| mordred | dstufft: yup. I remember that initial 8 hours of heated conversation very well and fondly | 00:00 |
|---|---|---|
| mordred | krotscheck: I had a delicious $something recently which made me think of you - but I had more of it and have promptly forgotten what it was | 00:00 |
| *** cdent has quit IRC | 00:01 | |
| krotscheck | mordred: As it should be. | 00:01 |
| * krotscheck is actually suspending his brewing for the forseeable future. | 00:01 | |
| pleia2 | that's what untappd is for | 00:01 |
| dstufft | mordred: I don't even remember what I was mad about, I jsut remember I was still at Nebula and being real mad about pbr | 00:01 |
| mordred | krotscheck: we have found the wall in our new apartment on which to do your pots and pans hanging board | 00:01 |
| clarkb | fungi: oh I need to report that the sake looks, smells, and mostly tastes like sake | 00:01 |
| jeblair | /topic Project Infrastructure | Packaging | Mind-altering substances | 00:01 |
| mordred | dstufft: I think you were mad because it's an abomination | 00:01 |
| clarkb | fungi: its very strong and a bit tart. Not my favorite sake but it seems to be drinkable | 00:02 |
| mordred | krotscheck: NOOOO! why?? | 00:02 |
| dstufft | mordred: now adays if I can convince y'all to move away from requirements.txt and into setup.cfg and maybe spend some time so pbr becomes not a requirement to run a sdist then I might even be convinced to start using it | 00:02 |
| krotscheck | mordred: I can’t take credit for that, it was a refinement on Julia Child’s kitchen. | 00:02 |
| mordred | krotscheck: we're going to credit you anyway | 00:02 |
| sigmavirus24 | dstufft: let's write some specs | 00:03 |
| krotscheck | mordred: It’s a pretty neat hack, especially for NYC-scale apartments :) | 00:03 |
| clarkb | dstufft: how do you avoid it as a requirement to sdist? I thought that as the whole point | 00:03 |
| mordred | clarkb: we have thoughts | 00:03 |
| jeblair | mordred, fungi: you may be interested and not at all surprised to note that according to the most recent issue of "Distiller (tm)" magazine, new york, california, and washington recorded the highest increase in craft distilleries from 2012-2013 | 00:04 |
| dstufft | clarkb: well you wouldn't get the "use pip to install crap instead of setuptools" bit | 00:04 |
| clarkb | I mean at that point you don't need pbr anymore | 00:04 |
| mordred | clarkb: not as a requirement to _generate_ an sdist | 00:04 |
| mordred | clarkb: as a requirement to consume sdist tarballs | 00:04 |
| krotscheck | mordred: I’m on meds that don’t do well with Alcohol. | 00:04 |
| mordred | clarkb: and the tl;dr is "emit a setup.py into the source tarball that has just the right bits" | 00:04 |
| mordred | krotscheck: fair | 00:04 |
| mordred | krotscheck: and eek | 00:05 |
| *** bswartz has joined #openstack-infra | 00:05 | |
| dstufft | yea what mordred said | 00:05 |
| clarkb | jeblair: but be careful a lot of craft distilleries farm out to someplace in like indiana | 00:05 |
| mordred | jeblair: I actually now live across the street from one | 00:05 |
| mordred | krotscheck: oh! I know - it actually made me think of both you and clarkb | 00:05 |
| mordred | we got to see ben starr's latest home brewing gear | 00:05 |
| krotscheck | Why, because we’re distillery snobs? :D | 00:05 |
| clarkb | mordred: I see so its an install time lack of pbr not sdist time lack | 00:05 |
| dstufft | clarkb: yea, I meant lack of pbr when installing from sdist | 00:06 |
| dstufft | there's already a lack of pbr when install from wheel | 00:06 |
| dstufft | I'm just bad at words | 00:06 |
| openstackgerrit | Travis Tripp proposed openstack/requirements: Remove old deprecated xstatic angular packages https://review.openstack.org/145096 | 00:07 |
| mordred | dstufft: finish metadata 2.0 and we're there, right? | 00:07 |
| dstufft | mordred: metadata 2.0 and then sdist 2.0, but sdist 2.0 is mostly just "take this crap from metadata 2.0, stick it in a tarball" | 00:07 |
| clarkb | krotscheck: if you want to give sake a try uwajimaya has koji that is useable but I don't think they have polished rice. I couldn't find any sake polished rice at least | 00:07 |
| clarkb | krotscheck: but FH Steinbarts down here appears to get rice and koji from momokawa and resells that so my next attempt will involve a trip over there | 00:08 |
| dstufft | I haven't felt super motivated to dig into metadata 2.0 yet | 00:08 |
| dstufft | nick split everything up into like 4 different PEPs | 00:08 |
| dstufft | and I haven't read it yet since he's done that | 00:08 |
| krotscheck | clarkb: Once I get back to brewing, that sounds like fun :) | 00:08 |
| dstufft | I had to limit myself ot PEP 440 until we got that nailed down | 00:08 |
