: do you haev an example of what mordred say isn't necessary?15:21
*** geertj has quit IRC15:21
anteayahe said the create line isn't necessary let me find the logs15:22
*** markmcclain has joined #openstack-infra15:22
*** ociuhandu has quit IRC15:22
*** marcoemorais has joined #openstack-infra15:22
*** praneshp has joined #openstack-infra15:23
anteaya starting around timestamp 2014-05-02T21:16:5015:24
*** matsuhas_ has quit IRC15:25
*** matsuhashi has joined #openstack-infra15:25
mriedemmengxd: hey, this powervc driver license is not correct
*** zhiyan is now known as zhiyan_15:26
*** zns has joined #openstack-infra15:26
dkranzclarkb: So I found it. Didn't realize we were on jenkins07 now :). But you can't really get there from zuul. When I click on the link, I see the jobs but it then immediately jumps to the gerrit review page, but with the jenkins url showing in the status bar. Strange.15:26
*** moted has quit IRC15:26
*** atiwari has joined #openstack-infra15:26
anteaya2014-05-02T21:18:52  <anteaya> so acl files for 2.8 do not need a create line for refs/tags/*15:26
anteaya2014-05-02T21:19:00  <anteaya> okay thanks15:26
anteaya2014-05-02T21:19:07  <mordred> anteaya: that's correct15:26
clarkbanteaya: ok I grok now.15:26
*** annegentle_ has quit IRC15:26
clarkbanteaya: the section is necessary, the specific create is not15:26
anteayaclarkb: so what am I missing?15:26
anteayait just needs the refs/tags/* header with pushSignedTag15:27
clarkbdkranz: uh I don't know how that could happen15:27
clarkbdkranz: the links on the zuul status page are to jenkins15:27
clarkbdkranz: and our jenkins shouldn't redirect you15:27
anteayabut I keep seeing the create line since it was necessary with 2.4 I guess15:27
clarkbanteaya: yup15:27
anteayaand I seem to be the only one commenting on them in patches with acl edits15:28
*** moted has joined #openstack-infra15:28
anteayaso I am doing a sanity check15:28
*** azneita has quit IRC15:28
clarkbanteaya: I think part of the issue is that people cargo cult and documentation probably needs an update15:28
clarkbanteaya: we should udpate all of the existing acls at once along with the docs15:28
*** wenlock has joined #openstack-infra15:28
*** geertj has joined #openstack-infra15:29
anteayaI think the docs may be updated, but not the other acls15:29
anteayalet me see if I can figure out a patch15:29
*** jaypipes has joined #openstack-infra15:29
clarkbdkranz: I am not able to reproduce that behavior15:29
fungito clarify tags and create perms, mordred's original patch to switch to the signedtag permission was also going to delete all create perms anywhere in any acl. i pointed out that some project use create on refs/* to be able to add branches, so he agreed for the time being to not delete create everywhere (but i think he did at least test and confirm that you don't need create on refs/tags/* to be able15:29
fungito push a tag)15:29
dkranzclarkb: I looked a little more. The region around each zuul entry on the page seems to connect to a set of jenkins links. But the actual blue underlined link is to gerrit.15:30
*** terryw has joined #openstack-infra15:30
dkranzclarkb: So it seems that if you click the link it both shows the jenkins links and navigates to gerrit at the same time :)15:31
clarkbdkranz: oh you need to expand the block15:31
anteayafungi: your patch has a handy find and replace command in the commit message thank you15:31
Alex_Gaynorclarkb, lifeless: So what's the next step in debugging the weird "testr runs no tests on py3" issue?15:31
clarkbdkranz: just click on the block and it will drop open15:31
anteayaI will work with that as a starting point and see if I can create a patch to remove the create line in refs/tags/*15:31
*** matsuhashi has quit IRC15:31
dkranzclarkb: Right, but there is zero visual indication that there  is such a block. I get it now though.15:31
anteayathen if we all test and agree it isn't required we can merged15:31
clarkbdkranz: agreed15:32
clarkbdkranz: but it is a tradeoff between way too much noise on that page and slightly bad UI15:32
clarkbAlex_Gaynor: lifeless indidcated discover finds no tests successfully15:32
clarkbAlex_Gaynor: you should figure out why discover would find no tests15:32
dkranzclarkb: Not complaining. I just hadn't looked at zuul in a bit since it changed.15:33
*** otherwiseguy has quit IRC15:33
sdaguedkranz: that's been like that for months :)15:33
*** changbl has quit IRC15:33
fungianteaya: i think what i would probably do is to write a quick ini parser in python which normalizes the acls to what we want, with deterministic ordering, dropping unnecessary things (i was going to do something along those lines as a throwaway script but just haven't found time). that would allow us to then build a list of checksums and identify identical acls, so we could combine them via15:33
fungireferences in projects.yaml15:33
fungianteaya: alog with that i was going to collapse/replace the two current infra core groups we have into one15:34
*** chandan_kumar has quit IRC15:34
fungiand then after those steps, it should be possible to more easily spot divergences in acls which need further alignment, and might allow for even more consolidation15:34
anteayafungi: your way sounds better15:35
clarkbAlex_Gaynor: actually I have py33 now and am curious I may take a quick look15:35
anteayalet me know when you have something up15:35
*** marcoemorais has quit IRC15:35
*** e0ne has joined #openstack-infra15:36
Alex_Gaynorclarkb: ^ not sure how to reconcile those two things15:38
Alex_Gaynorthe pdb sesssion is from inside of testrepository and is looking at the command that fails15:39
*** ramashri has joined #openstack-infra15:45
clarkbAlex_Gaynor: huh so it is still failing on that all?15:47
clarkbafter fixing my pip.conf to make it python3 safe I almost have discover running15:47
Alex_Gaynorclarkb: Heading to the office now, will be back in 30 or so15:48
*** changbl has joined #openstack-infra15:48
*** lcheng_ has joined #openstack-infra15:49
*** mrodden has joined #openstack-infra15:51
*** etoews has joined #openstack-infra15:55
*** atiwari has quit IRC15:55
anteayasome groups have -core, -ptl and a -milestone group15:57
anteayawhat are the conditions when a -milestone group should be created?15:57
clarkbAlex_Gaynor: its a filtration problem15:57
clarkbAlex_Gaynor: it will run functional and unit tests without the filter15:58
*** etoews_ has quit IRC15:58
clarkbnow to figure out why that would matter15:58
*** marun_afk is now known as marun16:01
*** zns has quit IRC16:01
*** zns has joined #openstack-infra16:02
clarkbAlex_Gaynor: also you are using py34. not sure if that matters16:02
*** spy1 is now known as antonym16:02
*** markmcclain has quit IRC16:02
Alex_Gaynorclarkb: it's the same symptoms. I'm assuming it's a Python 3 in general thing (but you know what they say about assumptions...)16:03
clarkbAlex_Gaynor: so long story short I think discover is fine16:03
clarkbbecause `testr run --parallel` in that venv works but `testr run --parallel tests.unit` doesn't16:04
*** pcrews has joined #openstack-infra16:05
clarkbAlex_Gaynor: and `testr run --parallel tests` runs one test16:06
*** gokrokve_ has joined #openstack-infra16:07
*** markmc has quit IRC16:08
*** wenlock has quit IRC16:09
*** Ajaeger has joined #openstack-infra16:11
*** gokrokve has quit IRC16:11
anteayado stackforge projects need an abandon line in the refs/heads/* stanza in the acl?16:12
clarkbanteaya: yes please16:12
*** Alexei_987 has quit IRC16:14
*** atiwari has joined #openstack-infra16:15
*** changbl has quit IRC16:16
clarkbAlex_Gaynor: I think the error yo usee in pdb is due to the variable nto being expanded properly?16:16
clarkbyeah if I edit .testr.conf pdb runs successfully16:17
clarkbwth is going on16:17
