| *** sdake has quit IRC | 00:01 | |
| *** jaypipes has quit IRC | 00:01 | |
| *** otherwiseguy has quit IRC | 00:01 | |
| *** dims_ has quit IRC | 00:03 | |
| *** zul has quit IRC | 00:06 | |
| *** dkliban has joined #openstack-infra | 00:07 | |
| *** rainya has quit IRC | 00:09 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Updated project list interface. https://review.openstack.org/110169 | 00:11 |
|---|---|---|
| *** vermarah has quit IRC | 00:12 | |
| *** Sukhdev has joined #openstack-infra | 00:12 | |
| *** balajin has joined #openstack-infra | 00:15 | |
| *** balajin is now known as dumbhead | 00:16 | |
| *** marcoemorais has quit IRC | 00:16 | |
| *** dkliban has quit IRC | 00:18 | |
| fungi | okay, the last job on jenkins05 just finished, so restarting nodepool real quick now | 00:19 |
| *** gangil has quit IRC | 00:20 | |
| ianw | clarkb / fungi : know you're dealing with issues, when time can i ping you on some things (can bring to meeting tomorrow as well) | 00:21 |
| ianw | https://review.openstack.org/#/c/108869/ (Move PATH set earlier) | 00:21 |
| ianw | ^ little fix for f20 error, don't think it's causing problems | 00:21 |
| ianw | https://review.openstack.org/103458 (Initial RHEL7/Centos7 support) & https://review.openstack.org/109906 (Add centos7 image for rax) | 00:22 |
| ianw | ^ any objections to this approach. i'm running some things by hand, but I'd like to know nodepool can do it | 00:22 |
| *** rlandy has quit IRC | 00:22 | |
| fungi | okay, so rackspace did in fact finally update the ticket, and mentioned that the host on which jenkins05 is resident was "missing network flows, causing packets to not be routed correctly" | 00:25 |
| *** michchap_ has quit IRC | 00:25 | |
| *** michchap has joined #openstack-infra | 00:25 | |
| *** yamamoto_ has joined #openstack-infra | 00:26 | |
| fungi | we can choose to take an outage and proactively migrate jenkins05 to another host with their help, if we want to avoid the possibility that this is a recurrent issue on the current host | 00:26 |
| *** craigbr has joined #openstack-infra | 00:26 | |
| fungi | i'll add it to tomorrow's agenda and we can discuss when more people are around | 00:26 |
| *** jergerber has quit IRC | 00:26 | |
| fungi | jenkins05 nodes are properly getting marked used when they start running jobs now, so we should be in the clear | 00:27 |
| *** yamamoto has quit IRC | 00:28 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 00:29 | |
| *** wenlock has quit IRC | 00:29 | |
| *** sdake has joined #openstack-infra | 00:29 | |
| *** sdake has quit IRC | 00:29 | |
| *** sdake has joined #openstack-infra | 00:29 | |
| *** dustins has joined #openstack-infra | 00:31 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 00:32 | |
| *** dustins has quit IRC | 00:34 | |
| *** julim has joined #openstack-infra | 00:38 | |
| *** dmsimard_away is now known as dmsimard | 00:38 | |
| *** julim has quit IRC | 00:38 | |
| *** julim has joined #openstack-infra | 00:40 | |
| *** bdpayne has quit IRC | 00:40 | |
| *** dkliban has joined #openstack-infra | 00:42 | |
| *** marcoemorais has joined #openstack-infra | 00:43 | |
| *** homeless has quit IRC | 00:43 | |
| *** hemna is now known as hemna_ | 00:47 | |
| *** achanda has quit IRC | 00:52 | |
| *** gyee has quit IRC | 00:52 | |
| *** dkliban has quit IRC | 00:55 | |
| *** otter768 has joined #openstack-infra | 00:55 | |
| *** dmsimard is now known as dmsimard_away | 00:57 | |
| *** camunoz has quit IRC | 01:00 | |
| *** stevemar has joined #openstack-infra | 01:01 | |
| *** zns has quit IRC | 01:05 | |
| *** dimtruck is now known as zz_dimtruck | 01:05 | |
| *** yamamoto_ has quit IRC | 01:06 | |
| *** yaguang has joined #openstack-infra | 01:06 | |
| *** otter768 has quit IRC | 01:07 | |
| *** zz_dimtruck is now known as dimtruck | 01:07 | |
| *** otter768 has joined #openstack-infra | 01:11 | |
| anteaya | fungi: I am not sure if i will be present at tomorrow's meeting but based on data thus far I'm for migrating jenkins05 | 01:11 |
| *** banix has joined #openstack-infra | 01:15 | |
| *** camunoz has joined #openstack-infra | 01:19 | |
| *** markwash_ has quit IRC | 01:23 | |
| *** aysyd has quit IRC | 01:30 | |
| *** Sukhdev has quit IRC | 01:33 | |
| *** daya_k has joined #openstack-infra | 01:34 | |
| *** xianghui has joined #openstack-infra | 01:36 | |
| redrobot | SergeyLukjanov you around? | 01:39 |
| *** rfolco has joined #openstack-infra | 01:41 | |
| *** rwsu has quit IRC | 01:42 | |
| *** camunoz has quit IRC | 01:43 | |
| *** baohua has joined #openstack-infra | 01:46 | |
| *** jyuso has joined #openstack-infra | 01:46 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/config: Add experimental Centos7 job https://review.openstack.org/110178 | 01:46 |
| *** yamamoto_ has joined #openstack-infra | 01:47 | |
| *** gangil has joined #openstack-infra | 01:47 | |
| *** nosnos has joined #openstack-infra | 01:48 | |
| *** amotoki has joined #openstack-infra | 01:49 | |
| *** otherwiseguy has joined #openstack-infra | 01:49 | |
| *** amotoki has quit IRC | 01:51 | |
| *** camunoz has joined #openstack-infra | 01:56 | |
| *** craigbr has quit IRC | 01:58 | |
| *** shashankhegde has joined #openstack-infra | 01:58 | |
| fungi | okay, i've nova deleted 145 orphaned nodepool instances. more than half were in hpcloud-b5 | 02:01 |
| *** otherwiseguy has quit IRC | 02:01 | |
| fungi | and zuul seems to have caught up with the allocation demand now | 02:02 |
| *** stevemar has quit IRC | 02:03 | |
| *** otter768 has quit IRC | 02:04 | |
| *** otter768 has joined #openstack-infra | 02:06 | |
| *** craigbr has joined #openstack-infra | 02:08 | |
| *** bdpayne has joined #openstack-infra | 02:11 | |
| *** yamahata has quit IRC | 02:16 | |
| *** otherwiseguy has joined #openstack-infra | 02:20 | |
| *** asettle has quit IRC | 02:21 | |
| *** asettle has joined #openstack-infra | 02:21 | |
| *** masayuki_ has joined #openstack-infra | 02:29 | |
| *** masayukig has quit IRC | 02:29 | |
| *** otherwiseguy has quit IRC | 02:29 | |
| *** masayukig has joined #openstack-infra | 02:34 | |
| *** dimtruck is now known as zz_dimtruck | 02:35 | |
| *** otherwiseguy has joined #openstack-infra | 02:35 | |
| *** angdraug has quit IRC | 02:38 | |
| *** shashankhegde has quit IRC | 02:39 | |
| *** shashankhegde has joined #openstack-infra | 02:39 | |
| *** Ryan_Lane has quit IRC | 02:43 | |
| *** rainya has joined #openstack-infra | 02:44 | |
| *** unicell has joined #openstack-infra | 02:47 | |
| *** penguinRaider__ has quit IRC | 02:49 | |
| *** shashankhegde has quit IRC | 02:51 | |
| *** praneshp has joined #openstack-infra | 02:51 | |
| *** penguinRaider__ has joined #openstack-infra | 02:52 | |
| *** rainya has quit IRC | 02:54 | |
| *** praneshp_ has joined #openstack-infra | 02:54 | |
| *** rainya has joined #openstack-infra | 02:54 | |
| *** praneshp has quit IRC | 02:56 | |
| *** praneshp_ is now known as praneshp | 02:56 | |
| *** dkliban has joined #openstack-infra | 02:56 | |
| *** stevemar has joined #openstack-infra | 03:00 | |
| *** harlowja is now known as harlowja_away | 03:07 | |
| *** harlowja_away is now known as harlowja | 03:08 | |
| *** craigbr has quit IRC | 03:13 | |
| *** otherwiseguy has quit IRC | 03:15 | |
| *** terryw has joined #openstack-infra | 03:15 | |
| *** nosnos has quit IRC | 03:20 | |
| *** shashankhegde has joined #openstack-infra | 03:21 | |
| *** lttrl has quit IRC | 03:22 | |
| *** terryw has quit IRC | 03:23 | |
| *** otherwiseguy has joined #openstack-infra | 03:23 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Move PATH set earlier https://review.openstack.org/108869 | 03:24 |
| *** otter768 has quit IRC | 03:25 | |
| *** banix has quit IRC | 03:32 | |
| *** mestery_ has joined #openstack-infra | 03:34 | |
| *** mestery has quit IRC | 03:36 | |
| *** craigbr has joined #openstack-infra | 03:37 | |
| *** shashankhegde has quit IRC | 03:38 | |
| *** markwash has joined #openstack-infra | 03:41 | |
| *** mestery_ has quit IRC | 03:46 | |
| *** stevemar has quit IRC | 03:46 | |
| *** stevemar has joined #openstack-infra | 03:46 | |
| *** markwash has quit IRC | 03:47 | |
| *** craigbr has quit IRC | 03:47 | |
| *** flaviof_zzz is now known as flaviof | 03:49 | |
| *** amotoki has joined #openstack-infra | 03:55 | |
| *** mestery has joined #openstack-infra | 03:56 | |
| *** mestery_ has joined #openstack-infra | 03:59 | |
| *** mestery has quit IRC | 04:02 | |
| *** otherwiseguy has quit IRC | 04:04 | |
| *** otherwiseguy has joined #openstack-infra | 04:05 | |
| *** asettle has quit IRC | 04:06 | |
| *** alexandra has joined #openstack-infra | 04:06 | |
| *** alexandra is now known as asettle | 04:06 | |
| *** Sukhdev has joined #openstack-infra | 04:07 | |
| *** mestery_ has quit IRC | 04:11 | |
| *** armax has quit IRC | 04:15 | |
| *** chandankumar has joined #openstack-infra | 04:15 | |
| lifeless | jeblair: mordred: btw http://kanboard.net/ crossed my desk recently (as a opensource trello-alike) - I haven't tested it yet | 04:16 |
| *** nosnos has joined #openstack-infra | 04:16 | |
| greghaynes | khannnboard | 04:16 |
| greghaynes | im sure no one has ever made that joke | 04:16 |
| *** otherwiseguy has quit IRC | 04:19 | |
| *** mestery has joined #openstack-infra | 04:21 | |
| *** armax has joined #openstack-infra | 04:22 | |
| *** david-lyle has joined #openstack-infra | 04:25 | |
| *** wenlock has joined #openstack-infra | 04:27 | |
| *** Ryan_Lane has joined #openstack-infra | 04:29 | |
| *** Ryan_Lane has joined #openstack-infra | 04:29 | |
| *** Ryan_Lane has quit IRC | 04:30 | |
| *** timrc is now known as timrc-afk | 04:31 | |
| *** armax has quit IRC | 04:33 | |
| *** Ryan_Lane has joined #openstack-infra | 04:34 | |
| *** yfried has quit IRC | 04:35 | |
| *** craigbr has joined #openstack-infra | 04:39 | |
| *** HenryG is now known as HenryG_afk | 04:45 | |
| *** morganfainberg is now known as morganfainberg_Z | 04:59 | |
| tchaypo | r1chardj0n3s: i assume you're going to be in brisbane? | 05:00 |
| *** david-lyle has quit IRC | 05:00 | |
| *** afazekas is now known as __afazekas | 05:00 | |
| r1chardj0n3s | Yep tchaypo | 05:00 |
| *** david-lyle has joined #openstack-infra | 05:00 | |
| tchaypo | cool. I've been trying out devpi as a replacement for pypi-mirror | 05:01 |
| tchaypo | May want to talk to you about the wheel caching | 05:03 |
| tchaypo | heh. | 05:03 |
| * tchaypo was reading a mailing list post and thinkging "gosh this guy makes sense" | 05:03 | |
| tchaypo | and then realised it was my own email.. | 05:03 |
| tchaypo | but i think we can probably get away with using devpi as a proxy, and a pip wheel cache + greghaynes suggestion of a bind mount | 05:04 |
| *** david-lyle has quit IRC | 05:05 | |
| *** david-lyle has joined #openstack-infra | 05:05 | |
| *** otherwiseguy has joined #openstack-infra | 05:06 | |
| *** craigbr has quit IRC | 05:07 | |
| StevenK | tchaypo: The pypi element does a bind mount already, I think | 05:07 |
| *** rainya has quit IRC | 05:10 | |
| tchaypo | it mounts ~/.cache/image-create/pypi/mirror | 05:11 |
| tchaypo | and i can see whls inside that directory | 05:12 |
| *** david-lyle has quit IRC | 05:12 | |
| dstufft | are they nice wheels | 05:12 |
| tchaypo | i don't know. what does a nice wheel look like? | 05:12 |
| *** david-lyle has joined #openstack-infra | 05:12 | |
| dstufft | dunno :) | 05:13 |
| StevenK | tchaypo: http://www.salestrailers.com.au/media/catalog/product/cache/1/image/9df78eab33525d08d6e5fb8d27136e95/r/t/rt1_mag_wheel_14inch.jpg | 05:14 |
| StevenK | Like that | 05:14 |
| * StevenK ducks | 05:14 | |
| tchaypo | + pip install -U 'setuptools>=1.0' | 05:14 |
| tchaypo | http://localhost:3141/root/pypi/+simple/setuptools/ uses an insecure transport scheme (http). Consider using https if localhost:3141 has it available | 05:14 |
| tchaypo | Downloading/unpacking setuptools>=1.0 from http://localhost:3141/root/pypi/+f/5b7/b07029ad2285d/setuptools-5.4.1-py2.py3-none-any.whl#md5=5b7b07029ad2285d1cbf809a8ceaea08 | 05:14 |
| tchaypo | Using download cache from /tmp/pip/http%3A%2F%2Flocalhost%3A3141%2Froot%2Fpypi%2F%2Bf%2F5b7%2Fb07029ad2285d%2Fsetuptools-5.4.1-py2.py3-none-any.whl | 05:14 |
| tchaypo | it looks to me like we're actually already using wheels from that repo | 05:14 |
| tchaypo | greghaynes: clarkb: sounds to me like our new reccomendation is that people should install devpi and use the pypi element? | 05:14 |
| StevenK | tchaypo: How did bandersnatch break for you? I'm having no issues with it, aside from the use of ~100GiB of disk | 05:15 |
| StevenK | However, that fileserver has 1.6TiB of disk free, so I can't complain too much | 05:16 |
| tchaypo | i hadn't run it for about a month... the first few runs took hours, and as you found, each run only brings it incrementally closer to up-to=date | 05:16 |
| tchaypo | so i have to run it over and over and over and it takes an indeterminate amou nt of time to catch up and I can never tell how close it is to being up-to-date | 05:16 |
| *** david-lyle has quit IRC | 05:17 | |
| tchaypo | and 100Gi is a lot of disk | 05:17 |
| StevenK | On an SSD laptop, sure | 05:17 |
| tchaypo | on any laptop | 05:17 |
| StevenK | If you have 1TiB spinning rust, you might be able to swing it | 05:18 |
| tchaypo | but why? | 05:18 |
| tchaypo | 98% of the files in there are never going to be used | 05:19 |
| StevenK | tchaypo: Offline builds, but a up-to-date devpi might help too | 05:19 |
| tchaypo | the only time an up-to-date devpi won't help is if there's a new requirement not in cache | 05:20 |
| tchaypo | and even then devpi will get the updated package immediately with minimum bandwidth, bandersnatch would have to download umpteen gigs of unrelated changes to packages I don't care about | 05:20 |
| * tchaypo is thinking about how to work wandering around europe with minimal connectivity | 05:20 | |
| *** shashankhegde has joined #openstack-infra | 05:20 | |
| StevenK | tchaypo: I find my mirror syncs roughly 300 packages every morning, but it doesn't tell me how large they were. | 05:21 |
| StevenK | If bandersnatch also logged size of the objects it was fetching, I could actually work it out | 05:22 |
| r1chardj0n3s | tchaypo: sorry, I was in transit for a bit there. Yes, I'd love to talk devpi/etc at pyconau. I'm even giving a talk about devpi etc :) | 05:22 |
| dstufft | most people aren't going to run bandersnatch on a laptop unless they spend a lot of time with no connectivity and the need to install random crap | 05:26 |
| *** afazekas has joined #openstack-infra | 05:26 | |
| *** harlowja is now known as harlowja_away | 05:26 | |
| lifeless | dstufft: oh hai, did you just name me by usecase ? | 05:27 |
| dstufft | lifeless: quite possibly ;P | 05:27 |
| StevenK | Haha | 05:27 |
| r1chardj0n3s | also there's the thing about devpi caching all the packages including externals, unlike bandersnatch which can't mirror the externals | 05:29 |
| dstufft | death to externals | 05:30 |
| dstufft | :D | 05:30 |
| r1chardj0n3s | good luck with that :D | 05:30 |
| lifeless | deatttt | 05:30 |
| lifeless | h | 05:31 |
| dstufft | I'll figure out a way sooner or later | 05:31 |
| dstufft | I'm incorrigable | 05:31 |
| r1chardj0n3s | hah. you weren't in the conversations I was in at EP :P | 05:31 |
| *** Sukhdev has quit IRC | 05:31 | |
| dstufft | r1chardj0n3s: heh, maybe not, but my speciality is beating my head against a wall until I succeed :D | 05:32 |
| r1chardj0n3s | I suspect the only thing you'll get out of this is a sore head :( | 05:32 |
| lifeless | r1chardj0n3s: what was the conversation | 05:33 |
| *** ildikov has quit IRC | 05:34 | |
| r1chardj0n3s | lifeless: not just one but several. the desire to host packages externally is very strong. | 05:34 |
| dstufft | r1chardj0n3s: the desire by whom | 05:34 |
| dstufft | because not many are hosted externally | 05:34 |
| r1chardj0n3s | on the other hand, witness https://pypi.python.org/pypi/egenix-mx-base/ | 05:34 |
| dstufft | hoyl crap, MAL uploaded? | 05:35 |
| r1chardj0n3s | it's an installer | 05:35 |
| r1chardj0n3s | because he's just so annoyed with the current shit he has to put up with | 05:35 |
| r1chardj0n3s | so he just works around the issue | 05:36 |
| *** bdpayne has quit IRC | 05:36 | |
| *** shashankhegde has quit IRC | 05:36 | |
| dstufft | MAL was on board with PEP 470 he said ;P | 05:38 |
| dstufft | actually, is this new at all r1chardj0n3s? according to this the copyright on this file is 2000 | 05:40 |
| r1chardj0n3s | yes, it's new (at least to pypi) | 05:41 |
| *** lukego has joined #openstack-infra | 05:43 | |
| dstufft | egenix-mx-base has existed for a long time, since 2002 according to journals, is the web installer part new? | 05:45 |
| r1chardj0n3s | yes | 05:45 |
| *** lukego has quit IRC | 05:47 | |
| Alex_Gaynor | ' | 05:50 |
| dstufft | hey | 05:57 |
| dstufft | MALs thing actually securely downloads stuff | 05:57 |
| dstufft | so people who use his thing are secure now when before they wern't | 05:57 |
| dstufft | I consider it a success | 05:57 |
| dstufft | :D | 05:57 |
| r1chardj0n3s | sure, but it can't be cached | 05:58 |
| r1chardj0n3s | it's a suboptimal solution for users | 05:58 |
| r1chardj0n3s | even if you consider it a success | 05:58 |
| dstufft | uncacheable is better than insecure | 05:59 |
| r1chardj0n3s | in your opinion :) | 05:59 |
| *** simonmcc has quit IRC | 06:00 | |
| *** skath has quit IRC | 06:00 | |
| r1chardj0n3s | I'm not saying your opinion is *wrong* just that sometimes users have different priorities ;) | 06:00 |
| dstufft | Well those users are wrong | 06:00 |
| Alex_Gaynor | oh my god, I've never seen so many trove classifiers | 06:00 |
| *** jraim has quit IRC | 06:00 | |
| r1chardj0n3s | Alex_Gaynor: just don't look :) | 06:00 |
| *** zhiyan has quit IRC | 06:00 | |
| Alex_Gaynor | Operating System :: BeOS | 06:00 |
| *** harlowja_at_home has joined #openstack-infra | 06:00 | |
| r1chardj0n3s | \o/ | 06:00 |
| dstufft | I wonder if he tests on BeOS | 06:01 |
| r1chardj0n3s | (it seemed like a good idea at the time, in 2003...) | 06:01 |
| *** ctracey has quit IRC | 06:01 | |
| r1chardj0n3s | hell, *trove* seemed like a good idea in 2003. heh. | 06:02 |
| dstufft | computers seemed like a good idea in 1946 and look at where we are now | 06:03 |
| *** Longgeek has joined #openstack-infra | 06:04 | |
| *** serverascode has quit IRC | 06:04 | |
| *** jraim__ has joined #openstack-infra | 06:05 | |
| *** zhiyan has joined #openstack-infra | 06:05 | |
| *** skath has joined #openstack-infra | 06:06 | |
| *** ctracey has joined #openstack-infra | 06:07 | |
| StevenK | dstufft: I think I may have to put that quote in my .signature | 06:07 |
| *** simonmcc_ has joined #openstack-infra | 06:08 | |
| *** otherwiseguy has quit IRC | 06:09 | |
| openstackgerrit | A change was merged to openstack-infra/infra-manual: Add "Gerrit IRC Notifications" section to infra manual https://review.openstack.org/107322 | 06:09 |
| *** lukego has joined #openstack-infra | 06:10 | |
| *** serverascode has joined #openstack-infra | 06:12 | |
| *** yfried has joined #openstack-infra | 06:19 | |
| *** simonmcc_ has quit IRC | 06:22 | |
| *** skath has quit IRC | 06:22 | |
| *** serverascode has quit IRC | 06:22 | |
| *** pcrews has quit IRC | 06:22 | |
| *** ctracey has quit IRC | 06:22 | |
| *** zhiyan has quit IRC | 06:23 | |
| tchaypo | so the other advantage that I see with devpi over pypi-mirror | 06:24 |
| *** jraim__ has quit IRC | 06:24 | |
| tchaypo | is that even with my recent hack, it seems as though having a file:/// index is a use case that isn't something developers think about in as much detail as the more common case of a http:// index | 06:24 |
| r1chardj0n3s | devpi also has --set-cfg which makes everything a lot easier ;) | 06:25 |
| r1chardj0n3s | (and --always-set-cfg) | 06:25 |
| tchaypo | devpi seems lightweight enough that it feels like we should be able to spin up a devpi server just as easily as the pypi-mirror element | 06:25 |
| tchaypo | and where we previously did lots of munging to discover all the dependencies and make sure they were cached in pypi-mirror's folder - with devpi we can just point at the devpi server, 'pip install -r global-requirements.txt', done. | 06:26 |
| tchaypo | put the wheels in a useful place for later re-use and throw away the venv that triggered the download | 06:27 |
| *** zhiyan has joined #openstack-infra | 06:28 | |
| *** jraim__ has joined #openstack-infra | 06:28 | |
| r1chardj0n3s | yep. if the devpi is a local one (serving a single system) then wheels can be uploaded to a staging index in it | 06:28 |
| *** ctracey has joined #openstack-infra | 06:29 | |
| r1chardj0n3s | supporting multiple system types in a single devpi is something I want to see but needs a teensy little bit of support from wheel/pip which I'm hoping to see soon | 06:29 |
| *** skath has joined #openstack-infra | 06:29 | |
| *** simonmcc_ has joined #openstack-infra | 06:29 | |
| *** serverascode has joined #openstack-infra | 06:30 | |
| *** ildikov has joined #openstack-infra | 06:31 | |
| *** che-arne has quit IRC | 06:33 | |
| *** _nadya_ has joined #openstack-infra | 06:34 | |
| *** chandankumar has quit IRC | 06:35 | |
| *** harlowja_at_home has quit IRC | 06:38 | |
| *** MickeyO has joined #openstack-infra | 06:39 | |
| AJaeger | Alex_Gaynor, could you review https://review.openstack.org/#/c/109905/ , please? | 06:39 |
| *** pblaho has joined #openstack-infra | 06:42 | |
| *** Ryan_Lane1 has joined #openstack-infra | 06:51 | |
| *** Ryan_Lane1 has quit IRC | 06:52 | |
| *** yamamoto_ has quit IRC | 06:57 | |
| *** zns has joined #openstack-infra | 06:57 | |
| *** flaviof is now known as flaviof_zzz | 06:58 | |
| *** wenlock has quit IRC | 07:16 | |
| *** mpaolino has joined #openstack-infra | 07:16 | |
| *** akamyshnikova has quit IRC | 07:17 | |
| *** jcoufal-mobile has joined #openstack-infra | 07:18 | |
| *** cnesa6 has joined #openstack-infra | 07:18 | |
| *** samy__ has joined #openstack-infra | 07:22 | |