| *** rlandy has quit IRC | 00:10 | |
| fungi | clarkb: too awesome--i'm envious! worth noting that most sake brewers dillute with about 10% water before bottling, and that harsher sakes (for example more acidic ones) benefit more from heating to slightly above human body core temperature (39-40C) | 00:11 |
| *** banix has quit IRC | 00:12 | |
| clarkb | jeblair: I have reviewed 115187 | 00:13 |
| clarkb | fungi: ya apparently you ferment to ~18% abv then water down to ~15% | 00:14 |
| openstackgerrit | Travis Tripp proposed openstack/requirements: Angular xstatic for Version 1.3.7 https://review.openstack.org/142869 | 00:14 |
| fungi | clarkb: jeblair: i toured a brewery in nc over the holidays which was starting its own distilling and the distillery master was the one giving our tour. he said their biggest challenge was getting new oak aging barrels because there's a serious shortage and the large established distilleries get the favored pick of them already | 00:14 |
| clarkb | jeblair: I am +2, there appear to be a few places where E125 and E126 are fixed but we ignore at least 125. Not sure if that is -1 worthy to you | 00:14 |
| *** ryanpetrello has quit IRC | 00:15 | |
| ianw | anyone familiar with why pbr is throwing "Exception: Versioning for this project requires either an sdist tarball, or access to an upstream git repository. Are you sure that git is installed?" | 00:15 |
| ianw | under tox | 00:15 |
| clarkb | ianw: I think that may be a broken version of pbr that was released under the setuptools time crunch | 00:15 |
| *** dims__ has joined #openstack-infra | 00:15 | |
| ianw | it seems to be running git not in a git repo and failing (from strace) | 00:15 |
| clarkb | ianw: double check you have latest? | 00:15 |
| ianw | clarkb: ok, let me double check here... | 00:16 |
| ianw | clarkb: this is just running tox with oslo.config, so https://github.com/openstack/oslo.config/blob/master/requirements.txt | 00:17 |
| clarkb | ianw: is it possibly an existing tox venv? try tox -r ? | 00:18 |
| ianw | yeah, deleted .tox | 00:18 |
| *** wenlock has quit IRC | 00:18 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/system-config: Add support for running Zuul on Trusty https://review.openstack.org/141518 | 00:18 |
| *** dims_ has joined #openstack-infra | 00:19 | |
| clarkb | asselin: does that apache config work on 2.2 too? | 00:20 |
| asselin | clarkb, seems to, yes | 00:20 |
| *** andreykurilin has quit IRC | 00:20 | |
| clarkb | cool | 00:20 |
| asselin | clarkb, I just restarted apache, and zuul status page loads fine. | 00:21 |
| clarkb | asselin: on precise? | 00:21 |
| asselin | yes | 00:21 |
| asselin | both :) | 00:21 |
| *** dangers is now known as dangers_away | 00:21 | |
| *** ivar-lazzaro has quit IRC | 00:22 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add to rally non voting unstable performance job https://review.openstack.org/145098 | 00:22 |
| asselin | clarkb, I'm not 100% sure about the caching on 14.04 however. I see the status.json get cached, but the response in the access log is still 200 | 00:22 |
| *** dims__ has quit IRC | 00:23 | |
| *** dims_ has quit IRC | 00:24 | |
| openstackgerrit | Merged openstack-dev/pbr: Use a wheelhouse in the integration test https://review.openstack.org/145029 | 00:25 |
| *** wuhg has joined #openstack-infra | 00:25 | |
| ianw | clarkb: still remains even with master pbr | 00:26 |
| *** ivar-lazzaro has joined #openstack-infra | 00:26 | |
| ianw | it works in the gate, so it's obviously something local / fedora related i guess | 00:26 |
| *** sarob has quit IRC | 00:27 | |
| *** markmcclain has quit IRC | 00:27 | |
| clarkb | ianw maybe the checks pbr does fail with git because git version differences? | 00:28 |
| *** teran has joined #openstack-infra | 00:28 | |
| fungi | or is this tox running on fedora's source for oslo.config such that the egg-info is missing (similar to the earlier fedora+pbr gertty packaging discussion)? | 00:29 |
| ianw | clarkb / fungi : full error is http://paste.openstack.org/show/155669/ | 00:30 |
| fungi | worth noting that error message is the one pbr regurgitates if it can't find a .git dir and can't find any egg-info installed into the virtualenv | 00:31 |
| fungi | (and hasn't been fed version info from an envvar) | 00:31 |
| clarkb | ya it looks like UNKNOWN-0.0.0.zip is the thing installed | 00:32 |
| *** ChuckC has quit IRC | 00:32 | |
| *** teran has quit IRC | 00:32 | |
| *** david-lyle has joined #openstack-infra | 00:33 | |
| ianw | 15474 execve("/usr/bin/git", ["git", "rev-parse", "--git-dir"], [/* 58 vars */] <unfinished ...> | 00:33 |
| *** Masahiro has joined #openstack-infra | 00:33 | |