*** _nadya_ has quit IRC16:18
*** lakshmiS has joined #openstack-infra16:20
*** andreaf has joined #openstack-infra16:21
openstackgerritDirk Mueller proposed a change to openstack-dev/hacking: Set version of master tree to 0.9.0
*** kgriffs is now known as kgriffs|afk16:28
*** thingee_ is now known as thingee16:29
*** changbl has joined #openstack-infra16:29
*** derekh_ has quit IRC16:30
*** dstanek is now known as dstanek_zzz16:31
*** Longgeek has quit IRC16:32
AjaegerHi infra-team, the database-api project  can get retired now - its content has been moved to the trove repository. What is needed to close it?16:33
*** UtahDave has joined #openstack-infra16:36
anteayaAjaeger: so it was alive just under 4 months16:36
*** homeless has quit IRC16:36
fungiAjaeger: add it to the infra meeting agenda so we can discuss moving it into the attic16:37
fungiwe've already got a few there being relocated to the openstack-attic namespace16:38
clarkbAlex_Gaynor: I see the bug16:39
*** homeless has joined #openstack-infra16:39
clarkbAlex_Gaynor: I will haev a link and a possible fix16:39
Ajaegerfungi: Added - but to the wrong section. Will move ;)16:41
clarkbAlex_Gaynor: taht clsure consumes the filters iterable the first time include is called16:41
clarkbAlex_Gaynor: subsequent calls iterate an empty iterable and noop16:41
clarkbI think this must be a difference in map behavior python2 vs python316:41
clarkbiirc map gives you a list in python2 but an iterable in python3 \o/16:41
clarkbfilters = list(map(re.compile, self.test_filters)) is the fix I think16:42
clarkblifeless: ^16:42
clarkbI feel like I accomplished something this morning16:43
msabramoyep, python3 map gives you an object rather than a list16:43
*** jpich has quit IRC16:44
msabramo❯ python2.7 -c 'print(type(map(lambda x: x, [1])))'16:44
msabramo<type 'list'>16:44
msabramo❯ python3.3 -c 'print(type(map(lambda x: x, [1])))'16:44
msabramo<class 'map'>16:44
*** marcoemorais has joined #openstack-infra16:45
clarkblist(somemapobject) seems to work fine16:45
clarkbI'm guessing map isn't much more special than an iterable16:45
*** crc32 has joined #openstack-infra16:45
clarkbthe type not the function16:46
*** bhuvan has joined #openstack-infra16:49
*** davidlenwell_ is now known as davidlenwell16:49
mkodererhi folks, QA will have a meet-up on sunday before the summit16:53
mkodererif somebody is interested or ML16:53
openstackgerritA change was merged to openstack-infra/devstack-gate: Add heat-cfntools, heat-templates to PROJECTS
*** andreykurilin_ has quit IRC16:55
*** yfried has joined #openstack-infra16:56
*** gyee has joined #openstack-infra16:56
anteayamkoderer: thanks, what time are you gathering?16:56
Ajaegerclarkb: I took the liberty to include you explicitely in the translation discussion on the i18n mailing list - if I shouldn't do that in the future, please tell me16:56
anteayathe joint board/tc meeting is 3-5 which I am planning on attending16:56
mkodereranteaya: not fixed yet.. I will try to reserve a table around 8pm I guess16:56
fungimkoderer: i'll be there in spirit, but more corporeally i'll probably be at the foundation staff/board dinner16:57
Ajaegerclarkb: I welcome your feedback in that discussion ;)16:58
clarkbAjaeger: its ok. actually on that topic I dont really have an answer. I think we need to figure out if incomplete translations are helpful to users16:58
mkodererfungi: I see.. yep it's not easy to find a empty slot :)16:58
fungithough maybe i can just drink with the board and then bail early to meet up with qa peeps16:58
*** melwitt has joined #openstack-infra16:58
Ajaegerclarkb: yep, let's see...16:58
clarkbI dont properly speak a second language so I am not sure myself16:58
*** markmcclain has joined #openstack-infra16:58
Ajaegerclarkb: English is my second language, not sure whether I speak it properly ;)16:59
fungiAjaeger: english is my first (and effectively only) language and i'm not sure whether i speak it properly16:59
clarkbits worth noting I dont think any project translates by default16:59
clarkbso users have to be interested in the translations. horizon is special because the user is the end user not sysadmin17:00
*** e0ne has quit IRC17:00
anteayamkoderer: kk thanks17:00
*** harlowja_away is now known as harlowja17:01
anteayaI try to have no plans around the planned things, but it is nice to be aware this is happening17:01
Alex_Gaynorclarkb: ahhhhhhh. nice find! Are you/lifeless working on a patch, or should I?17:04
clarkbI can do it, but I havent done it yet17:04
clarkbits just that one line change17:04
clarkbwrapping map() in list()17:05
clarkbprobably should add a test too17:05
Alex_Gaynorclarkb: Sounds good -- I don't really know how to launchpad/bzr17:06
clarkblifeless will accept patch files17:06
Ajaegerclarkb: yeah, I see clear value for Horizon translation but others like nova is a complete different story17:06
clarkbAjaeger yup17:06
*** Guest51066 has joined #openstack-infra17:07
*** andreaf has quit IRC17:07
*** ramashri has quit IRC17:08
*** aconrad has joined #openstack-infra17:09
*** arnaud has joined #openstack-infra17:10
*** kgriffs|afk is now known as kgriffs17:11
*** Guest51066 is now known as annegentle17:11
anteayahi arnaud17:13
arnaudcould I get core reviews for ?17:13
anteayathat is the glance specs repo17:13
*** mestery_ has joined #openstack-infra17:14
*** mestery has quit IRC17:14
*** UtahDave has quit IRC17:15
*** mestery_ is now known as mestery17:16
*** terryw is now known as otherwiseguy17:16
anteayaI am going to get some sunshine17:16
*** UtahDave has joined #openstack-infra17:17
*** amcrn has joined #openstack-infra17:21
*** gokrokve_ has quit IRC17:21
*** kgriffs is now known as kgriffs|afk17:21
*** jamespage_ has joined #openstack-infra17:21
*** gondoi is now known as zz_gondoi17:27
*** dstanek_zzz is now known as dstanek17:29
*** zul has quit IRC17:30
bnemecAnybody know why I can't vote on (or a bunch of other patches in that repo)17:32
Ajaegerbnemec: Neither can I vote - not evven +1/-117:33
AjaegerI see "Submit TypeRULE_ERROR"17:33
Ajaegersomething is broken with the patch17:33
bnemecYeah, that would be my guess too.17:35
bnemecWould just like to be able to do something with those because they're cluttering up the tripleo open reviews.17:35
clarkbI bet there is a draft17:35
clarkbso stop using drafts17:35
clarkbnext release of gerrit we can disable them17:37
bnemecOkay, looks like I can still abandon it, so I might see if anyone would object to that.  Those changes need to go away anyway.17:37
*** kgriffs|afk is now known as kgriffs17:37
bnemecclarkb: Thanks for the explanation17:37
*** Ryan_Lane has joined #openstack-infra17:37
*** ominakov_ has joined #openstack-infra17:38
*** rcarrillocruz has joined #openstack-infra17:40
*** ominakov has quit IRC17:41
*** rcarrill` has quit IRC17:42
*** jamespage_ has quit IRC17:43
*** ominakov_ has quit IRC17:45
*** dizquierdo has quit IRC17:47
fungiclarkb: from last night's scrollback, you talked about cleaning out references to pbr's testr setup target... is that basically just making sure we get rid of references to it in tox configs of all projects (and possibly slave scripts/job configs where we've missed removing it too)?17:47
fungiwell, s/getting rid of/replacing it with test/17:47
*** hogepodge has joined #openstack-infra17:48
*** krotscheck has quit IRC17:49
*** annegentle has quit IRC17:49
* clarkb wears down the ssd in his laptop with `gertty -d`17:49
fungiground and polished smooth17:49
*** krotscheck has joined #openstack-infra17:50
clarkbfungi: ya, projects that do `python testr` use the distutils hook in testrepository whihc lifeless doesn't like17:50
*** reed has joined #openstack-infra17:50
clarkbfungi: `python test` in pbr'd projects will run the hook in pbr which is preferable because we can release fixes for that and so on17:50
*** thedodd has quit IRC17:50
openstackgerritClark Boylan proposed a change to stackforge/gertty: Removed closed changes from unreviewed list.