| *** jcoufal-mobile has quit IRC | 07:23 | |
| *** flaper87|afk is now known as flaper87 | 07:26 | |
| *** samy has joined #openstack-infra | 07:27 | |
| *** samy has left #openstack-infra | 07:27 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 07:28 | |
| *** __afazekas has quit IRC | 07:29 | |
| *** roeyc has joined #openstack-infra | 07:30 | |
| *** jistr has joined #openstack-infra | 07:30 | |
| *** samy__ has quit IRC | 07:33 | |
| *** luqas has joined #openstack-infra | 07:33 | |
| *** luqas has quit IRC | 07:35 | |
| *** samy_ has joined #openstack-infra | 07:36 | |
| *** samy__ has joined #openstack-infra | 07:36 | |
| *** samy__ has left #openstack-infra | 07:38 | |
| *** luqas has joined #openstack-infra | 07:39 | |
| *** jgallard has joined #openstack-infra | 07:39 | |
| *** stevemar has quit IRC | 07:46 | |
| *** ihrachyshka has joined #openstack-infra | 07:46 | |
| *** jp_at_hp has joined #openstack-infra | 07:46 | |
| *** tnovacik has joined #openstack-infra | 07:47 | |
| *** ihrachyshka has quit IRC | 07:50 | |
| *** gangil has quit IRC | 07:50 | |
| *** ihrachyshka has joined #openstack-infra | 07:51 | |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-infra/gerritlib: Python3 support https://review.openstack.org/110226 | 07:53 |
| openstackgerrit | Chmouel Boudjnah proposed a change to openstack-infra/gerritlib: Python3 support https://review.openstack.org/110226 | 07:55 |
| *** samy_ has quit IRC | 07:55 | |
| *** samy_ has joined #openstack-infra | 07:56 | |
| *** jpich has joined #openstack-infra | 07:56 | |
| *** samy__ has joined #openstack-infra | 07:56 | |
| *** jistr_ has joined #openstack-infra | 07:58 | |
| *** jistr_ has quit IRC | 07:59 | |
| *** samy_ has left #openstack-infra | 08:00 | |
| *** cnesa6 has quit IRC | 08:01 | |
| *** samy__ has quit IRC | 08:07 | |
| *** andreykurilin has joined #openstack-infra | 08:09 | |
| *** jlibosva has joined #openstack-infra | 08:18 | |
| openstackgerrit | Christian Berendt proposed a change to openstack-infra/config: Niceness checks for docs should vote https://review.openstack.org/109559 | 08:19 |
| *** e0ne has joined #openstack-infra | 08:20 | |
| *** praneshp has quit IRC | 08:22 | |
| *** jamielennox is now known as jamielennox|away | 08:23 | |
| jp_at_hp | So can anyone here help me with oslo.rootwrap? Do the infra team release that to pypi? | 08:24 |
| lifeless | we can check the zuul config to see | 08:25 |
| lifeless | jp_at_hp: see modules/openstack_project/files/zuul/layout.yaml in openstack-infra/config | 08:26 |
| lifeless | jp_at_hp: - name: openstack/oslo.rootwrap | 08:26 |
| lifeless | ... | 08:26 |
| lifeless | - name: publish-to-pypi | 08:26 |
| lifeless | so when a tag is pushed to oslo.rootwrap, that will trigger a release to pypi | 08:26 |
| jp_at_hp | Sure - it was more to ask why pypi.org has 1.3.0.0a1 or whatever and pypi.openstack.org hasn't mirrored that version... | 08:26 |
| lifeless | oh | 08:27 |
| *** e0ne has quit IRC | 08:27 | |
| lifeless | so I suspect thats fallout from the great 'omg alphas don't work with pip well' experiment from a couple weeks back | 08:27 |
| jp_at_hp | I was wondering if the pypi mirroring did not mirror alpha versions for some reason. Because global-requirements specifies that version constraint. | 08:27 |
| *** e0ne has joined #openstack-infra | 08:28 | |
| jp_at_hp | ah - just checked the update merge date, it was last night at 8pm, so probably waiting for a mirror run to pull in... | 08:29 |
| jp_at_hp | ta lifeless | 08:30 |
| *** andreaf has joined #openstack-infra | 08:32 | |
| *** hashar has joined #openstack-infra | 08:35 | |
| *** e0ne has quit IRC | 08:40 | |
| *** cnesa6 has joined #openstack-infra | 08:42 | |
| *** leopoldj has joined #openstack-infra | 08:42 | |
| *** hashar_ has joined #openstack-infra | 08:49 | |
| *** Alexei_987 has joined #openstack-infra | 08:50 | |
| *** ZZelle has quit IRC | 08:51 | |
| *** ZZelle has joined #openstack-infra | 08:51 | |
| *** andreykurilin has quit IRC | 08:53 | |
| *** hashar has quit IRC | 08:54 | |
| *** cnesa7 has joined #openstack-infra | 08:56 | |
| *** cnesa6 has quit IRC | 08:56 | |
| *** hashar_ is now known as hashar | 09:00 | |
| openstackgerrit | Dina Belova proposed a change to openstack-infra/config: New python-opentsdbclient repo https://review.openstack.org/108630 | 09:03 |
| *** luqas has quit IRC | 09:03 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Merger: support for pruning remotes https://review.openstack.org/92141 | 09:03 |
| *** ociuhandu has joined #openstack-infra | 09:03 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Merger: support for finding branches in remote https://review.openstack.org/92161 | 09:03 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Add accessor for Repo._initialized https://review.openstack.org/97487 | 09:03 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: cloner to easily clone dependent repositories https://review.openstack.org/70373 | 09:03 |
| hashar | ^^^ lame manual rebase | 09:04 |
| *** cnesa7 has quit IRC | 09:08 | |
| *** e0ne has joined #openstack-infra | 09:12 | |
| *** yamamoto_ has joined #openstack-infra | 09:13 | |
| *** yamamoto_ has quit IRC | 09:15 | |
| *** zehicle_at_dell has quit IRC | 09:21 | |
| *** amalagon has joined #openstack-infra | 09:21 | |
| *** e0ne has quit IRC | 09:24 | |
| openstackgerrit | Dina Belova proposed a change to openstack-infra/config: Add job to run Tempest dsvm on MongoDB backend for Ceilometer https://review.openstack.org/110245 | 09:30 |
| *** e0ne has joined #openstack-infra | 09:32 | |
| *** derekh_ has joined #openstack-infra | 09:33 | |
| openstackgerrit | Dina Belova proposed a change to openstack-infra/devstack-gate: Enable overriding of Ceilometer storage backend https://review.openstack.org/110247 | 09:35 |
| *** rlandy has joined #openstack-infra | 09:37 | |
| *** Alexei_987 has quit IRC | 09:37 | |
| openstackgerrit | Dina Belova proposed a change to openstack-infra/config: Add job to run Tempest dsvm on MongoDB backend for Ceilometer https://review.openstack.org/110245 | 09:38 |
| *** unicell has quit IRC | 09:39 | |
| *** daya_k has quit IRC | 09:39 | |
| *** luqas has joined #openstack-infra | 09:40 | |
| *** unicell has joined #openstack-infra | 09:42 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 09:43 |
| *** enikanorov_ has joined #openstack-infra | 09:45 | |
| *** enikanorov has quit IRC | 09:46 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 09:50 |
| *** Slower has quit IRC | 10:00 | |
| *** dannyla has joined #openstack-infra | 10:07 | |
| *** jcoufal-mobile has joined #openstack-infra | 10:07 | |
| *** bauzas has joined #openstack-infra | 10:08 | |
| *** jcoufal-mobile has quit IRC | 10:08 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Enable py33 gating on Ceilometer https://review.openstack.org/110252 | 10:11 |
| *** banix has joined #openstack-infra | 10:11 | |
| *** jgallard has quit IRC | 10:15 | |
| *** ryanpetrello has quit IRC | 10:16 | |
| *** alexpilotti has joined #openstack-infra | 10:19 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Use python3-jobs template where possible https://review.openstack.org/110256 | 10:21 |
| *** ryanpetrello has joined #openstack-infra | 10:21 | |
| *** marzif has joined #openstack-infra | 10:22 | |
| *** DuncanT- has joined #openstack-infra | 10:32 | |
| SergeyLukjanov | redrobot, I'm here atm | 10:34 |
| *** xianghui has quit IRC | 10:36 | |
| AJaeger | SergeyLukjanov, is there a special reason you gave your +2 but did not approve the *-specs publishing jobs? https://review.openstack.org/#/c/106100 | 10:36 |
| *** pcm_ has joined #openstack-infra | 10:36 | |
| *** jyuso has quit IRC | 10:36 | |
| *** yaguang has quit IRC | 10:37 | |
| SergeyLukjanov | AJaeger, hey | 10:37 |
| SergeyLukjanov | AJaeger, there is no specs.openstack.org DNS entry | 10:37 |
| *** pcm__ has joined #openstack-infra | 10:37 | |
| AJaeger | SergeyLukjanov, understood. Is there anything that I can do there - or who can? | 10:38 |
| SergeyLukjanov | AJaeger, I think that I should be able to do it, but I have no credentials for managing o.o dns | 10:39 |
| SergeyLukjanov | :) | 10:39 |
| *** DuncanT- has left #openstack-infra | 10:39 | |
| AJaeger | I see, let me add a note to the patch that DNS needs to be created. | 10:39 |
| AJaeger | and I'll ask around later who can do it | 10:39 |
| SergeyLukjanov | AJaeger, oh, and I just see one more concern | 10:40 |
| SergeyLukjanov | AJaeger, probably it was already discussed | 10:40 |
| AJaeger | speak up ;) | 10:40 |
| SergeyLukjanov | AJaeger, your patch will publish specs to specs.o.o for all projects | 10:40 |
| AJaeger | SergeyLukjanov, yes | 10:40 |
| SergeyLukjanov | AJaeger, including the non-official (stackforge) ones | 10:40 |
| AJaeger | in directories {github-org}/project/ | 10:40 |
| SergeyLukjanov | AJaeger, but we're not doing it with docs.o.o | 10:41 |
| *** pcm_ has quit IRC | 10:41 | |
| SergeyLukjanov | AJaeger, was this question already rised? | 10:41 |
| AJaeger | anteaya raised this in the review and fungi and stevemar discused here as well - see fungi's comment | 10:41 |
| *** banix has quit IRC | 10:41 | |
| AJaeger | We can create a top-level page that differentiates these projects... | 10:41 |
| *** cnesa7 has joined #openstack-infra | 10:42 | |
| AJaeger | right now fuel-specs and tacker-specs are part of that series. | 10:43 |
| AJaeger | If you think we should remove them from there for now, I can change the patch. | 10:43 |
| SergeyLukjanov | AJaeger, I'm not against it, just remember that we're publishing to o.o only official projects | 10:44 |
| SergeyLukjanov | (at least for docs.o.o) | 10:45 |
| AJaeger | Let me double check with fungi et all later today... | 10:45 |
| AJaeger | SergeyLukjanov, I added a review comment. | 10:47 |
| AJaeger | SergeyLukjanov, one more favor: Could you put https://review.openstack.org/#/c/108464/ in your review queue, please? | 10:47 |
| SergeyLukjanov | AJaeger, looking | 10:50 |
| AJaeger | thanks, SergeyLukjanov ! | 10:52 |
| *** HenryG_afk is now known as HenryG | 10:53 | |
| rcarrillocruz | SergeyLukjanov: hi Sergey, if you are not too busy at this moment, i'd be grateful if i could get a core review on https://review.openstack.org/#/c/109816/ | 10:56 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/nodepool: Build images using diskimage-builder https://review.openstack.org/46482 | 10:56 |
| SergeyLukjanov | rcarrillocruz, it's a "bit" larger than AJaeger's one, so, I've added it to my review list | 10:57 |
| rcarrillocruz | sure, thx | 10:57 |
| AJaeger | yeah, a tiny bit larger ;) | 10:57 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Setup a private gerrit instance for security reviews https://review.openstack.org/47937 | 10:59 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix upstream_translation_update_manuals https://review.openstack.org/108464 | 11:01 |
| zaro | rcarrillocruz: did you need something? | 11:04 |
| *** asettle has quit IRC | 11:05 | |
| *** pblaho has quit IRC | 11:06 | |
| *** alexandra_ has joined #openstack-infra | 11:06 | |
| *** denis_makogon has quit IRC | 11:06 | |
| *** salv-orlando has quit IRC | 11:06 | |
| *** banix has joined #openstack-infra | 11:08 | |
| openstackgerrit | A change was merged to openstack-infra/config: removed py3 tests from refstack-client https://review.openstack.org/109443 | 11:12 |
| *** cdub has joined #openstack-infra | 11:13 | |
| *** dmellado has joined #openstack-infra | 11:13 | |
| *** lttrl has joined #openstack-infra | 11:14 | |
| *** pcm__ has quit IRC | 11:14 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Publish tooz documentation and gate on docs https://review.openstack.org/110263 | 11:14 |
| yolanda | SergeyLukjanov, thx for reviewing | 11:15 |
| *** pcm_ has joined #openstack-infra | 11:15 | |
| openstackgerrit | A change was merged to openstack-infra/config: convert to two level recheckwatchbot.yaml https://review.openstack.org/108719 | 11:15 |
| *** zns has quit IRC | 11:16 | |
| cnesa7 | Hi all, we have a blocking bug on the HyperV CI, devstack fails during neutron database creation. stack trace here -> http://paste.openstack.org/show/88896/. Is there a workaround? | 11:17 |
| *** salv-orlando has joined #openstack-infra | 11:19 | |
| *** krtaylor has quit IRC | 11:23 | |
| rcarrillocruz | zaro: hey | 11:23 |
| *** _nadya_ has quit IRC | 11:23 | |
| rcarrillocruz | i've refactored the artifact upload scripts | 11:23 |
| rcarrillocruz | instead of having 'jenkinsci-upload.sh, mavencentral-upload.sh, pypi-*-upload.sh', i created a generic artifact-upload.sh script | 11:24 |
| rcarrillocruz | curl --data-binary does not work for uploading stuff to nexus | 11:24 |
| rcarrillocruz | i could only make it work with 'curl -T' | 11:24 |
| rcarrillocruz | i pinged you since i saw you created mavencentral-upload.sh file , from git history... | 11:25 |
| rcarrillocruz | did 'curl -X PUT --data-binary' ever work for Maven Central? | 11:27 |
| rcarrillocruz | i'd be surprised, since Maven Central is also based on Nexus afaik... | 11:27 |
| openstackgerrit | Nicholas Randon proposed a change to openstack-infra/config: Add firewall traffic on port 27410 to 27510. https://review.openstack.org/108737 | 11:27 |
| *** banix has quit IRC | 11:28 | |
| *** denis_makogon has joined #openstack-infra | 11:28 | |
| *** salv-orlando has quit IRC | 11:28 | |
| zaro | rcarrillocruz: sounds good. i don't remember if it worked or not. scripted isn't actually used anymore. the maintainer of our only java project that needs to be uploaded to maven central decided to use the maven deploy plugin instead. | 11:30 |
| zaro | rcarrillocruz: i do remember testing it and it did work for me. | 11:30 |
| rcarrillocruz | yep, didn't find any real job using that | 11:30 |
| rcarrillocruz | i think the issue is with curl | 11:31 |
| zaro | rcarrillocruz: have you tried testing the mavencentral-upload.sh script as-is to see if you can upload anything? | 11:31 |
| rcarrillocruz | and i'm doing tests with curl on precise (where it will be run) | 11:31 |
| rcarrillocruz | doesn't work | 11:31 |
| rcarrillocruz | i mean, i can't upload to Central, as i don't have credentials for it | 11:31 |
| rcarrillocruz | but i tried using that script to upload to a local nexus | 11:31 |
| rcarrillocruz | and it didn't work, i get a 500 on Nexus side | 11:32 |
| *** GheRiver1 is now known as GheRivero | 11:32 | |
| openstackgerrit | Julien Danjou proposed a change to openstack-infra/config: Publish tooz documentation and gate on docs https://review.openstack.org/110263 | 11:32 |
| *** Longgeek has quit IRC | 11:32 | |
| *** Longgeek has joined #openstack-infra | 11:32 | |
| rcarrillocruz | tracing curl output there's some encoding issues when using --data-binary | 11:32 |
| rcarrillocruz | again, using curl on precise version, i found also a blog post in the Internet pointing in that direction, let me dig | 11:32 |
| zaro | well i know the jenkins-upload.sh works, but i think that's going to an artifactory server. | 11:32 |
| rcarrillocruz | http://blog.killtheradio.net/tricks-hacks/curl-cli-not-sending-full-file-data-when-using-data-binary/ | 11:33 |
| rcarrillocruz | zaro: the change is https://review.openstack.org/#/c/109816/ if you want to have a look | 11:34 |
| rcarrillocruz | note i haven't removed prior upload artifact , just pushing the script itself | 11:34 |
| zaro | rcarrillocruz: have you tried curl outside of the script. just a simple curl command by itself? | 11:35 |
| rcarrillocruz | when it lands and if all is ok, we can then remove other artifact upload scripts and change jjb jobs to use the new script | 11:35 |
| rcarrillocruz | zaro: yes | 11:35 |
| rcarrillocruz | i even have a trace output from a working curl with -T vs a non-working curl with --data-binary | 11:35 |
| *** pblaho has joined #openstack-infra | 11:36 | |
| zaro | rcarrillocruz: that definately worked for me to nexus | 11:36 |
| zaro | rcarrillocruz: i believe you can create your own account on nexus | 11:37 |
| zaro | rcarrillocruz: i mean on maven central | 11:37 |
| rcarrillocruz | how long it takes? i read that you need to open a jira ticket with them | 11:37 |
| *** luqas has quit IRC | 11:38 | |
| rcarrillocruz | would you be able to do a quick 'curl -T foo.jar' ? | 11:38 |
| zaro | rcarrillocruz: anyways, why change a broken script with another one that doens't work? shouldn't we fix it first? | 11:38 |
| rcarrillocruz | mine works, i've just refactored the existing upload artifact scripts into one for a better code reuse, and switched to 'curl -T' , instead of 'curl --data-binary' | 11:39 |
| rcarrillocruz | and also allow parameterization | 11:40 |
| zaro | rcarrillocruz: so what doesn't work again? | 11:40 |
| rcarrillocruz | the 'curl --data-binary' to nexus | 11:40 |
| rcarrillocruz | the parameterization allows to use arbitrary artifact URLs | 11:41 |
| zaro | rcarrillocruz: ohh so 'curl -T' to nexus works? | 11:41 |
| rcarrillocruz | yes, it does | 11:41 |
| rcarrillocruz | curl -T == curl --upload-file btw | 11:41 |
| rcarrillocruz | and it works for both binaries and text files | 11:41 |
| zaro | rcarrillocruz: ohh, cool. then we should just go with that. | 11:41 |
| rcarrillocruz | the paramterization is also good, cos with the new script you can just pass a URL to a local Nexus . If you don't pass any URL for a 'java-package' upload, it defaults to Maven Central | 11:42 |
| zaro | rcarrillocruz: yes, you are right about the asking for nexus account. it's a little involved and IIRC it took about a week to setup an account. | 11:42 |
| rcarrillocruz | i.e. i wrote the script to be as close to what we have right now, yet being more flexible for downstream consumption | 11:43 |
| *** salv-orlando has joined #openstack-infra | 11:43 | |
| zaro | i'll take a look at your change. nice work! | 11:47 |
| rcarrillocruz | appreciate, thx :-) | 11:47 |
| *** BobBall has quit IRC | 11:48 | |
| *** salv-orlando_ has joined #openstack-infra | 11:52 | |
| *** salv-orlando has quit IRC | 11:52 | |
| *** salv-orlando_ is now known as salv-orlando | 11:52 | |
| sweston | krtaylor, anteaya third party terms are here https://review.openstack.org/#/c/109800/ | 11:53 |
| *** _nadya_ has joined #openstack-infra | 11:54 | |
| *** BobBall has joined #openstack-infra | 11:54 | |
| *** yfried has quit IRC | 11:55 | |
| *** Longgeek has quit IRC | 11:55 | |
| *** yfried_ has joined #openstack-infra | 11:55 | |
| *** Longgeek has joined #openstack-infra | 11:55 | |
| *** weshay has joined #openstack-infra | 11:56 | |
| *** zns has joined #openstack-infra | 12:02 | |
| *** zns has joined #openstack-infra | 12:03 | |
| openstackgerrit | Davide Guerri proposed a change to openstack-infra/config: Add git-upstream as new project to stackforge https://review.openstack.org/109292 | 12:04 |
| *** tnovacik has quit IRC | 12:05 | |
| *** hashar has quit IRC | 12:07 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add rally-dsvm jobs to gate pipeline in rally https://review.openstack.org/107259 | 12:07 |
| openstackgerrit | A change was merged to openstack-infra/config: Add pep8 check/gate for stackforge/python-tackerclient https://review.openstack.org/107410 | 12:08 |
| openstackgerrit | A change was merged to openstack-infra/nodepool: Use correct provider in test-case https://review.openstack.org/109184 | 12:08 |
| dguerri`afk | SergeyLukjanov, can I edit http://ci.openstack.org/stackforge.html to add the additional configuration needed for configuring the IRC bot? | 12:09 |
| *** dguerri`afk is now known as dguerri | 12:10 | |
| *** ArxCruz has joined #openstack-infra | 12:12 | |
| *** dkranz has joined #openstack-infra | 12:15 | |
| SergeyLukjanov | dguerri, the sources of this page located in https://github.com/openstack-infra/config/blob/master/doc/source/stackforge.rst | 12:16 |
| dguerri | ta | 12:16 |
| *** miqui has joined #openstack-infra | 12:18 | |
| dguerri | SergeyLukjanov if I understand it well, adding a channel to channels.yaml will result in the definition of new OPs for that channel and will define the mask for IRC users not listed in that file. Is that accurate? | 12:18 |
| *** cdub has quit IRC | 12:18 | |
| *** tnovacik has joined #openstack-infra | 12:20 | |
| *** pdmars has joined #openstack-infra | 12:22 | |
| *** jgallard has joined #openstack-infra | 12:27 | |
| *** baohua has quit IRC | 12:27 | |
| *** tnovacik has quit IRC | 12:28 | |
| *** jlibosva has quit IRC | 12:28 | |
| *** jistr has quit IRC | 12:29 | |
| *** ihrachyshka_ has joined #openstack-infra | 12:30 | |
| *** aysyd has joined #openstack-infra | 12:30 | |
| rcarrillocruz | hmm | 12:31 |
| rcarrillocruz | zaro: re: my comment on the script change | 12:31 |
| rcarrillocruz | would you say that ALL pypi jobs that upload either tarballs or wheels first download the artifact from tarballs.openstack.org ? | 12:31 |
| rcarrillocruz | i.e. is it always tarballs.openstack.org? | 12:31 |
| *** ihrachyshka_ has quit IRC | 12:31 | |
| rcarrillocruz | cos if it does, i would just put that as the default argument | 12:31 |
| *** ihrachyshka_ has joined #openstack-infra | 12:32 | |
| *** ihrachyshka has quit IRC | 12:32 | |
| *** jistr has joined #openstack-infra | 12:33 | |
| rcarrillocruz | if I grep for 'tarball-site' in jjb projects.yaml, i only get 'tarballs.openstack.org' value | 12:33 |
| openstackgerrit | Ekaterina Chernova proposed a change to openstack-infra/config: Update murano jenkins job https://review.openstack.org/110277 | 12:33 |
| rcarrillocruz | it's a quasi hardcode value ! :-) | 12:33 |
| *** jlibosva has joined #openstack-infra | 12:34 | |
| *** maxbit has joined #openstack-infra | 12:35 | |
| *** adalbas has joined #openstack-infra | 12:41 | |
| *** che-arne has joined #openstack-infra | 12:42 | |
| *** tnovacik has joined #openstack-infra | 12:43 | |
| openstackgerrit | Nicholas Randon proposed a change to openstack-infra/config: Add firewall traffic on port 27410 to 27510. https://review.openstack.org/108737 | 12:43 |
