Monday, 2013-11-11

*** whoops has joined #openstack-infra00:03
*** whoops has quit IRC00:09
openstackgerritSean Dague proposed a change to openstack-infra/devstack-gate: trigger devstack-gate test as early as possible
openstackgerritSean Dague proposed a change to openstack-infra/devstack-gate: Extract git_fetch_at_ref
openstackgerritSean Dague proposed a change to openstack-infra/devstack-gate: Extract setup_project as a single function
*** wenlock has quit IRC00:23
*** whoops has joined #openstack-infra00:49
sdaguelifeless: you'll be happy to know I just sorted out enabling this - in my python-mode - hopefully cuts down on comment spelling issues00:49
clarkbsdague: I take it your flight toe the east coast wasn't turned around and sent back to HKG?00:50
sdagueI had the cathay pacific flight, and we were all good00:51
sdaguehonestly, I'm already sleeping better at home than I did any night in HK, so will be curious to see how long the flip back takes00:54
sdagueclarkb: your flights work out fine?00:55
lifelesssdague: \o/00:56
lifelesssdague: thank you!00:56
*** Ryan_Lane has quit IRC00:57
*** Ryan_Lane has joined #openstack-infra00:57
sdagueheh, no worries. I did poke fun a little on twitter about it. But all in good fun.00:58
*** matsuhashi has joined #openstack-infra01:02
*** whoops has quit IRC01:06
*** nosnos has joined #openstack-infra01:07
clarkbsdague: yes my flights were fine got home last night then fell asleep for 13 hours01:17
*** nosnos has quit IRC01:17
morganfainberghallo infra.  how's everyone post summit?01:17
*** nosnos has joined #openstack-infra01:17
clarkbmorganfainberg: tired01:17
clarkbsee 13 hours of sleep. That really helped though01:18
morganfainbergclarkb, i'll second that one.  i am still a little jetlagged but not too bad01:18
sdagueyeh, I got about 8 hrs total last night, though in a couple chunks.01:18
morganfainbergsdague, ouch.01:19
sdagueI think the adrenaline and copious tea kept me running fine on so little sleep at summit01:19
sdaguemorganfainberg: that's 50% more than any night that I got in HK :)01:19
morganfainbergsdague, 50% more is a big impovement! :)01:19
morganfainbergi know i drank far too much coffee and tea there.01:20
morganfainbergit was a mistake to fly out on friday night.  should have gotten another night of sleep prior to the long haul flight.01:20
morganfainbergnext time i will know01:20
clarkbthe tea was really good01:20
morganfainbergclarkb, ++ drank so much of it.01:21
sdagueyeh, the tea was really good01:22
sdaguesomeone needs to figure out what that was so I can get some :)01:22
morganfainbergsdague, if you do find out, please share!01:22
morganfainbergoverall, this has been one of the best summits i've been to.  very pleased with the results (and directions)01:25
clarkb++ I felt like the sessions I went to were very productive01:25
morganfainbergat least for keystone the amount of things we got done ad-hoc was amazing as well.01:25
clarkband now I need to drink a real beer and watch football01:26
clarkbmordred: we missed the sounders timbers games :( they looked like good ones01:26
morganfainbergclarkb, i did the real beer thing on the way back from the airport!01:27
morganfainbergi couldn't even wait an hour from when i landed01:27
clarkbmorganfainberg: I think I somehow completely missed you at the summit, oh well01:29
morganfainbergclarkb, perhaps.  I don't remeber running across you there either.01:30
morganfainbergclarkb, i was camping in the dev loung almost 100% of time time i wasn't in sessions.01:30
clarkbmorganfainberg: I was in sessions almost 100% of the time :) I spent one time slot in the dev lounge01:31
clarkbthat is the life of an infra person01:31
morganfainbergclarkb, ah yeah.01:31
*** markwash has joined #openstack-infra01:31
morganfainbergclarkb, well atlanta isn't that far off, i'll make sure not to miss the people i missed this time there!01:32
clarkbmorganfainberg: same here01:32
clarkbfungi: ping me when you are ready to start tackling the havana/folsom stuff (eg tuesday) and we can do that together (to make sure we get it in correctly)01:32
morganfainbergactually, i think i also totally missed lifeless too01:33
lifelessmorganfainberg: hi01:44
lifelessmorganfainberg: 5000 people, hard to say hi to everyone :)01:44
morganfainberglifeless, hi there! next summit i need to not miss the infra people i have on my short-list to say hi to in person01:45
morganfainbergyeah.  i hear ya01:45
*** whoops has joined #openstack-infra01:49
*** senk has joined #openstack-infra01:59
*** sarob has joined #openstack-infra02:00
*** yaguang has joined #openstack-infra02:00
lifelessmorganfainberg: maybe we should have a 'Meet the Infra' session ?02:04
morganfainberglifeless, might be a good idea!  you know, without you guys OS would be a very different place.02:04
morganfainbergannndddd 8 % batter on the laptop... i need to get home catch everyone later02:05
*** ljjjustin has joined #openstack-infra02:07
lifelessmorganfainberg: I can't take much credit for Infra - thanks for counting me as one though!02:11
*** bingbu has joined #openstack-infra02:16
*** sarob has quit IRC02:19
*** sarob has joined #openstack-infra02:19
*** sarob_ has joined #openstack-infra02:20
*** sarob has quit IRC02:23
*** Walton_ has joined #openstack-infra02:29
*** jcooley_ has quit IRC02:30
*** jcooley_ has joined #openstack-infra02:32
*** whoops has quit IRC02:32
*** Walton_ is now known as whoops02:32
*** Alssi has joined #openstack-infra02:35
*** jcooley_ has quit IRC02:39
*** jcooley_ has joined #openstack-infra02:40
*** jcooley_ has quit IRC02:44
*** boris-42 has joined #openstack-infra02:45
*** dcramer_ has quit IRC02:45
*** sdake has joined #openstack-infra02:46
*** sdake has quit IRC02:46
*** sdake has joined #openstack-infra02:46
*** ljjjustin has quit IRC02:49
*** boris-42 has quit IRC02:52
*** DennyZhang has joined #openstack-infra02:54
*** sarob_ has quit IRC02:56
*** sarob has joined #openstack-infra02:57
*** dcramer_ has joined #openstack-infra02:58
*** ljjjustin has joined #openstack-infra03:01
*** sarob has quit IRC03:01
*** dstanek has joined #openstack-infra03:08
*** matsuhashi has quit IRC03:19
*** matsuhashi has joined #openstack-infra03:19
*** matsuhashi has quit IRC03:24
*** dims has quit IRC03:25
*** matsuhashi has joined #openstack-infra03:31
*** dstanek has quit IRC03:33
*** whoops has quit IRC03:34
*** dstanek has joined #openstack-infra03:34
*** dstanek has quit IRC03:35
*** senk has quit IRC03:35
*** DennyZhang has quit IRC03:35
*** dstanek has joined #openstack-infra03:35
*** senk has joined #openstack-infra03:36
*** senk has quit IRC03:37
*** matsuhashi has quit IRC03:57
*** matsuhashi has joined #openstack-infra03:57
*** matsuhashi has quit IRC04:02
*** sarob has joined #openstack-infra04:07
*** ian_ott has joined #openstack-infra04:12
*** sarob has quit IRC04:12
*** mriedem has quit IRC04:12
*** ArxCruz has joined #openstack-infra04:17
*** ArxCruz_ has quit IRC04:19
*** dcramer_ has quit IRC04:23
*** dstanek has quit IRC04:28
*** slong has quit IRC04:28
*** dcramer_ has joined #openstack-infra04:36
*** ljjjustin has quit IRC04:40
clarkbdstufft: any reason for using elasticsearch over pyelasticsearch in warehouse?04:43
clarkbwe are using pyelasticsearch and sdague has had to do some cleaning around it, curious if that is a common problem04:43
*** dstanek has joined #openstack-infra04:44
clarkbdoesn't look like elasticsearch would play nice with our proxy method though04:45
*** chandankumar has joined #openstack-infra04:45
clarkbwhich is unfortunate that everyone expects elasticsearch to just be open to whatever you have talking to it (fine for pypi)04:45
*** talluri has joined #openstack-infra04:47
*** senk has joined #openstack-infra04:50
*** dcramer_ has quit IRC04:51
*** Ryan_Lane has quit IRC04:51
*** nosnos_ has joined #openstack-infra04:57
*** markwash has quit IRC04:57
openstackgerritA change was merged to openstack-infra/config: Add missing running-your-own hidden toctree.