clarkband now back to patching testr17:51
* clarkb learns him a bzr17:51
Ajaegerclarkb, fungi: Until recently jobs with negative reviews expired after a week - I haven't seen this for a few weeks now. Is that a concise change?17:52
dhellmannhi, folks, I need a little help debugging an error with an oslo.i18n test job:
clarkbAjaeger: I think it is fallout from the gerrit upgrade17:52
clarkbAjaeger: and is a bug17:52
dhellmannI think that just means the oslo.i18n repo wasn't checked out somewhere, but I'm not sure why that would be or what controls that.17:53
Ajaegerclarkb: shall I file one?17:53
fungiAjaeger: do you have an example of one with negative reviews and no activity open for more than two weeks?17:53
fungior was it only a week? i don't remember17:53
*** SumitNaiksatam has joined #openstack-infra17:53
Ajaegerfungi: - last change 22nd april17:54
*** thedodd has joined #openstack-infra17:54
*** amcrn_ has joined #openstack-infra17:54
Ajaegerfungi: I think it was a week17:54
*** homeless has quit IRC17:54
*** _nadya_ has joined #openstack-infra17:54
Ajaegerfungi: Just opened one of those emails, it contains "code review expired after 1 week of no activity after a negative review"17:54
fungiregardless, yeah that's more than 2 weeks17:54
*** gokrokve has joined #openstack-infra17:54
fungibug, sure17:55
*** mestery has quit IRC17:55
AjaegerOk, will file a bug report17:55
*** mestery has joined #openstack-infra17:55
clarkbdhellmann: we may need to add oslo.i8ln to the d-g PROJECTS var17:55
dhellmannclarkb: ah, yeah, I'll bet that's it17:55
clarkbdhellmann: doesn't look like setup workspace copied the repo out of the cache since it isn't in the PROJECTS list17:55
fungiAjaeger: clarkb: i see the (a?) bug already...17:55
dhellmannclarkb: ok, I'll go review my checklist and then do that17:56
dhellmannclarkb: thanks17:56
clarkbdhellmann: np17:56
Ajaegerfungi: so no need for a bug report, ok will stop ;)17:56
clarkbfungi: that was fast17:56
clarkbninja debugging17:56
fungiAjaeger: clarkb: the cron job tries to run 'python /usr/local/bin/expire-old-reviews launchpadsync /home/gerrit2/.ssh/launchpadsync_rsa' and /home/gerrit2/.ssh/launchpadsync_rsa dne17:56
*** amcrn has quit IRC17:56
fungiprobably never puppeted and so didn't end up on the new machine17:56
AjaegerArgh ;(17:57
fungichecking now to confirm what we need to do to get it set up properly there17:57
Ajaegerfungi: that was quick!17:57
*** mrodden has quit IRC17:57
openstackgerritDoug Hellmann proposed a change to openstack-infra/devstack-gate: add oslo.i18n to project list
*** vponomaryov has quit IRC17:59
openstackgerritDoug Hellmann proposed a change to openstack-infra/devstack-gate: add oslo.i18n to project list
*** Sukhdev has joined #openstack-infra17:59
fungiyup, we have a script_key_file variable we set the filename in, but never actually create that file, so i need to get it ported into hiera from the old server17:59
fungipatch on the way18:00
sandywalshhey infra-gods ... our stacktach repo was missing a tox.ini ... so we can't do any reviews. Can we freshen up the repo from ?18:01
uvirtbotLaunchpad bug 1317607 in testrepository "testr test filters fail to apply under python3" [Undecided,New]18:01
clarkbsandywalsh: just push a tox.ini18:01
Alex_Gaynorclarkb: thanks so much18:01
sandywalshclarkb: to where? I thought it all had to go through the review process?18:01
clarkbsandywalsh: yup via gerrit18:01
dhellmannsandywalsh: if you submit a review with a tox.ini, it will be there when the job tries to use it18:01
fungisandywalsh: you review a change which adds tox.ini18:02
*** yfried has quit IRC18:02
sandywalshah, true ... lemme try that ... thanks18:02
dhellmannclarkb: would appreciate your eyes on ^^ as well as the oslo-specs import ( when you have a few minutes18:02
*** UtahDave has quit IRC18:03
clarkbdhellmann: d-g change approved18:04
clarkbcan look at specs when I finsih this testr thing18:04
dhellmannclarkb: thanks!18:05
*** homeless has joined #openstack-infra18:05
*** james_li has joined #openstack-infra18:07
*** liam__ has joined #openstack-infra18:07
*** liam__ has quit IRC18:07
*** davidlenwell has quit IRC18:07
*** davidlenwell has joined #openstack-infra18:09
*** mriedem has quit IRC18:09
*** wenlock has joined #openstack-infra18:09
*** Ng_ has joined #openstack-infra18:11
*** Ng_ has quit IRC18:12
*** mriedem has joined #openstack-infra18:14
*** _nadya_ has quit IRC18:14
*** zul has joined #openstack-infra18:14
*** mwagner_ has joined #openstack-infra18:14
*** markmcclain has quit IRC18:14
*** bhuvan has quit IRC18:15
clarkbhas anyone run testrepository unittests recently?18:16
clarkbsdague: ^18:16
*** rcarrillocruz has quit IRC18:17
*** bhuvan has joined #openstack-infra18:17
*** bhuvan has quit IRC18:17
*** bhuvan has joined #openstack-infra18:17
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Puppet the key used for expiring old reviews
*** rcarrillocruz has joined #openstack-infra18:17
clarkbI know I fgigured this out before but I have created two virtualenvs one for python2.7 and one for python3.3 and running `make` in either results in AttributeError: 'module' object has no attribute 'tests'18:17
*** mrodden has joined #openstack-infra18:17
anteayafungi: line 5 is missing a comma:
*** morganfainberg_Z is now known as morganfainberg18:22
fungianteaya: thanks. the change is premature anyway because i started digging in hiera and it looks like the keys are already in there just under other names18:22
clarkblooks like my test woes were a dependency thing + one other thing I worked around >_>18:25
anteayaso have you figured it out yet?18:25
clarkbnope all test dependency issues ok good tests are running18:26
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Puppet the key used for expiring old reviews
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Puppet the key used for expiring old reviews
*** bhuvan_ has joined #openstack-infra18:27
fungianteaya: fixed your comment (forgot to do it on the previous patch)18:27
*** darraghb has quit IRC18:27
*** Ng has quit IRC18:27
*** yfried has joined #openstack-infra18:28
clarkbAlex_Gaynor: now with a patch18:29
*** bhuvan has quit IRC18:29
*** Ng has joined #openstack-infra18:30
*** markmcclain has joined #openstack-infra18:31
Alex_Gaynorclarkb: awesome18:32
*** homeless has quit IRC18:33
*** _nadya_ has joined #openstack-infra18:34
anteayafungi: +118:35
openstackgerritBen Nemec proposed a change to openstack-infra/config: Add dib-utils project
Ajaegeranteaya: thanks!18:36
marunclarkb: I think the neutron functional job should be made voting, but I'm not sure how to verify the failure rate.18:38
marunclarkb: How do I ensure that failures are due to code rather than the job?18:39
clarkbmarun: you can use logstash to filter all failures against known bugs18:40
marunclarkb: er, is there an easy way to do that?18:41
*** thedodd has quit IRC18:41
clarkbmarun: I'm not aware of an easy way :) sdague may have ideas though18:42
clarkbI think he spends more time than anyone doing analysis like that18:42
marunsdague: ^ The neutron functional job seems stable to me but I'd like to be sure before proposing it become voting.18:42
marunclarkb: ok, thanks :)18:42
marunsdague: making it voting would impact devstack since it relies on it for setup.18:43
openstackgerritDoug Hellmann proposed a change to openstack/requirements: Add oslo.db and oslo.i18n to project list
*** otherwiseguy has quit IRC18:45
clarkbdhellmann: oslo-specs review should show up shortly once gertty syncs18:46
dhellmannclarkb: thanks! (what's gertty?)18:46
fungidhellmann: console-based gerrit client built on top of its rest api18:47
dhellmannfungi: is that something you guys have built?18:47
fungijeblair's written the bulk of it, though others have started adding changes18:47
clarkbdhellmann: its mostly jeblair and its still pretty rough around the edges, but I find it works really well for sitting down and doing reviews all at once18:47
clarkbbut it is async and offline (good for airplanes) whihc means after I enter a review it takes a little while to catch up18:48
fungidhellmann: imap mail client like batch local/offline operations18:48