| ianw | 15474 write(2, "fatal: Not a git repository (or "..., 142) = 142 | 00:33 |
| clarkb | I think the error is before that | 00:34 |
| clarkb | when it sdists oslo.config | 00:34 |
| clarkb | see if python setup.py sdist looks sane? | 00:35 |
| *** pc_m has quit IRC | 00:35 | |
| *** david-ly_ has joined #openstack-infra | 00:36 | |
| fungi | ahh, yeah that zipfile must lack the metadata. if you're sdisting from not-a-git-repo then you need to set an envvar to tell it what version to use, right? | 00:37 |
| ianw | this is a git repo though, just a normal checkout | 00:37 |
| ianw | [iwienand@fedora19 oslo.config]$ python setup.py sdist | 00:38 |
| ianw | /usr/lib64/python2.7/distutils/dist.py:267: UserWarning: Unknown distribution option: 'pbr' | 00:38 |
| fungi | ianw: oh, there's a .git dir in the cwd? | 00:38 |
| *** hdd has quit IRC | 00:38 | |
| *** Masahiro has quit IRC | 00:38 | |
| *** otter768 has quit IRC | 00:38 | |
| fungi | ianw: ouch. is that ancient pre-fork distutils? | 00:38 |
| fungi | ianw: what virtualenv version do you have? | 00:39 |
| ianw | ok, "yum install python-pbr" solves this... | 00:39 |
| fungi | fedora19 might be in the right timeframe to have a virtualenv without a vendored setuptools | 00:39 |
| *** btran has quit IRC | 00:39 | |
| ianw | this is actually f21, that hostname just stuck from before upgrade | 00:40 |
| *** hdd has joined #openstack-infra | 00:40 | |
| fungi | got it. so... does oslo.config use system site packages in the tox config? | 00:40 |
| fungi | system-installed pbr should have no bearing on what happens inside a virtualenv under normal circumstances | 00:40 |
| *** david-ly_ has quit IRC | 00:41 | |
| ianw | yeah, i mean that's the idea of tox right :) | 00:41 |
| * fungi bets you have virtualenv 1.9 or earlier | 00:42 | |
| fungi | virtualenv 1.10 was the first to use new setuptools | 00:42 |
| ianw | [iwienand@fedora19 oslo.config]$ yum info python-virtualenv | 00:43 |
| ianw | Version : 1.11.6 | 00:43 |
| fungi | huh, another theory shot down in its infancy | 00:43 |
| *** ociuhandu has quit IRC | 00:43 | |
| *** amitgandhinz has joined #openstack-infra | 00:43 | |
| clarkb | is it possible that the sdist is happening outside the venv? | 00:44 |
| clarkb | so that the tox sdist step is unaffected by all that? | 00:44 |
| ianw | clarkb: yeah, that's my guess | 00:44 |
| fungi | yeah, no sitepackages=true in oslo.config's tox.ini either | 00:45 |
| ianw | .tox/log/tox-0.log has | 00:46 |
| ianw | /usr/lib64/python2.7/distutils/dist.py:267: UserWarning: Unknown distribution option: 'pbr' | 00:47 |
| ianw | warnings.warn(msg) | 00:47 |
| ianw | which is the root cause | 00:47 |
| ianw | but of course you never see that... :( | 00:47 |
| *** amitgandhinz has quit IRC | 00:48 | |
| *** ryanpetrello has joined #openstack-infra | 00:49 | |
| openstackgerrit | Andrew Bramley proposed openstack/requirements: Avoid using pexpect version 3.3 https://review.openstack.org/145104 | 00:50 |
| *** achanda has joined #openstack-infra | 00:50 | |
| *** achanda has quit IRC | 00:55 | |
| *** achanda has joined #openstack-infra | 00:56 | |
| ianw | it works with usedevelop=True | 00:56 |
| *** Masahiro has joined #openstack-infra | 00:56 | |
| *** dannywilson has quit IRC | 00:57 | |
| *** markmcclain has joined #openstack-infra | 00:59 | |
| *** achanda has quit IRC | 01:01 | |
| *** shashankhegde has joined #openstack-infra | 01:01 | |
| *** ChuckC has joined #openstack-infra | 01:03 | |
| *** sputnik13 has quit IRC | 01:05 | |
| *** hdd has quit IRC | 01:05 | |
| zaro | fungi: thanks. i thought review-dev's db was trove db, forgot that it's just local. | 01:06 |
| clarkb | ok I have confirmed that the changes made to www repo make my dev puppet stuff seem to work | 01:11 |
| clarkb | I am now going to try and move them over to using gerrit as the upstream | 01:11 |
| clarkb | and will spin up a www-dev host once that happens | 01:11 |
| clarkb | I still need a databse dump though | 01:11 |
| clarkb | which I can probably cheat to get but I Figure I should do this like a normal person | 01:11 |
| *** weshay has joined #openstack-infra | 01:13 | |
| *** markmcclain has quit IRC | 01:15 | |
| *** baoli has joined #openstack-infra | 01:15 | |
| *** teran has joined #openstack-infra | 01:16 | |
| *** teran_ has joined #openstack-infra | 01:17 | |
| *** unicell has joined #openstack-infra | 01:17 | |
| *** markmcclain has joined #openstack-infra | 01:18 | |
| *** teran has quit IRC | 01:21 | |
| *** salv-orlando has quit IRC | 01:21 | |
| *** teran_ has quit IRC | 01:21 | |
| *** markmcclain has quit IRC | 01:21 | |
| *** banix has joined #openstack-infra | 01:24 | |
| *** dims__ has joined #openstack-infra | 01:24 | |