*** sarob has joined #openstack-infra05:00
*** nosnos has quit IRC05:00
*** matsuhashi has joined #openstack-infra05:01
*** dcramer_ has joined #openstack-infra05:06
openstackgerritA change was merged to openstack-infra/config: Precache Ironic and Ironic client
*** jcooley_ has joined #openstack-infra05:11
*** markwash has joined #openstack-infra05:11
*** marun has joined #openstack-infra05:17
mordredfungi, jeblair, clarkb: btw - since the patch queue is so ludicrously long, I'm +2/+A'ing trivial things- like the 7 line doc patch from fungi and the two above05:17
openstackgerritA change was merged to openstack-infra/config: Flush Gerrit caches after fixing E-mail dupes
clarkbmordred: go for it, I was going to do similar but got distracted by football and laundry05:19
mordredclarkb: also, small changes related to stackforge projects seem to fit the bill too :)05:19
clarkbmordred: the timbers got absolutely destroyed by RSL tonight too05:19
openstackgerritA change was merged to openstack-infra/config: Add coverage jobs to tempest and nova experimental queue
clarkbalso I think the media is seriously confused about what stackforge is05:21
openstackgerritA change was merged to openstack-infra/config: Allow fuel-core to create branches in fuel repos
*** jcooley_ has quit IRC05:25
*** jcooley_ has joined #openstack-infra05:26
openstackgerritA change was merged to openstack-infra/config: Add ironic translation jobs.
*** jcooley_ has quit IRC05:31
*** sarob has quit IRC05:31
*** sarob has joined #openstack-infra05:32
*** jcooley_ has joined #openstack-infra05:32
*** sarob has quit IRC05:36
dstufftclarkb: elasticsearch is the official client is why I usd it05:37
clarkbgotcha thanks05:37
openstackgerritA change was merged to openstack-infra/config: Make pypy and python33 gates non-voting for solum project
openstackgerritA change was merged to openstack-infra/jeepyb: New headers argument for github.Issue.Issue
dstufftclarkb: it has a requests and urllib3 transport though, you can probably use a proxy with it05:38
dstufftI use TLS and basic auth for ours though05:38
*** nicedice_ has quit IRC05:38
clarkbI think pur proxy may be too aggresice killing connections05:40
anteayaI see I am not the only one awake05:41
*** jcooley_ has quit IRC05:42
*** jcooley_ has joined #openstack-infra05:43
*** dstanek has quit IRC05:44
*** slong has joined #openstack-infra05:46
anteayaclarkb: why do you say that?05:47
clarkbanteaya: which thing I said?05:47
openstackgerritA change was merged to openstack-infra/config: Upgrade puppetlabs-haproxy to fix concat module.
openstackgerritA change was merged to openstack-infra/config: Upgrade puppetlabs-postgresql to fix concat module
clarkbmordred ^ may break git.o.o and postgres unittests :) I dont expect it to though05:48
*** nosnos has joined #openstack-infra05:55
*** nosnos_ has quit IRC05:55
*** matsuhashi has quit IRC05:56
anteayathat our proxy may be too agressive killing connections05:56
* anteaya reading backscroll05:56
anteaya<sdague> the review anti-pattern thread makes me wonder if we should really have a gerrit docs tree for some of these project wide policy / culture things05:56
anteayasdague: +105:57
anteayaI'm glad to read that a few of you feel there will be progress on neutron tests in the gate by icehouse-105:57
clarkbanteaya: there are timeouts and they bother things that try to pool persistent connections05:58
anteayaI am not that optimistic but I will be glad to be wrong05:58
anteayaclarkb: ah okay05:58
anteayaso increase the timeouts?05:58
*** matsuhashi has joined #openstack-infra05:59
*** senk has quit IRC06:00
anteayaand wow, what a horrible flight experience fungi :(06:00
*** ljjjustin has joined #openstack-infra06:10
anteayathis is a patch to improve the neutron testing environment, it replaces mox with mock:
anteayaI know it is a lot to ask because eveyone just got back and is tired, but if anyone familiar with the mox to mock transition has some pity on me, your comments would be most appreciated06:11
anteayaAlex_Gaynor: ^ if you have time06:16
*** afazekas has quit IRC06:17
*** guohliu has joined #openstack-infra06:28
*** sarob has joined #openstack-infra06:42
*** sarob has quit IRC06:47
anteayawhere are reviewstats published currently?06:50
clarkbanteaya: those?06:51
anteayayup, thanks06:53
anteayathe repo is under -infra in git so I was thinking it was on one of our servers now06:53
clarkbis it in infra?06:54
*** nosnos has quit IRC06:54
*** xeyed4good1 has joined #openstack-infra06:54
clarkboh hey it is06:55
clarkbI can't keep up06:55
anteayaneither can I06:55
*** matsuhashi has quit IRC06:56
*** nosnos_ has joined #openstack-infra06:56
*** matsuhashi has joined #openstack-infra06:57
*** matsuhashi has quit IRC06:57
*** matsuhashi has joined #openstack-infra06:57
*** pblaho has joined #openstack-infra07:05
*** matsuhashi has quit IRC07:08
*** yolanda has joined #openstack-infra07:09
*** matsuhashi has joined #openstack-infra07:09
*** matsuhas_ has joined #openstack-infra07:12
*** matsuhashi has quit IRC07:12
*** davidhadas has quit IRC07:14
anteayadis-heartening to see patches merged to neutron with gate-tempest-devstack-vm-neutron-large-ops failing07:23
clarkbanteaya: I am not sure it ever passed07:23
clarkbjog0: would know07:23
anteayaokay thanks, I will look forward to jog0's rendition of the history of this test07:26
anteayajust collecting stats and impressions07:26
anteayasome people are doing a lot of work07:26
anteayaso that is good to see07:27
anteayaas in some people in addition to markmclain and salvatore07:28
*** matsuhas_ has quit IRC07:32
*** sileht has joined #openstack-infra07:32
*** matsuhashi has joined #openstack-infra07:32
*** tkammer has joined #openstack-infra07:54
*** davidhadas has joined #openstack-infra08:00
*** markwash has quit IRC08:08
*** davidhadas_ has joined #openstack-infra08:13
*** davidhadas has quit IRC08:16
*** sdake has quit IRC08:18
*** shardy has joined #openstack-infra08:20
*** markwash has joined #openstack-infra08:21
*** matsuhashi has quit IRC08:29
*** davidhadas_ has quit IRC08:29
*** matsuhashi has joined #openstack-infra08:30
*** marun has quit IRC08:31
*** marun has joined #openstack-infra08:32
*** nosnos_ has quit IRC08:32
*** nosnos has joined #openstack-infra08:32
*** matsuhashi has quit IRC08:34
*** matsuhashi has joined #openstack-infra08:37
*** jcooley_ has quit IRC08:38
*** jcooley_ has joined #openstack-infra08:38
*** jpich has joined #openstack-infra08:41
openstackgerritVladimir Kozhukalov proposed a change to openstack-infra/config: Add Fuel-Provision project to Stackforge
*** jcooley_ has quit IRC08:43
*** markwash has quit IRC08:48
*** nosnos has quit IRC08:52
*** nosnos has joined #openstack-infra08:52
*** flaper87 has quit IRC08:57
*** flaper87 has joined #openstack-infra08:57
openstackgerritA change was merged to openstack-infra/reviewstats: Fix misspelled contributor user name in Ironic
*** tkammer has quit IRC09:00
*** markwash has joined #openstack-infra09:01
*** mattymo has joined #openstack-infra09:02
*** derekh has joined #openstack-infra09:14
*** afazekas has joined #openstack-infra09:19
*** davidhadas has joined #openstack-infra09:26
*** jcooley_ has joined #openstack-infra09:28
mordredanteaya: <-- there's a follow up to the above neutron patch that removes stubout09:32
openstackgerrityolanda.robla proposed a change to openstack-infra/config: Remove extra chars in launchpad bug links
*** bingbu has quit IRC09:48
*** boris-42 has joined #openstack-infra10:00
*** jcooley_ has quit IRC10:02
*** ArxCruz has quit IRC10:03
*** davidhadas_ has joined #openstack-infra10:05
*** davidhadas has quit IRC10:07
*** guohliu has quit IRC10:09
*** dstanek has joined #openstack-infra10:13
*** dstanek has quit IRC10:18
*** yaguang has quit IRC10:21
openstackgerritNikita Konovalov proposed a change to openstack-infra/storyboard: Pagination added to task lists
*** davidhadas has joined #openstack-infra10:34
*** davidhadas_ has quit IRC10:37
*** markmc has joined #openstack-infra10:37
*** lcestari has joined #openstack-infra10:38
*** ljjjustin has quit IRC10:40
*** nosnos has quit IRC10:41
*** nosnos has joined #openstack-infra10:41
*** rfolco has joined #openstack-infra10:43
*** ArxCruz has joined #openstack-infra10:43
*** nosnos_ has joined #openstack-infra10:55
*** jcooley_ has joined #openstack-infra10:58
*** boris-42 has quit IRC10:58
*** nosnos has quit IRC10:59
*** michchap has joined #openstack-infra10:59
*** nosnos_ has quit IRC11:06
*** nosnos has joined #openstack-infra11:06
*** matsuhashi has quit IRC11:14
*** matsuhashi has joined #openstack-infra11:14
*** BobBallAway is now known as BobBall11:14
*** matsuhashi has quit IRC11:15
*** matsuhas_ has joined #openstack-infra11:15
*** adalbas has joined #openstack-infra11:21
*** johnthetubaguy has joined #openstack-infra11:24
*** matsuhas_ has quit IRC11:25
*** matsuhashi has joined #openstack-infra11:26
*** pcm_ has joined #openstack-infra11:27
*** matsuhashi has quit IRC11:30
*** rfolco has quit IRC11:31
*** flaper87 is now known as flaper87|afk11:31
*** matsuhashi has joined #openstack-infra11:31
*** jcooley_ has quit IRC11:32
*** katyafervent has joined #openstack-infra11:33
*** katyafervent has quit IRC11:33
ttxfungi: fwiw I couldn't find your email with the Horizon PTL voters list11:45
ttxfungi: so please re-send if possible11:45
openstackgerritArx Cruz proposed a change to openstack-infra/config: Adding nodepool compatibility with Fedora.