dhellmannclever idea, I'll have to give that a try18:48
openstackgerritA change was merged to openstack-infra/config: Scale back on tripleo testing
fungiclarkb: how well does it handle pushing review votes when there are new patchsets since the last time you sync'd?18:49
fungii haven't used it enough yet to find out18:49
clarkbfungi: I haven't run into that18:49
clarkbMy gertty log is up to 31MB will have to restart it here without the debug logging18:50
dhellmannI gave the link to markmcclain, since he's on airplanes more than he's on the ground I thought he'd be a good beta user :-)18:50
clarkbdhellmann: I think jeblair is trying to not promise a whole lot18:50
dhellmannclarkb: open source, right? :-)18:51
openstackgerritAnita Kuno proposed a change to stackforge/gertty: Expand the .gitignore file to ignore .egg files
anteayahey!! I can see stuff now18:53
*** jerryz has joined #openstack-infra18:53
*** arnaud has quit IRC18:55
Ryan_Lanehas anyone checked to see if the search issues are gone in the wiki since the maintenance the other day?18:57
*** xdmeng has joined #openstack-infra18:57
*** mengxd has quit IRC18:57
fungiRyan_Lane: i just checked now, and at least the index seems to have been updated18:58
*** whoops has joined #openstack-infra18:58
fungithings which i removed months ago have (in the past few days) ceased showing up in wiki search results18:58
anteayaRyan_Lane: well done18:58
Ryan_LaneI had completely rebuilt the index18:58
Ryan_Lanelet me know if anything fails18:58
*** zz_gondoi is now known as gondoi18:58
fungiwill do, and thanks!18:58
Ryan_Laneturns out you need to run maintenance scripts when you update the search plugin ;)18:59
Ryan_Lanewhich I didn't do last time18:59
openstackgerritA change was merged to openstack-infra/config: Create glance-specs git repo
*** james_li has quit IRC19:00
*** wenlock has quit IRC19:00
openstackgerritDoug Hellmann proposed a change to openstack-infra/config: Create an oslo-specs repository
clarkbArxCruz: want to follow up with a change that fixes zeromq too?19:01
*** jgrimm has quit IRC19:01
ArxCruzclarkb: sure :)19:01
*** mrodden has quit IRC19:03
*** james_li has joined #openstack-infra19:04
*** openstackgerrit has quit IRC19:04
*** openstackgerrit has joined #openstack-infra19:06
*** amcrn_ has quit IRC19:07
*** homeless has joined #openstack-infra19:07
*** primeministerp has quit IRC19:09
*** andreaf has joined #openstack-infra19:10
*** andreaf has quit IRC19:10
*** andreaf has joined #openstack-infra19:11
*** james_li has quit IRC19:11
*** Ng has quit IRC19:12
*** arnaud has joined #openstack-infra19:17
zaroclarkb: i think that's what jeblair wanted in 88432, maybe should just wait for him to comment.19:17
clarkbzaro: ok19:17
*** _nadya_ has quit IRC19:18
clarkbreed: can you add an etherpad to for your session?19:18
reedclarkb, I sure can, thanks for the poke19:19
*** etoews_ has joined #openstack-infra19:19
krtaylorclarkb, for the 3rd party session, are you set?19:20
clarkbkrtaylor: I think so19:20
krtaylorclarkb, I saw the etherpad for it, glad to see some of the topics19:20
clarkbkrtaylor: feel free to add stuff to the therpad19:21
krtayloralot of the stuff I was wanting to see discussed19:21
krtaylorI'll go through my etherpad and cross reference19:21
*** arnaud has quit IRC19:21
*** etoews has quit IRC19:23
fungiclarkb: zaro: do you have any insight into whether my jenkins publisher assumptions on about ftp remove-prefix vs scp keep-hierarchy are correct, or whether i need to set up some test jobs to confirm?19:25
clarkbfungi: keep hierarchy will flatten if set to false iirc19:27
fungibascially we want to scp x/** (which contains x/y/z) to the site and get y/z there19:27
clarkbfungi: so foo/bar/baz and foo/zort get copied as /baz and /zort19:27
fungiother ideas as to how to do that with the scp publisher?19:27
clarkbfungi: use the remove prefix as before19:27
fungii didn't see that as an option for the scp publisher19:28
clarkboh you are switching to scp19:28
*** Ng has joined #openstack-infra19:28
clarkbI missed that /me thinks more19:28
fungiyeah, that was the suggestion was that we should do scp instead of ftp now19:28
clarkbfungi: jeblair said no19:28
fungiwith scp we lose remove-prefix it seems19:28
clarkbhe wants to do it all at once rather than piecemeal19:28
fungithen i need to recreate all the sites on the 8 jenkins masters again19:29
clarkbbecause apparently the reason we have ci.o.o in a weird place is this was attempted piecemeal forever ago and never finished19:29
fungiokay, will do19:29
clarkbfungi: well its certainly worth talking to him more about19:29
clarkbespecially if the work is done. I will keep looking at how to make this work with scp19:29
fungiwell, i can redo it with ftp, just don't want the docs team waiting on me any longer than they already have been19:30
clarkboh right I remember. We have worked around this my copying/moving files19:30
sdaguemarun: what's the question exactly?19:30
clarkbif you have files at foo/bar/** but don't want foo/bar/** you mv foo/bar ./19:30
fungiif this isn't straightforward to solve with the scp publisher and we'd rather just add more ftp sites for now and switch them later, then i'll just redo these bits19:30
clarkbthen copy them19:30
clarkbfungi: I prefer scp myself, but defer to jeblair since he was around the first time this was attempted19:31
fungisure, makes sense19:31
*** thedodd has joined #openstack-infra19:31
*** wenlock has joined #openstack-infra19:32
*** Ng_ has joined #openstack-infra19:32
*** Ng has quit IRC19:32
*** annegentle has joined #openstack-infra19:33
*** Ng_ is now known as Ng19:33
*** geertj has quit IRC19:35
*** annegentle has quit IRC19:35
clarkbya so the scp plugin really doesn't have a way to do it without that mv trick above19:36
*** annegentle has joined #openstack-infra19:36
clarkbits not just a jjb limitation19:36
*** annegentle has quit IRC19:36
*** mrodden has joined #openstack-infra19:36
*** annegentle has joined #openstack-infra19:36
*** annegentle has quit IRC19:37
marunsdague: how do I know if it's time for the check-neutron-dsvm-functional job to become voting19:37
marunsdague: (and gating)19:37
openstackgerritJoe Gordon proposed a change to openstack-infra/devstack-gate: Add more fixed ips
openstackgerritK Jonathan Harker proposed a change to openstack-infra/config: Move openstack-specific config out of ::jenkins
*** annegentle has joined #openstack-infra19:39
*** geertj has joined #openstack-infra19:40
*** wenlock has quit IRC19:40
marunsdague: I'd like to start refactoring the unit test suite into unit and functional test, but it won't make sense to merge anything until the functional job is gating.19:40
sdaguemarun: the following query will give you one hit per run -
sdagueif you click on build_status -> score you can get the current pass / fail ratio, which looks like about 8% failure19:44
sdaguethe real question is whether those failures are actually catching real bugs or are fail negatives19:44
lifelessclarkb: nice find19:44
sdagueI typically go through and spot check19:44
*** mrodden1 has joined #openstack-infra19:45
*** mrodden has quit IRC19:47
openstackgerritJoe Gordon proposed a change to openstack-infra/config: Add neutron-large-ops-testing job
*** _nadya_ has joined #openstack-infra19:49
*** krotscheck has quit IRC19:50
*** krotscheck has joined #openstack-infra19:51
openstackgerritJeremy Stanley proposed a change to openstack-infra/config: Publish API/CLI/SDK site to developer.o.o
*** _nadya_ has quit IRC19:53
marunsdague: ok19:53
clarkblifeless once I figured out pdb isnt installed by virtualenv it was easy19:54
marunsdague: is there a way to correlate failures between jobs?19:55
marunsdague: i.e. filter out functional job failures that coincide with unittest failures?19:55
*** lakshmiS has quit IRC19:55
*** otherwiseguy has joined #openstack-infra19:56
*** rcarrill` has joined #openstack-infra19:57
*** markmcclain has quit IRC19:58
*** marun is now known as maru_afk19:58
*** rcarrill` has left #openstack-infra19:58
*** rcarrillocruz has quit IRC19:58
*** thomasem has quit IRC19:58
*** thomasem has joined #openstack-infra19:58
*** thomasem has quit IRC19:58
*** Sukhdev has quit IRC19:59
*** thomasem has joined #openstack-infra19:59
*** gondoi is now known as zz_gondoi20:00
openstackgerritDoug Hellmann proposed a change to stackforge/gertty: Alternate colors for light-background terminals