| jeblair | clarkb: yeah, i think it's best not to cheat | 01:26 |
| *** patrickeast has quit IRC | 01:27 | |
| *** teran has joined #openstack-infra | 01:27 | |
| *** doug-fish has left #openstack-infra | 01:28 | |
| *** banix has quit IRC | 01:29 | |
| fungi | doubleplus. we should expect those install instructions to work for anyone, or else they're useless for their intended purpose | 01:31 |
| clarkb | indeed | 01:31 |
| *** ivar-lazzaro has quit IRC | 01:34 | |
| * fungi needs to go collapse for a few hours. hopefully tomorrow will prove more productive | 01:35 | |
| *** baoli has quit IRC | 01:36 | |
| clarkb | fungi: get better | 01:36 |
| fungi | trying. you too | 01:36 |
| fungi | and thanks | 01:37 |
| *** baoli has joined #openstack-infra | 01:37 | |
| anteaya | fungi: yes get better | 01:37 |
| *** mtanino has quit IRC | 01:38 | |
| *** baoli has quit IRC | 01:43 | |
| *** markmcclain has joined #openstack-infra | 01:45 | |
| *** nosnos has joined #openstack-infra | 01:50 | |
| *** yaguang has joined #openstack-infra | 01:50 | |
| *** yamahata has quit IRC | 01:51 | |
| *** yaguang has quit IRC | 01:51 | |
| *** yaguang has joined #openstack-infra | 01:51 | |
| *** baoli has joined #openstack-infra | 01:52 | |
| *** cpallares has quit IRC | 01:53 | |
| *** amitgandhinz has joined #openstack-infra | 01:53 | |
| *** amitgandhinz has quit IRC | 01:54 | |
| *** amitgandhinz has joined #openstack-infra | 01:54 | |
| *** fandi has quit IRC | 01:55 | |
| *** bdpayne has quit IRC | 01:56 | |
| *** mriedem1 has joined #openstack-infra | 01:56 | |
| *** mriedem has quit IRC | 01:57 | |
| *** markmcclain has quit IRC | 01:57 | |
| *** adalbas has quit IRC | 01:57 | |
| *** rkukura has quit IRC | 01:57 | |
| *** otter768 has joined #openstack-infra | 01:58 | |
| *** rwsu has quit IRC | 02:03 | |
| *** ChuckC has quit IRC | 02:07 | |
| *** SumitNaiksatam has quit IRC | 02:10 | |
| *** thingee has quit IRC | 02:10 | |
| *** amitgandhinz has quit IRC | 02:13 | |
| *** ryanpetrello has quit IRC | 02:14 | |
| *** dims__ has quit IRC | 02:16 | |
| *** asettle is now known as asettle-appt | 02:17 | |
| *** ChuckC has joined #openstack-infra | 02:18 | |
| *** jyuso has quit IRC | 02:22 | |
| *** gyee has quit IRC | 02:25 | |
| *** rkukura has joined #openstack-infra | 02:28 | |
| *** amitgandhinz has joined #openstack-infra | 02:31 | |
| *** baoli_ has joined #openstack-infra | 02:37 | |
| *** dims__ has joined #openstack-infra | 02:38 | |
| *** baoli__ has joined #openstack-infra | 02:39 | |
| *** baoli has quit IRC | 02:41 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 02:42 | |
| *** baoli_ has quit IRC | 02:42 | |
| *** kaisers1 has joined #openstack-infra | 02:44 | |
| *** kaisers has quit IRC | 02:45 | |
| *** yjiang5 has left #openstack-infra | 02:46 | |
| *** jyuso has joined #openstack-infra | 02:54 | |
| *** ddieterly has joined #openstack-infra | 02:56 | |
| *** shashankhegde has quit IRC | 03:00 | |
| *** hdd has joined #openstack-infra | 03:04 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 03:04 | |
| *** weshay has quit IRC | 03:07 | |
| *** koolhead17 has joined #openstack-infra | 03:07 | |
| *** salv-orlando has joined #openstack-infra | 03:07 | |
| *** amitgandhinz has quit IRC | 03:07 | |
| *** salv-orlando has quit IRC | 03:10 | |
| *** salv-orlando has joined #openstack-infra | 03:10 | |
| *** koolhead_ has joined #openstack-infra | 03:13 | |
| *** dims__ has quit IRC | 03:14 | |
| *** koolhead17 has quit IRC | 03:15 | |
| *** marun has quit IRC | 03:15 | |
| *** salv-orlando has quit IRC | 03:16 | |
| *** salv-orlando has joined #openstack-infra | 03:16 | |
| *** bdpayne has joined #openstack-infra | 03:17 | |
| *** nelsnelson has quit IRC | 03:24 | |
| *** jerryz has joined #openstack-infra | 03:32 | |
| *** nosnos has quit IRC | 03:34 | |
| *** bdpayne has quit IRC | 03:34 | |
| *** ryanpetrello has joined #openstack-infra | 03:38 | |
| openstackgerrit | Praveen Kumar proposed openstack-infra/jenkins-job-builder: Add example to demonstrate flow plugin for job-template https://review.openstack.org/143250 | 03:39 |
| *** david-lyle has quit IRC | 03:39 | |
| *** SumitNaiksatam has joined #openstack-infra | 03:40 | |
| *** baoli__ has quit IRC | 03:40 | |
| *** david-lyle has joined #openstack-infra | 03:45 | |
| *** david-lyle has quit IRC | 03:46 | |
| *** david-lyle has joined #openstack-infra | 03:48 | |
| *** salv-orlando has quit IRC | 03:54 | |
| *** ryanpetrello has quit IRC | 03:55 | |
| *** mfink has joined #openstack-infra | 03:55 | |
| *** Longgeek has joined #openstack-infra | 03:56 | |
| *** hdd has quit IRC | 03:57 | |
| *** Longgeek has quit IRC | 03:57 | |
| *** dannywilson has joined #openstack-infra | 03:58 | |