| *** amalagon has quit IRC | 12:47 | |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for grenade ssh protocol banner bug 1349617 https://review.openstack.org/110164 | 12:50 |
| uvirtbot | Launchpad bug 1349617 in nova "test_volume_boot_pattern fails in grenade with "SSHException: Error reading SSH protocol banner[Errno 104] Connection reset by peer"" [Undecided,New] https://launchpad.net/bugs/1349617 | 12:50 |
| *** nosnos has quit IRC | 12:51 | |
| *** mrmartin has joined #openstack-infra | 12:51 | |
| *** cdub has joined #openstack-infra | 12:53 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for nova-network floating IPs bug 1349590 https://review.openstack.org/110142 | 12:53 |
| uvirtbot | Launchpad bug 1349590 in nova "nova-network appears to lock up while associating floating ip's" [High,Confirmed] https://launchpad.net/bugs/1349590 | 12:53 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for grenade ssh protocol banner bug 1349617 https://review.openstack.org/110164 | 12:54 |
| uvirtbot | Launchpad bug 1349617 in nova "test_volume_boot_pattern fails in grenade with "SSHException: Error reading SSH protocol banner[Errno 104] Connection reset by peer"" [Undecided,New] https://launchpad.net/bugs/1349617 | 12:54 |
| *** sballe has joined #openstack-infra | 12:55 | |
| openstackgerrit | Davide Guerri proposed a change to openstack-infra/config: Complete the instruction to add a new IRC channel https://review.openstack.org/110285 | 12:56 |
| *** ihrachyshka__ has joined #openstack-infra | 12:56 | |
| *** mfink has quit IRC | 12:57 | |
| *** ihrachyshka_ has quit IRC | 12:59 | |
| *** jlibosva has quit IRC | 13:00 | |
| *** jlibosva has joined #openstack-infra | 13:00 | |
| *** jistr has quit IRC | 13:01 | |
| openstackgerrit | Davide Guerri proposed a change to openstack-infra/config: Complete the instruction to add a new IRC channel https://review.openstack.org/110285 | 13:01 |
| *** jgrimm has joined #openstack-infra | 13:02 | |
| *** jistr has joined #openstack-infra | 13:03 | |
| *** pblaho_ has joined #openstack-infra | 13:03 | |
| *** pblaho has quit IRC | 13:03 | |
| *** luqas has joined #openstack-infra | 13:04 | |
| *** nelsnelson has joined #openstack-infra | 13:05 | |
| *** dmsimard_away is now known as dmsimard | 13:06 | |
| *** dustins has joined #openstack-infra | 13:07 | |
| *** zns has quit IRC | 13:08 | |
| *** amalagon has joined #openstack-infra | 13:09 | |
| *** craigbr has joined #openstack-infra | 13:11 | |
| *** miqui has quit IRC | 13:14 | |
| *** krtaylor has joined #openstack-infra | 13:14 | |
| *** daya_k has joined #openstack-infra | 13:18 | |
| *** samy has joined #openstack-infra | 13:20 | |
| *** samy_ has joined #openstack-infra | 13:20 | |
| *** samy is now known as Guest90717 | 13:20 | |
| *** zns has joined #openstack-infra | 13:20 | |
| *** zehicle_at_dell has joined #openstack-infra | 13:21 | |
| *** dkranz has quit IRC | 13:23 | |
| *** dguerri is now known as dguerri`afk | 13:25 | |
| openstackgerrit | gordon chung proposed a change to openstack-infra/config: Create git repo and jenkins jobs for oslo.middleware https://review.openstack.org/109349 | 13:27 |
| *** mfink_ has joined #openstack-infra | 13:27 | |
| *** stevemar has joined #openstack-infra | 13:28 | |
| *** dguerri`afk is now known as dguerri | 13:29 | |
| *** masayuki_ has quit IRC | 13:30 | |
| *** mfink__ has joined #openstack-infra | 13:30 | |
| *** mfink_ has quit IRC | 13:30 | |
| *** dannyla has quit IRC | 13:30 | |
| *** mfink_ has joined #openstack-infra | 13:30 | |
| *** signed8bit has joined #openstack-infra | 13:31 | |
| *** dims has joined #openstack-infra | 13:31 | |
| *** pdmars_ has joined #openstack-infra | 13:34 | |
| *** baohua has joined #openstack-infra | 13:35 | |
| *** juzuluag has joined #openstack-infra | 13:35 | |
| *** pdmars has quit IRC | 13:35 | |
| *** dkranz has joined #openstack-infra | 13:35 | |
| phschwartz | so I just had an interesting thing happen. I left a browser window open with my submission for a talk in Paris, and safari auto refreshed the page and I just got an email saying thank you for submitting it. | 13:39 |
| phschwartz | Should have been closed since yesterday so not sure why it just allowed that on the refresh | 13:40 |
| *** bknudson has joined #openstack-infra | 13:41 | |
| *** armax has joined #openstack-infra | 13:42 | |
| *** hashar has joined #openstack-infra | 13:43 | |
| fungi | phschwartz: open a lp bug against openstack-org | 13:44 |
| fungi | phschwartz: todd will probably want to know about it | 13:45 |
| phschwartz | fungi: Will do that, just hope my talk that I submitted yesterday doesn't get disqual'ed for them thinking it was resubmitted or changed today | 13:45 |
| fungi | phschwartz: right, also a good reason to reach out to them ;) | 13:45 |
| SergeyLukjanov | fungi, I've added specs.o.o CNAME for statis.o.o for future specs publishing feature | 13:45 |
| SergeyLukjanov | AJaeger, ^^ | 13:45 |
| fungi | SergeyLukjanov: awesome, thanks! | 13:46 |
| *** daya_k has quit IRC | 13:46 | |
| phschwartz | fungi: I will open the bug, what is Todd's nick in irc? | 13:46 |
| fungi | phschwartz: when he's around, he's toddmorey | 13:46 |
| fungi | phschwartz: he's also todd@openstack.org | 13:46 |
| SergeyLukjanov | fungi, the last concern I have is that for docs.o.o we're only publishing openstack/* projects, but for for specs we're trying to publish any projects including stackforge/* | 13:47 |
| SergeyLukjanov | jeblair, mordred, clarkb, AJaeger ^^ | 13:47 |
| fungi | SergeyLukjanov: right, the specs.o.o site is not under the management of the docs team. i'm kindof indifferent as to whether stackforge projects get their specs there as long as they're clearly namespaced and don't use openstack documentation theming/branding/logos | 13:48 |
| *** lnxnut has joined #openstack-infra | 13:48 | |
| *** dmellado has quit IRC | 13:48 | |
| fungi | SergeyLukjanov: for example, we publish the same document sources on git.openstack.org | 13:48 |
| SergeyLukjanov | fungi, yeah, that's fair | 13:49 |
| *** UtahDave has joined #openstack-infra | 13:49 | |
| *** UtahDave has left #openstack-infra | 13:49 | |
| fungi | SergeyLukjanov: that said, others very well may disagree, so it bears some discussion with the group | 13:49 |
| *** samy_ has left #openstack-infra | 13:50 | |
| SergeyLukjanov | fungi, do you think we should wait for some time before approving it? | 13:50 |
| *** yfried_ has quit IRC | 13:50 | |
| *** homeless has joined #openstack-infra | 13:51 | |
| fungi | SergeyLukjanov: we probably want to make sure any contentious/political concerns over stackforge spec publication are addressed first, so more review time may help with that | 13:51 |
| SergeyLukjanov | fungi, ack | 13:51 |
| *** pcrews has joined #openstack-infra | 13:52 | |
| fungi | SergeyLukjanov: for example, are the docs jobs for the stackforge specs generating content with openstack logos? i haven't checked, but i probably should and may readjust my vote accordingly | 13:52 |
| SergeyLukjanov | fungi, they do if use oslosphinx | 13:52 |
| fungi | SergeyLukjanov: right, i think at one point we discussed with dhellmann the feasibility of having different branding payloads for stackforge documents, but i don't think any of the 'forgers ever stepped up to work out the details | 13:53 |
| *** zz_dimtruck is now known as dimtruck | 13:54 | |
| *** prad has joined #openstack-infra | 13:56 | |
| fungi | the counterargument is that we have openstack branding/logo on http://git.openstack.org/cgit/stackforge/gertty/ | 13:56 |
| *** flaviof_zzz is now known as flaviof | 13:56 | |
| fungi | but we may be drawing a (fuzzy) line between branding a service provided by the project infrastructure and branding content/documents themselves | 13:57 |
| *** otherwiseguy has joined #openstack-infra | 14:01 | |
| *** laxity_ has quit IRC | 14:02 | |
| fungi | there are a few stackforge/.*-specs projects with no requirements specified at all (some are still empty projects), but the following currently install oslosphinx: fuel-specs, openstack-chef-specs, solum-specs, tacker-specs | 14:02 |
| *** laxity_ has joined #openstack-infra | 14:03 | |
| fungi | according to http://git.openstack.org/cgit/openstack/oslosphinx/tree/README.rst there's a special "incubating" theme available already, so maybe one of those projects wants to step up and make an "ecosystem" theme or similar | 14:03 |
| openstackgerrit | Ricardo Carrillo Cruz proposed a change to openstack-infra/config: Refactor the various artifact upload scripts into one https://review.openstack.org/109816 | 14:05 |
| *** samy_ has joined #openstack-infra | 14:05 | |
| *** eharney has quit IRC | 14:06 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/jeepyb: Allow create/update projects with description into Gerrit https://review.openstack.org/110009 | 14:07 |
| *** unicell has quit IRC | 14:08 | |
| *** e0ne has quit IRC | 14:09 | |
| *** unicell has joined #openstack-infra | 14:09 | |
| *** unicell has quit IRC | 14:09 | |
| *** unicell has joined #openstack-infra | 14:09 | |
| *** Guest90717 has quit IRC | 14:09 | |
| *** pblaho_ has quit IRC | 14:09 | |
| *** craigbr has quit IRC | 14:09 | |
| fungi | phschwartz: what's the current best path for us to get deleting+error state instances terminated? we've got 63 in dfw, 26 in iad and 13 in ord | 14:10 |
| rcarrillocruz | would be nice to have a link on status.openstack.org/zuul changes that point you to the pertinent jenkins build console outputt | 14:10 |
| *** chandankumar has joined #openstack-infra | 14:11 | |
| fungi | rcarrillocruz: what do you mean? doesn't it already? | 14:11 |
| fungi | rcarrillocruz: something other than what you get by clicking the names of the running jobs? | 14:11 |
| phschwartz | fungi: Until we get our next release out (behind because fo the ironic release) it still has to be done via the admin interface or db to reset the task_state. | 14:11 |
| phschwartz | fungi: Our next release has the upstream fix that autocorrects task_state on the failure in it. | 14:12 |
| fungi | phschwartz: so just open a support case and attach the per-region lists of uuids? | 14:12 |
| *** e0ne has joined #openstack-infra | 14:12 | |
| fungi | if so, easy enough | 14:12 |
| phschwartz | fungi: I am pulling up the error'ed instances now and will get our csops looking at it :) should be faster then a ticket | 14:12 |
| *** jistr has quit IRC | 14:12 | |
| fungi | phschwartz: oh, okay--thanks bunches! | 14:13 |
| marzif | SergeyLukjanov, many thanks, 100% effective (thumbsup) | 14:13 |
| *** jistr has joined #openstack-infra | 14:13 | |
| rcarrillocruz | oh, as hovering over them didn't change the pointer never cared to click | 14:14 |
| rcarrillocruz | nice! | 14:14 |
| *** izinovik has left #openstack-infra | 14:14 | |
| yolanda | hi, i have a set of changes to allow setting project descriptions on gerrit: https://review.openstack.org/109957 and https://review.openstack.org/110009 | 14:14 |
| yolanda | can i get some reviews? | 14:14 |
| rcarrillocruz | fungi: would you say that all pypi upload jobs grab the artifacts from tarballs.openstack.org ? | 14:15 |
| rcarrillocruz | the pypi-{tarball, upload}-upload.sh scripts accept a URL argument | 14:15 |
| rcarrillocruz | but it seems that in the end we use tarballs site for everything, unless i'm missing something... | 14:15 |
| rcarrillocruz | i'll check yolanda | 14:16 |
| openstackgerrit | Ivan Kolodyazhny proposed a change to openstack-infra/config: Make 'jenkins-job-builder' repo configurable https://review.openstack.org/99190 | 14:16 |
| jeblair | fungi, SergeyLukjanov: i was thinking we would only publish openstack specs to specs.openstack.org | 14:16 |
| *** dhellmann has quit IRC | 14:17 | |
| *** chuckC has joined #openstack-infra | 14:18 | |
| fungi | jeblair: okay, fair enough... i'm indifferent on that | 14:18 |
| *** krtaylor has quit IRC | 14:18 | |
| *** niedbalski has quit IRC | 14:18 | |
| *** niedbalski has joined #openstack-infra | 14:18 | |
| yolanda | thx | 14:19 |
| rcarrillocruz | yolanda: the description.replace is to get rid of the "" you pass in the command line, right? | 14:19 |
| *** krtaylor has joined #openstack-infra | 14:19 | |
| fungi | rcarrillocruz: yes. to my knowledge we only parameterized that to make it easier to adjust on the backend when needed | 14:19 |
| fungi | rcarrillocruz: (in reference to your tarballs.o.o question) | 14:19 |
| yolanda | rcarrillocruz, to escape slashes if you pass them as a part of the description | 14:20 |
| rcarrillocruz | fungi: ok, just found odd that the URL is parameterized for pypi upload jobs, but it was not for maven/jenkins upload (they point always to upstream maven central/ jenkinsci respectively) | 14:20 |
| *** dhellmann has joined #openstack-infra | 14:20 | |
| openstackgerrit | afazekas proposed a change to openstack-infra/devstack-gate: Neutron default tenant network type vxlan https://review.openstack.org/105569 | 14:21 |
| rcarrillocruz | fungi: i refactored the upload artifacts scripts here: https://review.openstack.org/#/c/109816/7 , we need to parameterize the maven central location so arbitrary nexus servers (maven central is a nexus after all) can be used | 14:21 |
| openstackgerrit | afazekas proposed a change to openstack-infra/devstack-gate: multi-node dg setup https://review.openstack.org/106043 | 14:21 |
| yolanda | rcarrillocruz, to escape quotes, i mean | 14:22 |
| *** ildikov has quit IRC | 14:22 | |
| *** _nadya_ has quit IRC | 14:23 | |
| *** _nadya_ has joined #openstack-infra | 14:24 | |
| *** ihrachyshka__ is now known as ihrachyshka | 14:24 | |
| fungi | rcarrillocruz: looking. i've got a spec brewing to add a lot of extra machinery to the release artifact production and upload toolchain, so i'm interested to see where it might be simplified | 14:25 |
| rcarrillocruz | oh | 14:28 |
| *** dangers_away is now known as dangers | 14:28 | |
| rcarrillocruz | nice | 14:28 |
| fungi | rcarrillocruz: at a conceptual level, we want to separate artifact generation (which includes archiving them to an openstack webserver like tarballs.o.o) from artifact publication to formal repositories (pypi, jenkins-ci, maven nexus...) | 14:29 |
| *** yfried_ has joined #openstack-infra | 14:29 | |
| fungi | rcarrillocruz: since the build step for any of those artifacts almost always has the capability of running arbitrary source code, and we want the servers with the upload credentials for those external services to not coexist on the same worker (or even the same jenkins master environment) as workers running arbitrary code | 14:30 |
| *** david-lyle has joined #openstack-infra | 14:30 | |
| *** dims has quit IRC | 14:31 | |
| *** david-ly_ has joined #openstack-infra | 14:31 | |
| fungi | rcarrillocruz: that why, for example, you'll see our pypi uploads are a two-step process where step 1 is to create the sdist/wheel and upload them to tarballs.o.o, then step 2 is to retrieve them from tarballs.o.o and upload them to pypi.python.org | 14:31 |
| *** rainya has joined #openstack-infra | 14:31 | |
| *** rainya has quit IRC | 14:32 | |
| *** banix has joined #openstack-infra | 14:32 | |
| *** rainya has joined #openstack-infra | 14:32 | |
| *** wenlock has joined #openstack-infra | 14:32 | |
| *** david-ly_ is now known as david-lyle_ | 14:33 | |
| rcarrillocruz | fungi: how will the "build jenkins" communicate with "privileged jenkins" ? scp with keys? or do you have something else in mind? | 14:33 |
| *** baohua has quit IRC | 14:33 | |
| fungi | rcarrillocruz: they already do this today. zuul coordinates all that | 14:34 |
| *** yfried_ has quit IRC | 14:34 | |
| *** amotoki has quit IRC | 14:34 | |
| fungi | rcarrillocruz: i was explaining how our current upload process works | 14:34 |
| *** david-lyle has quit IRC | 14:35 | |
| fungi | ultimately i think we actually want to start treating tarballs.o.o in a more official manner as well, have the build worker upload the artifact to a swift container which isn't published from tarballs.o.o, then have a separate job or jobs retrieve those artifacts and sign them before putting them on tarballs.o.o and pypi.p.o or wherever else | 14:35 |
| rcarrillocruz | oh | 14:36 |
| rcarrillocruz | signing | 14:36 |
| rcarrillocruz | do you have the spec pushed yet? | 14:36 |
| rcarrillocruz | i'm interested on your work, because i'm doing build/upload artifact stuff at work | 14:37 |
| fungi | not yet, but my point wasn't so much about the signing in this case. more that your upload script seems to assume that the uploads aren't staged between the build process on one machine and the upload process on another | 14:37 |
| openstackgerrit | Ivan Kolodyazhny proposed a change to openstack-infra/config: Make 'puppetmaster_host' configurable https://review.openstack.org/99041 | 14:38 |
| fungi | the pypi-tarball and pypi-wheel entries there are using the $URL variable in reverse from how jenkins-plugin|java-package do | 14:40 |
| rcarrillocruz | my script is just a refactor of the jenkinsci/mavencentral/pypi upload scripts, i saw a lot of common code and lacked parameterization... | 14:40 |
| *** zehicle_at_dell has quit IRC | 14:40 | |
| *** datsun180b has joined #openstack-infra | 14:40 | |
| rcarrillocruz | yep, that's correct. jenkinsci/maven use the URL as a target repo, whereas pypi use the URL param as source repo | 14:41 |
| *** zehicle_at_dell has joined #openstack-infra | 14:41 | |
| rcarrillocruz | i can split it into SOURCE_URL and TARGET_URL if you think that may be better for what you have in mind | 14:41 |
| fungi | rcarrillocruz: i need to read back through the original upload jobs for jenkins plugins and java packages, but i believe we may want to pull more shell out of the job definitions for those since i think they also do the same pattern as the pypi upload | 14:42 |
| *** david-lyle_ is now known as david-lyle | 14:43 | |
| *** dguerri is now known as dguerri`afk | 14:43 | |
| *** eharney has joined #openstack-infra | 14:43 | |
| rcarrillocruz | fungi: one note about jenkins/java packages upload. the curl command that used -X PUT --data-binary do not work for me, i switched to use 'curl -T' | 14:43 |
| *** chandankumar has quit IRC | 14:43 | |
| rcarrillocruz | not related to our conversation, but just in case you wonder when you reach to that point | 14:44 |
| openstackgerrit | Ivan Kolodyazhny proposed a change to openstack-infra/config: Make 'puppetmaster_host' configurable https://review.openstack.org/99041 | 14:44 |
| fungi | rcarrillocruz: yeah, i saw that earlier discussion | 14:44 |
| *** dguerri`afk is now known as dguerri | 14:45 | |
| openstackgerrit | Steve Martinelli proposed a change to openstack-infra/config: Publish specs jobs for *-specs https://review.openstack.org/106100 | 14:45 |
| openstackgerrit | Alexei Kornienko proposed a change to openstack/requirements: Unbind version of pyzmq https://review.openstack.org/106497 | 14:45 |
| fungi | rcarrillocruz: here's a good example... http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/maven-plugin-jobs.yaml#n25 | 14:46 |
| fungi | rcarrillocruz: you'll note we have two templates there. the {name}-localrepo-upload jobs get called at the end of the build process to stick the resulting jarfile on tarballs.o.o | 14:47 |
| *** MarkAtwood has joined #openstack-infra | 14:47 | |
| fungi | rcarrillocruz: then the {name}-mavencentral-upload jobs get called on a different worker which has the upload credentials, and it downloads the jar from tarballs.o.o before calling the upload script | 14:47 |
| rcarrillocruz | yeah | 14:48 |
| rcarrillocruz | so, how you envision doing this differntly ? | 14:48 |
| rcarrillocruz | instead of tarballs.o.o plain swift , plus the signing? | 14:48 |
| fungi | rcarrillocruz: more thinking in context of your refactor, not so much the signing improvements, if you're going to put all that logic into one script then it probably needs to absorb the shell fragments we have in these job templates. or else two scripts (which i think would more cleanly illustrate the separation logic) where one is for putting things on tarballs.o.o and the other is for getting | 14:49 |
| fungi | things from tarballs.o.o and uploading them to various other places | 14:50 |
| rcarrillocruz | gotcha | 14:50 |
| rcarrillocruz | let me work on this | 14:50 |
| rcarrillocruz | thx | 14:50 |
| openstackgerrit | Steve Martinelli proposed a change to openstack-infra/config: Publish specs jobs for *-specs https://review.openstack.org/106100 | 14:50 |
| *** datsun180b has quit IRC | 14:50 | |
| fungi | your current refactored script has the download+upload logic for python packages but only the upload logic for other things, and then reuses common options/variables for different purposes | 14:51 |
| *** datsun180b has joined #openstack-infra | 14:51 | |
| fungi | those phases still more or less exist in my long-term plan, it's just that the destinations change slightly, we add a trust chain between the steps, and tarballs.o.o gains a little more credibility. also we should be moving tarballs.o.o itself into a swift container, which will solve some of our current upload atomicity issues with branch tarballs | 14:54 |
| rcarrillocruz | i think it's then a good time to rename the script itself, as it won't be just for uploading | 14:55 |
| rcarrillocruz | i.e. artifactrepo_upload.sh | 14:56 |
| *** yamamoto has joined #openstack-infra | 14:56 | |
| fungi | yeah, i'm 100% in favor of our scripts having clear names describing their purposes | 14:57 |
| *** chandankumar has joined #openstack-infra | 14:58 | |
| *** jedimike has joined #openstack-infra | 15:00 | |
| *** dguerri is now known as dguerri`afk | 15:01 | |
| *** morganfainberg_Z is now known as morganfainberg | 15:01 | |
| *** dguerri`afk is now known as dguerri | 15:01 | |
| *** james_li has joined #openstack-infra | 15:03 | |
| *** craigbr has joined #openstack-infra | 15:04 | |
| *** MickeyO has quit IRC | 15:07 | |
| *** MarkAtwood has quit IRC | 15:07 | |