*** markmcclain has joined #openstack-infra20:01
*** markmcclain1 has joined #openstack-infra20:03
*** zz_gondoi is now known as gondoi20:06
*** markmcclain has quit IRC20:06
*** whoops has quit IRC20:06
*** dizquierdo has joined #openstack-infra20:07
*** julim has quit IRC20:08
*** grantbow has quit IRC20:09
*** grantbow has joined #openstack-infra20:09
fungithat's ^ how you can tell people who spent more time on serial terminals vs workstations ;)20:10
sdaguemaru_afk: not easily20:12
dhellmannfungi: ?20:13
fungidhellmann: preference for dark backgrounds vs light backgrounds20:13
dhellmannfungi: ah, I get eye strain staring at white on black20:13
fungixterm defaulted to black text on white background going back to the beforetimes20:14
dhellmannfungi: but I have a weird hypersensitivity to green, too, so I used to turn the brightness all the way down on the crts20:14
fungidhellmann: oh, yeah, you probably liked the amber terminals better20:14
dhellmannfungi: yep, 20 years ago I was a light on dark guy, not so much any more20:14
dhellmannamber was the best20:14
fungithose were definitely easier on the eyes20:14
dhellmannI can read those green screens on settings where most people can't even tell they are on -- it's like a lame super power20:15
anteayaI blowed up gertty:
anteayaI selected a patch to open and gertty stopped20:16
*** geertj has quit IRC20:16
*** arborism has joined #openstack-infra20:16
anteayawhat I was looking for apparently didnt' exist20:16
fungigood news everyone! we can *actually* push acls to gerrit which look like (confirmed pushing through the ssh api on review-dev). it doesn't store them that way on disk, but means we can keep them more readable and editable (no stupid indents, hard tabs, parseable by configparser)20:17
anteayadhellmann: ha ha ha20:17
anteayasound like a reddit thread20:17
*** dstanek is now known as dstanek_zzz20:18
*** derekh_ has joined #openstack-infra20:18
anteayafungi: nice20:19
anteayafungi: though you have said we don't need [project] status = active anymore, correct?20:19
fungianteaya: right, this was just a raw acl i pulled from one of the test projects on review-dev20:19
fungiignore the contents, point was the formatting20:20
anteayathe formatting looks nice20:20
fungidropping indents and adding blank lines separating sections, like a typical ini file20:20
fungii understand why gerrit does it, because they're emulating .git/config defaults, but our devs are far more used to editing normal inifiles than gitconfig files20:21
openstackgerritA change was merged to openstack-dev/hacking: Update pointer to HACKING.rst in the readme
openstackgerritA change was merged to openstack-dev/hacking: Add link to readme at bottom of
*** jgrimm has joined #openstack-infra20:26
dimskrtaylor, getting 404 for urls posted by "IBM PowerKVM Testing" (ex:
krtaylordims, thanks, I'll look into it20:28
krtaylorArxCruz, ^^^20:28
*** Ajaeger has left #openstack-infra20:30
*** arborism is now known as amcrn20:30
*** etoews has joined #openstack-infra20:33
*** dangers_away is now known as dangers20:34
openstackgerritBen Nemec proposed a change to openstack-infra/config: Remove tripleo cross-testing with oslotest
*** etoews_ has quit IRC20:34
*** dstanek_zzz is now known as dstanek20:39
clarkbfungi: with that acl format does the resulting git commit in refs/meta/config use the format gerrit wants?20:41
clarkbfungi: I ask because the way manage-projects knows when to update gerrit is with git diff20:41
fungioh... hrm20:41
fungiclarkb: actually... it DOES seem to leave it in the format you push20:44
*** dkranz has quit IRC20:44
funginow checking to see what happens if i modify it from the webui20:44
fungimodifying it from the webui puts the indents back20:46
clarkbfungi: thats probably not too terrible as webui edits would be overwritten by manage-projects anyways20:47
clarkbso whitespace is a small thing in that case20:47
fungiwonder if that's a concession we can live with (presumably we only push these from m-p normally, and if it reapplies the same acl reformatted then not too bad_20:47
fungiyeah, agreed20:47
asselinHi, I'm trying to create a nodepool image with eth0 and eth1. I update the prepare_node script to set it up. However, after the vm is booted from the image, the new file /etc/network/interfaces.d/eth1.cfg  is either missing or empty.  Am I missing something?20:47
*** lcostantino has quit IRC20:48
*** dprince has quit IRC20:48
fungiasselin: is it possible that your cloud is doing some sort of file injection or overriding that via cloud-init?20:48
fungiasselin: basically, you've confirmed that at boot the node gets something different than is on the filesystem of the template instance?20:49
asselinfungi,  not sure where that would be. My cloud is a locally installed openstack stable/icehouse via
*** dstanek is now known as dstanek_zzz20:49
fungiamusingly, i don't know enough about openstack itself to guess what could be going on at boot to affect that, other than the things i've seen our providers do to new instances20:50
asselinfungi, yes, I'm stepping through the nodepool code and paused the code after the image is created. I can see the eth1 file there in the image-create-vm, but not in the manually booted vm using that new image.20:51
asselinit's as if nova wipes out everything except eth0 on boot?20:53
fungiasselin: my guess would be that the cloud-init utility is installed and doing something at boot to try to configure your networking. there are a few places cloud-init can look for configuration including in /etc, on a configdrive, user data injection, et cetera20:53
fungiasselin: look for the cloud-init log in /var/log on the broken node20:53
fungiif cloud-init is messing with things at boot time, i think it logs what it did20:53
asselinfungi, thanks I will look there20:54
*** arnaud has joined #openstack-infra20:55
*** thomasem has quit IRC20:59
ArxCruzdims: krtaylor  I'm doing a recheck, I believe the problem is fixed now, please, let me know if you can't access the logs20:59
ArxCruzI'm heading to home now20:59
ArxCruzback soon20:59
*** thomasem has joined #openstack-infra21:00
*** ArxCruz has quit IRC21:02
morganfainbergclarkb, for the apache-services tempest we need to wait for the next time zuul receives a HUP? or is this something I configured wrong? (not worried if it's just wait for next zuul config reload)21:03
morganfainbergclarkb, i just want to make sure i didn't do anything wrong. re:
clarkbmorganfainberg: looking21:03
*** ramashri has joined #openstack-infra21:03
clarkbmorganfainberg: it says its there21:04
mordredmorning all21:05
mordreddid I miss anything fun?21:05
morganfainbergclarkb, hm, not seeing it on the status page.21:05
clarkbmorganfainberg: but I agree I don't see it running21:05
mordredclarkb, fungi: I made a presentation today that I _think_ should get its own branch - but I'd be open to feedback on that21:05
morganfainbergclarkb, ok, let me take a closer look and i'll see if anything stands out21:06
clarkbI am looking too, you have made me curious21:06
morganfainbergclarkb, :)21:06
morganfainbergclarkb i think i see it21:06
clarkbdo tell21:06
morganfainbergprobably should be ^(?!stable/(havana|icehouse)).*$21:07
anteayahi mordred21:07
morganfainbergclarkb, !? vs ?! in the regex21:07
clarkbmorganfainberg: oh yup21:07
anteayamordred: how did your presentation go?21:07
morganfainbergclarkb, doh!21:07
clarkbif you propose that I can merge21:07
*** homeless has quit IRC21:07
morganfainbergclarkb, will do :)21:07
*** lcheng_ has quit IRC21:07
clarkbmordred: have it posted somewhere so we can look?21:07
clarkbI suppose a description of the talk would work too if you have a summary on a conference page21:08
mordredclarkb: I don't - it was an internal HP conference21:08
mordredbut I really like what I did with the slides21:08