| *** asettle-appt is now known as asettle | 04:01 | |
| *** dannywilson has quit IRC | 04:02 | |
| *** radez is now known as radez_g0n3 | 04:02 | |
| *** Longgeek has joined #openstack-infra | 04:03 | |
| *** mriedem1 has quit IRC | 04:04 | |
| *** naohirot has joined #openstack-infra | 04:04 | |
| *** otter768 has quit IRC | 04:05 | |
| *** ddieterly has quit IRC | 04:06 | |
| *** bdpayne has joined #openstack-infra | 04:06 | |
| *** sputnik13 has joined #openstack-infra | 04:07 | |
| *** bdpayne has quit IRC | 04:09 | |
| *** yamahata has joined #openstack-infra | 04:11 | |
| openstackgerrit | Ghanshyam Mann proposed openstack-infra/project-config: Move Nova V2.1 job from experimental to check/gate https://review.openstack.org/145123 | 04:12 |
| openstackgerrit | Ghanshyam Mann proposed openstack-infra/project-config: Move Nova V2.1 job from experimental to check/gate https://review.openstack.org/145123 | 04:14 |
| *** dims__ has joined #openstack-infra | 04:15 | |
| *** zz_dimtruck is now known as dimtruck | 04:15 | |
| *** dims__ has quit IRC | 04:20 | |
| *** marun has joined #openstack-infra | 04:20 | |
| *** amotoki has joined #openstack-infra | 04:21 | |
| *** signed8bit has quit IRC | 04:23 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 04:24 | |
| *** david-lyle is now known as david-lyle_afk | 04:25 | |
| *** Alexandra_ has joined #openstack-infra | 04:26 | |
| *** asettle is now known as asettle1 | 04:26 | |
| *** Alexandra_ is now known as asettle | 04:27 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Bring up Fedora 21 nodes on rackspace https://review.openstack.org/140857 | 04:27 |
| *** asettle1 is now known as Alexandra_ | 04:27 | |
| *** nosnos has joined #openstack-infra | 04:29 | |
| *** ddieterly has joined #openstack-infra | 04:37 | |
| *** ddieterly has quit IRC | 04:42 | |
| *** sputnik13 has quit IRC | 04:44 | |
| *** Ryan_Lane has quit IRC | 04:47 | |
| *** pcrews has quit IRC | 04:50 | |
| *** ddieterly has joined #openstack-infra | 04:56 | |
| *** ddieterly has quit IRC | 04:57 | |
| *** ddieterly has joined #openstack-infra | 04:58 | |
| *** sdake has quit IRC | 04:59 | |
| *** ddieterly has quit IRC | 05:02 | |
| *** yfried has quit IRC | 05:07 | |
| *** e0ne has joined #openstack-infra | 05:09 | |
| *** e0ne has quit IRC | 05:11 | |
| *** koolhead_ has quit IRC | 05:13 | |
| *** zul has quit IRC | 05:21 | |
| *** k4n0 has joined #openstack-infra | 05:24 | |
| *** salv-orlando has joined #openstack-infra | 05:25 | |
| *** k4n0 has quit IRC | 05:25 | |
| *** salv-orlando has quit IRC | 05:28 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/project-config: Add experimental Fedora 21 devstack job https://review.openstack.org/140889 | 05:29 |
| *** salv-orlando has joined #openstack-infra | 05:30 | |
| *** hdd has joined #openstack-infra | 05:41 | |
| *** dimtruck is now known as zz_dimtruck | 05:50 | |
| *** Longgeek has quit IRC | 05:51 | |
| *** ildikov has quit IRC | 05:54 | |
| *** sandywalsh has quit IRC | 05:55 | |
| *** mrmartin has joined #openstack-infra | 05:56 | |
| *** sandywalsh has joined #openstack-infra | 05:57 | |
| *** ddieterly has joined #openstack-infra | 05:57 | |
| *** ddieterly has quit IRC | 06:01 | |
| *** punith has joined #openstack-infra | 06:02 | |
| *** Abhishek has joined #openstack-infra | 06:04 | |
| *** dims__ has joined #openstack-infra | 06:04 | |
| *** otter768 has joined #openstack-infra | 06:06 | |
| *** Abhishek has quit IRC | 06:07 | |
| punith | hi all, | 06:07 |
| punith | root@cimaster:/# telnet 127.0.0.1 4730 Trying 127.0.0.1... Connected to 127.0.0.1. Escape character is '^]'. | 06:08 |
| *** Murad has joined #openstack-infra | 06:08 | |
| *** Abhishek has joined #openstack-infra | 06:08 | |
| punith | my greaman is not showing the workers | 06:08 |
| *** Longgeek has joined #openstack-infra | 06:09 | |
| *** otter768 has quit IRC | 06:11 | |
| *** salv-orlando has quit IRC | 06:11 | |
| *** salv-orlando has joined #openstack-infra | 06:13 | |
| *** jyuso has quit IRC | 06:14 | |
| *** jyuso has joined #openstack-infra | 06:15 | |
| *** shashankhegde has joined #openstack-infra | 06:20 | |
| *** shashankhegde has quit IRC | 06:20 | |
| *** salv-orlando has quit IRC | 06:20 | |
| *** zz_dimtruck is now known as dimtruck | 06:20 | |
| *** salv-orlando has joined #openstack-infra | 06:22 | |
| *** yfried has joined #openstack-infra | 06:25 | |
| *** avozza is now known as zz_avozza | 06:27 | |
| *** yfried has quit IRC | 06:30 | |
| *** yfried_ has joined #openstack-infra | 06:30 | |
| *** stevemar2 has joined #openstack-infra | 06:31 | |
| *** dimtruck is now known as zz_dimtruck | 06:32 | |
| *** zz_avozza is now known as avozza | 06:33 | |
| *** stevemar has quit IRC | 06:34 | |