| *** MarkAtwood has joined #openstack-infra | 15:08 | |
| *** chandankumar has quit IRC | 15:08 | |
| *** jistr has quit IRC | 15:11 | |
| *** shashankhegde has joined #openstack-infra | 15:12 | |
| vhoward | alright, in a rathole and could use some help, using pbuilder to build up an ubuntu 12_04 precise and trying to build a source package for neutron using pdebuild using the 2014.1.1 orig.tar and debian from launchpad | 15:15 |
| vhoward | keeps failing for me looking for some dependency in python-simplejson python-simplejson : Depends: python:any (>= 2.7.3-0ubuntu2~) but it is not installable | 15:16 |
| vhoward | if anyone would like to message me for more info i could really use another set of eyes, tried tons of stuff just wnat to build out my dependencies basically | 15:16 |
| fungi | vhoward: you probably want to ask the ubuntu openstack packagers | 15:17 |
| *** timrc-afk is now known as timrc | 15:17 | |
| clarkb | 15:18 | |
| clarkb | er sorry | 15:18 |
| clarkb | in other news it is 8:17am and almost 25C | 15:18 |
| * clarkb melts | 15:18 | |
| openstackgerrit | Timur Sufiev proposed a change to openstack-infra/config: Create merlin repo https://review.openstack.org/110325 | 15:18 |
| *** craigbr has quit IRC | 15:18 | |
| fungi | vhoward: this channel is where we discuss community infrastructure (our wiki, listserv, irc bots, code review system, testing and validation automation, et cetera) | 15:19 |
| vhoward | fungi: sorry about the spam then thought i had the right place | 15:19 |
| *** MarkAtwood has quit IRC | 15:19 | |
| openstackgerrit | Sergey Skripnick proposed a change to openstack-infra/devstack-gate: Add option to edit ceilometer notification topics https://review.openstack.org/104123 | 15:20 |
| fungi | vhoward: no worries--i know it can get confusing. you probably want these people: https://wiki.ubuntu.com/ServerTeam/OpenStack | 15:20 |
| *** craigbr has joined #openstack-infra | 15:20 | |
| *** vhoward has left #openstack-infra | 15:21 | |
| *** vhoward has joined #openstack-infra | 15:21 | |
| fungi | vhoward: i think jamespage does hang out in here (he's one of the ubuntu packagers), so he might be able to point you to the appropriate forum | 15:22 |
| phschwartz | any infra core interested in taking a look at https://review.openstack.org/#/c/109723/ ? We are using it internal here at Rax and pushed it up as we think it will help others using JJB | 15:28 |
| *** shashankhegde has quit IRC | 15:29 | |
| jamespage | vhoward, hello | 15:30 |
| * jamespage reads | 15:30 | |
| jamespage | vhoward, lets discuss in #openstack-packaging | 15:30 |
| *** adalbas has quit IRC | 15:31 | |
| *** zns has quit IRC | 15:32 | |
| *** zul has joined #openstack-infra | 15:34 | |
| bauzas | just tried the Gerrit REST API, but there is still not CORS header | 15:35 |
| *** che-arne has quit IRC | 15:35 | |
| bauzas | why couldn't it be activated for the Gerrit API ? | 15:35 |
| bauzas | I can understand for the regular UI, but here we talk about a REST one :) | 15:36 |
| fungi | bauzas: is it an option in gerrit we haven't enabled? | 15:36 |
| bauzas | fungi: sounds like | 15:36 |
| bauzas | fungi: I have to look at Gerrit | 15:36 |
| *** hashar has quit IRC | 15:36 | |
| fungi | bauzas: a link to the configuration option for it in gerrit's 2.8 documentation would be helpful | 15:37 |
| bauzas | fungi: but by looking at the API doc, doesn't seem that Gerrit prevents this | 15:37 |
| bauzas | fungi: sure, will look at and provide a review if I fint | 15:37 |
| bauzas | find it | 15:37 |
| *** yamamoto has quit IRC | 15:37 | |
| fungi | bauzas: thanks | 15:37 |
| fungi | bauzas: though it's worth noting that the authentication for gerrit api calls is distinct from the authentication for its webui, so the risk is minimized | 15:39 |
| *** zns has joined #openstack-infra | 15:39 | |
| *** rkukura has quit IRC | 15:39 | |
| bauzas | fungi: I was thinking about it, because most of the tools I found are doing ssh connections | 15:39 |
| bauzas | fungi: even by not doing Paramiko calls | 15:39 |
| bauzas | fungi: IMHO, calling a REST API would be much cleaner for doing searches :-) | 15:40 |
| bauzas | fungi: as Gerrit is providing it | 15:40 |
| fungi | bauzas: agreed. bear in mind that if it's our tools you're looking at, most of them were written while we were running a version of gerrit which only had an ssh api | 15:40 |
| bauzas | fungi: anyway, for my needs, I'll batch it and generate a static file | 15:40 |
| *** james_li has quit IRC | 15:40 | |
| *** gangil has joined #openstack-infra | 15:41 | |
| fungi | the rest api has only been available to us for a few months now | 15:41 |
| bauzas | fungi: yeah, I know | 15:41 |
| bauzas | fungi: I appreciated the move to 2.8 :-) | 15:41 |
| openstackgerrit | Julien Marin-Frisonroche proposed a change to openstack-infra/git-review: Improve windows support for git-review --setup https://review.openstack.org/110331 | 15:41 |
| AJaeger | SergeyLukjanov, thanks! Will read the comments later... | 15:41 |
| *** che-arne has joined #openstack-infra | 15:42 | |
| *** rwsu has joined #openstack-infra | 15:42 | |
| *** adalbas has joined #openstack-infra | 15:44 | |
| *** yamamoto has joined #openstack-infra | 15:45 | |
| *** ihrachyshka has quit IRC | 15:46 | |
| *** dguerri is now known as dguerri`afk | 15:47 | |
| *** lttrl has quit IRC | 15:48 | |
| *** HenryG_ has joined #openstack-infra | 15:48 | |
| *** james_li has joined #openstack-infra | 15:49 | |
| *** dguerri`afk is now known as dguerri | 15:49 | |
| *** vermarah_ has joined #openstack-infra | 15:50 | |
| *** amcrn has joined #openstack-infra | 15:51 | |
| *** HenryG has quit IRC | 15:52 | |
| *** amotoki has joined #openstack-infra | 15:52 | |
| *** HenryG_ is now known as HenryG_afk | 15:52 | |
| *** HenryG_afk is now known as HenryG | 15:52 | |
| *** rkukura has joined #openstack-infra | 15:52 | |
| *** pdmars_ has quit IRC | 15:53 | |
| *** hashar has joined #openstack-infra | 15:54 | |
| *** che-arne has quit IRC | 15:55 | |
| *** che-arne has joined #openstack-infra | 15:55 | |
| *** pdmars has joined #openstack-infra | 15:56 | |
| *** IvanBerezovskiy has left #openstack-infra | 15:57 | |
| *** skolekonov has quit IRC | 15:57 | |
| *** esker has joined #openstack-infra | 15:58 | |
| *** yamamoto has quit IRC | 15:58 | |
| *** marcoemorais has joined #openstack-infra | 16:03 | |
| *** markwash has joined #openstack-infra | 16:03 | |
| *** hashar is now known as hasharMeeting | 16:05 | |
| *** samy_ has quit IRC | 16:06 | |
| *** e0ne has quit IRC | 16:06 | |
| *** trinaths has joined #openstack-infra | 16:08 | |
| *** melwitt has joined #openstack-infra | 16:08 | |
| *** dims has joined #openstack-infra | 16:08 | |
| *** denis_makogon has quit IRC | 16:10 | |
| rcarrillocruz | fungi: what do you think about having three different scripts, artifact-build/artifact-stash/artifact-publish ? | 16:11 |
| rcarrillocruz | i'm thinking that's more future proof, because if you add a sign stage at some point it's just a matter of adding an 'artifact-sign' script | 16:12 |
| rcarrillocruz | instead of cramming build AND stash into just one script | 16:12 |
| rcarrillocruz | ? | 16:12 |
| rcarrillocruz | i could also write an artifact-release script that does the whole artifact lifecycle of build/stash/publish, for convenience | 16:12 |
| fungi | rcarrillocruz: well, the artifact build is already mostly scripted in various places (and can't really be collapsed easily because it's extremely toolchain-dependent). otherwise that's basically the idea i expressed in my review comment | 16:13 |
| *** pcm_ has left #openstack-infra | 16:14 | |
| *** melwitt has quit IRC | 16:14 | |
| *** melwitt has joined #openstack-infra | 16:14 | |
| *** pcm_ has joined #openstack-infra | 16:15 | |
| *** shashankhegde has joined #openstack-infra | 16:15 | |
| *** markmcclain has joined #openstack-infra | 16:15 | |
| pcm_ | dhellmann: ping | 16:15 |
| fungi | rcarrillocruz: effectively your refactor could be to combine various existing upload scripts and job shell blocks into a tarballs.o.o stash script and an external services upload script | 16:15 |
| *** gyee has joined #openstack-infra | 16:16 | |
| fungi | rcarrillocruz: but i wouldn't recommend altering the current artifact build scripts because they're already mostly necessarily distinct | 16:16 |
| rcarrillocruz | fungi: other question: why the artifact upload stage is done as a JJB builder script. wouldn't make sense to have it as a JJB publisher ? | 16:17 |
| *** ildikov has joined #openstack-infra | 16:18 | |
| rcarrillocruz | we have a 'war' JJB publisher already | 16:18 |
| *** banix has quit IRC | 16:18 | |
| fungi | rcarrillocruz: i think so that we can properly capture job "failure" as "i failed to upload this thing" | 16:19 |
| rcarrillocruz | hmm, ok | 16:19 |
| rcarrillocruz | thx | 16:19 |
| redrobot | SergeyLukjanov hi! I was wondering if you could review https://review.openstack.org/#/c/109398/ again. I think you may have missed that openstack-barbican was already in the channels.yaml file | 16:19 |
| fungi | rcarrillocruz: as opposed to when a publisher dies/aborts which jenkins reports as a status of "unstable" instead of "failure" | 16:20 |
| rcarrillocruz | ah, nice | 16:20 |
| rcarrillocruz | 'unstable' | 16:20 |
| rcarrillocruz | :-) | 16:20 |
| *** mrmartin has quit IRC | 16:22 | |
| *** melwitt has quit IRC | 16:23 | |
| *** melwitt has joined #openstack-infra | 16:23 | |
| *** melwitt has quit IRC | 16:26 | |
| *** melwitt has joined #openstack-infra | 16:26 | |
| *** melwitt has quit IRC | 16:27 | |
| *** bdpayne has joined #openstack-infra | 16:30 | |
| *** melwitt has joined #openstack-infra | 16:30 | |
| *** zns has quit IRC | 16:31 | |
| *** samy has joined #openstack-infra | 16:32 | |
| clarkb | there I have properly cherrypicked the hash seed changes for stable maint | 16:32 |
| *** melwitt has quit IRC | 16:32 | |
| *** samy is now known as Guest50025 | 16:33 | |
| *** marcoemorais has quit IRC | 16:33 | |
| *** _nadya_ has quit IRC | 16:34 | |
| *** marcoemorais has joined #openstack-infra | 16:34 | |
| fungi | excellent | 16:34 |
| clarkb | fungi: mordred jeblair jhesketh SergeyLukjanov I would like to consider setting a date for eg 2 weeks from now for the tox upgrade on our side | 16:35 |
| clarkb | I feel like this is a race I can't win if I try to test and update everyone | 16:35 |
| clarkb | instead better to just give warning | 16:35 |
| clarkb | and do my best | 16:36 |
| fungi | clarkb: so all the official projects have static hash seed changes proposed and backported to supported stable branches now? or only ones where you noticed obvious failure most of the time? | 16:36 |
| clarkb | only the ones that I got to | 16:36 |
| clarkb | but for the most part yes | 16:36 |
| fungi | ahh, okay cool | 16:36 |
| *** yamahata has joined #openstack-infra | 16:36 | |
| clarkb | and I have sent mail about the situation and people seem to be aware of it | 16:37 |
| fungi | 2 weeks seems perfectly reasonable to me in that case | 16:37 |
| *** zns has joined #openstack-infra | 16:38 | |
| *** chuckC has quit IRC | 16:38 | |
| *** crc32 has joined #openstack-infra | 16:41 | |
| *** jedimike has quit IRC | 16:41 | |
| *** afazekas has quit IRC | 16:44 | |
| dims | fungi, where can i check if i am part of oslo-ptl? (i have trouble cutting a oslo.utils release - http://paste.openstack.org/show/88956/) | 16:44 |
| anteaya | clarkb: sounds like a good idea to me | 16:44 |
| *** Longgeek_ has joined #openstack-infra | 16:45 | |
| anteaya | dims: you aren't https://review.openstack.org/#/admin/groups/148,members | 16:45 |
| clarkb | cool looks like we got marun's neutron functional tests working yesterday | 16:46 |
| * clarkb feels like he has accomplished things at this meetup | 16:46 | |
| *** shashankhegde has quit IRC | 16:46 | |
| *** Ryan_Lane1 has joined #openstack-infra | 16:46 | |
| anteaya | dims: so yeah either get dhellmann or markmc to add you or change the tagging permission on oslo.utils: http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/gerrit/acls/openstack/oslo.utils.config | 16:47 |
| anteaya | clarkb: yay | 16:47 |
| dims | ah thanks. will get dhellmann to add me | 16:48 |
| *** Longgeek has quit IRC | 16:48 | |
| clarkb | on to the next thing | 16:49 |
| *** dims has quit IRC | 16:49 | |
| fungi | dims: yeah... you can find the group membership link at https://review.openstack.org/#/admin/projects/openstack/oslo.utils,access | 16:49 |
| *** dims has joined #openstack-infra | 16:50 | |
| clarkb | adam_g: greghaynes: so tl;dr we need to make a job that does juno nova baremetal to juno ironic using grenade | 16:50 |
| *** crc32 has quit IRC | 16:50 | |
| clarkb | tis will require updates to devstack gate to do a release X to release X with different services update | 16:50 |
| greghaynes | so I hear | 16:51 |
| *** adalbas has quit IRC | 16:51 | |
| *** marun has joined #openstack-infra | 16:51 | |
| anteaya | that job sounds like it needs a sprint all to itself | 16:51 |
| clarkb | anteaya: maybe | 16:52 |
| *** pblaho_ has joined #openstack-infra | 16:52 | |
| *** reed has joined #openstack-infra | 16:53 | |
| *** marzif has quit IRC | 16:54 | |
| *** bdpayne has quit IRC | 16:54 | |
| *** banix has joined #openstack-infra | 16:55 | |
| *** james_li has quit IRC | 16:56 | |
| *** Longgeek_ has quit IRC | 16:56 | |
| *** Longgeek has joined #openstack-infra | 16:57 | |
| *** che-arne has quit IRC | 16:57 | |
| *** zns has quit IRC | 16:57 | |
| *** leopoldj has quit IRC | 16:57 | |
| *** luqas has quit IRC | 16:58 | |
| *** Longgeek_ has joined #openstack-infra | 16:58 | |
| *** james_li has joined #openstack-infra | 16:59 | |
| *** che-arne has joined #openstack-infra | 16:59 | |
| *** zns has joined #openstack-infra | 16:59 | |
| *** ZZelle_ has joined #openstack-infra | 16:59 | |
| *** che-arne|2 has joined #openstack-infra | 16:59 | |
| *** craigbr has quit IRC | 16:59 | |
| *** craigbr_ has joined #openstack-infra | 16:59 | |
| *** che-arne|2 has quit IRC | 16:59 | |
| *** gokrokve has joined #openstack-infra | 17:00 | |
| *** rainya has quit IRC | 17:00 | |
| *** trinaths has quit IRC | 17:01 | |
| *** trinaths has joined #openstack-infra | 17:01 | |
| *** Longgeek has quit IRC | 17:01 | |
| *** jpich has quit IRC | 17:03 | |
| pcm_ | anteaya: ping | 17:03 |
| *** hasharMeeting has quit IRC | 17:03 | |
| anteaya | pcm_: I'm here | 17:04 |
| anteaya | yolanda: okay looking better, in 800x600 slide 18 needs some love | 17:04 |
| pcm_ | anteaya: Have a question... I'm getting tox failure to install the new oslo.config package (for neutron) | 17:05 |
| anteaya | and I think slide 16 too, the text continues off the right hand side of the page | 17:05 |
| pcm_ | anteaya: that was just upstreamed. | 17:05 |
| anteaya | pcm_: well we are moving to a new version of tox | 17:05 |
| anteaya | pcm_: can you paste your failure? | 17:05 |
| pcm_ | anteaya: I think it was bug 106592 | 17:05 |
| uvirtbot | Launchpad bug 106592 in ubiquity "install bug (dup-of: 106591)" [Undecided,Invalid] https://launchpad.net/bugs/106592 | 17:05 |
| uvirtbot | Launchpad bug 106591 in ubiquity "install bug" [Undecided,Invalid] https://launchpad.net/bugs/106591 | 17:05 |
| anteaya | ubiquity? perhaps we have the wrong bug number | 17:06 |
| *** praneshp has joined #openstack-infra | 17:06 | |
| pcm_ | anteaya: yeah, let me check again... | 17:06 |
| * anteaya stands by | 17:06 | |
| pcm_ | anteaya: Gee... that's what I have https://review.openstack.org/#/c/106592/4 | 17:07 |
| *** rainya has joined #openstack-infra | 17:07 | |
| *** adalbas has joined #openstack-infra | 17:07 | |
| *** rainya has quit IRC | 17:08 | |
| clarkb | 106592 is a change not a bug. just a noun mapping mismatch :) | 17:08 |
| anteaya | pcm_: I'm lost | 17:08 |
| anteaya | pcm_: lets talk about you | 17:08 |
| anteaya | what are you seeing that said "Hey I need to talk to infra." | 17:08 |
| *** rainya has joined #openstack-infra | 17:09 | |
| pcm_ | anteaya: sure. Let me try to rephrase... | 17:09 |
| * anteaya waits | 17:09 | |
| pcm_ | anteaya: I updated my Neutron repo, and it has a new requirements.txt file | 17:09 |
| anteaya | cool | 17:09 |
| pcm_ | anteaya: I did "tox -e py27 -r --notest" | 17:09 |
| pcm_ | It fails not being able to install oslo.config 1.4.0.0a3, which was recently added to requirements and pushed to neutron. | 17:10 |
| *** derekh_ has quit IRC | 17:10 | |
| pcm_ | So, I'm wondering what to do? | 17:11 |
| anteaya | pcm_: honour me with a stacktrace? | 17:11 |
| *** Longgeek_ has quit IRC | 17:11 | |
| pcm_ | anteaya: Sure... http://paste.openstack.org/show/88960/ | 17:11 |
| * anteaya feels honoured | 17:11 | |
| anteaya | thanks | 17:11 |
| marun | clarkb: functional job is now working (thanks to you), so the patch that switches to the new tox job is ready for approval: https://review.openstack.org/#/c/104039/ | 17:11 |
| *** Longgeek has joined #openstack-infra | 17:11 | |
| *** jgallard has quit IRC | 17:12 | |
| * anteaya wonders how much significance to give: [pbr] Processing SOURCES.txt | 17:12 | |
| anteaya | warning: LocalManifestMaker: standard file '-c' not found | 17:12 |
| clarkb | marun: cool. it has my +2. maybe fungi can look at it quickly. if not I can approve in a bit | 17:12 |
| fungi | pcm_: what does '/opt/stack/neutron/.tox/py27/bin/pip --version' say? | 17:12 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Update core team for tempest and qa-specs https://review.openstack.org/109787 | 17:12 |
| pcm_ | anteaya: I tried removing .tox directory, but I get the same error every time. When I look at the "change" in gerrit, it shows it approved and seems to have a Jenkins failure (didn't look at it). | 17:12 |
| *** Longgeek_ has joined #openstack-infra | 17:13 | |
| pcm_ | pip 1.4.1 from /opt/stack/neutron/.tox/py27/lib/python2.7/site-packages (python 2.7) | 17:13 |
| *** Longgeek_ has quit IRC | 17:13 | |
| openstackgerrit | Matt Kovacs proposed a change to openstack-infra/config: Adding project saml-generator https://review.openstack.org/110378 | 17:13 |
| fungi | pcm_: odd... that version should understand python wheels | 17:14 |
| anteaya | pcm_: that patch has been approved but never merged due to the jenkins failure | 17:14 |
| *** e0ne has joined #openstack-infra | 17:15 | |
| fungi | pcm_: does /opt/stack/neutron/.tox/py27/tmp/pseudo-home/.pip/pip.log mention which urls it tried to download from? | 17:15 |
| * pcm_ wondering why my requirements.txt has oslo.config 1.4.0.0a3 | 17:15 | |
| pcm_ | me checking | 17:16 |
| *** crc32 has joined #openstack-infra | 17:17 | |
| fungi | in theory it should be finding it from http://pypi.python.org/packages/py2.py3/o/oslo.config/oslo.config-1.4.0.0a3-py2.py3-none-any.whl | 17:17 |
| *** Longgeek has quit IRC | 17:17 | |
| anteaya | pcm_: because that is what is in the neutron repo: http://git.openstack.org/cgit/openstack/neutron/tree/requirements.txt#n26 | 17:17 |
| pcm_ | fungi: for oslo.config it says https://pypi.python.org/simple/oslo.config/ | 17:17 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for bug 1342086 https://review.openstack.org/107032 | 17:17 |
| uvirtbot | Launchpad bug 1342086 in nova "DBDeadlock in delete_instance" [Undecided,Fix released] https://launchpad.net/bugs/1342086 | 17:17 |
| anteaya | is that the url parsing thing | 17:17 |
| pcm_ | You can see it at the tail end of the paste log above | 17:18 |
| *** shashankhegde has joined #openstack-infra | 17:18 | |
| anteaya | where if it can't parse hyphens correctly it reverts to looking in a parent directory | 17:18 |
| fungi | oh! i think old pip (e.g. the 1.4.1 you're using) won't find that without adding the --use-wheel option | 17:18 |
| clarkb | fungi: correct | 17:18 |
| fungi | https://pip.pypa.io/en/1.4.1/usage.html | 17:18 |
| clarkb | you had to say that for it to click for me | 17:18 |
| clarkb | but ya you need 1.5.x to use wheels | 17:19 |
| clarkb | by default | 17:19 |
| fungi | pcm_: so you either need to tweak the tox.ini to support pip 1.4 or you need to use pip >=1.5 | 17:19 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Fix the coverage job in tox https://review.openstack.org/109789 | 17:19 |
| fungi | pcm_: chances are you're getting pip 1.4.1 vendored in by an older virtualenv release | 17:19 |
| *** bdpayne has joined #openstack-infra | 17:20 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack/requirements: Add new library oslo.utils https://review.openstack.org/110380 | 17:20 |
| pcm_ | fungi: so I modify tox.ini to do this --use-wheels or use a newer pip? | 17:21 |
| *** jaypipes has joined #openstack-infra | 17:21 | |
| anteaya | yolanda: where did you get the blank index page for your patch? https://review.openstack.org/#/c/109570/2 I can't figure out where it came from | 17:21 |
| *** e0ne has quit IRC | 17:22 | |
| openstackgerrit | Matt Riedemann proposed a change to openstack/requirements: Update to python-neutronclient 2.3.6 https://review.openstack.org/110382 | 17:22 |
| fungi | pcm_: yep. for upstream ci we pre-install latest pip on our test servers | 17:22 |
| pcm_ | fungi: Can I just update pip to the latest? Would a simple way be to make sure devstack repo is latest and then do a stack to ensure the latest pip (I think it uninstalls and installs the latest) | 17:23 |
| pcm_ | fungi: or should I just manually update it? | 17:24 |
| dstufft | inside of a tox environment the version of pip is controlled by the version of virtualenv that is installed | 17:24 |
| fungi | pcm_: yes, i believe devstack does a similar dance but only if it finds that pip is not present or was installed from system packages | 17:24 |
| *** vladan has quit IRC | 17:25 | |
| pcm_ | fungi: Confused now... so on my system pip=1.5.6, but the virtualenv has 1.4.1. | 17:25 |
| fungi | looks like we also currently have virtualenv==1.11.6 installed on our job workers | 17:26 |
| *** dmsimard is now known as dmsimard_away | 17:26 | |
| fungi | to dstufft's point | 17:26 |
| *** gmurphy has quit IRC | 17:26 | |