sdagueclarkb: I think it passed for a minute, but it hasn't for a long time11:59
*** yamahata has joined #openstack-infra12:07
*** xchu has joined #openstack-infra12:08
mordredmorning sdague12:10
mordredmorning ttx12:10
*** michchap has quit IRC12:10
mordredttx: I had a storyboard thought, btw12:11
*** michchap has joined #openstack-infra12:11
mordredttx: instead of a pure django app12:11
mordredttx: what if we made a storyboard restful service using pecan/wsme just like all of the other openstack services12:11
mordredttx: and then used the horizon part of horizon (which is a library to help create django apps on top of rest apis)12:11
*** michchap has quit IRC12:12
mordredttx: to build the web app to talk to the restful service12:12
*** michchap has joined #openstack-infra12:12
*** rfolco has joined #openstack-infra12:13
mordredwith folks like tripleo and solum wanting to include the CI story, it seems like having a rest service in the cloud that knows how to provide ticket services would be useful12:13
mordredwhich could have a default local impl12:13
mordredand potential plugins to connect it to things like existing jira whatnot12:13
mordredttx: it's also possible that I'm making things too hard - but it feels like we have a lot of pieces of the story for this inside of openstack already12:14
*** dstanek has joined #openstack-infra12:15
mordredttx: also, if olaph wanted to start hacking on storyboard in some spare time, what would be a good next step?12:15
sdaguemorning mordred12:16
mordredsdague: how's the jetlag?12:16
sdagueactually not as bad as I expected12:16
sdagueI'm waking up at the right times here naturally, now I just have to get the going to bed at the right times worked out12:16
sdaguemordred: I really like the idea of storyboard as rest service first, UI second12:17
mordredsdague: awesome. I'm glad that doesn't sound lke crazypants12:17
mordredI betcha it would not take a TON of effort to take the current code and port it to that format12:18
sdagueyeh, agreed. You'd probably just want to start from scratch with the basic models that seem good and build a rest stack on them12:19
openstackgerritArx Cruz proposed a change to openstack-infra/devstack-gate: Adding support to Fedora and Red Hat
sdagueDoes pecan do nested resources? I feel like that was an open question in the room that I can't remember from that oslo session12:19
*** dstanek has quit IRC12:20
mordredI do not know12:21
mordredArxCruz: hey! didn't you already upload that patch?12:22
ArxCruzmordred: that's a different one12:23
mordredArxCruz: ok. cool12:23
ArxCruzmordred: tempest was failing on my jenkins because I'm using fedora.12:24
ArxCruzthe postgres and mysql path in fedora are different from the ubuntu path12:24
mordredArxCruz: I am very much in favor of devstack growing fedora support12:24
ArxCruzmordred: :)12:24
ArxCruzmordred: it took me several hours to figure out that the problem was in this check hope you approve this path :)12:25
mordredArxCruz: I like the path - I just -1'd you on a couple of small things12:26
ArxCruzmordred: sure, just tell me and I make the changes :)12:27
mordredArxCruz: I put them on the review - I'm trying to be better about doing code review there instead of just telling peope in IRC :)12:27
mordredotherwise, my review stats look like I just like everything...12:28
ArxCruzmordred: hehe, sure, I will wait your review then :)12:28
*** jcooley_ has joined #openstack-infra12:28
mordredArxCruz: it should be there now12:28
openstackgerritArx Cruz proposed a change to openstack-infra/devstack-gate: Adding an option to use qpid instead of rabbit or zeromq
ArxCruzmordred: okay, here's another one :)12:29
sdaguemordred: so it occurs to me that as ArxCruz heads down this journey of making devstack-gate cross platform... we might try to figure out if there was a sane way to share the functions file from devstack, which has a ton of cross distro bits in it12:30
mordredsdague: yes12:30
mordredsdague: well, in general, I'd love for devstack-gate to do less12:30
mordredsdague: also, I think d-g can always assume the existence of at least one devstack git repo12:31
mordredso sourcing some files might not be out of bounds12:31
mordredsdague: for instance:
mordredsdague: I think honestly that the above ^^ could be potentially moved into devstack to be a thing that runs12:32
mordredif you run and after it's done the configured database does not have log files, then it's a good indication that something went wrong12:32
mordredin fact12:32
*** DennyZhang has joined #openstack-infra12:33
sdagueyeh, let me ping dtroyer about it once he gets home, I think we'd need to clean things a little in functions to make it fully safe to source outside of devstack / grenade. A few too many variables look to be leaked out of it unintentionally12:33
ArxCruzmordred: i have a question, can you please enlighten me ?12:34
mordredArxCruz: shoot!12:34
ArxCruzmordred:  post in
*** obiat has joined #openstack-infra12:35
ArxCruzmordred: for documentation purpose :)12:35
sdaguemordred: yeh, though that check was there as a specific bit of double book keep accounting, because we actually accidentally made the pg job run mysql for a bit due to a mismatch between d-g and devstack12:35
sdagueand it took us a week to realize it12:35
sdagueat which point pg was broken12:36
mordredwell then. it does make sense to have that in dg then12:36
sdagueyeh, I'll come back around once we sort out ArxCruz's patch and explain the reason for that piece of code. I realize there is super important context about why it shouldn't be pulled out, which is only in a few people's head (not a good place for it)12:37
sdaguedon't want to cause a merge conflict for no good reason first12:37
openstackgerritA change was merged to openstack-infra/config: Relax silence threshold for ConfBridge
ttxmordred: yes, we need to make some hard calls on the storyboard architecture early on12:40
openstackgerritA change was merged to openstack-infra/config: Update default location for astdatadir for asterisk
ttxmordred: a REST + Horizon thing would definitely fit withing the openstack story... although I'd have to look into it more closely to see if the interface would play ball12:41
mordredit _should_12:41
ttxmordred: I'll admit not to have look into horizon code AT ALL12:41
mordredI learned a good deal more about it this cycle12:41
mordredand at the summit12:42
mordredtl;dr - there are two pieces12:42
mordredone is called horizon, although that part is essentially a library or "django app" that helps write django apps that talk to restful services12:42
ttxmy only concern on the technology base is to be appealing to the openstack devs... python/django everyone knows, and Pecan/WSME/Horizon a lot of openstack dudes know12:42
mordredthere is another part called openstack_dashboard which is a django application that uses horizon to build the thing we think of as horizon12:43
ttxso I'm pretty fine with either. REST + strong JS client... slightly less attractive12:43
*** talluri has quit IRC12:43
mordredttx: no, I think we need to have the web framework be strongly django12:44
mordredbut strong js in the UI of that seems like a good idea12:44
*** talluri has joined #openstack-infra12:44
mordredand is something that horizon itself can benefit from perhaps12:44
ttxto answer your last question, not sure where he should focus. He can pick an idea on the etherpad list of TODOs12:45
mordredttx: where's the etherpad?12:45
ttxbut that might be a bit useless if we scrap the whole paradigm12:45
ttxso ideally we would nail that first12:45
mordredI doubt we'll scrap the whole paradigm12:45
ttxi.e. better focus on feature than on UI12:46
mordredI think we can take what you have and basically split the underlying model out12:46
mordredwe could get a v1 up and going so that we can use it12:46
openstackgerritArx Cruz proposed a change to openstack-infra/devstack-gate: Adding support to Fedora and Red Hat
mordredand make a pecan/wsme+horizon a v2 thing12:46
ArxCruzmordred: I think it's better now :)12:46
mordredand if the _data_ model is the same, and the UI is using bootstrap12:46
mordredwe shoudl be able to do that without killing ourselves12:47
mordredolaph: ^^
sdaguethe other great idea about going REST first, is it would naturally mean multiple UIs, as people could write their own bits that make sense for their context12:47
ttxmordred: also in case you missed it: Nikita Konovalov proposed a change to openstack-infra/storyboard: Pagination added to task lists
mordredsdague: yes!12:47
mordredlike "storyboard list" and "storyboard create" command line ui's :)12:48
ttxso it looks like our Russian friends are interested to play with it too12:48
*** talluri has quit IRC12:48
mordredyes. Sergey told me that they were going to jump in12:49
mordredI think we should schedule a weekly storyboard meeting12:49
ttxyes yes, just too many things to do postsummit !=12:49
mordredwell, none of my new hires will start before Dec 212:49
ttxNow I just need to work on my "happily ignore stuff" program to get more time to do useful work12:54
openstackgerritA change was merged to openstack-infra/nodepool: add introduction for README document