mordredclarkb: perhaps I should just propse a patch to the current overview talk and you guys can tell me if you think it's a new talk or a reworking of overview21:08
openstackgerritMorgan Fainberg proposed a change to openstack-infra/config: Correct filter for apache-services temptest
clarkbmordred: ++21:09
*** gondoi is now known as zz_gondoi21:09
fungimordred: sounds great21:09
*** mwagner_ has quit IRC21:10
*** thomasem has quit IRC21:11
*** aysyd has quit IRC21:12
*** andreykurilin_ has joined #openstack-infra21:12
morganfainbergclarkb, thanks!21:13
fungiheading out to dinner, but will be back soon21:13
anteayaenjoy dinner21:14
openstackgerritA change was merged to openstack-infra/config: Correct filter for apache-services temptest
*** oomichi has joined #openstack-infra21:16
*** signed8bit has quit IRC21:16
*** mbacchi has quit IRC21:18
openstackgerritK Jonathan Harker proposed a change to openstack-infra/config: Move openstack-specific config out of ::jenkins
openstackgerritBen Nemec proposed a change to openstack-infra/reviewstats: Add tripleo-specs to tripleo project
*** dstanek_zzz is now known as dstanek21:23
openstackgerritMonty Taylor proposed a change to openstack-infra/publications: Made a branch for OpenStack Process Overview
harlowjaqq, just checking, is it known that returns 'Internal Server Error' :-/21:24
harlowjafor example,
clarkbharlowja: yes, kill your cookies21:24
harlowjathx Clabbe21:24
clarkbI think when the DB changed the cookie stuff got lost or corrupted21:24
clarkbthankfully killing them client side makes things happy21:25
jesusaurusfwiw yesterday paste.o.o gave me an error when i tried to make a private paste, i unchecked private and then it worked fine21:25
anteayaprivate paste seems to work for me21:26
harlowjaclarkb ya, worked, just knocked off the cookie21:26
*** e0ne has joined #openstack-infra21:27
clarkbfungi: Alex_Gaynor FYI I am manually triggering nodepool image builds for nodes that need new images to get the update run unittests scropt21:27
mordredclarkb, fungi: there you go ^^ sdague you too, since I stole some things from you21:28
mordred(well, the change several lines up)21:28
clarkbstarting with py3k precise nodes as they hit the problem21:28
sdaguemordred: man, and didn't even give me credit in the commit? :)21:28
mordredsdague: I put you on the title page ...21:28
sdagueoh, so you did. +121:29
clarkbgertty checkout is git review -d on steroids21:29
morganfainbergclarkb, i see an apache-services tempest on a keystoneclient change! woot! :)21:31
morganfainbergclarkb, you use linux as a desktop? os x?21:31
morganfainbergclarkb, i'm having issues with gertty doing odd things on os x21:31
clarkbmorganfainberg: linux21:32
morganfainbergclarkb, ah ok, if i figure out the gertty bugs for os x i'll propose fixes / log bugs on it21:32
*** derekh_ has quit IRC21:32
*** doug-fish has left #openstack-infra21:32
clarkbmordred: your jobs launched per day graph looks old. It doesn't say 15k21:32
clarkbmorganfainberg: good excuse to play with storyboard too21:33
clarkbkrotscheck: I filed a gertty bug today and storyboard is shiny21:33
morganfainbergclarkb, that was my thought :)21:33
morganfainbergi looked at storyboard... i ... can't wait for that to be ready.21:33
clarkbmordred: so hpcloud doesn't have a precise image anymore21:35
clarkbmordred: are we supposed to use "partner images" now? what does that mean?21:36
mordredclarkb: I _think_ partner images are from canonical - but I can ask?21:36
mordredclarkb: we should finish my nodepool changes so we can stop caring21:37
*** moted has quit IRC21:38
clarkbmordred: jesusaurus figured it out they were renamed to add (deprecated) to the nae21:39
clarkbmordred: yes we should but our images are already more than a week old and I double nodepool dib will happen before summit or during summit21:39
clarkband I really don't like have 3 week old images21:39
mordredclarkb: yah21:39
*** smarcet has quit IRC21:39
clarkbmordred: fungi: so we can use the deprecated images or the partner images21:39
clarkbif we decide to use partner images we can also use trusty21:39
clarkbbut as a consumer of partner images we have no way of knowing who isrooting us21:40
*** rlandy has quit IRC21:40
*** michchap has quit IRC21:40
clarkbat least before that was relatively clear21:40
clarkbalso we need to build new images today to fix a bug21:40
mordredjesusaurus: any chance you're on internal chat at HP and can ask someone on the nova team who the "partner" is?21:41
jesusaurusmordred: yeah, i can ask21:41
mordredjesusaurus: thanks. I'm not properly set up for that ATM21:41
clarkbfwiw I don't think using the deprecated image is too terrible since our image process updates the base image21:42
jesusaurusclarkb: right, the image is identical to the one we were using yesterday, they just appended "(deprecated)" to the name21:42
clarkbya ok21:42
*** moted has joined #openstack-infra21:43
clarkbso worst case I update the name to say (deprecated)21:43
asselinfungi, FYI, there wasn't anything in the log files. I managed to boot both eth0 and eth1 using nova boot --nic net-id=<> -nic net-id=<> command option. However doing this would require changed to nodepool.21:43
*** xdmeng has quit IRC21:44
*** mengxd has joined #openstack-infra21:44
asselinfungi, so I'll look at setting up eth1 from within the jenkins job which should be simpler. just need to get sudo access....21:44
*** lcostantino has joined #openstack-infra21:45
openstackgerritMonty Taylor proposed a change to openstack-infra/publications: Made a branch for OpenStack Process Overview
mordredclarkb: that should get us to nodepool at least21:46
sdagueclarkb / mordred: how are you guys feeling about adding in the custom dashboard I wrote up. The wayward reviews section has come up in a couple of contexts this week21:47
mordredsdague: I think it's a great idea21:48
*** masayukig has joined #openstack-infra21:48
clarkbI am not opposed but currently juggling the image builds21:49
anteayaI think added your dashboard would be good, sdague21:50
sdagueok, someone in project bootstrappers needs to do it21:50
mordredsdague: do you have the finalized-ish dashboard finished?21:50
jesusaurusmordred: the ubuntu partner images are coming directly from canonical21:51
*** andreaf_ has joined #openstack-infra21:51
mordredjesusaurus: awesome. that's great to know21:51
mordredclarkb: ^^21:51
jesusaurusmordred: in order to become a partner there are documents that must be signed21:51
sdaguemordred: - yes, the "Version 2 now actually working"21:51
*** UtahDave has joined #openstack-infra21:54
*** andreaf has quit IRC21:54
*** mriedem has quit IRC21:56
mordredclarkb: what if I just give sdague PB on review.o.o and let him push that dashboard himself? I'm not sure I'm going to run the commands _better_ myself21:56
lifelesspeanut butter?21:57
mordredlifeless: you are so good at the random drop-in21:57
lifelessmordred: I know, right?21:58
*** terryw has joined #openstack-infra21:58
*** otherwiseguy has quit IRC22:00
*** dkranz has joined #openstack-infra22:02
*** e0ne has quit IRC22:04
*** e0ne has joined #openstack-infra22:05
*** zul has quit IRC22:05
mordredclarkb: also - re: publications - I think in the animation when the keystone change fails we should freeze the animation for the other changes22:06
mordredso that we can talk about the reason we're moving it is because the changes in front of it are still running22:06
*** markmcclain1 has quit IRC22:06
mordredand then, once we've reparented, we can continue the lines on the other changes too22:06
mordredI'm saying that here because I have not figured out that code yet - and in case anyone ELSE here knows how it works and agrees with me ...22:07
* mordred looks around22:07
*** dkranz has quit IRC22:09
*** dkranz has joined #openstack-infra22:10
*** pdmars has quit IRC22:11
clarkbmordred: that code is ugh :P22:11
clarkbbut yes I think I agree22:11
clarkbas far as sdague perms goes thats keys to the kingdom... not that I don't trust sdague22:11
*** dkranz has quit IRC22:13
*** marcoemorais has quit IRC22:13
openstackgerritClark Boylan proposed a change to openstack-infra/config: HPCloud Precise image has a new name.