| *** dims__ has quit IRC | 06:38 | |
| *** avozza is now known as zz_avozza | 06:46 | |
| *** koolhead17 has joined #openstack-infra | 06:48 | |
| *** zz_avozza is now known as avozza | 06:50 | |
| *** garyk has joined #openstack-infra | 06:52 | |
| *** reed has quit IRC | 06:54 | |
| *** avozza is now known as zz_avozza | 06:57 | |
| *** ddieterly has joined #openstack-infra | 06:57 | |
| Abhishek | Hi all, | 06:59 |
| Abhishek | can anyone help with the issue of german not showing any workers while doing telnet | 06:59 |
| *** ddieterly has quit IRC | 07:02 | |
| *** e0ne has joined #openstack-infra | 07:02 | |
| *** jamielennox is now known as jamielennox|away | 07:03 | |
| grue_pm | Abhishek: is it actually configured to run on that port? | 07:04 |
| *** Abhishek_ has joined #openstack-infra | 07:06 | |
| *** Abhishek_ has quit IRC | 07:07 | |
| *** Abhishek has quit IRC | 07:08 | |
| *** vigneshvar has joined #openstack-infra | 07:08 | |
| clarkb | did you enter 'status' on that connection? | 07:09 |
| grue_pm | that'd help too | 07:09 |
| clarkb | connecting via telnet wont report info until you query it | 07:09 |
| *** Abhishek has joined #openstack-infra | 07:09 | |
| grue_pm | from bash you can use | 07:10 |
| grue_pm | # (echo status ; sleep 0.1) | netcat 127.0.0.1 4730 | 07:10 |
| grue_pm | or | 07:10 |
| grue_pm | # (echo workers ; sleep 0.1) | netcat 127.0.0.1 4730 | 07:10 |
| grue_pm | apparently | 07:10 |
| Abhishek | thanks for the suggestion, I'll try and see if its working fine or not | 07:12 |
| *** e0ne has quit IRC | 07:18 | |
| *** hdd has quit IRC | 07:25 | |
| *** Abhishek has quit IRC | 07:33 | |
| *** Abhishek has joined #openstack-infra | 07:35 | |
| *** koolhead17 has quit IRC | 07:36 | |
| *** armax has quit IRC | 07:36 | |
| *** asettle has quit IRC | 07:37 | |
| *** koolhead17 has joined #openstack-infra | 07:38 | |
| *** fabo_ is now known as fabo | 07:41 | |
| *** fabo has quit IRC | 07:43 | |
| *** fabo has joined #openstack-infra | 07:43 | |
| *** stevemar2 has quit IRC | 07:44 | |
| *** ildikov has joined #openstack-infra | 07:46 | |
| *** chlong has quit IRC | 07:49 | |
| *** chlong has joined #openstack-infra | 07:49 | |
| *** chlong has quit IRC | 07:51 | |
| *** sarob has joined #openstack-infra | 07:52 | |
| *** chlong has joined #openstack-infra | 07:53 | |
| *** belmoreira has joined #openstack-infra | 07:56 | |
| *** ddieterly has joined #openstack-infra | 07:57 | |
| *** zz_avozza is now known as avozza | 08:00 | |
| *** daya_k has joined #openstack-infra | 08:01 | |
| *** ddieterly has quit IRC | 08:02 | |
| *** teran has quit IRC | 08:02 | |
| *** jedimike has joined #openstack-infra | 08:05 | |
| *** arxcruz has joined #openstack-infra | 08:05 | |
| *** mrunge has joined #openstack-infra | 08:06 | |
| punith | hi guys i'm setting up CI for cinder | 08:06 |
| punith | but i'm getting this error in zuul | 08:06 |
| punith | ERROR zuul.IndependentPipelineManager: Unable to find change queue for project openstack/cinder | 08:06 |
| punith | any fixes ? | 08:06 |
| punith | thanks | 08:06 |
| *** otter768 has joined #openstack-infra | 08:07 | |
| *** fandi has joined #openstack-infra | 08:08 | |
| *** amuller has joined #openstack-infra | 08:09 | |
| *** hashar has joined #openstack-infra | 08:10 | |
| openstackgerrit | Mohammad Hanif proposed openstack-infra/project-config: Adding edge VPN project to stackforge https://review.openstack.org/145160 | 08:10 |
| *** otter768 has quit IRC | 08:11 | |
| *** Abhishek has quit IRC | 08:13 | |
| *** afazekas has joined #openstack-infra | 08:14 | |
| *** salv-orlando has quit IRC | 08:15 | |
| *** salv-orlando has joined #openstack-infra | 08:15 | |
| *** Abhishek has joined #openstack-infra | 08:19 | |
| *** jedimike has quit IRC | 08:19 | |
| anteaya | punith: we are having a third-party meeting in #openstack-meeting right now, it is an ideal place to ask questions about anything dealing with third party matters | 08:23 |
| *** zz_dimtruck is now known as dimtruck | 08:24 | |
| *** sarob has quit IRC | 08:27 | |
| *** belmoreira has quit IRC | 08:27 | |
| *** cnesa has joined #openstack-infra | 08:30 | |
| *** cnesa2 has joined #openstack-infra | 08:31 | |
| *** dimtruck is now known as zz_dimtruck | 08:33 | |
| openstackgerrit | Mohammad Hanif proposed openstack-infra/project-config: Adding edge VPN project to stackforge https://review.openstack.org/145160 | 08:38 |
| *** MaxV has joined #openstack-infra | 08:40 | |
| *** Abhishek has left #openstack-infra | 08:45 | |
| *** omrim has joined #openstack-infra | 08:48 | |
| *** che-arne has quit IRC | 08:49 | |
| *** jistr has joined #openstack-infra | 08:58 | |
| *** jpich has joined #openstack-infra | 08:59 | |