| pcm_ | fungi: I have 1.10.1 | 17:26 |
| *** sarob has quit IRC | 17:26 | |
| dstufft | you want 1.11.x | 17:26 |
| dstufft | for pip 1.5 | 17:26 |
| fungi | right, i think virtualenv 1.10.1 installs pip 1.4.1 in your venvs | 17:27 |
| dstufft | yes | 17:27 |
| fungi | which will not be happymaking without --use-wheel | 17:27 |
| pcm_ | fungi: dstufft: thanks will update. | 17:27 |
| mestery | I have a question for someone in infra: How can I create a milestone for python-neutronclient here: https://launchpad.net/python-neutronclient | 17:27 |
| mestery | I can create releases, but I don't see how to create milestones so I can assign bugs, etc. | 17:28 |
| anteaya | dims: congratulations on the oslo.utils release | 17:28 |
| dims | anteaya, thanks! | 17:28 |
| *** flaviof has quit IRC | 17:28 | |
| *** sarob has joined #openstack-infra | 17:28 | |
| anteaya | mestery: permissions are on python-neutronclient: http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/gerrit/acls/openstack/python-neutronclient.config | 17:29 |
| *** gmurphy has joined #openstack-infra | 17:29 | |
| mestery | anteaya: Thanks! I think I figured it out: I have to create a series, then pull milestones from that. | 17:29 |
| fungi | anteaya: he's talking about lp permissions not gerrit permissions | 17:29 |
| anteaya | mestery: and you are a member of the neutron-milestone group: https://review.openstack.org/#/admin/groups/55,members | 17:29 |
| anteaya | ah | 17:29 |
| anteaya | my gerrit searching is useless for lp | 17:30 |
| anteaya | mestery: cool, happy milestoning | 17:30 |
| mestery | anteaya: Thanks! :) | 17:30 |
| fungi | and yes you create the milestones relative to a specific series in lp | 17:30 |
| pcm_ | anteaya: fungi: dstufft: clarkb: Thanks for the help! | 17:31 |
| *** amuller has joined #openstack-infra | 17:31 | |
| *** vladan has joined #openstack-infra | 17:31 | |
| *** Ajaeger1 has joined #openstack-infra | 17:32 | |
| *** david-lyle has quit IRC | 17:33 | |
| *** roeyc has quit IRC | 17:33 | |
| *** david-lyle has joined #openstack-infra | 17:33 | |
| *** david-lyle has quit IRC | 17:35 | |
| *** sballe has quit IRC | 17:36 | |
| *** david-lyle has joined #openstack-infra | 17:36 | |
| openstackgerrit | A change was merged to openstack-infra/config: Fix neutron func job to use gate-specific tox env https://review.openstack.org/104039 | 17:36 |
| *** harlowja_away is now known as harlowja | 17:36 | |
| anteaya | pcm_: glad you are on your way again | 17:37 |
| *** markwash_ has joined #openstack-infra | 17:37 | |
| fungi | marun: clarkb: ^ should be live shortly | 17:37 |
| clarkb | danke | 17:37 |
| *** markwash has quit IRC | 17:39 | |
| *** markwash_ is now known as markwash | 17:39 | |
| *** mwagner_lap has quit IRC | 17:39 | |
| *** chandankumar has joined #openstack-infra | 17:41 | |
| *** david-lyle has quit IRC | 17:44 | |
| *** david-lyle has joined #openstack-infra | 17:44 | |
| *** cody-somerville has joined #openstack-infra | 17:45 | |
| *** dmsimard_away is now known as dmsimard | 17:47 | |
| *** arosen has joined #openstack-infra | 17:48 | |
| *** eharney has quit IRC | 17:48 | |
| krotscheck | Anyone want to Workflow +1 this? We’ve got lots of +2’s. | 17:48 |
| krotscheck | https://review.openstack.org/#/c/97521/ | 17:48 |
| *** david-lyle has quit IRC | 17:49 | |
| krotscheck | (And by lots I mean 2) | 17:49 |
| * clarkb prepares to use all the tether bw by starting gertty | 17:51 | |
| anteaya | ha ha ha | 17:51 |
| anteaya | I should too | 17:52 |
| anteaya | though I'm betting it will just blow up | 17:52 |
| anteaya | since as jeblair says it isn't ready, don't use it | 17:52 |
| jeblair | krotscheck: i never saw a response to my question: why? i don't object, per se, but i don't think we should treat that as automatic | 17:52 |
| *** trinaths has quit IRC | 17:52 | |
| krotscheck | jeblair: Ah, got it. Things get buried in the comment feed, no? | 17:53 |
| jeblair | krotscheck: heh, yep. especially if i don't '-1' because i don't want to seem unduly negative. :/ | 17:53 |
| jeblair | krotscheck: basically, if you and/or nikita say "oh, yeah totally there's adminy stuff he needs to do" i'm all for it, but if it's just for parity, then i'd say let's wait until there's a real need | 17:54 |
| rcarrillocruz | hey | 17:54 |
| rcarrillocruz | could anyone please point me where's the source code for the wheel/tarball jjb publishers? | 17:55 |
| rcarrillocruz | i can't seem to find them anywhere | 17:55 |
| *** htrmeira has joined #openstack-infra | 17:55 | |
| anteaya | hey I made gertty go boom: http://paste.openstack.org/show/88970/ | 17:55 |
| anteaya | what do I do to fix it? | 17:55 |
| clarkb | rcarrillocruz: modules/openstack_project/files/jenkins_job_builder/config/pypi-jobs.yaml iirc | 17:55 |
| *** SumitNaiksatam has quit IRC | 17:56 | |
| anteaya | I screwed up something with sqlalchemy | 17:56 |
| jeblair | anteaya: remove your db file (~/.gertty.db) and start over | 17:56 |
| anteaya | kk | 17:56 |
| krotscheck | jeblair: Responded with a comment- My take on it is less “he needs to do admin things” and more “If something breaks in admin land, it would help to be an admin to debug what happened” | 17:57 |
| *** signed8bit is now known as signed8bit_ZZZzz | 17:57 | |
| anteaya | very pretty welcome page | 17:57 |
| anteaya | not sure if I got that the first time | 17:57 |
| *** Alexei_987 has joined #openstack-infra | 17:58 | |
| jeblair | krotscheck: cool, thx | 17:58 |
| jeblair | anteaya: you did not :) | 17:58 |
| rcarrillocruz | clarkb: i guess you mean '{name}-pypi-wheel-upload' . that's indeed in python-jobs, for uploading the wheel to pypi. But I'm interested in the bits that upload the wheel/tarball to tarballs.openstack.org, right after it's built | 17:58 |
| krotscheck | An alternative to that would be to build out db synchronization so that we can have local copies of all the prod data to test against. | 17:58 |
| fungi | rcarrillocruz: http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/python-jobs.yaml#n226 | 17:59 |
| fungi | rcarrillocruz: it very well may be that at this stage, we don't even need that part extracted into a separate script because it's straightforward enough in the publisher | 18:00 |
| anteaya | jeblair: well it is a great addition then | 18:00 |
| clarkb | rcarrillocruz: the tarball job in python-jobs.yaml | 18:00 |
| clarkb | rcarrillocruz: it uploads to tarballs.o.o | 18:00 |
| *** _nadya_ has joined #openstack-infra | 18:01 | |
| clarkb | sdague: did 107599 fail due to your change? eg did that influence the large ops job? | 18:01 |
| clarkb | jogo: ^ | 18:01 |
| fungi | rcarrillocruz: so your refactor could just be the one script which retrieves things from tarballs.o.o and uploads them to various services (so just combining the various upload scripts with the builder bits which do the downloading in the jar/war upload jobs) | 18:01 |
| *** marcoemorais has quit IRC | 18:02 | |
| *** marcoemorais has joined #openstack-infra | 18:02 | |
| *** sarob has quit IRC | 18:02 | |
| fungi | rcarrillocruz: in which case the change you have now is fairly close, aside from not covering some bits of shell we currently have in the java and jenkins-ci upload jobs | 18:03 |
| *** sarob has joined #openstack-infra | 18:03 | |
| fungi | and realigning the meaning of your command-line options and internal variables to be consistent | 18:03 |
| *** eharney has joined #openstack-infra | 18:04 | |
| fungi | on an unrelated note, does anyone else shudder when they see the "welcome new contributor" response on a change to add a new project? | 18:05 |
| sdague | clarkb: I don't know, haven't looked at that one | 18:06 |
| openstackgerrit | Matt Kovacs proposed a change to openstack-infra/config: Adding project saml-generator https://review.openstack.org/110378 | 18:06 |
| sdague | clarkb: did you see my reviews on the d-g multinode patch | 18:07 |
| clarkb | not yet | 18:07 |
| jogo | clarkb: I think 107599 failed because of a bug 1349590 | 18:07 |
| jogo | sdague: ^ | 18:07 |
| uvirtbot | Launchpad bug 1349590 in nova "nova-network appears to lock up while associating floating ip's" [High,Confirmed] https://launchpad.net/bugs/1349590 | 18:07 |
| anteaya | fungi: every time | 18:07 |
| *** david-lyle has joined #openstack-infra | 18:07 | |
| *** sarob has quit IRC | 18:07 | |
| sdague | because I think that instead of calling out devstack functions, we should copy those needed functions into d-g | 18:07 |
| *** jamielennox|away is now known as jamielennox | 18:07 | |
| sdague | otherwise I think we could break some stuff | 18:08 |
| clarkb | sdague: it should be cogated | 18:08 |
| clarkb | but pulling out is fine too | 18:08 |
| *** SumitNaiksatam has joined #openstack-infra | 18:08 | |
| sdague | yeh, but devstack has branches | 18:09 |
| rcarrillocruz | fungi: thanks, i was getting lost with variable expansion trail, digging the hole from projects.yaml all the way down | 18:09 |
| sdague | and d-g does not | 18:09 |
| sdague | so the functions called will be different code depending on branch | 18:09 |
| sdague | which I think is error prone, and a kind of coupling we don't want | 18:09 |
| clarkb | sdague: did you see my comments in a previous patchset? | 18:09 |
| clarkb | oh good point | 18:09 |
| sdague | I don't think so, sorry, I just dove in on the current patch | 18:10 |
| *** e0ne has joined #openstack-infra | 18:10 | |
| *** holden247 has joined #openstack-infra | 18:10 | |
| clarkb | I think we said similar thinsg about pushing the iptables stuff into devstack | 18:10 |
| clarkb | basically that is required to do multinode so make devstack do it | 18:10 |
| *** markmcclain has quit IRC | 18:12 | |
| *** marcoemorais has quit IRC | 18:12 | |
| *** marcoemorais has joined #openstack-infra | 18:12 | |
| *** marcoemorais has quit IRC | 18:13 | |
| *** marcoemorais has joined #openstack-infra | 18:14 | |
| *** che-arne has quit IRC | 18:15 | |
| *** markmcclain has joined #openstack-infra | 18:16 | |
| *** markmcclain1 has joined #openstack-infra | 18:17 | |
| *** banix has quit IRC | 18:19 | |
| *** markmcclain has quit IRC | 18:21 | |
| *** ociuhandu has quit IRC | 18:24 | |
| *** juzuluag has quit IRC | 18:24 | |
| *** markwash has quit IRC | 18:26 | |
| *** marcoemorais has quit IRC | 18:26 | |
| *** mpaolino has quit IRC | 18:26 | |
| *** jlibosva has quit IRC | 18:26 | |
| *** marcoemorais has joined #openstack-infra | 18:26 | |
| clarkb | jeblair: fungi jhesketh SergeyLukjanov mordred 104386, I think we should go ahead with the change as is unless someone wants to look at the string and change the type | 18:28 |
| clarkb | iirc using text for everything will just make those parameters render in a larger box which is fine by me | 18:28 |
| *** banix has joined #openstack-infra | 18:28 | |
| *** e0ne has quit IRC | 18:30 | |
| clarkb | fungi: I am going to approve changes to git-review without your +2 if that is ok with you | 18:32 |
| clarkb | fungi: the test suite is pretty legit now so I trust it | 18:32 |
| fungi | clarkb: sounds great to me | 18:32 |
| *** e0ne has joined #openstack-infra | 18:32 | |
| openstackgerrit | imain proposed a change to openstack-infra/config: Back to running docker tempest tests on Fedora. https://review.openstack.org/110409 | 18:33 |
| *** sdake has quit IRC | 18:33 | |
| *** sdake has joined #openstack-infra | 18:33 | |
| *** vladan has quit IRC | 18:34 | |
| clarkb | fungi: 108146 is an easy one | 18:37 |
| clarkb | man -l ftw | 18:37 |
| openstackgerrit | A change was merged to openstack/requirements: Update openstack-doc-tools to at least 0.17 https://review.openstack.org/109905 | 18:38 |
| nibalizer | jeblair: if at any point you want to poke hiera or puppet3, let me know | 18:38 |
| *** jcoufal has joined #openstack-infra | 18:39 | |
| *** vladan has joined #openstack-infra | 18:42 | |
| *** zns has quit IRC | 18:44 | |
| *** zns has joined #openstack-infra | 18:44 | |
| openstackgerrit | A change was merged to openstack-infra/git-review: Enabled hacking checks H305 and H307 https://review.openstack.org/107576 | 18:46 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Stop using select-mirror in slave scripts https://review.openstack.org/106107 | 18:46 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Clean up select-mirror backward compatibility https://review.openstack.org/106112 | 18:46 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Delete select-mirror and all the args https://review.openstack.org/103355 | 18:46 |
| *** yfried_ has joined #openstack-infra | 18:48 | |
| *** Sukhdev has joined #openstack-infra | 18:49 | |
| *** gmurphy has quit IRC | 18:50 | |
| openstackgerrit | A change was merged to openstack-infra/git-review: Avoid a stacktrace when no reviews are pending https://review.openstack.org/103178 | 18:50 |
| *** markwash has joined #openstack-infra | 18:50 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 18:51 | |
| *** lukego has quit IRC | 18:51 | |
| reed | does anybody know why Smokestack seems to be last updated 3 months ago? https://ask.openstack.org/en/question/43662/what-happened-to-httpssmokestackopenstackorg/ | 18:52 |
| clarkb | reed: I think dprince decided it wans't needed anymore? | 18:53 |
| fungi | reed: dprince might know if you see him around | 18:53 |
| clarkb | we are testing on fedora now and bobball is testing xen | 18:53 |
| clarkb | the only thing we don't quite cover is the package building that happens before testing | 18:53 |
| reed | clarkb, fungi: the wiki page makes it look like it's a bigger effort than just dprince... is that the case or is this a personal project of dan? | 18:54 |
| *** markwash_ has joined #openstack-infra | 18:54 | |
| clarkb | I thought it was basically just dan doing a thing | 18:54 |
| fungi | reed: i think there might have been one or two other hatters working on it from time to time, but mostly just him yeah | 18:54 |
| *** markwash has quit IRC | 18:55 | |
| *** markwash_ is now known as markwash | 18:55 | |
| reed | fungi, ok, I'm answring that question then | 18:55 |
| *** banix has quit IRC | 18:55 | |
| *** jistr has joined #openstack-infra | 18:56 | |
| reed | do we use it for anything? | 18:56 |
| *** pblaho_ is now known as pblaho | 18:56 | |
| fungi | pleia2: is reviewday still tied at all to the results on smokestack.org? | 18:56 |
| openstackgerrit | A change was merged to openstack-infra/git-review: Fix groff warnings with manpage https://review.openstack.org/108146 | 18:56 |
| *** david-lyle has quit IRC | 18:57 | |
| *** david-lyle has joined #openstack-infra | 18:57 | |
| *** gmurphy has joined #openstack-infra | 18:57 | |
| *** david-lyle has quit IRC | 18:59 | |
| *** david-ly_ has joined #openstack-infra | 18:59 | |
| *** ramashri has joined #openstack-infra | 19:00 | |
| *** sarob has joined #openstack-infra | 19:00 | |
| fungi | meeting time! | 19:00 |
| *** markwash has quit IRC | 19:01 | |
| jeblair | yay! | 19:01 |
| *** sarob_ has joined #openstack-infra | 19:02 | |
| *** whoops has joined #openstack-infra | 19:04 | |
| *** sarob has quit IRC | 19:05 | |
| *** esker has quit IRC | 19:05 | |
| *** esker has joined #openstack-infra | 19:05 | |
| *** david-ly_ is now known as david-lyle | 19:05 | |
| *** craigbr_ has quit IRC | 19:06 | |
| *** sarob_ has quit IRC | 19:07 | |
| openstackgerrit | Davide Guerri proposed a change to openstack-infra/config: Complete the documentation needed to add a new IRC channel https://review.openstack.org/110285 | 19:07 |
| *** hashar has joined #openstack-infra | 19:08 | |
| *** craigbr has joined #openstack-infra | 19:09 | |
| *** esker has quit IRC | 19:10 | |
| *** tnovacik has quit IRC | 19:12 | |
| openstackgerrit | Ken Giusti proposed a change to openstack-infra/config: Add the Apache Qpid AMQP 1.0 messaging libraries to all slaves. https://review.openstack.org/110431 | 19:13 |
| *** shashankhegde has quit IRC | 19:14 | |
| *** hashar has quit IRC | 19:14 | |
| *** mwagner_lap has joined #openstack-infra | 19:15 | |
| *** esker has joined #openstack-infra | 19:17 | |
| *** esker has quit IRC | 19:18 | |
| *** masayukig has quit IRC | 19:21 | |
| pcm_ | anteaya: ping | 19:21 |
| *** chandankumar has quit IRC | 19:24 | |
| *** kgiusti has joined #openstack-infra | 19:25 | |
| *** flaviof_zzz has joined #openstack-infra | 19:26 | |
| *** sarob has joined #openstack-infra | 19:26 | |
| *** masayukig has joined #openstack-infra | 19:27 | |
| *** andreykurilin has joined #openstack-infra | 19:28 | |
| *** rainya has quit IRC | 19:28 | |
| *** rainya has joined #openstack-infra | 19:29 | |
| *** derekh_ has joined #openstack-infra | 19:30 | |
| *** bauzas has quit IRC | 19:36 | |
| *** amotoki has quit IRC | 19:38 | |
| *** unicell has quit IRC | 19:38 | |
| anteaya | pcm_: ahoy | 19:41 |
| *** holden247 has quit IRC | 19:42 | |
| *** banix has joined #openstack-infra | 19:43 | |
| yolanda | hi anteaya, i used the template branch | 19:43 |
| *** dims has quit IRC | 19:47 | |
| anteaya | yolanda: ah | 19:48 |
| *** openstackgerrit has quit IRC | 19:48 | |
| *** bauzas has joined #openstack-infra | 19:49 | |
| *** otherwiseguy has quit IRC | 19:51 | |
| *** lcheng_ has joined #openstack-infra | 19:52 | |
| *** otherwiseguy has joined #openstack-infra | 19:52 | |
| yolanda | anteaya, something wrong there? | 19:54 |
| anteaya | yolanda: not at all, I have to create a slide deck | 19:55 |
| *** ramashri has quit IRC | 19:55 | |
| anteaya | and wanted to copy you | 19:55 |
| anteaya | and didn't know about the template branch until you told me, so thank you | 19:55 |
| *** arnaud has joined #openstack-infra | 19:56 | |
| *** markmcclain1 has quit IRC | 19:59 | |
| *** markmcclain has joined #openstack-infra | 20:00 | |
| *** amuller has quit IRC | 20:00 | |
| fungi | anteaya: you probably want to read the readme in the master branch | 20:01 |
| *** mikal_ is now known as mikal | 20:01 | |
| fungi | oh, actually it doesn't mention the template branch | 20:01 |
| anteaya | fungi: thanks, I can do that again as well | 20:02 |
| anteaya | oh, well it is possible I read things and miss stuff | 20:02 |
| *** sarob__ has joined #openstack-infra | 20:02 | |
| fungi | yeah, there' | 20:02 |
| fungi | s a good todo item | 20:02 |
| mordred | dguerri: yeah - we didn't make it through the whole agenda - but we can certainly pick things up in here | 20:03 |
| *** jp_at_hp has quit IRC | 20:03 | |
| dguerri | cool | 20:03 |
| dguerri | thanks | 20:03 |
| *** e0ne has quit IRC | 20:03 | |
| dguerri | I think there are a couple of bullets before mine, though | 20:03 |
| fungi | anteaya: so yeah, http://git.openstack.org/cgit/openstack-infra/publications/tree/README.rst?h=template is what i was remembering, but i agree that's not discoverable | 20:04 |
| jeblair | they're also at the top of the agenda for next week | 20:04 |
| anteaya | fungi: thanks | 20:04 |
| dguerri | anteaya, about https://review.openstack.org/109292 | 20:05 |
| dguerri | This is about adding a new tool to Stackforge | 20:05 |
| anteaya | dguerri: I'm on really bad wifi on a plane | 20:06 |
| *** jistr has quit IRC | 20:06 | |
| *** sarob__ has quit IRC | 20:07 | |
| anteaya | just state what you want to accomplish so others can offere their thoughts and they will do so as they are able | 20:07 |
| clarkb | krotscheck: can you see my comment in 109069 | 20:07 |
| clarkb | krotscheck: if you think that isn't necessary I can be convinced to approve as is | 20:07 |
| anteaya | since some are in the next meeting currently in progress | 20:07 |
| yolanda | clarkb, looking at your reviews, problem is that there are 2 changes that should be together: | 20:07 |
| dguerri | anteaya ack | 20:07 |
| clarkb | krotscheck: mostly worried we may sneak in new stuff between options checking and skipping a project | 20:07 |
| clarkb | yolanda: if key in dict: is functionall equivalent to if key in dict.keys() right? | 20:08 |
| yolanda | https://review.openstack.org/109957 and https://review.openstack.org/110009 | 20:08 |
| dguerri | so, as I said this is about including a new tool in Stackforge. Tool name is git-upstream | 20:08 |
| krotscheck | clarkb: Safety third! Gimme a sec, I’ll update it. | 20:08 |
| clarkb | yolanda: oh that | 20:08 |
| dguerri | basically git-upstream can be used to keep in sync a mirror with local patches with upstream OpenStack repos. It has been used for a while in HP and a some time ago we decided to polish it and release it as FOSS (Apache 2.0 license) | 20:08 |
| clarkb | yolanda: ok I see now | 20:08 |
| yolanda | clarkb, not now, because i added description there | 20:08 |
| dguerri | There has been a concern about the name of the tool (anteaya) as it could create confusion with git-review. | 20:08 |
| clarkb | yolanda: but its still project: description | 20:08 |
| dguerri | After a discussion with anteaya here, we agreed to add a clear statement about the main use case of git-upstream (in the documentation for the tool itself and in commit message for the above review). | 20:09 |
| clarkb | yolanda: ok I will need to review the gerritlib change | 20:09 |
| yolanda | clarkb, before that it was a list, [project1, project2, project3} | 20:09 |
| yolanda | now it's {project1:desc1, project2:desc2} | 20:09 |
| dguerri | So, who could have a look to this review? https://review.openstack.org/109292 :) | 20:09 |
| dguerri | please | 20:10 |
| clarkb | yolanda: ya but in python if you say if key in dict: that checks if the key is in the dict right? | 20:10 |
| *** _nadya_ has quit IRC | 20:10 | |
| yolanda | clarkb, yes, is that faster? | 20:10 |
| krotscheck | clarkb: Updated, better? https://review.openstack.org/#/c/109069/3/jeepyb/cmd/manage_projects.py | 20:10 |
| yolanda | i don't mind changing that, sure | 20:10 |
| anteaya | dguerri: I think the question you need to ask is if anyone feels there will be confusion betwen your tool and git-review given your new toold description | 20:10 |
| clarkb | yolanda: yup because it uses the hash rather than searching a list | 20:11 |
| dguerri | anteaya I mentioned that | 20:11 |