openstackgerritA change was merged to openstack-infra/nodepool: Add more details to developer setup in README
openstackgerritA change was merged to openstack-infra/config: Add gerrit notifications to #solum
openstackgerritA change was merged to openstack-infra/config: Remove use of undefined console-log-post publisher
*** flaper87|afk is now known as flaper8712:57
*** matsuhashi has quit IRC13:01
*** jcooley_ has quit IRC13:01
*** matsuhashi has joined #openstack-infra13:02
*** matsuhashi has quit IRC13:04
*** matsuhashi has joined #openstack-infra13:04
*** CaptTofu has quit IRC13:05
*** CaptTofu has joined #openstack-infra13:05
*** dkranz has joined #openstack-infra13:06
*** openstackgerrit has quit IRC13:09
*** CaptTofu has quit IRC13:09
*** openstackgerrit has joined #openstack-infra13:10
*** davidhadas_ has joined #openstack-infra13:13
mordredneeds a patch to be able to enable rackspace performance flavors13:13
*** davidhadas has quit IRC13:15
*** dstanek has joined #openstack-infra13:15
openstackgerritMonty Taylor proposed a change to openstack-infra/nodepool: Add ability to filter on flavor name
*** nosnos_ has joined #openstack-infra13:19
*** nosnos_ has quit IRC13:20
*** dstanek has quit IRC13:20
*** nosnos_ has joined #openstack-infra13:20
*** nosnos has quit IRC13:21
*** talluri has joined #openstack-infra13:21
*** nosnos_ has quit IRC13:24
openstackgerritA change was merged to openstack-infra/config: Add anvil py26/py27 testing
*** krtaylor has quit IRC13:26
openstackgerritA change was merged to openstack-infra/config: Add congress project to stackforge
*** dcramer_ has quit IRC13:29
*** dstanek has joined #openstack-infra13:30
*** dstanek has quit IRC13:31
*** dstanek has joined #openstack-infra13:32
*** dstanek has quit IRC13:32
openstackgerritA change was merged to openstack-infra/config: Add puppet unit tests against Puppet 3.2/3.3
mordredsdague: can you look at ?13:38
mordredsdague: the change itself looks fine, but I don't want to land it if it's not ok to land13:39
mordredoh - wait. it's in the experimental queue13:40
*** jcoufal has joined #openstack-infra13:41
*** obiat has quit IRC13:42
openstackgerritA change was merged to openstack-infra/config: Switch neutron full job to run in parallel
ArxCruzmordred: it fails, is this an issue with openstack infra or was my patch ?13:44
mordredArxCruz: looks like a tempest race:13:45
mordredI noticed tempest failed, I think you hit bug(s): and
mordredArxCruz: ^^13:45
ArxCruzmordred: what should I do in this case?13:45
mordredyou may want to reverify with one of those13:45
mordredso, leave a comment "reverify bug 1218391"13:45
*** hashar has joined #openstack-infra13:46
openstackgerritA change was merged to openstack-infra/os-loganalyze: fix typo
mordredsdague: you in stable-maint? there's a bunch of repos, like requirements, where I'm starting to feel like infra should be in the stable reviewer list13:48
ArxCruzmordred: done13:48
*** thomasem has joined #openstack-infra13:49
ArxCruzmordred: may I have your +2 on ?13:50
sdaguemordred: no, I'm not in stable-maint13:52
mordredsdague: I mainly just noticed because there is a patch to set the .gitreview default branch and I cannot land it13:52
sdagueright, we should figure out how the stable-maint team wants to handle that13:52
sdaguettx: actually... pending requirements changes should be one of those things which is a standing item on the project meeting, so they don't get lost13:53
ttxsdague: or at least a standing item in some meeting... choosing which depends a bit on which program we would attach "requirements" to13:55
ttxif it's attached to relmgt then yes, the project/release meeting would make the most sense13:55
*** dstanek has joined #openstack-infra13:56
ttxI could see it land in oslo or infra though.13:56
sdagueI think it's really release management, because the reason's we'd -1 a req add13:57
*** zoresvit has quit IRC13:57
ttxyes, although several projects care about dep convergence. I wouldn't mind if it was attached to relmgt, if the other potential homes would agree13:58
mordredI agree13:58
mordredI think it should belong to relmgt13:58
ttxit's not as if it wasn't a three-legged duck already13:58
mordredmmm. duck is tasty13:59
mordredthis is sad13:59
sdagueI think we need to just agree that relmgt program is different, and build some different rules for it13:59
ttxthree legs means more duck feet to eat, too13:59
mordred"Make tempest-devstack-vm-neutron-large-ops voting13:59
mordredapparently, the job used to work13:59
*** dims has joined #openstack-infra13:59
*** zoresvit has joined #openstack-infra13:59
sdagueyeh, I'd wait for jog0 to get grounded again. He's got context on that14:00
ttxmordred: I don't remember if fungi was going back to work this week, do you ?14:00
sdaguettx: I thought yes, but he was on the flight that got turned around over japan14:01
*** weshay has joined #openstack-infra14:01
sdagueso completely out of pocket today IIRC from last night's irc14:01
ttxsdague: sounds ike I missed all the fun. What was that about ?14:01
sdagueone of the flights back to the east coast of the US apparently had some issue 4 hrs in14:02
sdagueand they didn't have authority to land in japan, so they had to go back to HK14:02
sdaguefungi was on that flight, not sure who else14:03
*** xchu has quit IRC14:03
*** davidhadas_ has quit IRC14:05
*** davidhadas has joined #openstack-infra14:05
*** dkranz has quit IRC14:05
*** yaguang has joined #openstack-infra14:07
*** julim has joined #openstack-infra14:08
mordredttx: yes, he is. but he's probably not going to be useful today14:08
mordreddue to that flight14:08
*** dcramer_ has joined #openstack-infra14:08
openstackgerritA change was merged to openstack-infra/config: Touch site.pp after git updates.
mordredwow. ok14:12
*** yaguang has quit IRC14:12
mordredthat took forever14:12
openstackgerritA change was merged to openstack-infra/config: Add Fuel-Provision project to Stackforge
mordredI have cleared my review queue14:12
*** cody-somerville has joined #openstack-infra14:12
mordredI've also landed a bunch of trivial changes, so hopefully other infra review folks will not have quite as long on the queue14:13
*** matsuhashi has quit IRC14:13
Shrewsmordred: great job on the keynote. Wish I could've watched it live14:13
mordredShrews: thanks!14:14
*** mriedem has joined #openstack-infra14:19
*** dkranz has joined #openstack-infra14:20
*** mfer has joined #openstack-infra14:21
*** xeyed4good1 has quit IRC14:21
*** dkliban has joined #openstack-infra14:21
*** openstackgerrit has quit IRC14:22
*** openstackgerrit has joined #openstack-infra14:23
*** SergeyLukjanov has joined #openstack-infra14:25
openstackgerritA change was merged to openstack-infra/devstack-gate: don't UNDO_REQUIREMENTS after we update from g-r
*** matsuhashi has joined #openstack-infra14:25
mordredcody-somerville: in the scrollback, ttx and I were discussing a few storyboard topics14:29
mordredcody-somerville: most notably a) need for a weekly IRC meeting, as we're about to have a whole new team of folks14:30
mordredcody-somerville: and b) the idea that we should potentially split it into a REST server project using pecan/wsme like the other openstack projects, and then make the web dashboard a horizon-based rest-consuming web app (horizon used here in the proper form to talk about the django library)14:30
*** mestery has quit IRC14:31
mordredI _think_ we could defer b) to be a v2 thing if in looking at that it seems like an extreme amount of work14:31
mordredbut the idea of being able to have a python-storyboardclient that works like the other projects do is potentially pretty cool14:31
*** mestery has joined #openstack-infra14:31
cody-somervillethe first part of b was entirely what I was thinking as well14:32
cody-somervilleit has the added benefit of making the latter less of a commitment14:32
cody-somervilleas it means we can easily throw away UI and can freely iterate14:32
cody-somervilleor even provide paradigm-specific UIs14:32
mordredalso means that potentially it could grow into a service that could be deployed in an openstack - say with different data storage backends14:33
mordredshould an operator want to integrate an issue reporting thing14:33
cody-somervilleYes sir! :)14:33
* ttx looks up semantic ui14:34
mordred(that's not thing #1 - but I know that robert already wants tripleo clouds to contain an infra, so it stands to reason that an integrated task management wouldn't be crazy)14:34
cody-somervilleLot's of great lessons learnt with OpenStack, I say we leverage all that14:34
mordredcool. I figured we'd be on the same page there :)14:34
*** blamar has joined #openstack-infra14:35
*** rnirmal has joined #openstack-infra14:37
*** jcoufal has quit IRC14:38
mordredcody-somerville: another benefit - we know how to unittest rest servers and data models really well14:39
cody-somervillemordred: ttx: So besides IRC meetings and such which can happen in parallel, I was thinking first steps are to come up with proper problem statement signed off by all stakeholders and then what I like to call a Value-Context-Performance-Hypotheses (VCPH) pack which will help us (and outsides) understand the value of the project, the context of the problem, how we can measure success, and our hypothesis (ie. storyboard).14:39