| anteaya | jhesketh: sorry trianths had a question | 09:00 |
| *** dtantsur|afk is now known as dtantsur | 09:00 | |
| jhesketh | sure, I can try :-) | 09:00 |
| anteaya | and now he left irc | 09:00 |
| anteaya | so basically he pointed out that keystone has an oslo.db requirement: http://git.openstack.org/cgit/openstack/oslo.db/ | 09:01 |
| *** Abhishek has joined #openstack-infra | 09:01 | |
| anteaya | and that there is only 1.0.3 on pypi.python.org | 09:01 |
| anteaya | sorry wrong link | 09:02 |
| anteaya | the keystone requirement: http://git.openstack.org/cgit/openstack/keystone/tree/requirements.txt#n24 | 09:02 |
| anteaya | so he is at a loss as to where the correct version of oslo.db should be coming from | 09:02 |
| *** jcoufal has joined #openstack-infra | 09:03 | |
| *** jcoufal has quit IRC | 09:03 | |
| morganfainberg | how is oslo.db 1.1.0 in requirements/global reqs and not in pypi? | 09:03 |
| anteaya | morganfainberg: good question | 09:03 |
| *** jedimike has joined #openstack-infra | 09:05 | |
| anteaya | and it sounds like we don't have a fast and ready answer | 09:06 |
| anteaya | I am willing to pick this up again (or read scrollback if anyone else is interested) after a few more hours sleep | 09:07 |
| morganfainberg | ++ sleep is good | 09:07 |
| anteaya | jhesketh: sorry to dump and run, thanks for the offer I'll read when I return | 09:07 |
| jhesketh | that's okay | 09:07 |
| anteaya | morganfainberg: how are you not asleep? | 09:07 |
| jhesketh | I'm don't have a quick answer | 09:08 |
| jhesketh | but I'll take a closer look after dinner | 09:08 |
| morganfainberg | anteaya, uhm... insomnia. | 09:08 |
| anteaya | jhesketh: hmmm, okay, thank you, enjoy dinner | 09:08 |
| *** derekh has joined #openstack-infra | 09:08 | |
| anteaya | morganfainberg: redshift | 09:08 |
| *** ihrachyshka has joined #openstack-infra | 09:08 | |
| morganfainberg | and i have a trigger to ping on people saying "keystone" | 09:08 |
| anteaya | stop looking at blue light | 09:08 |
| morganfainberg | redshift gives me a massive headache | 09:08 |
| anteaya | morganfainberg: I see that | 09:08 |
| anteaya | morganfainberg: does it | 09:08 |
| morganfainberg | yep. splitting | 09:08 |
| anteaya | I'm out of suggestions then, but sleep soon | 09:09 |
| morganfainberg | i was in a dark room staring at the ceiling. i'll go to sleep soon enough | 09:09 |
| morganfainberg | but heard my phone go off. reminded me to set it to quiet mode ;) | 09:09 |
| anteaya | ah sorry to wake you | 09:10 |
| morganfainberg | don't be | 09:10 |
| morganfainberg | my fault for not setting quiet-time in my phone post holidays :) | 09:10 |
| morganfainberg | annnnd with insomnia good reminder :) | 09:10 |
| anteaya | quiet-time | 09:11 |
| morganfainberg | yeah automatically squelch sounds/notifications between 11pm and 7am | 09:11 |
| anteaya | good idea | 09:11 |
| anteaya | squelch squelch | 09:11 |
| morganfainberg | lol | 09:11 |
| anteaya | my cat doesn't even get up now | 09:11 |
| *** mpaolino has joined #openstack-infra | 09:11 | |
| anteaya | she just lies on the bed and waits for me to return | 09:12 |
| anteaya | she knows she isn't getting fed | 09:12 |
| * anteaya departs | 09:12 | |
| *** yamahata has quit IRC | 09:12 | |
| *** yamahata has joined #openstack-infra | 09:13 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 09:17 | |
| *** Ala has joined #openstack-infra | 09:21 | |
| *** salv-orlando has quit IRC | 09:22 | |
| *** teran has joined #openstack-infra | 09:24 | |
| *** zz_dimtruck is now known as dimtruck | 09:25 | |
| *** andreaf has joined #openstack-infra | 09:25 | |
| *** Masahiro has quit IRC | 09:27 | |
| *** Masahiro has joined #openstack-infra | 09:27 | |
| *** salv-orlando has joined #openstack-infra | 09:27 | |
| *** ihrachyshka has quit IRC | 09:30 | |
| *** Masahiro has quit IRC | 09:31 | |
| *** hashar has quit IRC | 09:31 | |
| *** amotoki has quit IRC | 09:31 | |
| *** Hal has joined #openstack-infra | 09:32 | |
| *** jlibosva has joined #openstack-infra | 09:32 | |
| *** pblaho has joined #openstack-infra | 09:32 | |
| *** Hal is now known as Guest21046 | 09:33 | |
| *** ZZelle has quit IRC | 09:34 | |
| *** ZZelle has joined #openstack-infra | 09:34 | |
| *** dimtruck is now known as zz_dimtruck | 09:34 | |
| *** avozza is now known as zz_avozza | 09:35 | |
| *** hashar has joined #openstack-infra | 09:35 | |
| *** Masahiro has joined #openstack-infra | 09:37 | |
| *** Masahiro has quit IRC | 09:39 | |
| *** ociuhandu has joined #openstack-infra | 09:40 | |
| *** ssam2 has joined #openstack-infra | 09:41 | |
| *** BobBall_AWOL is now known as BobBall | 09:41 | |
| *** cnesa2 has quit IRC | 09:47 | |