| clarkb | krotscheck: yup thanks | 20:11 |
| yolanda | clarkb, ok, i can just change it, just let me know your feedback also about the other parts of the change | 20:11 |
| anteaya | I think your new description does a good job differenciating itself but I would like to hear the opinons of others | 20:11 |
| dguerri | yes, that's the main question, thanks :) | 20:11 |
| clarkb | yolanda: ya I need t oreview gerritlib first. I missed that | 20:11 |
| yolanda | clarkb, i wanted to add some dependency between each project but i didn't find an easy way, maybe to file a comment on it as anteaya suggested | 20:12 |
| yolanda | is there any other way on gerrit to declare dependencies between projects? | 20:12 |
| dguerri | please keep in mind that even if we would love to see this tool in openstack-infra, we are proposing it to Stackforge atm. | 20:12 |
| *** derekh_ has quit IRC | 20:12 | |
| jesusaurus | yolanda: no :( the best you can do is comment about it in the commit message | 20:12 |
| fungi | yolanda: usually i explain cross-project change sequencing in the commit messages | 20:13 |
| *** amalagon has quit IRC | 20:13 | |
| yolanda | fungi, jesusaurus, not filing an inline comment but adding that into the commit message itself? | 20:13 |
| fungi | yolanda: yes | 20:14 |
| jesusaurus | yolanda: yeah, thats what i do | 20:14 |
| yolanda | one new thing learned :) | 20:14 |
| fungi | yolanda: it's also helpful information to end up in the commit log because it helps people who may be continuously deploying those projects know about the lockstep upgrade need | 20:14 |
| *** _nadya_ has joined #openstack-infra | 20:15 | |
| *** ramashri has joined #openstack-infra | 20:15 | |
| yolanda | fungi, makes sense | 20:15 |
| fungi | dguerri: my biggest concern with it being in openstack-infra is that it effectively encourages downstreams to carry forks of openstack projects (though your more recent readme does a good job of reinforcing upstream contribution of changes and only using it for short-term divergence) | 20:16 |
| fungi | dguerri: i'm far less concerned about it being in stackforge however | 20:16 |
| dguerri | fungi: we are convinced that git-upstream will encourage people to stay up-to-date, avoiding forks. | 20:16 |
| dguerri | that is the way we are using it | 20:17 |
| dguerri | and that is the reason why it was developed | 20:17 |
| fungi | dguerri: just keep in mind that we have lots and lots of changes to review and are even further behind than usual, and changes adding projects to stackforge are a low review priority for us (we need to prioritize keeping openstack running), so four days in review is actually not that long | 20:17 |
| ianw | dguerri: feedback from someone who had never heard of this till now -- a small sample cli session in the README.md for all those 7 bullet points might help | 20:18 |
| *** _nadya_ has joined #openstack-infra | 20:19 | |
| *** dims has joined #openstack-infra | 20:19 | |
| dguerri | fungi: absolutely, no problem with that... I am raising the issue here because of the anteaya's concern | 20:19 |
| anteaya | dguerri: which I thank you for you are doing a great job | 20:20 |
| dguerri | ianw: good idea. There is a CLI session in USAGE.md I will add an example for each bullets | 20:20 |
| dguerri | anteaya thanks :) | 20:20 |
| *** _nadya_ has quit IRC | 20:20 | |
| yolanda | so clarkb, i updated that, and updated commit message to reflect dependency on gerritlib | 20:21 |
| clarkb | yolanda: thank you | 20:22 |
| *** david-lyle has quit IRC | 20:22 | |
| *** david-lyle has joined #openstack-infra | 20:22 | |
| *** rlandy has quit IRC | 20:22 | |
| *** dhellman_ has joined #openstack-infra | 20:22 | |
| dguerri | fungi: a misuse is possible for every tool (even git itself, of course). Things could be perceived worse for git-upstream as it contains the phrase "locally carried patches" | 20:24 |
| dguerri | fungi: but I think that for big companies that are using openstack those are unavoidable.. and are not about trade secrets or the like | 20:26 |
| fungi | dguerri: yep. as i said it's not as much of a concern for me now, but hp had a history of running old and heavily patched openstack, so when the tool you use to make that anti-pattern possible was initially proposed as something we could maintain within openstack-infra, i was skeptical | 20:26 |
| dguerri | fungi exactly! | 20:27 |
| *** david-lyle has quit IRC | 20:27 | |
| dguerri | fungi: we are now using git-upstrema to avoid that | 20:27 |
| dguerri | it has been created exactly for that readon | 20:27 |
| dguerri | *reason | 20:27 |
| *** david-lyle has joined #openstack-infra | 20:27 | |
| dguerri | anyonw has concern about the name of the tool? | 20:27 |
| fungi | yeah, the tool seems fine to me. i'm still unconvinced that it's a fit for being an openstack-infra project, but it's perfectly okay to have something like that in stackforge i think | 20:28 |
| *** david-lyle has quit IRC | 20:28 | |
| *** david-lyle has joined #openstack-infra | 20:29 | |
| asselin | hi, not sure what happened, but I can no longer stop zuul. any tips? No process in pidfile '/var/run/zuul/zuul.pid' found running; none killed. | 20:29 |
| yolanda | anteaya, thx to you for all the help :) | 20:29 |
| asselin | even after rebooting | 20:29 |
| yolanda | i just listed the branches and discovered that, maybe we could file that in some place about "How to create publications2 | 20:30 |
| yolanda | "How to create publications" | 20:30 |
| *** lnxnut has quit IRC | 20:30 | |
| *** rainya has quit IRC | 20:30 | |
| anteaya | yolanda: my pleasure, thanks for your persistence | 20:30 |
| fungi | yolanda: i agree the readme in the template branch is great and also not very discoverable | 20:31 |
| anteaya | yolanda: something to helpf folks like me | 20:31 |
| yolanda | i also had some valuable help from pleia2, because i started doing the slides in a powerpoint | 20:31 |
| *** flaviof_zzz is now known as flaviof | 20:31 | |
| *** rainya has joined #openstack-infra | 20:31 | |
| fungi | yolanda: the readme in the master branch should probably direct people to the template branch | 20:31 |
| anteaya | ah | 20:31 |
| ianw | dguerri: only concern is that "upstream" is the main remote tracking branch used to track, well, upstream -- as suggested by https://help.github.com/articles/fork-a-repo (which i assume is a big reference in general) | 20:31 |
| *** zns has quit IRC | 20:31 | |
| ianw | so you've got "git pull upstream master" and "git upstream pull" ... | 20:32 |
| ianw | gets confusing | 20:32 |
| *** zns has joined #openstack-infra | 20:32 | |
| yolanda | anteaya, as all the slides i did in the past were done using PowerPoint, in my mind i was thinking on some power point template that was exported in some way to the final html | 20:32 |
| *** rainya has quit IRC | 20:33 | |
| *** lcheng_ has quit IRC | 20:33 | |
| *** shashankhegde has joined #openstack-infra | 20:34 | |
| *** rainya has joined #openstack-infra | 20:34 | |
| ianw | dguerri: maybe if something abstract, a play on quilt ... it seems to be angled as a more git-centric version of that | 20:35 |
| anteaya | yolanda: ah | 20:36 |
| * anteaya wants an infra-core world tour t-shirt | 20:37 | |
| anteaya | the fanboi size | 20:37 |
| clarkb | yolanda: ok reviewed the gerritlib change | 20:37 |
| *** kgiusti has quit IRC | 20:37 | |
| *** SumitNaiksatam has quit IRC | 20:37 | |
| clarkb | yolanda: can you see my comment and let me know what you think? | 20:37 |
| pleia2 | fungi: I have a review outstanding with those instructions | 20:37 |
| fungi | pleia2: excellent! i've no doubt i will run across it soon ;) | 20:38 |
| pleia2 | https://review.openstack.org/#/c/98906/ | 20:38 |
| pleia2 | :) | 20:38 |
| dguerri | ianw: I think the main issue about the name was for the git- prefix which is needed by any git subcommand | 20:38 |
| anteaya | pleia2: you should be resting | 20:38 |
| pleia2 | jeblair: re: away - I'll probably be back half day on thursday, post surgery pain still making it tricky to be at computer for more than a little bit | 20:38 |
| pleia2 | anteaya: yes :) | 20:38 |
| dguerri | ianw: upstream could be confusing for a novice... but definitely not for the target audience of these kind of tools | 20:38 |
| fungi | pleia2: stop computering | 20:38 |
| jeblair | clarkb: i got a bunch of emails from rax saying the hosts that logstash 06,16,14 are on ar having problems | 20:39 |
| *** dmsimard is now known as dmsimard_away | 20:39 | |
| anteaya | pleia2: it isn't even a week yet | 20:39 |
| fungi | jeblair: clarkb: mordred: SergeyLukjanov: jhesketh: i was going to raise the question briefly in today's meeting if we got time, but any opinions on whether we should take a maintenance window and ask rackspace to migrate jenkins05 to a new host? | 20:39 |
| clarkb | jeblair: cool. anything I can do to help with that? | 20:39 |
| pleia2 | right right, back to resting! :) | 20:39 |
| jeblair | pleia2: yes, rest! no rush! | 20:39 |
| clarkb | fungi: Ithink we can do that. its relatively easy to take a node out of hte rotation | 20:39 |
| fungi | jeblair: any possibility that's the same host jenkins05 was having issues on? | 20:39 |
| clarkb | fungi: and if we get more reliablity then +2 | 20:39 |
| fungi | jeblair: did they elaborate on what kind? | 20:40 |
| clarkb | oh that would be interseting | 20:40 |
| jeblair | fungi: no and no | 20:40 |
| fungi | poop | 20:40 |
| clarkb | oh it does have missing graphs in cacti | 20:40 |
| clarkb | whcih could be network tuoble | 20:40 |
| yolanda | clarkb, i did the conditional because other values could be boolean or numeric types | 20:40 |
| clarkb | which could be the same issue | 20:40 |
| jeblair | "At this time we have not yet been able to restore access to your cloud server 'logstash-worker14.openstack.org'." | 20:40 |
| *** eharney has quit IRC | 20:40 | |
| clarkb | yolanda: oh hrm | 20:40 |
| jeblair | sounds serious though | 20:40 |
| fungi | ouch | 20:40 |
| clarkb | jeblair: fwiw we can nuke those nodes and start over if it helps | 20:40 |
| clarkb | jeblair: there is no real data on them | 20:41 |
| fungi | unlike the elasticsearch nodes | 20:41 |
| clarkb | and even then its the volumes that matter | 20:41 |
| yolanda | i think description is the only string value that api supports, let me double check | 20:41 |
| ianw | dguerri: right, just saying "upstream" is a pretty big name-space to use, since its such a common term. also i know i've written similar tools, so there's lots of approaches (we used to have this weird collection of sub-modules, etc) | 20:41 |
| fungi | point | 20:41 |
| clarkb | yolanda: ok I will +2 | 20:41 |
| clarkb | yolanda: if that becomes a problem we can worry about it then | 20:41 |
| yolanda | cool | 20:41 |
| fungi | on jenkins05, i'll go ahead and find out what sort of turnaround time is feasible and then plan to put it into a graceful shutdown well in advance of the scheduled migration window | 20:42 |
| yolanda | clarkb, https://gerrit-documentation.googlecode.com/svn/Documentation/2.6/cmd-set-project.html . So description is the only value that subject to have quotes there | 20:42 |
| dguerri | ianw ack... I will think about that :) | 20:42 |
| dguerri | ianw: thanks for the feedback | 20:43 |
| clarkb | yolanda: thanks | 20:43 |
| yolanda | clarkb, fungi, so i was expecting that when jeepyb changes are merged, if gerritlib change is not merged tests will be failing, but it would depend on the nature of the tests taken | 20:44 |
| yolanda | can we ensure that gerritlib changes are merged before jeepyb to don't break anything? | 20:44 |
| *** dguerri is now known as _dguerri | 20:45 | |
| *** _dguerri is now known as dguerri`afk | 20:45 | |
| *** pblaho has quit IRC | 20:46 | |
| fungi | yolanda: yes, as long as the commit message states that dependency (which it sounds like it does) then whoever is babysitting the approval on those changes should hopefully at least check puppetdb.openstack.org between them to ensure the first one is applied before approving the second | 20:46 |
| *** juzuluag has joined #openstack-infra | 20:47 | |
| yolanda | fungi, yes, i commented the dependency on jeepyb changes of gerritlib | 20:48 |
| *** Sukhdev has quit IRC | 20:48 | |
| *** juzuluag has quit IRC | 20:48 | |
| JayF | openstackgerrit appears to be gone | 20:53 |
| fungi | JayF: i'll hunt it down and correct | 20:53 |
| yolanda | pleia2, i hope you recover well | 20:54 |
| JayF | Thanks fungi | 20:54 |
| *** lukego has joined #openstack-infra | 20:55 | |
| *** openstackgerrit has joined #openstack-infra | 20:59 | |
| *** dkranz has quit IRC | 21:02 | |
| jeblair | fungi: ticket about jenkins05 just updated | 21:02 |
| jeblair | not urgent, as it turns out | 21:02 |
| SergeyLukjanov | jeblair, fungi , clarkb, mordred, folks, I'm sorry - I've been unable to attend today's infra meeting | 21:03 |
| jeblair | fungi: so it sounds like we don't _need_ to migrate 05? do you still want to anyway? | 21:03 |
| *** sarob_ has joined #openstack-infra | 21:03 | |
| anteaya | SergeyLukjanov: everything okay? | 21:03 |
| anteaya | SergeyLukjanov: or are you finally needed to actually get some sleep in your life? | 21:03 |
| anteaya | the amount you work is astounding | 21:04 |
| SergeyLukjanov | anteaya, yup, everything is ok, lack of time and a bunch of housekeeping duties ;) | 21:04 |
| anteaya | SergeyLukjanov: how are the renovations coming along? | 21:05 |
| SergeyLukjanov | anteaya, we've finally moved to the new place last wednesday | 21:05 |
| *** david-lyle has quit IRC | 21:05 | |
| anteaya | yay!! | 21:05 |
| anteaya | how is it? | 21:05 |
| SergeyLukjanov | anteaya, it's awesome, especcially after the long time of making it from scratch :) | 21:06 |
| *** david-lyle has joined #openstack-infra | 21:06 | |
| SergeyLukjanov | especially* | 21:06 |
| fungi | jeblair: my only concern is that they basically identified the symptom on the host but didn't imply they knew the cause and since it's happened twice now it could presumably happen again | 21:06 |
| anteaya | SergeyLukjanov: woohooo | 21:07 |
| anteaya | house party at your place next time we are all in russia | 21:07 |
| arosen | Hi, I'm following jaypipes blog here ( http://www.mirantis.com/blog/setting-external-openstack-testing-system-part-1/) . When i submit a patch to review.openstack.org to the sandbox repo I see the following message in /var/log/zuul/debug.log | 21:07 |
| SergeyLukjanov | anteaya, sure, but I'm afraid that it'll be not so soon as I want :) | 21:07 |
| anteaya | where are you, moscow? st petersburg? | 21:07 |
| arosen | but the results never post bacK: WARNING zuul.Scheduler: Project openstack-dev/sandbox not found | 21:07 |
| anteaya | SergeyLukjanov: yeah, I hear that | 21:07 |
| *** aysyd has quit IRC | 21:07 | |
| SergeyLukjanov | anteaya, saratov, it's a much smaller city :) | 21:07 |
| jeblair | fungi: okay, works for me | 21:07 |
| *** weshay has quit IRC | 21:07 | |
| *** sarob_ has quit IRC | 21:07 | |
| anteaya | SergeyLukjanov: ah, I will have to look that up | 21:08 |
| fungi | jeblair: and given the current state of our tools, when a jenkins master disappears off the network for some period of time, we end up with stuck jobs in zuul, nodepool going crazy allocating ready nodes which it doesn't know are getting used (once the server reappears), et cetera | 21:08 |
| arosen | I was wondering if anyone might know off hand. I saw that someone else had this issue on the mailing list but they never reached a solution. | 21:08 |
| fungi | jeblair: so unfortunately somewhat disruptive | 21:08 |
| anteaya | arosen: so yesterday at the third party meeting joa linked us to his notes he created: https://etherpad.openstack.org/p/Setting_up_a_3rd-party_CI___Issues_encountered | 21:09 |
| anteaya | arosen: about setting up a ci using jaypipes notes | 21:09 |
| *** zul has quit IRC | 21:09 | |
| anteaya | arosen: something might be helpful there | 21:10 |
| clarkb | anteaya: is 106495 the multinode spec you wanted peopel to read? | 21:10 |
| *** david-lyle has quit IRC | 21:10 | |
| anteaya | clarkb: yes | 21:10 |
| clarkb | ok | 21:10 |
| *** david-lyle has joined #openstack-infra | 21:11 | |
| clarkb | I think maybe we should force those people to talk to sdague and afazekas | 21:11 |
| clarkb | because I think 95% of the work is done | 21:11 |
| anteaya | clarkb: awesome | 21:11 |
| clarkb | and just make sure everyone is still on the same page | 21:11 |
| anteaya | and yes having a meeting would be good | 21:11 |
| clarkb | I also would like to put a vote in the hat for not requiring specs for new types of testing | 21:11 |
| clarkb | at least not in infra | 21:11 |
| *** zul has joined #openstack-infra | 21:11 | |
| *** flaper87 is now known as flaper87|afk | 21:12 | |
| clarkb | we supply the nodes, what and how people test things is up to them IMO | 21:12 |
| anteaya | I mentioned at the neutron meeting yesterday and they were keen to know progress | 21:12 |
| anteaya | perhaps they should send a delgate to the -qa meeting? | 21:12 |
| clarkb | anteaya: there is a d-g change up that passes nova network and neutron currenktly | 21:12 |
| clarkb | it needs refinement but its working | 21:12 |
| anteaya | really? | 21:12 |
| clarkb | yup | 21:12 |
| anteaya | okay well I am the one that insisted they write the spec | 21:13 |
| sdague | anteaya: yeh, it's passing neutron and n-net in the last rev | 21:13 |
| anteaya | believeing it would seed the qa/infra mulitnode work, and I was wrong | 21:13 |
| *** yamamoto has joined #openstack-infra | 21:13 | |
| clarkb | anteaya: well i think it may be a good exercise for them to do as maybe a neutron spec | 21:13 |
| anteaya | so this is on me, my fault | 21:13 |
| anteaya | sure, so do comment, knowing this was me driving them to create an infra-spec | 21:14 |
| *** yamamoto has quit IRC | 21:14 | |
| anteaya | sdague: yay | 21:14 |
| clarkb | an exception may be dhellmann's spec to do cross project testing because well cross project makes stuff weird | 21:14 |
| *** yamamoto has joined #openstack-infra | 21:14 | |
| anteaya | sdague: should I send them to your meeting on thursday? | 21:14 |
| openstackgerrit | Nicholas Randon proposed a change to openstack-infra/config: Add firewall traffic on port 27410 to 27510. https://review.openstack.org/108737 | 21:14 |
| *** david-lyle has quit IRC | 21:15 | |
| sdague | anteaya: I'm likely not to be in the meeting as on the road | 21:15 |
| * anteaya notes she is at half power on her laptop battery | 21:15 | |
| *** dguerri`afk is now known as _dguerri | 21:15 | |
| anteaya | sdague: kk | 21:15 |
| anteaya | I'll just ask mestery to read a bit of backscroll | 21:16 |
| anteaya | mestery: the last few minutes | 21:16 |
| * mestery lurks while in #openstack-meeting in the project meeting | 21:16 | |
| mestery | anteaya: got it | 21:16 |
| *** eharney has joined #openstack-infra | 21:17 | |
| *** dustins has quit IRC | 21:17 | |
| mestery | anteaya clarkb: If this is about multi-node, I'll plan to attend the QA meeting for discussions. | 21:17 |
| clarkb | mestery: it is, tl;dr is afazekas has it working | 21:18 |
| clarkb | mestery: we need to clean it up but its mostly there | 21:18 |
| mestery | clarkb: Yay! Very cool! | 21:18 |
| *** cdub has quit IRC | 21:18 | |
| *** mmaglana has joined #openstack-infra | 21:20 | |
| *** Ajaeger1 has quit IRC | 21:21 | |
| *** yamamoto has quit IRC | 21:22 | |
| clarkb | dhellmann: I have a comment on your spec for cross project unittests that I would like your thoughts on but I am +2 | 21:23 |
| *** dkliban has quit IRC | 21:23 | |
| openstackgerrit | A change was merged to openstack-infra/gerritlib: Update pbr version https://review.openstack.org/103320 | 21:23 |
| mmaglana | hi. i'd like to remove aviator from stackforge since there's a crucial part of its CI workflow that i haven't set up yet and the repo in stackforge is no longer in-sync with https://github.com/aviator/aviator. I don't know when I'll be able to work on the missing piece so, to avoid any confusion, i'd like to remove the one in stackforge for now. what are the steps i need to take? | 21:23 |
| anteaya | arosen: okay so on the process to request a gerrit ci account, this patch is up and hasn't merged yet: https://review.openstack.org/#/c/101013/ | 21:23 |
| anteaya | clarkb: feel like a +A? https://review.openstack.org/#/c/101013/ | 21:24 |
| anteaya | would be nice to get this in | 21:24 |
| anteaya | this one isn't causing issues | 21:24 |
| arosen | anteaya: okay so you want me to add this to my email i sent? | 21:24 |
| clarkb | anteaya: basically making it eaiser to construct the name? | 21:24 |
| clarkb | mmaglana: that is a tricky question. we don't remove things, but do have an "attic" for stuff | 21:25 |
| arosen | anteaya: I'm using my normal gerrit account right now till the ci one gets approved. | 21:25 |
| clarkb | mmaglana: usually thoguh if the intent is to go back to using a thing I would argue that the repo be left as is | 21:25 |
| anteaya | arosen: sure, just state the company name for the account and what you are verifying, bluebprint links are great | 21:25 |
| anteaya | clarkb: yes | 21:25 |
| anteaya | arosen: understood | 21:25 |
| redrobot | SergeyLukjanov ping | 21:26 |
| mmaglana | clarkb: i see. i just want to make sure that there's no confusion as to which one is the de facto upstream repo while aviator repo is still in this transition phase. | 21:26 |
| *** lukego has quit IRC | 21:26 | |
| anteaya | mmaglana: do you need support for the ci workflow? you are welcome to attend the third party meetings | 21:26 |
| jesusaurus | zaro: you around? im confused about project ownership in gerrit. is the Project Owners group special somehow? how does gerrit know what group owns a project? | 21:26 |
| clarkb | mmaglana: well there shouldn't be a transition phase :) once you are on strackforge that is what you should use :P | 21:26 |
| clarkb | redrobot: do you have a specific question? | 21:27 |