mordredsounds great to me - although my simple answer to P is "has openstack moved off of launchpad" :)14:40
*** alexpilotti has joined #openstack-infra14:41
mordredbut yes - and also documentation of such things is a great thing to have for projects with larger sets of devs14:41
*** SergeyLukjanov has quit IRC14:41
cody-somervilleIt's important for us to capture the context of why so that people who aren't familiar can go "ah, I see why they did that". I'm positive people with maybe a vested interest in Launchpad will demand this of us and our argument will look sillier if we don't have the facts to back it.14:41
*** dkranz has quit IRC14:42
*** dcramer_ has quit IRC14:44
ttxcody-somerville: semantic ui looks interesting, I'll have to dig more14:46
*** blamar has quit IRC14:46
*** jcooley_ has joined #openstack-infra14:46
ttx(syntax looks sane and the collection of items is pretty decent)14:46
*** dkranz has joined #openstack-infra14:55
*** dprince has joined #openstack-infra14:56
*** katyafervent has joined #openstack-infra15:00
fungiclarkb: working on the -folsom/+havana patches are on my to do list for tomorrow as well now, thanks!15:00
*** jpeeler1 is now known as jpeeler15:01
*** dkranz has quit IRC15:02
ttxfungi: didn't receive your Horizon PTl voting list. Won't set up the vote until tomorrow afternoon though15:04
*** davidhadas has quit IRC15:08
fungittx: i've bounced you another copy of the horizon electorate list, but if memory serves we ran into this before. i think rackspace may be choking on/killing the attachment. i'll gzip it and send another that way15:08
cody-somervillemordred: ttx: Once we have sign off on those two documents, my plan is to prepare a report and presentation on what I feel will need to be done during the initial discovery phase. That'll consist of breaking down the problem into an issue tree so we that can identify the constituent parts - and of my goodness it's snowing out - so that we can start delegating parts of the puzzle for people to break into action items. From t15:09
cody-somervillehere, we'll plot our action items on a 2x2 prioritization matrix (impact vs. need) and come up with a plan with a defined scope. After which we'll be able to supply to the decision makers an interim report and presentation and if we're in agreement we'll use the tentative mandate from that meeting to move forward with preparing the last bits of the discovery phase where we'll test and verify our hypothesis. The output of th15:09
cody-somervilleat will be a final report/presentation with an implementation plan and the data to back up the justification.15:09
cody-somervilleI hope this doesn't sound like too much pig-swattle (or whatever, lol) but I feel this project will be especially important to have properly documented for both the project, for new devs, and to justify the cost of investment for member companies helping to develop this.15:10
*** hashar has quit IRC15:10
cody-somervilleThere is nothing worse than a project with an existential crisis.15:11
sorenI'm confused. Why isn't this being merged?
fungittx: resent both ways. let me know which you receive15:12
*** zul has quit IRC15:13
soren...whereas this made it through that whole process in a matter of two minutes:
ttxcody-somerville, mordred: I fear you may be overthinking this. I expect storyboard to be a very inclusive project with a lot of pass-by contribs from existing developers. Your (well-thought) process seems to apply better to a self-contained team15:13
ttxi.e. some internal dev team15:13
*** miqui has joined #openstack-infra15:14
*** xeyed4good has joined #openstack-infra15:14
openstackgerritJenkins proposed a change to openstack/requirements: Updated from global requirements
ttxfungi: received the gz15:15
ttxcody-somerville, mordred: so I fear most people would just ignore the well-thought plan and just organically do stuff and get it reviewed... which would mean a lot of project management wasted effort15:16
*** dripton has joined #openstack-infra15:17
cody-somervillettx: I'm not suggesting coming up with a big project plan or expansive requirements document or anything like that. I'm talking about coming up with a one page report (with additional pages of supporting data as necessary) that clearly describes the problem, the context, the scope, and the expected end product in broad terms.15:18
*** dkranz has joined #openstack-infra15:18
*** wenlock has joined #openstack-infra15:19
sorenOh, mordred didn't hit the "approved" button on it. Got it.15:19
sorenIt's been so long since I've used this thing, I forgot what the process was. :(15:20
ttxcody-somerville: tried to come up with that when i started -- then I figured I should write a POC instead.15:20
*** jcooley_ has quit IRC15:20
ttxcody-somerville: so I welcome a clear project scope definition and mission statement15:20
mordredsoren: welcome back!15:20
mordredttx, cody-somerville: I'm going to agree with both of you real quick15:21
sorenmordred: Thank you!15:21
* soren goes to dinner15:21
*** krtaylor has joined #openstack-infra15:21
mordredI think if cody-somerville expands out into a few more documents and blueprints and whatnot15:21
cody-somervillettx: People are going to demand an explanation of why we're moving off Launchpad and why we're not using something off the shelf. I want to be able to provide that explanation by pointing at something and showing it's been signed off on.15:21
mordredthen the 'what should I work on now' becomes easier to people15:21
ttxcody-somerville: ack -- like I said, that's what I tried and failed to do, due to lack of time.15:22
ttxI had to throw code out there so that mordred would not make us use JIRA or something15:22
*** matsuhashi has quit IRC15:23
sorenttx: Much appreciated!15:23
mordredttx: I'm hurt that you'd think I'd ever make that choice15:23
*** matsuhashi has joined #openstack-infra15:23
* ttx likes to use JIRA and Xml as ultimate insults15:23
*** dcramer_ has joined #openstack-infra15:23
* ttx properly has worked enough for a holiday day today15:24
* fungi is caught up on scrollback, but needs to run the errands he couldn't get done yesterday because EUNITED15:26
Shrewsmordred, lifeless: I need testr to capture stdout in the logs. What is the standard way to capture that?15:27
*** senk has joined #openstack-infra15:27
*** matsuhashi has quit IRC15:28
mordredShrews: one sec - lemme find you a link15:28
*** sarob has joined #openstack-infra15:28
Shrewsmy google-foo failed me  :(15:28
*** sarob has quit IRC15:28
*** sarob has joined #openstack-infra15:29
*** xeyed4good has left #openstack-infra15:30
*** davidhadas has joined #openstack-infra15:30
*** senk has quit IRC15:33
*** senk has joined #openstack-infra15:33
*** sarob has quit IRC15:33
*** senk has quit IRC15:36
miquihello...anybody still run into this:
miquii am using gerrit 2.4.4-1415:37
*** cody-somerville has quit IRC15:38
*** datsun180b has joined #openstack-infra15:38
*** cody-somerville has joined #openstack-infra15:38
*** dstanek has quit IRC15:41
*** markwash has quit IRC15:42
*** zul has joined #openstack-infra15:46
Shrewsmordred: nm. managed to figure it out15:48
*** alcabrera has joined #openstack-infra15:49
*** blamar has joined #openstack-infra15:49
*** oubiwann has joined #openstack-infra15:50
*** rcleere has joined #openstack-infra15:51
*** jcooley_ has joined #openstack-infra15:51
*** CaptTofu has joined #openstack-infra15:53
*** dstanek has joined #openstack-infra15:54
*** ruhe has joined #openstack-infra15:56
ArxCruzfungi: ping15:56
*** UtahDave has joined #openstack-infra15:57
ArxCruzShrews: what did you do to capture? rfolco need to capture some testr output15:58
ArxCruzShrews: cool! thanks16:01
Shrewsno problem16:03
*** danger_fo_away is now known as danger_fo16:04
*** davidhadas has quit IRC16:09
*** dkranz has quit IRC16:10
*** DennyZhang has quit IRC16:15
*** afazekas has quit IRC16:15
*** sdake_ has quit IRC16:16
*** sdake has joined #openstack-infra16:17
*** rnirmal has quit IRC16:21
rfolcoShrews, ArxCruz: actually what I was looking for is how to grep the runtime info for each individual test on Tempest (I am running with tox)...16:23
*** dkranz has joined #openstack-infra16:23
*** dstanek has quit IRC16:24
clarkbmiqui: I don't think we have that problem. maybe try one of our wars?16:25
*** HenryG has joined #openstack-infra16:25
ArxCruzmordred: in I believe it's necessary a check in branch_data['debs']  and branch_data['rpms'] because If I ran this on my fedora image here, it will always have a branch_data['debs'] because it gets this from devstack_path/files/apts and it will always try to run apt-get install -d -y (although the script doesn't throw any error or exit because of this)16:27
*** dstanek has joined #openstack-infra16:27
clarkbrfolco: it should all end up in the subunit stream. you can pass that to stdout by passing --subunit to testr then you can pipe and grep/parse16:27
*** dstanek has quit IRC16:27
*** sarob has joined #openstack-infra16:28
mordredArxCruz: no - what I mean is16:30
mordredArxCruz: apt-get -y -d install16:31