clarkbmordred: fungi ^22:14
sdagueI'm happy *without* keys to the kingdom, just need someone with the perms to do the thing :)22:14
clarkbsdague: have a link to the thing you want added?22:14
clarkbsdague: pastebin or whatever22:15
sdague - L16 - L2822:15
clarkbsdague: what is a good filename for that? review-inbox?22:16
*** marcoemorais has joined #openstack-infra22:16
sdagueworks for me22:17
* clarkb relearns how to do this :)22:17
clarkbsdague: at,dashboards22:18
clarkbdo we want review-inbox to be in the same table as important-changes?22:19
clarkbit will look like important-changes\nimportant-changes-dashboard\nreview-inbox-dashboard22:19
pleia2so, I am going to miss summit :( been sick and dr says "no no no" to travel next week22:20
sdagueone sec, qa meeting22:20
*** dizquierdo has quit IRC22:20
pleia2"but it's important!" "no"22:20
*** kgriffs is now known as kgriffs|afk22:20
gothicmindfoodpleia2: bummer. (x 1,000)22:20
pleia2considered lying, but then she sneakily scheduled another appointment for next week ;)22:21
clarkbsilly doctors22:22
pleia2silly humanity22:22
*** rfolco has quit IRC22:23
*** dims has quit IRC22:23
*** jgrimm has quit IRC22:24
clarkbanteaya: zaro do you have the gerrit upgrade etherpad handy? my notes on gerrit dashboards are in that etherpad22:25
* clarkb is much more careful on production machine22:27
*** prad__ has left #openstack-infra22:27
*** marcoemorais has quit IRC22:29
*** arnaud has quit IRC22:30
*** marcoemorais has joined #openstack-infra22:30
*** bknudson has quit IRC22:31
sdaguepleia2: bummer :(22:31
anteayaclarkb: looking22:31
pleia2sdague: yeah :(22:32
clarkbanteaya: thanks. though I may have figured it out (I am looking at bash history on review-dev)22:32
*** maru_afk is now known as marun22:32
Alex_Gaynorclarkb: cool (re: nodepool)22:33
*** jhesketh has quit IRC22:33
anteayapleia2: no no no no22:33
anteayapleia2: I need you there22:33
pleia2anteaya: I know, I am very sad!22:34
anteayawho am I going to be a bad influence on22:34
anteayamordred: I have to travel to SF to see pleia2 before november22:34
anteayawe have to like, work on stufff, yeah that's it22:34
anteayaoh this is horrible news22:35
anteayaboth for the fact taht you won't be there22:35
*** mwagner_ has joined #openstack-infra22:35
anteayaand that something is going to keep you from coming22:35
*** e0ne has quit IRC22:35
pleia2yeah :\22:35
anteayaI'm so sad22:35
anteayaI'm not dealing with this well at all22:35
*** markmcclain has joined #openstack-infra22:36
pleia2aw :(22:36
* pleia2 hugs anteaya 22:36
anteayaI needed that22:37
clarkbsdague: I don't think this dashboard will work because the foreach=${project} won't do what you expect22:37
anteayawell at least you will have cats22:37
anteayayou need cats22:37
clarkbsdague: it will restrict all queries to All-Projects22:37
morganfainbergpleia2, boo for you not making it to the summit!22:39
*** terryw has quit IRC22:40
*** arnaud has joined #openstack-infra22:42
sdagueclarkb: so the value of it is actually when yuo call it on nova, for instance22:45
clarkbsdague: actually I see you would have to use the dashboard from the subprojects instead22:45
clarkbsdague: isn't it better to just consolidate for the user?22:45
clarkbmaybe I am overthinking this. I can push it as is and we can see how it works22:45
clarkbright but as a reviewer I review things in like 20 projects22:46
clarkbI don't want to open a different dashboard for each one22:46
clarkbI am also probably an outlier22:46
sdagueclarkb: I actually have different dashboards for groups of projects22:46
sdaguebecause they require different mindset22:46
anteayamorganfainberg: yeah, she said she is going to rest now22:48
clarkbsdague: last question. Do you want a limit:50 on wayward changes?22:48
clarkbsdague: they other two have a limit22:48
sdagueI actually didn't want one, at least to start22:48
sdaguewe can tweak it over time22:48
clarkbpushing nowish22:48
EmilienMsdague: not sure it's the right place, but I have a 500 on paste.o.o22:51
clarkbsdague: done22:52
clarkbEmilienM: clear your paste cookie22:52
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard: Stories can now be filtered by assignee_id
mattoliverauMorning all22:52
clarkbmattoliverau: ohai22:53
EmilienMclarkb: indeed22:53
clarkbmattoliverau: you are supposed to work with sdague to get devstack multinode testing working22:53
sdaguemikal hath decreed it22:53
clarkbmikal was kind enough to volunteer you22:53
mattoliverauclarkb: I just woke up to an IRC notification that Mikal was talking about me22:53
mattoliverauhe's good at that :P22:54
*** dstanek is now known as dstanek_zzz22:54
mattoliverausdague: Well have to meetup in Atlanta and walk me through your thoughts on the matter :)22:54
bnemecHere's another fun Jenkins loop:
anteayamorning mattoliverau22:55
clarkbbnemec: unabandon it22:55
sdaguemattoliverau: yes, I was even #actioned about that in the nova meeting :)22:55
mattoliveraumorning anteaya22:55
clarkbbnemec: and mark it WIP instead until we get the fix in22:55
clarkbfungi: speaking of where are we on that?22:55
bnemecclarkb: I can't mark it WIP.  That's one of the ones I have no voting at all on.22:55
clarkbbnemec: are you core?22:56
bnemec(also, it needs to be abandoned anyway)22:56
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Switched new story modal to use typeahead project completion
bnemecclarkb: Yes22:56
clarkbbnemec: core should be able to unabandon and WIP22:56
fungiclarkb: where are we on...22:56
bnemecI was just trying to be nice by asking the submitter to abandon rather than doing it forcefully.22:56
clarkbbnemec: right but the abandonment is causing the loop22:56
* fungi checks scrollback22:56
mattoliverauSo now I just need to wait for mikal to tell me himself :P He is already in the states, so our timezones dont actually mesh at the moment, but at worst case I'll run into him on Sat in Altanta.22:56
bnemecclarkb: I abandoned it to stop the loop.  It started after I commented.22:56
anteayabnemec: unabandon let tests stop, then abandon22:56
*** dkranz has joined #openstack-infra22:57
clarkbfungi I thought stopping the loops meant allowing zuul to give a vote22:57
clarkbbut it can't vote on abandoned changes so it loops22:57
bnemecI have no idea what happened here.  This is the weird draft change.22:58
*** andreaf_ has quit IRC22:58
* bnemec suddenly pictures a lightbulb :-)22:58
sdaguedraft change, burn it with FIRE!22:59
clarkbbnemec: you commented on patchset 2 so zuul testing patchset 222:59
clarkbbut it cant vote on patchset 2 because it isn't the latest commit22:59
fungiasselin: nodepool already supports setting the network ids since that's necessary for clouds using neutron networks (such as the tripleo providers we're interacting with). see the networks list at
bnemecWhy the heck did I comment on PS 2?22:59
clarkbso publish the draft or go drinking. I think those are the two options here (I think fungi knows of a third workaround though and we should hear what he has to say)22:59
clarkbbnemec: because you can't comment on PS323:00
clarkbbecause its a draft23:00
clarkband drafts should die in a fire23:00
*** gokrokve has quit IRC23:00
bnemecI'm feeling less and less polite about this chain of commits all of a sudden.  My abandon finger is feeling a bit itchy.23:00
clarkbbnemec: that will only make the looping worse23:01
clarkbso venture forth at your own risk23:01
bnemecIck.  I suppose I could push a new patch set to every one of them and then -2.23:02
*** markmcclain has quit IRC23:02
bnemecBut...soooo tedious.23:02
bnemecOh, wait, can I do that with the inline commit message editing?23:02
clarkbbnemec: that may work23:02
clarkbI haven't tested it23:02
* bnemec braces for another onslaught of Gerrit spam23:02
clarkbit may also tell you to go away because of that draft change23:02
clarkbdraft patchsets are evil23:03
*** mengxd has quit IRC23:03
fungiclarkb: the merge-failed loop is a new one on me...23:03
clarkbfungi: what about getting the requires stuff in?23:04
bnemecHa!  -2'd so hard.23:04
clarkbit looked like yesterday you and jeblair were working on it?23:04
greghaynesbnemec: success?23:04
greghaynesWas about to try and ping the author if no23:04
bnemecgreghaynes: Yes:
* fungi is still catching up on scrollback23:04
* bnemec gleefully leaves to -2 all the other drafts23:05
*** andreykurilin_ has quit IRC23:05
fungiworkaround for drafts is that i twiddle bits in the database to publish it for you23:05
JayFfungi: I was just bumped from the list for bounces again \o/23:05
*** andreykurilin_ has joined #openstack-infra23:05
clarkbfungi: I was trying to avoid that. I think bnemec found a workaround that is not terrible23:05
fungiJayF: that's awesome23:06
* JayF blames DMARC23:06
fungiclarkb: um, yeah the requires patch landed days ago and zuul got restarted23:06