| mmaglana | clarkb: :) i haven't been RTFMing properly. | 21:27 |
| clarkb | I think SergeyLukjanov is afkfor the night | 21:27 |
| SergeyLukjanov | clarkb, not yet) | 21:27 |
| SergeyLukjanov | redrobot, yup, I'm here | 21:27 |
| clarkb | SergeyLukjanov: oh hey tehre you are | 21:27 |
| redrobot | woot! | 21:27 |
| SergeyLukjanov | clarkb, /me on the project meeting | 21:27 |
| mmaglana | anteaya: what's missing in my workflow is a service that i have to set-up to run aviator tests in travis ci. | 21:27 |
| redrobot | SergeyLukjanov do you have a chance to look at https://review.openstack.org/#/c/109398/ again? | 21:28 |
| clarkb | mmaglana: are you trying to use travis with stackforge? I think the idea is you would use our CI system | 21:28 |
| redrobot | SergeyLukjanov I think you missed that the channel in question is already in the file you requested | 21:28 |
| *** shashankhegde has quit IRC | 21:29 | |
| fungi | jesusaurus: "project owner" is a group setting on a project, defaulting to the "administrators" group, and the "project owners" meta-group is basically an acl hook for "whatever group owns the project in question" | 21:29 |
| mordred | mmaglana: yah - travis is way less featureful than openstack ci ... and the main benefit to hosting with us is getting to use our CI system | 21:30 |
| *** shashankhegde has joined #openstack-infra | 21:30 | |
| mmaglana | clarkb: mordred: i figured that since the tests aren't really hitting an openstack instance (it uses cassettes made by the VCR gem), i would be needlessly running load on openstack ci. | 21:31 |
| *** _dguerri is now known as dguerri`afk | 21:31 | |
| jesusaurus | fungi: ah, so there is a permission called owner that you set to tell gerrit which group is the owner group? | 21:31 |
| fungi | yep | 21:32 |
| jesusaurus | cool, thanks | 21:32 |
| mordred | mmaglana: that's fine - the likelihood of you stressing us out is low ... and ultimately having a stackforge project get tested by travis and not us would be SUPER confusing to people | 21:32 |
| clarkb | jesusaurus: keep in mind owner gets to do lots of stuff | 21:32 |
| mordred | mmaglana: although I do appreciate you thinking of us in that way :) | 21:33 |
| clarkb | jesusaurus: so you may want to double check all of the various things that lets you do | 21:33 |
| mordred | mmaglana: you've also already got tests defined in our system that look to be about the same as your travis.yaml | 21:34 |
| SergeyLukjanov | redrobot, see it now, replaced with -2 | 21:34 |
| SergeyLukjanov | I mean +2 ;) redrobot | 21:34 |
| mordred | all that really needs done is something similar to your rvm matrix - but we do similar thigns for other projects | 21:34 |
| redrobot | SergeyLukjanov hehe... sounds good. thanks! | 21:35 |
| *** amalagon has joined #openstack-infra | 21:35 | |
| fungi | jesusaurus: though also don't confuse the "project owners" meta-group (which refers back to the owner group configured for a project) with the "owner" acl permission (which can be ref pattern specific) | 21:35 |
| mordred | mmaglana: so adding an rvm version matrix thing would likely be something that would benefit others too | 21:35 |
| mmaglana | mordred: i see. in that case, i'm going to need help later on with two things 1) how to set up the CI such that I can run parallel tests each running a different version of Ruby (1.8.7, 1.9.x, 2.0.x, 2.1.x) and 2) how to sync the stackforge aviator with the latest without all of the commits going through gerrit. i wouldn't be able to work on these right away, but when i have time, who should i contact? | 21:35 |
| mordred | mmaglana: any of us - just ping us in here ... I'd love to get your test matrix thing going - should be a piece of cake | 21:36 |
| fungi | mmaglana: how many commits? you could just propose them all as a patch series and then approve them, unless it's like dozens and dozens | 21:36 |
| mordred | mmaglana: as for the other thing, whenever you're ready to sync for real, just let us know | 21:36 |
| *** daya_k has joined #openstack-infra | 21:37 | |
| jesusaurus | clarkb: is there a specific big bad perm in particular you have in mind? | 21:37 |
| mmaglana | nice! thanks mordred. fungi: i think it's about 2~3 dozens. | 21:38 |
| *** lcheng_ has joined #openstack-infra | 21:38 | |
| jesusaurus | fungi: oh, if the owner acl permission is different, then where is the project owners permission set? | 21:38 |
| anteaya | arosen: thanks | 21:38 |
| clarkb | jesusaurus: https://review.openstack.org/Documentation/access-control.html#category_owner the big one is editing alcs | 21:38 |
| clarkb | jesusaurus: which we try to take away from people and push through review | 21:39 |
| clarkb | but create and delete branches may be another issue depending on situation | 21:39 |
| *** andreykurilin has quit IRC | 21:39 | |
| *** mwagner_lap has quit IRC | 21:39 | |
| *** stevebaker is now known as steveburt | 21:41 | |
| *** steveburt is now known as stevebaker | 21:41 | |
| anteaya | so from my understanding, once this patch is merged zaro wanted to discuss updating gerrit so that I have access to enhanced permissions for third party ci accounts: https://review.openstack.org/#/c/109797/ | 21:42 |
| anteaya | I hope I'm not putting words in zaro's mouth | 21:42 |
| fungi | jesusaurus: and https://review.openstack.org/Documentation/access-control.html#project_owners | 21:43 |
| anteaya | but also wanted to d to get some thoughts on timing | 21:43 |
| arosen | jaypipes: if you get a sec I'm having trouble getting jenkins to vote back on sandbox. I see events comming in though to sudo tail -f /var/log/zuul/debug.log | grep sandbox | 21:44 |
| arosen | 21:44 | |
| jeblair | anteaya: when it lands we should be able to upgrade review-dev, make sure it, works, then upgrade review | 21:44 |
| anteaya | kk, thanks | 21:44 |
| jeblair | anteaya: the downtime will be minimal, so we don't really have to worry about it too much. maybe do it on a friday afternoon or something | 21:45 |
| anteaya | great | 21:45 |
| jeblair | (we don't usually announce downtimes for things that are just service restarts, which is what this would be) | 21:45 |
| anteaya | glad this will be just a small plip | 21:46 |
| jaypipes | arosen: check the zuul-merger.log. | 21:46 |
| jaypipes | arosen: anything in there? | 21:46 |
| *** zaneb has joined #openstack-infra | 21:46 | |
| anteaya | feels like we are coming into land | 21:47 |
| anteaya | back in a bit | 21:47 |
| arosen | jaypipes: nope I don't have that log file created. | 21:47 |
| *** zul has quit IRC | 21:48 | |
| zaneb | I have a patch where Jenkins started the gate jobs on Sunday but there's no sign of it in the Zuul queue: https://review.openstack.org/#/c/106540/ | 21:48 |
| zaneb | any clues on how to kick it? | 21:48 |
| arosen | jaypipes: i also copied all these files you have here to /etc/ on my system https://github.com/jaypipes/os-ext-testing-data/tree/master/etc | 21:48 |
| zaneb | "reverify no bug" and reapprove did not help | 21:49 |
| jesusaurus | clarkb: fungi: thanks for the links | 21:49 |
| openstackgerrit | A change was merged to openstack-infra/config: Edits Third Party CI account request directions https://review.openstack.org/101013 | 21:49 |
| *** zul has joined #openstack-infra | 21:50 | |
| *** ramashri has quit IRC | 21:51 | |
| fungi | zaneb: that's https://storyboard.openstack.org/#!/story/147 | 21:53 |
| fungi | zaneb: leave a "recheck no bug" comment | 21:53 |
| jeblair | zaro: 91193 has 2x+2s; please ping an infra-core for approval when you are around and ready to monitor it | 21:53 |
| fungi | zaneb: i have a patch in zuul as of today which should alleviate this and help us pinpoint the underlying cause the next time we restart | 21:53 |
| zaneb | fungi: I left "reverify no bug" and nothing happened... does that not work any more? | 21:53 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 21:54 | |
| fungi | zaneb: reverify has needed a bug number since ages, but also reverify only works of there is an existing "jenkins" vote on your change | 21:54 |
| zaneb | ah, ok | 21:55 |
| *** signed8bit has quit IRC | 21:55 | |
| zaneb | there was a Jenkins vote at one point, but it got cleared when the gate jobs started | 21:56 |
| zaneb | fungi: so basically just never use "reverify" any more? | 21:56 |
| fungi | zaneb: basically that change got ejected from the gate when its parent failed to merge, zuul didn't leave a jenkins vote because of story #147, and the child change was subsequently not automatically reenqueued when the parent was put back in because there was now no vote on it | 21:56 |
| sdague | infra question, would there be objection to exposing the Rebase button for general use in gerrit? | 21:57 |
| *** zns_ has joined #openstack-infra | 21:57 | |
| sdague | and not just for patch owners | 21:57 |
| fungi | zaneb: "reverify NNNNNN" where NNNNNN is a bug number and there is an existing jenkins vote will work fine | 21:57 |
| zaneb | fungi: gotcha, thanks :) | 21:57 |
| *** dhellman_ has quit IRC | 21:57 | |
| fungi | zaneb: there is just effectively a hole in the enqueuing configuration right now where reverify and reapplying workflow +1 don't work if there's not a current jenkins vote at all (positive or negative) | 21:58 |
| fungi | however recheck works even if there is no current jenkins vote on the change | 21:58 |
| *** alexandra_ is now known as asettle | 21:59 | |
| *** zns has quit IRC | 21:59 | |
| jeblair | krotscheck: i may not have been clear before. we do not have a policy that just because you are core on a project that we run, you are granted superuser access to a running instance of that project | 21:59 |
| zaneb | am I right in thinking that recheck has basically the same effect now anyway? or is reverify still quicker when it works? | 21:59 |
| sdague | zaneb: they are basically the same | 22:00 |
| jeblair | krotscheck: instead, our policy is to be rather conservative in granting superuser access | 22:00 |
| *** chuckC has joined #openstack-infra | 22:00 | |
| zaneb | sdague: cool, that's what I thought. I'll just use recheck from now on :) | 22:01 |
| krotscheck | jeblair: I’m confused. I didn’t really think we did, my own thoughts on getting Nikita up there is that he’ll be more able to debug prod issues quickly should they arise. | 22:01 |
| *** sarob_ has joined #openstack-infra | 22:01 | |
| jeblair | krotscheck: your comment in 97521 seems more a justification of why storyboard-core should have superuser access, rather than indicating that there are actual admin tasks for him to perform | 22:03 |
| *** pdmars has quit IRC | 22:04 | |
| krotscheck | jeblair: At the moment, all of our admin tasks are handled by automation. I don’t expect that to change anytime soon. If the measure of “Who gets admin access” is “Hey there’s adminy things I need to do”, then shouldn’t nobody have access? | 22:05 |
| * krotscheck is cool with that too. | 22:05 | |
| *** sarob_ has quit IRC | 22:05 | |
| jeblair | krotscheck: i think that minimization should be the preferred approach, yes. and we're much more liberal about access to staging environments (eg, review-dev, jenkins-dev). | 22:07 |
| krotscheck | jeblair: So… we should have a dev environment? | 22:07 |
| daya_k | arosen, jaypipes: i had this exact issue, debugged for 2 days with sweston help. the zuul merger process keeps dying when attempting the merge | 22:08 |
| arosen | jaypipes: should something be running that creates zuul-merger.log file? | 22:08 |
| jamielennox | all, can i get some reviews on bumping keystoneclient bump to 0.10 in requirements: https://review.openstack.org/#/c/108832/ - needed for some work in keystonemiddleware | 22:08 |
| jeblair | krotscheck: seems like it | 22:08 |
| arosen | daya_k: did you follow Jays bog post? | 22:09 |
| *** banix has quit IRC | 22:09 | |
| daya_k | arosen, jaypipes: one of the issues was permissions on the /var/lib/zuul directories, the zuul user did not have write permissions in it to perform the merge | 22:09 |
| daya_k | arosen: yes | 22:09 |
| jaypipes | arosen: the zuul-merger daemon creates the zuul-merger.log | 22:09 |
| *** yamamoto has joined #openstack-infra | 22:09 | |
| *** cnesa7 has quit IRC | 22:10 | |
| daya_k | asrosen, jaypipes: yes, if you run zuul-merger -d (not daemon mode) you will see the exception, but not when you try to run it as a service | 22:10 |
| *** Guest83948 is now known as Adri2000 | 22:10 | |
| *** htrmeira has quit IRC | 22:10 | |
| *** Adri2000 has quit IRC | 22:11 | |
| *** Adri2000 has joined #openstack-infra | 22:11 | |
| arosen | daya_k: thanks just fixed that let me see if things post now. | 22:11 |
| daya_k | arosen, jaypipes: check zuul.conf for where the ssh and git directories are and make sure the zuul user has write permissions on them | 22:11 |
| *** dangers is now known as dangers_away | 22:11 | |
| *** lcheng_ has quit IRC | 22:12 | |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add query for grenade quota reserve deadlock bug 1350064 https://review.openstack.org/110481 | 22:12 |
| uvirtbot | Launchpad bug 1350064 in nova "Deadlock in quota reservations in security groups tests on old side of grenade (icehouse)" [Undecided,New] https://launchpad.net/bugs/1350064 | 22:12 |
| arosen | daya_k: got another issue: when i run zuul-merge it dies witH: ImportError: 'gitdb' could not be found in your PYTHONPATH | 22:13 |
| daya_k | jaypipes: i'm actually still facing this issue. once i got past the permission errors, i see some git fetch errors, and the merger daemon dies again, however i see the merger daemon continues to working without exceptions in non-daemon mode, | 22:13 |
| *** BadCub02 has quit IRC | 22:14 | |
| *** whoops has quit IRC | 22:14 | |
| arosen | daya_k: jaypipes pip installing that seems to fix that error for me. | 22:14 |
| sdague | jamielennox: can you provide the rational in the commit message about why we can't work with something lower? | 22:14 |
| jaypipes | daya_k: hmm... strange.. check to see you don't have old crud in /var/run/zuul | 22:14 |
| daya_k | arosen: correct, i also saw that, you need to add gitdb module | 22:14 |
| openstackgerrit | Maru Newby proposed a change to openstack-infra/config: Fix neutron-dsvm-api job https://review.openstack.org/110482 | 22:15 |
| daya_k | jaypipes: how can i clear old jobs, i do see it building up, and git fetch complains about those commits being absent. dont know how to clear the queue | 22:15 |
| jaypipes | daya_k: hmm good question! :) clarkb, any ideas? | 22:16 |
| openstackgerrit | Jamie Lennox proposed a change to openstack/requirements: Bump keystoneclient to 0.10 https://review.openstack.org/108832 | 22:16 |
| jamielennox | sdague: sufficient ^ ? | 22:16 |
| marun | clarkb: I've posted a patch for the experimental neutron api job that includes the same fixes made to the functional job. ^^ | 22:16 |
| mattoliverau | Morning all | 22:16 |
| asselin | speaking of zuul merger, can I get some reviews on this? should be quick and easy: https://review.openstack.org/#/c/109364/ | 22:17 |
| jeblair | nibalizer: can you review 99041? | 22:18 |
| *** HenryG is now known as HenryG_afk | 22:18 | |
| asselin | daya_k, did you see the same exception as in https://review.openstack.org/#/c/109364/ ? | 22:19 |
| daya_k | asselin: yes, i saw this, and imported the gitdb module, am seeing a new one now. let me look it up and pastebin in a min. just logging in | 22:20 |
| clarkb | daya_k: do you have tracebacks for when it dies? or is nothing written to the log file? | 22:20 |
| daya_k | clarkb: when i start as a service, nothing written to log file, when i start in non-d, i see a traceback, | 22:21 |
| asselin | daya_k, please test and review that patch...it should will install gitdb for you when it installs zuul | 22:21 |
| clarkb | daya_k: what is the error? | 22:21 |
| *** rainya has quit IRC | 22:22 | |
| daya_k | clarkb: multiple issues, seen so sorry, will be a bit all over the place, but here goes - http://paste.openstack.org/show/89031/ | 22:24 |
| nibalizer | jeblair: sure | 22:24 |
| daya_k | clarkb: this one was in merger.log when the zuul-merger process dies, but i dont always see it, | 22:25 |
| clarkb | daya_k: that looks like the merger is not able to connect to the gearman process | 22:25 |
| clarkb | the connection object is None rather than a socket | 22:25 |
| jeblair | nibalizer: re:102071 (beaker specs) that refers to a file that i can't find in puppet-nova; is there an outstanding change for it? | 22:25 |
| daya_k | clarkb: ok, and this one http://paste.openstack.org/show/89032/ i related to old jobs in the queue i think, i see it in merger-debug.log | 22:26 |
| fungi | arosen: daya_k: do you have more details on the traceback for the gitdb import error? zuul doesn't import it so that must be bubbling up from one of the other dependencies... wondering which one and how it was installed | 22:26 |
| sdague | jamielennox: good for me, thanks. +2 | 22:26 |
| daya_k | fungi: asselin posted a review on it above | 22:27 |
| fungi | oh, thanks daya_k | 22:27 |
| clarkb | daya_k: I am less sure about that one, but I would doulbe check that your merger can talk to gerrit so that it can find all the refs | 22:27 |
| *** rainya has joined #openstack-infra | 22:28 | |
| daya_k | clarkb: how can i check, my current status is that if i run with -d, it all works. | 22:28 |
| nibalizer | this is the nova change that will be the first to be able to run specs | 22:28 |
| nibalizer | https://review.openstack.org/#/c/102020/ | 22:28 |
| fungi | daya_k: asselin: it's unclear to me why you need to explicitly install gitpython. it should already get installed by the zuul module when it pip installs zuul, since zuul's requirements.txt has GitPython>=0.3.2.RC1 specified already | 22:28 |
| clarkb | daya_k: it may be an authentication thing. when you run it with -d the working dir does not change so it will use relative paths for ssh keys | 22:29 |
| clarkb | daya_k: when use run it in daemon mode it will cd / | 22:29 |
| *** rainya has quit IRC | 22:30 | |
| *** bknudson has quit IRC | 22:30 | |
| asselin | fungi, good question...I just ran into this again on a clean vm. Add that patch above gets it installed....?? | 22:30 |
| daya_k | clarkb: yeah, i has multiple issues with permissions earlier, the zuul user did not have any write permissions, but i've fixed that for the /var/lib/zuul directiries, and i can ssh, i have the ssh path set in zuul.conf | 22:30 |
| clarkb | daya_k: I dont think it is a perms thing but a path thing | 22:30 |
| jhesketh | Morning | 22:31 |
| *** ramashri has joined #openstack-infra | 22:31 | |
| clarkb | sdague: hey can you visit the ironic room? so we can talk about the things | 22:31 |
| clarkb | jhesketh: ohai | 22:31 |
| daya_k | fungi: one of the things to do is to add logging to the jaypipes install scripts :) i dont have the history unfortunately. novice error | 22:32 |
| jeblair | clarkb, sdague: i can phone if you need me to talk about the ironic things | 22:32 |
| *** prad has quit IRC | 22:33 | |
| sdague | clarkb: coming | 22:33 |
| clarkb | jeblair: long story short is nova wants ironic to test nova baremetal to ironic migration at juno release code level | 22:34 |
| daya_k | clarkb: the path in zuul.conf has /var/lib/zuul/ssh , and the key is in the location. can i check anything else? | 22:34 |
| *** jaypipes has quit IRC | 22:35 | |
| clarkb | daya_k: it isn't /var/lib/zuul/.ssh? that would be the main thing though is the key | 22:35 |
| asselin | fungi, do you think the command needs to be sudo pip install /opt/zuul?? | 22:35 |
| *** mfink_ has quit IRC | 22:35 | |
| clarkb | jeblair: so we are going to add a job to do that and modify d-g to do juno to juno with different service lists | 22:35 |
| clarkb | then only run that on branch = juno | 22:35 |
| fungi | asselin: no, puppet agent should already be running as root | 22:35 |
| clarkb | jeblair: I think we have it pretty well covered | 22:35 |
| daya_k | clarkb: no, not .ssh, only ssh , the private key is in the location, with rw for zuul user only | 22:36 |
| jeblair | clarkb: ok. i trust you. just letting you know i'm here if you need me. :) | 22:36 |
| fungi | asselin: it would be good to repeat the exercise (without your proposed patch) and see what puppet logs when it applies that manifest | 22:36 |
| jeblair | clarkb: i'll keep -1ing changes in config in the mean time. :) | 22:36 |
| daya_k | fungi, asselin: fwiw, when i ran the scripts, i ran with sudo , instead of su. | 22:37 |
| jeblair | SlickNik: can you review https://review.openstack.org/#/c/103172/ ? | 22:37 |
| clarkb | jeblair have fun :) | 22:37 |
| fungi | clarkb: daya_k is right, we puppet our keys into /var/lib/zuul/ssh (no hidden dotdir) | 22:37 |
| *** bauzas has quit IRC | 22:37 | |
| asselin | fungi, daya_k here's the puppet output before the patch http://paste.openstack.org/show/89038/ | 22:38 |
| *** mfer has joined #openstack-infra | 22:38 | |
| *** mfer has quit IRC | 22:39 | |
| fungi | asselin: err: /Stage[main]/Zuul/Exec[install_zuul]: Failed to call refresh: pip install /opt/zuul returned 1 instead of one of [0] at /root/config/modules/zuul/manifests/init.pp:122 | 22:39 |
| fungi | asselin: too bad it doesn't elaborate | 22:39 |
| *** mfink has joined #openstack-infra | 22:40 | |
| fungi | asselin: you might try repeating 'pip install /opt/zuul' (as root or prefixed with sudo) and see if it throws a specific error | 22:40 |
| asselin | fungi, ok | 22:40 |
| *** yamamoto has quit IRC | 22:41 | |
| *** shashankhegde has quit IRC | 22:41 | |
| *** jaypipes has joined #openstack-infra | 22:42 | |
| jeblair | fungi: mind if i delete the old review.o.o, as well as the mirror slaves (which are no longer in use)? | 22:42 |
| fungi | jeblair: should be safe now | 22:43 |
| jeblair | | 795ac71e-0a30-4a8a-8a7f-3afddac8d70d | review.openstack.org | ACTIVE | - | Running | public=2001:4800:780d:509:3bc3:d7f6:ff04:39f0, 198.101.231.251; private=10.179.224.234 | | 22:43 |
| fungi | jeblair: i take it we hit quota again | 22:43 |