mordredsimply downloads the package and stores it in the local apt cache16:31
mordredit does not install it16:31
ArxCruzoh, ok16:31
mordredwhereas yum install -y installs16:31
ArxCruzmordred: however, it always have the problem I mention16:31
ArxCruzboth branch_data debs and rpms will always be not null16:31
mordredArxCruz: yes - I believe we may need to mkae the logic there better as well16:31
ArxCruzand run_local will always execute both apt-get and yum16:32
rfolcoclarkb, ok thanks. Will take a look at --subunit option... not very familiar with tox though16:32
ArxCruzalthough I didn't saw any error from my side16:32
ArxCruzthe run_local don't return any failures for not find apt-get for example16:32
*** nicedice_ has joined #openstack-infra16:32
clarkbrfolco: this would be an arg to testr not tox (you can pass through with tox though using posargs which should already be configured)16:33
*** rnirmal has joined #openstack-infra16:33
*** sarob has quit IRC16:33
*** dprince has quit IRC16:36
*** dprince has joined #openstack-infra16:37
rfolcoclarkb, got it. thx++16:38
*** jcooley_ has quit IRC16:39
*** DennyZhang has joined #openstack-infra16:41
NobodyCammordred: May i quote you "it's like I spent a week with my hands replaced by eggplants"16:42
NobodyCamlol :)16:43
mordredNobodyCam: yes. always16:43
*** flaper87 is now known as flaper87|afk16:45
*** DennyZhang has quit IRC16:45
*** mrodden has quit IRC16:48
*** nonnoelio has joined #openstack-infra16:48
*** nonnoelio has quit IRC16:52
*** flaper87|afk is now known as flaper8716:58
*** annegentle has joined #openstack-infra17:01
*** mrodden has joined #openstack-infra17:03
*** jcooley_ has joined #openstack-infra17:03
*** pblaho has quit IRC17:05
*** acabrera has joined #openstack-infra17:05
*** tianst20 has quit IRC17:07
*** marun has quit IRC17:07
*** alcabrera has quit IRC17:08
*** rnirmal_ has joined #openstack-infra17:08
*** rnirmal has quit IRC17:09
*** rnirmal_ is now known as rnirmal17:09
*** rnirmal has quit IRC17:10
*** ruhe has quit IRC17:10
*** rnirmal has joined #openstack-infra17:10
*** pcrews_ has quit IRC17:12
*** davidhadas has joined #openstack-infra17:20
*** sarob has joined #openstack-infra17:30
*** beekneemech is now known as bnemec17:30
*** blamar has quit IRC17:32
*** jpich has quit IRC17:34
*** jcooley_ has quit IRC17:34
*** sarob has quit IRC17:34
*** jcooley_ has joined #openstack-infra17:35
openstackgerritSteve Leon proposed a change to openstack/requirements: Add designate dns client (python-designateclient) to requirements
jog0anteaya: large-ops does a nova boot --num-instances=150 or so17:36
jog0anteaya: so neutron is failing that somewhat simple test17:36
jog0anteaya: I haven't had a chance to hunt down folks to fix the neutron side of that yet, will do that now17:37
*** syerrapragada1 has quit IRC17:38
*** syerrapragada has joined #openstack-infra17:38
*** jcooley_ has quit IRC17:42
*** jcooley_ has joined #openstack-infra17:43
*** DennyZhang has joined #openstack-infra17:44
jog0anteaya: I filed a bug
*** jcooley_ has quit IRC17:49
*** markwash has joined #openstack-infra17:56
*** blamar has joined #openstack-infra17:57
*** derekh has quit IRC18:03
*** Ryan_Lane has joined #openstack-infra18:04
*** sarob has joined #openstack-infra18:05
*** pcrews_ has joined #openstack-infra18:09
*** vipul is now known as vipul-away18:10
*** sarob_ has joined #openstack-infra18:10
*** sarob has quit IRC18:10
*** Hefeweizen has quit IRC18:16
danger_fomordred:  re  -- sorry for the continued confusion; I confirmed directly with Denny that he had the start/end dates format wrong when he originally added himself to the gitdm config18:17
*** UtahDave has quit IRC18:18
*** dcramer_ has quit IRC18:19
*** flaper87 is now known as flaper87|afk18:19
*** gyee has joined #openstack-infra18:19
*** talluri_ has joined #openstack-infra18:21
*** dprince has quit IRC18:22
*** acabrera is now known as alcabrera18:22
*** briancline has quit IRC18:23
*** talluri has quit IRC18:23
*** blamar has quit IRC18:24
*** blamar has joined #openstack-infra18:24
*** CaptTofu has quit IRC18:32
*** CaptTofu has joined #openstack-infra18:32
*** vipul-away is now known as vipul18:35
*** johnthetubaguy has quit IRC18:40
*** whoops has joined #openstack-infra18:42
*** jerryz has joined #openstack-infra18:45
openstackgerritA change was merged to openstack/requirements: Add designate dns client (python-designateclient) to requirements
*** sdake_ has joined #openstack-infra18:53
openstackgerritJoe Gordon proposed a change to openstack-dev/hacking: Remove vim modeline sample
*** jergerber has joined #openstack-infra18:56
*** xeyed4good has joined #openstack-infra19:06
*** xeyed4good has left #openstack-infra19:06
*** DennyZhang has quit IRC19:10
*** hdd has joined #openstack-infra19:14
*** xeyed4good has joined #openstack-infra19:22
*** xeyed4good has left #openstack-infra19:22
yolandafungi, jeblair, i'm interested in work on some other bug, but not sure on what can be best to choose, any suggestion? i have expertise on gerrit, zuul, jenkins, jenkins-job-builder19:26
clarkbyolanda: have you looked at the list of bugs tagged low-hanging-fruit? those are generally a good place to look19:31
clarkband if none of those seem appropriate feel free to let us know. I am sure we have more that could be tagged that way that aren't19:32
yolandaclarkb, yes, i've looked at them, but I think i'd need more clarification about them19:33
*** mkoderer has quit IRC19:33
yolandaalso saw some of them with an old date, so not sure if that makes sense to grab them19:33
clarkbyolanda: any particular examples? I can do my best now to fill in details for a couple19:34
yolandawell, looking at the ones with low priority, because i don't have much time and don't want to block important ones. For example 1235335, 1233846 (if that's fixed upstream, only a backport needed?)19:35
clarkbyolanda: hmm ya, that particular one may need some discussion on how we want to tackle it19:38
clarkbyolanda: if we go the route of a new daemon we probably want that service to handle general bug updates too. We could add that as a meeting agenda item and discuss it with ttx and jeblair and fungi19:38
clarkb(I am talking about 123533519:39
lifelessShrews: got an answer?19:39
Shrewslifeless: yep. thx19:39
*** briancline has joined #openstack-infra19:39
clarkbyolanda: for 1233846, my comment is that someone (maybe us) should fix the upstream bug rather than fixing it on our fork19:39
clarkbyolanda: then when we upgrade gerrit we will get the fix for free. I don't believe it is currently fixed upstream and carrying another backport is probably less valuable than working to upgrade19:40
yolandaok, that should be a good one then19:40
yolandaalso i may work on this one 1234959, i can catch up with Diogo about that19:42
*** ArxCruz has quit IRC19:42
yolandai think it makes sense for us19:42
clarkbyolanda: yup JJB patches are always welcome19:43
clarkbparticularly when they are useful to mutliple groups of people19:43
yolandamm, i'll assign this and discuss with Diogo, so we are able to test it19:44
fungiArxCruz: pong (were you just pinging me about the fedora/rhel devstack tweaks above, or something else?)19:44
clarkbyolanda: sounds good19:44
*** chandankumar has quit IRC19:46
fungiyolanda: also, any bugs you think look interesting but may need clarification, feel free to just update the bug asking for more info. as to age, we try to get through refreshing them every couple months on a team bug review day, but the low/wishlist level bugs do have a tendency to get less refreshing19:46
jog0clarkb: as per the summit can be merged without a second +2 right ?19:54
jog0I ask becuase I missed the e-r session19:54
clarkbjog0: yup go for it19:54
jog0clarkb:  cool19:56
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add fingerprint for bug 1242777
openstackgerritA change was merged to openstack/requirements: Add django-bootstrap-form to requirements
*** alcabrera is now known as alcabrera|afk20:02
*** alexpilotti has quit IRC20:03
anteayamordred: awesome thank you for that patch20:04
*** rfolco has quit IRC20:05
anteayajog0: thanks for that bug report, if you have any suggestions for how I can work on identifying a fix, I am open to suggestions and will dig into it in the mean time20:06
jog0anteaya: I couldn't find when the job started failing20:07
jog0anteaya: but I think graphite should show you20:07
jog0anteaya: if you can find the day when the job started failing it shouldn't be too hard to isolate the offending patch20:08
jog0anteaya: thanks for doing this20:08
anteayajog0: I will need to keep hearing that20:09
anteayaand okay I will head over to graphite20:09
anteayaso if I understand correctly, large-ops did pass at one point and then started failing?20:10
anteayaso far I am not setting a devstack job in graphite labelled large-ops, might it be called something else?20:12
jog0anteaya: correct, it worked at one point20:13
*** melwitt has joined #openstack-infra20:14
jog0anteaya: hmm looking20:14
*** LinuxJedi_ is now known as LinuxJedi20:16
jog0anteaya: so it failed Oct 31st20:17
*** UtahDave has joined #openstack-infra20:17
anteayayes, thanks jog020:19