fungiso comments on abandoned changes shouldn't cause loops23:07
clarkbfungi: I see. So this is a new bug specifically with drafts23:07
clarkbis it possible to make mysql s/DRAFT/NEW/ in the database when rows are updated?23:07
fungiinstead the two new loops i've seen are comments on not-the-current patchset and comments on draft changes (which may actually be the same scenario)23:07
openstackgerritA change was merged to openstack-infra/devstack-gate: Add qpid config
openstackgerritA change was merged to openstack-infra/devstack-gate: add oslo.i18n to project list
clarkbfungi: well I thought requires fixed both23:08
clarkber fixed the first23:08
fungiclarkb: the change i want to propose, though i'm unsure how to write a zuul unit test for, is to have zuul require that the patchset being potentially triggered by the comment-added event is the current patchset23:08
*** arnaud has quit IRC23:09
clarkbfungi: or even if it just gracefully fails23:09
clarkbcouldnt vote? oh well23:09
fungiclarkb: "open==false" and "status==new" are change-wide, not patchset-specific (at least for the queries i've tried) so we probably need the extra conditional23:09
clarkbgerrit query has --current-patchset23:10
*** jhesketh has joined #openstack-infra23:10
clarkbso we can find it easily enough (assuming when a change is a draft change you get the correct number back)23:10
fungiright, i think the patchset number is also listed in the detail returned from querying the change, which zuul already does, so it has it in that data structure23:11
*** Alexandra_ has joined #openstack-infra23:11
fungithe current patchset number for the change, that is23:11
fungimorning jhesketh. saw your key signing question in scrollback, attempted to answer it in the ml thread (since i received several similar questions)23:12
jheskethfungi: ah cool, thanks23:13
fungiclarkb: so what's your take on breaking the current loop... should zuul always ignore comment-added events when the comment patchset is not the current patchset, or should that be configurable (per pipeline like require open and such)?23:14
fungii don't see an obvious use case for making it configurable23:14
clarkbfungi: I actually want to look at the zuul gerrit reporter. I think simplest thing is to just gracefully halt23:14
fungibut i could be missing something (usually am)23:14
fungiclarkb: well, we want zuul to be able to comment even if it can't vote. unless we're now revisiting that decision23:15
bnemecOkay, apparently my workaround only worked for that one patch. :-(23:15
clarkbfungi: it is able to comment if it can't vote23:15
*** lcostantino has quit IRC23:15
clarkbthats why it is problematic. It creates a ton of comments23:16
*** andreykurilin_ has quit IRC23:16
clarkbfungi: oh wait23:16
clarkbis it completely requeuing it?23:16
fungiclarkb: right, but maybe i'm misinterpreting your suggestion to "gracefully halt"23:16
fungiyes, it's reacting to its own comment23:16
clarkbdamnit for some reason I thought it was looping a little tighter than that23:17
*** etoews_ has joined #openstack-infra23:17
fungiwell, for merge failures it ends up being a very tight loop23:17
*** markmcclain has joined #openstack-infra23:17
clarkbbut I thoguth it ws looping in the reporter23:17
fungibecause it doesn't take it long to figure out and leave a comment23:17
*** andreykurilin_ has joined #openstack-infra23:18
clarkbmay just need to hardcode it23:18
clarkband fwiw older patchsets cannot merge so there is little value in testing them23:18
*** james_li has joined #openstack-infra23:19
clarkbwhich is why we kick them out of pipelines23:19
*** etoews has quit IRC23:19
clarkbso I would be in favor of a zuul event check that said if not latest patchset: continue23:19
fungiright, that's what i'm envisioning23:19
clarkbfungi: then the test for that would use the add event mechanism on an older patchset23:21
clarkbfungi: shouldn't be too terrible to construct. You add a change, add a second patchset to it. Let jobs flush. Then add a new event for patchset 1 and let things flush ensure old job count = current job count23:21
fungioh, though if the scheduler starts checking the current patchset value, i guess we'll need to start setting it on all fake changes23:21
clarkbya but that happens in one place iirc23:22
fungiisDraft: true23:23
*** marun is now known as maru_afk23:24
*** maru_afk has quit IRC23:25
*** marun has joined #openstack-infra23:25
clarkbfungi: where is that?23:25
clarkband is it visible to everyone?23:25
clarkbanyways I think I have a hack23:25
fungissh -p 29418 gerrit query --all-approvals --comments --commit-message --current-patch-set --dependencies --files --patch-sets --submit-records 7792423:26
fungithat's the api call zuul uses23:26
clarkbfungi: are you running it as zuul?23:26
fungias me23:26
fungiso i'm possibly seeing more23:26
clarkbI would double check with zuul creds23:26
*** marun has quit IRC23:26
*** marun has joined #openstack-infra23:27
*** andreykurilin_ has quit IRC23:27
*** jhesketh has quit IRC23:27
fungias for the automagic setting draft to new in the table, probably not. at least in 2.4.x there were several tables which needed touching to un-draft a change23:28
fungimaybe 2.8 is improved in that regard though23:28
*** mburned is now known as mburned_out23:28
openstackgerritKhai Do proposed a change to openstack-infra/config: puppetize installation of gerrit third party plugins
*** UtahDave has quit IRC23:29
lifelessI thought draft was going away?23:29
clarkblifeless: when gerrit 2.9 happens23:30
fungilifeless: as soon as we can make it happen23:30
fungiclarkb: so it looks like currentPatchSet.number will contain the latest patchset number23:32
fungisampling the event stream now to see where the comment-added patchset number ends up23:33
*** masayukig has quit IRC23:33
clarkbfungi: in zuul the Change class has self.is_current_patchset23:34
clarkbso we can just check that presumably23:34
fungioh! perfect23:34
clarkband it looks like the test suite already handles it properly so now I am trying to add a test23:34
fungiand yeah, comment-added events in the stream have patchSet.number if we end up needing it23:34
fungii do find it interesting that the gerrit webui won't show draft patchsets to my account but gerrit queries will show they were uploaded23:35
*** praneshp has quit IRC23:35
fungijust more fodder for "gerrit drafts are a failure"23:35
*** james_li has quit IRC23:36
openstackgerritKhai Do proposed a change to openstack-infra/config: puppetize installation of gerrit third party plugins
*** morganfainberg is now known as morganfainberg_Z23:38
*** jhesketh has joined #openstack-infra23:38
*** dstanek_zzz is now known as dstanek23:39
*** maxbit has joined #openstack-infra23:41
*** mrodden1 has quit IRC23:42
*** gokrokve has joined #openstack-infra23:43
*** praneshp has joined #openstack-infra23:44
clarkbfungi: any chance you can lookup what a comment-added json blob from gerrit looks like? I htink I have everything but the approvals section23:45
*** arnaud has joined #openstack-infra23:45
clarkbfungi: need an event for the test23:45
fungioh, i had one. just a sec23:45
*** thedodd has quit IRC23:47
fungi(trailing backslash at no extra charge)23:47
*** zns has quit IRC23:47
clarkbperfect thanks23:47
fungii really need a ~1cm free radius around my return key23:47
*** masayukig has joined #openstack-infra23:48
clarkbhuh that doesn't match their docs23:48
fungibecause that's te one key backspace won't undo in irc ;)23:48
clarkbwhich is lovely23:48
fungiyeah, that was captured straight from's stream23:48
*** morganfainberg_Z is now known as morganfainberg23:49
fungissh -p 29418 gerrit stream-events23:49
*** pcrews has quit IRC23:49
fungiso either their docs are lagging behind or their code is23:49
*** dstanek is now known as dstanek_zzz23:49
clarkbI think the docs23:50
openstackgerritMichael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Textareas now autoresize their height.
krotscheckclarkb: Finally got around to that thing you wanted.23:50
*** hogepodge has quit IRC23:51
StevenKfungi: If I miss Enter, I tend to hit Shift instead, which is handy.23:52
fungiStevenK: my problem is a partial miss where i hit \ or ] or ' at almost the same time as return but just barely before23:54
*** markmcclain has quit IRC23:54
clarkbkrotscheck: woot23:54
fungiat least my new keyboard doesn't have backspace and return adjacent. that was downright terrible23:54
clarkbfungi: so I can get it to pass my assertions but not a test suite level assertion23:54
clarkbfungi: I think I am going to push my code as is and people can hack it23:54
clarkbbut it fails23:54
clarkbactually hold on23:55
clarkbno that wasn't it23:56
*** atiwari has quit IRC23:56
*** flaper87 is now known as flaper87|afk23:58

Generated by 2.14.0 by Marius Gedminas - find it at!