| openstackgerrit | Salvatore Orlando proposed a change to openstack-infra/config: Make neutron full job voting (neutron gate only) https://review.openstack.org/88289 | 22:43 |
| openstackgerrit | A change was merged to openstack-infra/config: Adds addtional details for browsable logs for third party ci https://review.openstack.org/101227 | 22:43 |
| jeblair | i'm going to delete that one. everyone agreed that's the old one? :) | 22:43 |
| jeblair | fungi: nope, just cleaning up | 22:43 |
| fungi | rechecking | 22:43 |
| jeblair | fungi: (i'm about to approve the change to remove the mirror slaves from puppet, i thought i should also remove them actually) | 22:44 |
| fungi | looks safe to delete. no gerrit running, no logins in ages, uptime of many months | 22:44 |
| asselin | fungi, looks ok: http://paste.openstack.org/show/89039/ | 22:44 |
| fungi | jeblair: too awesome. i'm entirely in favor | 22:44 |
| jeblair | also i apparently like deleting long running jenkins slaves | 22:46 |
| fungi | you have a knack | 22:46 |
| jeblair | all done | 22:46 |
| fungi | asselin: oh, that doesn't tell us much other than all the requirements are present. was this on a fresh vm or one where you had already previously installed gitpython directly? | 22:46 |
| asselin | fungi, I uninstalled gitpython and re-ran it. I can try again from a clean vm. | 22:47 |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/config: First draft of enabling beaker tests to run https://review.openstack.org/102071 | 22:47 |
| asselin | fungi, do you know how to get more details out of puppet? | 22:47 |
| nibalizer | woot fixed | 22:47 |
| *** adalbas has quit IRC | 22:48 | |
| fungi | asselin: yeah, i suspect something's failing to install gitpython on the first run through puppet on a fresh instance for some reason. you can add stdout/stderr logging of commands which run in a puppet exec by adding an option to it in the manifest i think... digging up the syntax (unless nibalizer knows it off the top of his head) | 22:48 |
| fungi | asselin: logoutput => true, | 22:49 |
| nibalizer | yup | 22:49 |
| fungi | asselin: sticking that inside the install_zuul exec block should get you more details in the puppet output | 22:50 |
| asselin | fungi, ok I will try it out. thanks | 22:50 |
| *** bauzas has joined #openstack-infra | 22:51 | |
| nibalizer | does anyone know if this class is ever used? http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/simple_slave.pp | 22:52 |
| fungi | phschwartz: did you have any luck getting someone to delete our (currently 103) error nodes, or should i go ahead and open a support case? | 22:52 |
| nibalizer | i'd like to nuke it, if possible | 22:52 |
| fungi | nibalizer: i find no references to it at all in openstack-infra/config | 22:53 |
| *** jp_at_hp has joined #openstack-infra | 22:53 | |
| daya_k | clarkb, jaypipes, arosen: here's the exceptions i saw when the zuul user did not have proper permissions to /var/lib/zuul - http://paste.openstack.org/show/89042/ | 22:54 |
| arosen | fungi: sorry I stepped away. Here's the trace for gitdb from zuul http://codepad.org/aX2bZ2OW | 22:54 |
| clarkb | nibalizer it should have a comment | 22:54 |
| *** derekh_ has joined #openstack-infra | 22:54 | |
| clarkb | nibalizer it is a convenience class for humans | 22:55 |
| fungi | nibalizer: unless it's referenced in documentation somewhere outside of the config repo, which seems pretty unlikely, i can't imagine what would be using it now. looking for the last reference in the git log | 22:55 |
| arosen | daya_k: is your CI now able to pick up jobs from gerrit? | 22:55 |
| clarkb | fungi thats what it is for | 22:55 |
| clarkb | if you want to reproduce tests use that class | 22:55 |
| fungi | clarkb: oh, do you know where we document its existence? on the wiki? | 22:55 |
| daya_k | arosen: yes, if i run as non-daemon, something is still broken. | 22:55 |
| fungi | arosen: asselin got us further on the missing gitdb issue... for some reason the install_zuul exec in the zuul puppet module is not successfully installing dependencies | 22:56 |
| *** jgrimm has quit IRC | 22:56 | |
| clarkb | fungi I forget where | 22:57 |
| daya_k | fungi: is it also responsible for setting up permissions for zuul user? | 22:57 |
| fungi | clarkb: nibalizer: perhaps the comment in it could use some expanding on, at least, saying not to delete it even though it's not referenced by any other class | 22:57 |
| arosen | daya_k: are there any other places to look for errors on why it's not picking up the job? | 22:58 |
| *** jcoufal has quit IRC | 22:58 | |
| fungi | daya_k: the various file objects in the zuul puppet module set owner => zuul on all the directories i know we need to care about | 22:59 |
| daya_k | arosen: all the logs, merger.log, and merger-debug.log, also gearman should show merger in its worker status, telent 127.0.0.1 4730, followed by workers. let me paste my output for you. | 22:59 |
| nibalizer | clarkb: but if you want to reproduce a test case you have to apply openstack_project::server i think | 23:00 |
| nibalizer | since you need to simulate the same iptables and so on | 23:00 |
| daya_k | fungi: ok that part is also broken then. i had to manually set the permissions to get rid of above exceptions - http://paste.openstack.org/show/89042/ | 23:00 |
| arosen | daya_k: i can telnet fine to that. I also got the Success in jenkins when clicking the test button for Gearman. | 23:00 |
| fungi | daya_k: having the puppet apply output from a fresh run where you can reproduce the incorrect ownership/permissions condition would be great | 23:01 |
| *** sarob_ has joined #openstack-infra | 23:01 | |
| openstackgerrit | Craig Bryant proposed a change to openstack-infra/config: Add new chef cookbook-monasca-schema https://review.openstack.org/110493 | 23:01 |
| openstackgerrit | A change was merged to openstack-infra/config: Remove the mirror slaves https://review.openstack.org/103361 | 23:02 |
| daya_k | fungi: ok, will try that on a fresh node later in the day | 23:02 |
| daya_k | arosen: here's what my gearman shows - http://paste.openstack.org/show/89046/ besides this, i could capture the excpetions when i ran with -d, | 23:03 |
| *** zul has quit IRC | 23:03 | |
| arosen | daya_k: you're saying when you ran zuul-merger -d ? | 23:04 |
| daya_k | arosen: correct. because if you run it as a service, it just keeps dying without any logs | 23:04 |
| arosen | daya_k: what's the output of : http://paste.openstack.org/show/89046/ ? | 23:04 |
| arosen | zuul-merger -d as well? | 23:04 |
| *** arnaud has quit IRC | 23:04 | |
| *** alexpilotti has quit IRC | 23:05 | |
| arosen | daya_k: zuul-merger -d doesn't seem to output anything for me. I stopped the zuul-merger service before letting it run manually. | 23:06 |
| daya_k | arosen: if you do 'telnet 127.0.0.1 4730' then type 'status' and 'workers' you should see the above - there should be zuul 'merger' in there like i have showed | 23:06 |
| *** sarob_ has quit IRC | 23:06 | |
| arosen | daya_k: this is what i get then: http://paste.openstack.org/show/89047/ | 23:07 |
| fungi | looks like the merger is running and connected | 23:07 |
| clarkb | nibalizer no the point is you get 99% there without all thst extra iptables ntp etc | 23:07 |
| *** datsun180b has quit IRC | 23:07 | |
| daya_k | arosen: if its not giving an o/p, its running ok, only spews out exceptions. but why dont u see the jobs at all in gearman, maybe someone needs to check your layout.yaml ? | 23:08 |
| fifieldt_ | hi infra | 23:08 |
| fungi | hi fifieldt_ | 23:09 |
| fifieldt_ | how are y'all? | 23:09 |
| arosen | daya_k: i copied layout.yaml exactly from here: https://github.com/jaypipes/os-ext-testing-data/blob/master/etc/zuul/layout.yaml | 23:09 |
| daya_k | fungi: i could always see the build jobs, in the gearman output even when i was having htese issues | 23:09 |
| fifieldt_ | can I get a quick assist with populating the enterprise-wg mailing list? As it's administered by someone else (Carol), I don't have the password. I have a list of addresses to dump into the mass subscribe box :) | 23:10 |
| arosen | daya_k: where did you get our layout.yaml from? | 23:10 |
| fungi | fifieldt_: eww... they're going to opt-in confirm though, right? | 23:10 |
| fifieldt_ | yup, no problems there | 23:11 |
| openstackgerrit | A change was merged to openstack-infra/config: Initial RHEL7/Centos7 support https://review.openstack.org/103458 | 23:11 |
| fifieldt_ | I also have a record of Carols' consent I can forward via email :) | 23:11 |
| arosen | daya_k: my ci account hasn't been approved though yet so i'm just using my ssh creds that i use to push patches to gerrit. I hope that doesn't matter. | 23:11 |
| fungi | fifieldt_: can you be more specific about why carol, as the list admin, isn't able to take care of it? | 23:11 |
| fifieldt_ | she lacks the necessary knowledge and I offered to assist | 23:11 |
| fungi | arosen: shouldn't matter. the only thing you won't be able to do is leave verify votes with it | 23:12 |
| fungi | fifieldt_: okay. does she lack the necessary knowledge to admin that mailing list, or does she need a co-administrator? | 23:12 |
| daya_k | arosen: i also got it from the same location, but have edited slightly for actions. i think the ci account will matter, its used by the zuul system to pull patches and run merges. | 23:12 |
| reed | the good old 'mailman is user friendly but picky about who his friends are' :) | 23:12 |
| arosen | fungi: okay. So I guess I'd still see a build run. Haven't gotten that to happen yet. | 23:13 |
| fifieldt_ | fungi, all I know is she asked for help in this instance | 23:13 |
| reed | fifieldt_, I'd suggest you to take over the list management | 23:13 |
| daya_k | clarkb: here's what i just got http://paste.openstack.org/show/89048/ | 23:13 |
| reed | fifieldt_, because mailman *is* tricky :) | 23:14 |
| *** dims has quit IRC | 23:14 | |
| fungi | fifieldt_: i'm with reed on that, at least volunteer as a list co-administrator. otherwise you'll be coming to us with more requests along similar lines | 23:14 |
| fifieldt_ | I volunteer :) | 23:14 |
| reed | fungi, right, he is volunteering... now he needs the password :) | 23:14 |
| reed | I suspect that if fifieldt_ came here to ask for it, he's tried to gather it from Carol already and failed | 23:15 |
| fungi | fifieldt_: the next thing to check is whether carol had the wherewithal to change the initial password. hold on and i'll get it to you | 23:15 |
| reed | oops :) | 23:15 |
| *** yamahata_ has quit IRC | 23:15 | |
| *** dims has joined #openstack-infra | 23:16 | |
| *** dims is now known as Guest13918 | 23:16 | |
| *** Guest13918 is now known as dims_ | 23:17 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add centos7 image for rax https://review.openstack.org/109906 | 23:17 |
| daya_k | arosen: so you should be able to 'ssh <ci-acct>@review.openstack.org -i /var/lib/zuul/ssh/id_rsa -p 29418 | 23:17 |
| daya_k | arosen: otherwise its not going to work | 23:17 |
| arosen | daya_k: yup that worked | 23:18 |
| *** dims_ has left #openstack-infra | 23:20 | |
| arosen | daya_k: how did you know you needed to change the layout.yaml file? | 23:20 |
| *** rainya has joined #openstack-infra | 23:20 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add f20 experimental job to tempest https://review.openstack.org/104878 | 23:20 |
| *** dims_ has joined #openstack-infra | 23:20 | |
| *** bknudson has joined #openstack-infra | 23:20 | |
| daya_k | arosen: not for fixing these issues, i edited it to run my customized job, and also not review '-1' till things start working. let me paste it for you. | 23:21 |
| jeblair | ttx: can you review 105895 please? | 23:21 |
| *** rainya has quit IRC | 23:22 | |
| arosen | daya_k: gotcha, I'm only trying to replicate what jaypipes has in his blog at this point. | 23:22 |
| *** rainya has joined #openstack-infra | 23:22 | |
| daya_k | arosen: you should see something like http://paste.openstack.org/show/89049/ in the merger log files | 23:22 |
| *** dimtruck is now known as zz_dimtruck | 23:23 | |
| arosen | daya_k: all i see is; 2014-07-29 16:06:28,499 DEBUG zuul.MergeServer: Starting merge listener | 23:23 |
| arosen | 23:23 | |
| *** banix has joined #openstack-infra | 23:23 | |
| daya_k | arosen: dont know what your issue might be, other than what ive suggested. gearman output on the lcient itselfi seems to be lacking.. | 23:23 |
| daya_k | arosen: will post layout.yaml. restart zuul, zuul-merger and jenkins with these, and check | 23:24 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix the constant failure of puppet on cacti https://review.openstack.org/104989 | 23:24 |
| daya_k | arosen: can u drop me an email - daya_k@yahoo.com. sweston sent me instructions on the sequence that i can forward | 23:24 |
| *** bknudson has quit IRC | 23:25 | |
| *** rainya has quit IRC | 23:27 | |
| *** rainya has joined #openstack-infra | 23:27 | |
| *** dkehn_ has joined #openstack-infra | 23:27 | |
| jeblair | unreviewed_config_changes-=43 | 23:28 |
| fungi | once 105895 is approved, i'm totally using "recheck no story" | 23:29 |
| jeblair | ha! | 23:29 |
| daya_k | arosen: here's my layout.yaml http://paste.openstack.org/show/89051/, u should probably leave the dsvm-tempest-full tied to sandbox till things start working for you | 23:29 |
| arosen | daya_k: ak okay. So I just rebooted the box and now I'm getting error in merge.log. I guess I didn't restart the right thing after changing layout.yaml | 23:29 |
| arosen | daya_k: http://paste.openstack.org/show/89053/ Looks like there is a bug in zuul/merger/merger.py too | 23:30 |
| fungi | 105895 also reminds me that we can probably rip out all the recheckwatch code now | 23:30 |
| *** james_li has quit IRC | 23:31 | |
| *** dkehnx has quit IRC | 23:31 | |
| *** bdpayne has quit IRC | 23:31 | |
| arosen | looks like it's not using my ssh key | 23:32 |
| fungi | arosen: that doesn't look like a bug in the merger. looks like you haven't accepted/puppeted the gerrit host key into your /var/lib/zuul/ssh/authorized_keys file | 23:32 |
| daya_k | arosen: yes, but dont know if that one was tied to the permissions issues i was seeing. the exception i get now, is different | 23:32 |
| *** ZZelle_ has quit IRC | 23:32 | |
| fungi | arosen: note that the error from paramiko is about the host key, not your key | 23:33 |
| openstackgerrit | A change was merged to openstack-infra/config: Adds a statement regarding infra disabling ci accounts https://review.openstack.org/105333 | 23:33 |
| *** craigbr has quit IRC | 23:33 | |
| openstackgerrit | A change was merged to openstack-infra/config: Use job templates and add docs build to python-jenkins https://review.openstack.org/105348 | 23:33 |
| fungi | er, s/paramiko/git/ | 23:34 |
| *** bknudson has joined #openstack-infra | 23:34 | |
| *** yfried_ has quit IRC | 23:34 | |
| *** zns_ has quit IRC | 23:34 | |
| *** Sukhdev has joined #openstack-infra | 23:35 | |
| arosen | fungi: Hrm do you know off hand how to do that or could you point me in the right direction? | 23:35 |
| daya_k | fungi: should this also be created in /var/lib/zuul/ssh ? i dont see the file created either | 23:35 |
| fungi | arosen: as for the traceback from the merger, it's arguably a zuul bug that repo is only defined within a try but then returned outside the try block: http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/merger/merger.py#n61 | 23:36 |
| arosen | fungi: yea probably should raise something instead. | 23:36 |
| *** banix has quit IRC | 23:37 | |
| daya_k | fungi: i was getting this exception when the dir had root permissions only, not for zuul user, i've fixed now, but no authorized_keys being created now. | 23:37 |
| *** otherwiseguy has quit IRC | 23:37 | |
| fungi | arosen: daya_k: sorry, brainbug. i mean to type /var/lib/zuul/ssh/known_hosts | 23:38 |
| daya_k | fungi:ok, that one is in /home/zuul/.ssh for me | 23:38 |
| *** bdpayne has joined #openstack-infra | 23:39 | |
| fungi | arosen: though for whatever reason, yeah, we install it there instead... http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/zuul_prod.pp#n65 | 23:39 |
| fungi | where daya_k said | 23:39 |
| daya_k | fungi: any idea why .git/objects is throwing exceptions related to permissions, is some other user (not zuul) trying to create these? | 23:40 |
| fungi | the zuul::init class doesn't install that file since its contents are fairly openstack-specific (though we could in theory pass it in through a variable), so instead it's being managed in the openstack_project::zuul class | 23:40 |
| *** banix has joined #openstack-infra | 23:41 | |
| *** bauzas has quit IRC | 23:41 | |
| fungi | daya_k: in theory the zuul merger process would be running as the zuul user, and shouldn't be able to create those files as anyone else... unless perhaps you tried running the zuul merger daemon yourself as root or with sudo rather than invoking it as the zuul user (or from the initscript where that would generally be specified) | 23:41 |
| fungi | in which case it might have created/updated some repo clones as root and then later when you tried running as zuul again it was all fouled up | 23:42 |
| openstackgerrit | A change was merged to openstack/requirements: Add xstatic-packaged JavaScript libraries for Horizon https://review.openstack.org/97793 | 23:42 |
| daya_k | fungi: it actually started working only after i acciedentally ran as root, and then created the directory structure under /var/lib/zuul/git thats how sweston and i discovered the issue | 23:42 |
| arosen | fungi: awesome! It worked :) https://review.openstack.org/#/c/110484/ | 23:43 |
| fungi | the contents of the /var/lib/zuul/git directory are ephemeral and can be deleted if necessary | 23:43 |
| daya_k | fungi: but now, running as zuul user, with permissions set for zuul:zuul to 755 for /var/lib/zuul on down for everything | 23:43 |
| daya_k | fungi : even with these set, i'm getting http://paste.openstack.org/show/89048/ | 23:44 |
| fungi | arosen: next order of business. fix your layout.yaml so you don't leave "starting xxx jobs" messages on changes when you begin testing | 23:44 |
| *** HenryG_afk is now known as HenryG | 23:45 | |
| *** banix has quit IRC | 23:46 | |
| *** derekh_ has quit IRC | 23:46 | |
| fungi | daya_k: what does 'ls -dl /var/lib/zuul/git/openstack/neutron/.git/objects' say the permissions/ownership are on it? | 23:46 |
| daya_k | fungi : drwxrwxr-x 253 zuul zuul 4096 Jul 29 20:37 /var/lib/zuul/git/openstack/neutron/.git/objects | 23:47 |
| arosen | fungi: hrm that line does that in layout.yaml? Also seems like recheck isn't quite working yet | 23:47 |
| fungi | daya_k: and zuul-merger has been restarted since you fixed the permissions on /var/lib/zuul? | 23:48 |
| arosen | fungi: my layout file is exactly this: https://github.com/jaypipes/os-ext-testing-data/blob/master/etc/zuul/layout.yaml | 23:48 |
| daya_k | fungi: yes, many times :) | 23:48 |
| arosen | daya_k: I did: su zuul ; then ssh review.openstack.org -p <balh> to get the known host file populated for the zuul user. | 23:49 |
| daya_k | fungi: in fact zuul-merger dies when i start it as a service everytime, i only see this excpetion when i run it with -d | 23:49 |
| fungi | daya_k: are you running with -d as the zuul account on your system, or with some other account? | 23:49 |
| daya_k | fungi: zuul account | 23:50 |
| daya_k | to start service, i use root account | 23:50 |
| *** trinaths has joined #openstack-infra | 23:50 | |
| fungi | daya_k: very odd... does 'find /var/lib/zuul/git/ -not -user zuul' return anything? | 23:53 |
| daya_k | fungi: is it ok if some of the underlying files pulled under the /var/lib/zuul/.. yes i was just going to say that | 23:53 |
| *** dkehn_ is now known as dkehnx | 23:53 | |
| daya_k | under /var/lib/zuul/git/openstack/neutron, theres a few files with root perms. | 23:53 |
| daya_k | fungi : setup.cfg | 23:53 |
| fungi | daya_k: yeah, you probably want to 'sudo chown -R zuul:zuul /var/lib/zuul/git' to make sure it's all owned by the right user:group | 23:54 |
| fungi | daya_k: or just delete everything in /var/lib/zuul/git | 23:54 |
| *** trinaths has quit IRC | 23:54 | |
| daya_k | and lots of files for the cmds u asked | 23:55 |
| asselin | fungi: forgot to take out the install of gitpyton, but there's still errors. I will try again: http://paste.openstack.org/show/89055/ | 23:55 |
| *** banix has joined #openstack-infra | 23:55 | |
| fungi | daya_k: also, when starting zuul and zuul-merger daemons as root, you're running the initscripts for them to do that, right? not calling the python executables directly? | 23:56 |
| daya_k | fungi: ok, i'll rm and restart zuul-merger | 23:56 |
| asselin | daya_k, which repo are you using for 3rd party ci? | 23:56 |
| daya_k | fungi: service zuul start/reload | 23:56 |
| daya_k | fungi: although while debugging, i might have used the python cmd, was trying many things | 23:56 |
| fungi | daya_k: okay, that should be safe unless you've got a weird initscript which isn't switching to the correct user | 23:57 |
| asselin | arosen, to fix the start messages, commend out the 2 "start" blocks | 23:57 |
| fungi | the service command should be safe i mean. calling the zuul daemon python executables directly as someone other than your zuul user is probably going to cause headaches later | 23:57 |
| daya_k | fungi : when it had accidentally started working, i ran zuul-merger -d as root user, then all the files etc. got created, until then it was dying | 23:58 |
| *** banix has quit IRC | 23:58 | |
| arosen | asselin: awesome thanks! | 23:59 |
| asselin | arosen, lines 21-23 https://github.com/jaypipes/os-ext-testing-data/blob/master/etc/zuul/layout.yaml#L21 | 23:59 |
| fungi | asselin: aha! looks like we may be missing the libxml headers when python lxml is being installed by git | 23:59 |
| *** garyh has quit IRC | 23:59 | |
| asselin | arosen, didn't succeed to submit a pull request on that.... | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