anteayaI need to get better at both graphite and weechat20:19
anteayabut I will start will checking on what merged Oct 31st20:20
jog0anteaya: it took me a while to figure out where we hide things in graphite20:21
*** blamar has quit IRC20:21
anteayagreat, it will take me a bit - but I will get it20:23
openstackgerritChuck Short proposed a change to openstack/requirements: Add HACKING.rst
*** blamar has joined #openstack-infra20:23
anteayaif I am seeing this correctly, there were two patches merged to neutron on October 31st, one added some comments to a file and the other: edited neutron/openstack/common/log.py20:24
anteayawould a logging error bring down large-ops?20:25
jog0anteaya: I don't think so, but you can just try a git-revert and see what jenkins/gerrit/zuul says20:26
anteayajog0: I have never done that before, do you know where i might find an example of how to do that?20:28
*** yolanda has quit IRC20:30
fungianteaya: git revert 1506f26b6563f5b8e69482cb14ad84d00565f6da20:31
jog0anteaya: what fungi  said20:31
fungianteaya: then 'git commit -a --amend' so that a change-id header gets added to the revert commit20:31
fungiand add some info to the commit message about why you're reverting it20:31
fungithen git review it like any other commit20:31
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Typo serious=>set
anteayaI'll give it a shot20:38
*** dcramer_ has joined #openstack-infra20:38
*** gfa has quit IRC20:42
*** gfa has joined #openstack-infra20:42
*** sarob_ has quit IRC20:43
*** sarob has joined #openstack-infra20:43
anteayais there any way of testing this git revert locally prior to submitting it?20:44
fungican you explain what you mean by "testing" it?20:45
fungiyou can use 'git show' to see the commit, and confirm that it does the inverse of (e.g. backs out) the 1506f26 commit20:46
jog0anteaya: you could set up a devstack and run the large-ops test if you want20:46
jog0but its easier to push it up instead20:46
jog0and mark as WIP20:46
fungiright, the simulating section of the openstack-infra/devstack-gate readme file is as close as you'll probably get to manually recreating what the large-ops job does, but that's overkill unless you really need to test in private for some reason20:47
anteayaWIP it is20:48
*** sarob has quit IRC20:48
annegentlewas anyone at the "future of design summits" session who has a recap?20:53
* annegentle also reads
fungiwe tried to capture all of it there20:54
*** hogepodge has joined #openstack-infra20:54
anteayathe biggest point of discussion was the timing of the design summit and the conference20:55
fungithe biggest tl;dr is that we'll skew it with conference mon-thu and design tue-fri20:55
anteayawe had to make a decision about both Atlanta and Paris due to contracts for venues for Paris20:55
annegentlefungi: anteaya: ok, that helps20:55
fungiand try to pack more of the technical plenaries into monday20:55
annegentlemuch excitement around later night?20:56
fungino, i think that was still status quo20:56
fungilots of people are doing meetings in the evenings, so it makes conflicts with late-running design sessions more likely20:56
annegentlefungi: oh do people want room provided for hacking into the night?20:56
annegentlefungi: anteaya: any discussion around the types of meeting we had with the boot camp for infra and docs?20:57
annegentlefungi: anteaya: as in, does anyone else plan to do those ongoing?20:58
fungii don't think we talked about hackerspaces open into the evenings, more just that scheduling design sessions after 6pm was probably not great20:58
annegentlefungi: oh ok20:58
*** markmc has quit IRC20:58
fungiit does sound like groups expect to also do mid-cycle sprints/hackathons. we've been talking about another one for infra, probably more goal-oriented rather than training-focused though20:59
annegentlefungi: ok. Ours was sort of expected to be training but then we did more of team meeting and getting to know each other, I think.20:59
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Remove fixed bugs from queries.yaml
fungilike we pick a bigger project which benefits from several of us beating it into shape together in person, and then invite anyone who wants to pitch in20:59
fungibut yeah, the main point to come out of the future of summits discussion was the conference/design offset by a day21:00
fungialso possibly setting up dedicated project-specific bof breakout spaces for getting things done together during holes people may have in their design summit schedules during the day21:02
notmynameif I had a static real-hardware cluster to do openstack tests, I'd be concerned that per-commit tests would get backed up in a queue. so if I did nightly testing, is there a place other than the openstack wiki (or my own website) where I could public test results? any suggestions on how to close the feedback loop there?21:02
notmynamejeblair: ^21:02
*** dcramer_ has quit IRC21:04
*** pcrews_ has quit IRC21:04
*** CaptTofu has quit IRC21:04
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Fix functional tests
*** CaptTofu has joined #openstack-infra21:05
mordrednotmyname: hey21:07
mordrednotmyname: jeblair is out this week21:07
notmynameah ok then :-)21:07
mordrednotmyname: at the moment, the only 'good' feedback loop we've found for periodic jobs is an IRC bot or email21:07
notmynamekk. I can think of much more horrible ones21:08
notmynamemordred: are there hooks already set up where I could post to an openstack bot, if it had my key?21:09
mordrednotmyname: not really - at the moment you'd have to write your own21:09
mordred(it's possible that clarkb or fungi might have better ideas here)21:09
*** CaptTofu has quit IRC21:09
notmynamemordred: ok, no worries either way, but I'm happy to use infrastructure that already exists21:10
mordrednotmyname: yah- seems like the srt of thing we should have21:11
notmynamewell, only after the 2nd person wants it :-)21:11
openstackgerritA change was merged to openstack-infra/elastic-recheck: Typo serious=>set
*** dstufft has quit IRC21:13
*** luisg has quit IRC21:14
*** julim has quit IRC21:14
fungiwe have a need to do some sort of periodic reporting as well as a place to put post-merge job results (for example, coverage reports). i think the backend is probably an uninteresting discussion (or rather an interesting discussion we've already had/are having), but the presentation--how to get people to pay attention and simultaneously be able to find the results they're interested in seeing--seems21:15
fungito have been our biggest issue over time21:15
*** dstufft has joined #openstack-infra21:15
*** julim has joined #openstack-infra21:16
*** dcramer_ has joined #openstack-infra21:17
fungiwe've been running bitrot and coverage jobs for ages, but typically nobody paid attention when results went out via e-mail, nor looked them up on the web until something was discovered broken through other avenues (and then merely used them to help narrow down the timeframe breakage started)21:17
anteayafungi: I just tried to git revert this patch: and got this error: is a merge but no -m option was given21:18
anteayaany hints?21:18
fungianteaya: ahh, you want to get revert the commit merged by c1577c5 rather than the merge commit itself21:19
anteayaI will look for that21:19
notmynamefungi: I like the idea of a bot, but if it's running "nightly", who's night and how is that reported for those who are currently asleep. (agreed on the fact that the communication is a hard part)21:19
fungiwhose night? ;)21:19
fungioh, you said that21:20
fungiyeah, agreed. it's a hard problem. if it were an easy problem, we'd have already solved it and moved on21:20
notmynamepossessive of who is whose? not who's? WTF English?21:20
*** alcabrera|afk is now known as alcabrera21:20
funginotmyname: i wasn't correcting you, just replying to you before i read everything you wrote21:21
fungifingers ahead of eyes i'm afraid21:21
notmynameheh. no worries. I was just surprised at that little detail :-)21:21
*** shardy is now known as shardy_afk21:22
fungiyeah, for learning to speak/read only one language, i sure picked a really bad one21:22
fungianteaya: so if you look at that url, you'll see it links two parent commits. one of those is presumably the one you want to revert21:23
anteayait was, I did a search using the log message and found the original commit, but using the parent would work too, thanks21:24
notmynamefungi: IRC bot that is quiet until there is a failing test and then proclaims it in a particular channel every hour until there is a passing run21:25
funginotmyname: a nagbot! i like it ;)21:25
lifelessshould I be able to vote on openstack/governance proposals ?21:26
lifeless(it is the TC voting area right?)21:26
notmynamelifeless: no. you are just a figurehead ;-)21:26
*** hdd has quit IRC21:26
* fungi checks the acl on it21:26
notmynamefungi: of course, you'd have to have a way to manually trigger a new test run (or whatever) so it could be quieted earlier than once a day21:26
*** vipul is now known as vipul-away21:27
fungi#nagbot shaddup!21:27
fungithough yeah, an irc bot with the power to run arbitrary things gets into dangerzone21:27
fungilifeless: so, to answer your question, it appears you should be able to -2/+221:28
notmynamefungi: no, something like "rerun test" would be sufficient. if it already has scheduling (ie do something every hour), then it would be easy to add "don't do something more than X times per unit time"21:28
*** luisg has joined #openstack-infra21:29
*** senk has joined #openstack-infra21:29
fungithat's granted to the tech-committee group, of which you are a member21:29
lifelessfungi: indeed, I can now, dunno why I couldn't before21:29
*** jhesketh_ has joined #openstack-infra21:29
sdaguedoes anyone have the link to upload presentations, I can't seem to find that email anywhere21:29
sdagueand I'd like to be a good citizen and get mine up21:30
fungilifeless: maybe someone pushed a draft patchset? that prevents you from being able to vote on the last visible patchset21:30
clarkbmordred: fungi: notmyname: I don't have many good ideas. this came up in HKG when talking to MS folks21:30
funginotmyname: right, having it schedule a (near-)immediate rerun would probably be easy to accomplish with minimal privileges21:31
*** sarob has joined #openstack-infra21:31
notmynamefungi: or even let anyone trigger it, but it only runs 2 times per every 10 minutes or something (in my mind, these are swift tests that swiftstack is hosting on a physical cluster in our lab)21:32
clarkbcould we do something like the experimental queue in zuul21:32
clarkbwhere the backend listens for specific gerrit events and reports back on the cahnge generating that event?21:32
clarkbthat would be human triggered21:33
fungithough it sounds like this is for tip-of-branch stuff rather than on proposed changes21:33
notmynameya. for me, I want cluster testing on real hardware of a known config. I can't scale that for every patch proposal, so I think I'd need something to run periodically21:36
*** sarob has quit IRC21:36
clarkbperhaps you could run jobs on branch/ref updated events then leave comments on the change post merge? (I don't actually like that idea as it doesn't solve the problem of when to run well or the reporting problem)21:38
*** vipul-away is now known as vipul21:43
sdagueclarkb: jog0's holding this one on your sanity check -
sdaguewhen you get a few21:51
fungifrom what i've seen, nobody actually reads comments on closed changes post-merge21:51
fungiperhaps for very large values of nobody/very small values of somebody21:51
fungibut as a means of publication, it won't reach much of an audience21:52
clarkbsdague: lgtm, but what about the queries internal to e-r. should we update those too?21:52
sdagueclarkb: probably21:54
sdaguebut I figured I'd clean those up in other refactoring once I'm unfogged from the flight21:54
clarkbsdague: wfm I will approve 55144 now21:54
openstackgerritA change was merged to openstack-infra/elastic-recheck: update queries to new schema
*** UtahDave has quit IRC21:59
*** pcrews has joined #openstack-infra22:01
*** mfer has quit IRC22:02
*** senk has left #openstack-infra22:02
*** adalbas has quit IRC22:08
*** MarkAtwood has joined #openstack-infra22:12
fungiheh... summit survey fail22:13
fungi"3. [...blah, blah, blah...] Please choose all that apply."22:14
fungiexcept... uh... it's a RADIO BUTTON22:14
fungichoose as many as you need, so long as you only need one22:15
*** pcm_ has quit IRC22:16
*** dkranz has quit IRC22:17
*** jergerber has quit IRC22:22
*** danger_fo is now known as danger_fo_away22:30
*** hogepodge has quit IRC22:31
*** sarob has joined #openstack-infra22:31
*** thomasem has quit IRC22:34
*** senk has joined #openstack-infra22:34
*** sarob has quit IRC22:36
*** yamahata has quit IRC22:37
*** alcabrera has quit IRC22:38
*** yamahata has joined #openstack-infra22:39
*** weshay has quit IRC22:40
*** senk has quit IRC22:43
*** hogepodge has joined #openstack-infra22:43
*** mriedem has quit IRC22:44
*** CaptTofu has joined #openstack-infra22:45
*** johnthetubaguy has joined #openstack-infra22:45
*** UtahDave has joined #openstack-infra22:46
*** johnthetubaguy has quit IRC22:50
*** loq_mac has joined #openstack-infra22:51
*** lcestari has quit IRC22:53
*** UtahDave has quit IRC22:53
*** michchap has quit IRC22:53
*** yamahata has quit IRC22:55
*** michchap has joined #openstack-infra22:56
*** sarob has joined #openstack-infra22:56
*** sarob has quit IRC22:57
*** sarob has joined #openstack-infra22:57
*** luisg has quit IRC23:00
*** luisg has joined #openstack-infra23:00
*** bodepd has quit IRC23:05
*** Hunner_ is now known as Hunner_tail23:07
*** UtahDave has joined #openstack-infra23:09
*** fifieldt has joined #openstack-infra23:11
*** bodepd has joined #openstack-infra23:13
*** rcleere has quit IRC23:19
*** vipul is now known as vipul-away23:20
ekarlsoare there plans to bump gerrit to latest soon ?23:25
openstackgerritJoe Gordon proposed a change to openstack-infra/elastic-recheck: Update query for bug 1225664
*** oubiwann has quit IRC23:26
jog0clarkb sdague: can one of you review ^23:26
*** sarob has quit IRC23:27
jog0even though says I can review it myself I would prefer not to23:27
clarkbekarlso: yes once we sort out WIP. there is one oustanding chamge to upstream that we need first23:27
*** sarob has joined #openstack-infra23:27
clarkbjog0 sure give me a minute23:27
*** vipul-away is now known as vipul23:27
*** UtahDave has quit IRC23:29
*** guitarza1 is now known as guitarzan23:31
*** UtahDave has joined #openstack-infra23:31
*** sarob has quit IRC23:32
clarkbjog0: approved. are we running out of disk on the slaves? we can allocate more to swift/glance/cinder if necessary23:32
clarkbIdeally we would sort out why the disk is being used up though23:32
*** datsun180b has quit IRC23:34
openstackgerritA change was merged to openstack-infra/elastic-recheck: Update query for bug 1225664
*** xchu has joined #openstack-infra23:34
jog0clarkb: I think its the swift loopback device23:36
jog0or at least that change didn't help23:36
clarkbtoday has been such a weird day. Went grocery shopping, now fighting off a nap23:38
jog0clarkb: ohh jetlag23:39
notmynamejog0: clarkb: what is the max object size setting in swift.conf?23:40
notmynamejog0: clarkb: then, look at the ring creation to see how many replicas are being used (IIRC, just one).23:41
jog0notmyname: yeah just one23:41
jog0notmyname: what is the config option name23:42
jog0for max object size23:43
sdaguejog0: so the output of elastic recheck summit session is self approving of individual query adds is socially acceptable23:43
clarkbdevstack grabs values from the default configs provided by projects then overrides some of them with iniset, I would need to track down whether or not that is changed23:43
*** rnirmal has quit IRC23:43
jog0sdague: yeah I am fine with that, just don't want to personally if I can help it23:43
jog0I like having a second pair of eyes on my stuff23:43
sdagueyep, fair, me too23:43
sdaguebut I realized you weren't in the room :)23:43
sdagueso I should share23:44
jog0sdague: thankfully the etherpad notes are fairly detailed23:44
*** rnirmal has joined #openstack-infra23:44
jog0so I figured that one out, but thanks23:44
clarkbnotmyname: jog0: looks like max_file_size?23:45
*** adalbas has joined #openstack-infra23:45
jog0notmyname: devstack doesn't touch any config option with the word 'max' in it23:45
notmynamejog0: oh, good. then it's trying to put 5GB files :-)23:46
notmynamejog0: clarkb:
clarkbnotmyname: which should be fine right?23:46
notmynamejog0: clarkb: well, if you don't have 5GB on the target system, it's not. actually some small multiple of that to account for the test putting objects that size and right under it23:47
jog0notmyname: ^ is the bug we are seeing23:47
clarkbnotmyname: so setting max_file_size to say 2GB would be safer?23:48
*** whoops has quit IRC23:48
clarkbor bumping the store to ~10GB?23:48
*** rnirmal has quit IRC23:50
*** oubiwann has joined #openstack-infra23:50
*** loq_mac has quit IRC23:54
clarkbFYI jenkinsci on github apparently had a derp. gearman-plugin repo there may have been affected23:54
mordredclarkb: aroo?23:54
clarkbthey appear to be fixing it, but worth knowing should something pop up23:55
clarkbzaro: if you want more reasons why I think the whole github <-> gerrit thing is silly ^23:56
clarkbzaro: I think that derp basically falls under what my existing argument is though. That it adds unnecessary complexity and confusion that results in problems :)23:56
fungii think we've seen several changes bumping max_file_size already over the past year. the most recent one we shunted over to start managing that via the default in devstack rather than overriding it with devstack-gate config23:56
notmynameclarkb: what's the current setting? 1GB? so how has it ever been working?23:56
clarkbfungi: I don't think that is bumping max_file_size but bumping the backend store instead23:58
clarkbnotmyname: 4GB is the backend size currently23:58
clarkbnotmyname: I am assuming that tests run serially don't run into problems23:58
clarkbnow with parallel tests we can hit swift with mutliple large requests concurrently which causes the failure23:58
*** mdenny has quit IRC23:59

Generated by 2.14.0 by Marius Gedminas - find it at!