| *** cnesa has quit IRC | 09:48 | |
| jedimike | running come tempest jobs, getting the error: ERROR: openstackclient.shell Exception raised: (python-keystoneclient 0.11.2.79.g07edcb5 (/opt/stack/new/python-keystoneclient), Requirement.parse('python-keystoneclient>=0.11.1'), set(['python-neutronclient', 'python-cinderclient', 'python-novaclient'])) | 09:49 |
| *** cnesa has joined #openstack-infra | 09:50 | |
| jedimike | could this be related to the new release of pip? I notice that the python-keystoneclient version is unusual... | 09:50 |
| *** Masahiro has joined #openstack-infra | 09:50 | |
| Abhishek | hi all, i have setup the CI , but the dsvm-tempest-full job is failing due to some failures on running tempest, can anyone help me to resolve thi issue | 09:52 |
| *** Masahiro has quit IRC | 09:53 | |
| *** andreykurilin has joined #openstack-infra | 09:53 | |
| *** andreykurilin has quit IRC | 09:57 | |
| *** Masahiro has joined #openstack-infra | 09:57 | |
| *** andreykurilin has joined #openstack-infra | 09:58 | |
| tchaypo | Is one of our CI regions a bit sad? looking at http://goodsquishy.com/downloads/tripleo-jobs.html i see quite a few instances where multiple tests are failing at the same moment, which sounds to me like something in the environment is sad | 10:00 |
| *** mkoderer_cloud has joined #openstack-infra | 10:00 | |
| *** zz_avozza is now known as avozza | 10:00 | |
| *** Masahiro has quit IRC | 10:00 | |
| *** alexpilotti has joined #openstack-infra | 10:02 | |
| jedimike | tchaypo, can you check the devstaklog.txt for me? We're seeing jobs fail with errors to do with parsing requirements ^^ - if you're seeing the same thing, we can help each other fix | 10:04 |
| jedimike | ERROR: openstackclient.shell Exception raised: (python-keystoneclient 0.11.2.79.g07edcb5 (/opt/stack/new/python-keystoneclient), Requirement.parse('python-keystoneclient>=0.11.1'), set(['python-neutronclient', 'python-cinderclient', 'python-novaclient'])) | 10:04 |
| *** jcoufal has joined #openstack-infra | 10:04 | |
| tchaypo | the first one i look at fails at: | 10:07 |
| tchaypo | 2015-01-06 01:45:54.689 | E: Unable to locate package nagios-plugins-all | 10:07 |
| *** otter768 has joined #openstack-infra | 10:07 | |
| *** Masahiro has joined #openstack-infra | 10:09 | |
| *** isviridov_away is now known as isviridov | 10:10 | |
| *** dims__ has joined #openstack-infra | 10:11 | |
| tchaypo | yep "E: Unable to locate package nagios-plugins-all” seems to be the common factor in most of the failures i’m seeing | 10:12 |
| *** otter768 has quit IRC | 10:12 | |
| tchaypo | and the ones where it’s not clearly that I can’t see a clear issue at all - logs are missing or aren’t detailed enough | 10:12 |
| tchaypo | Where are you seeing the requirement for 0.11.1? | 10:13 |
| *** Masahiro has quit IRC | 10:14 | |
| *** cnesa1 has joined #openstack-infra | 10:15 | |
| *** dims__ has quit IRC | 10:17 | |
| *** cnesa1 has left #openstack-infra | 10:19 | |
| *** cnesa1 has joined #openstack-infra | 10:20 | |
| *** cnesa1 has left #openstack-infra | 10:20 | |
| *** cnesa1 has joined #openstack-infra | 10:21 | |
| *** avozza is now known as zz_avozza | 10:22 | |
| *** zz_avozza is now known as avozza | 10:23 | |
| *** mpaolino has quit IRC | 10:24 | |
| *** e0ne has joined #openstack-infra | 10:25 | |
| *** zz_dimtruck is now known as dimtruck | 10:25 | |
| *** naohirot has quit IRC | 10:26 | |
| tchaypo | hrm. http://packages.ubuntu.com/search?suite=default§ion=all&arch=any&keywords=nagios-plugins-all&searchon=names doesn’t show a package called nagios-plugins-all, but there is one called nagios-plugins | 10:27 |
| *** mpaolino has joined #openstack-infra | 10:36 | |
| *** dimtruck is now known as zz_dimtruck | 10:37 | |
| *** jp_at_hp has joined #openstack-infra | 10:38 | |
| *** teran_ has joined #openstack-infra | 10:39 | |
| *** teran has quit IRC | 10:42 | |
| tchaypo | Okay, the nagios-plugins-all failures are all on https://review.openstack.org/#/c/145038/ which introduces that name, so I can ignore that. | 10:43 |
| *** che-arne has joined #openstack-infra | 10:45 | |
| tchaypo | jedimike: fwiw, I think that even with the new version of pip, 0.11.2.79.g07edcb5 is going to satisfy >=0.11.1 | 10:48 |
| tchaypo | but aside from that I can’t figure out what the problem is | 10:49 |
| jedimike | tchaypo, thanks :) I'm not sure what the problem with the nagois plugins might be either, but if I discover anything while I'm trying to fix the keystone client problem, I'll let you know | 10:49 |
| *** Longgeek has quit IRC | 10:51 | |
| *** Longgeek has joined #openstack-infra | 10:51 | |
| *** Abhishek has quit IRC | 10:51 | |
| *** yfried_ is now known as yfried|afk | 10:52 | |
| *** koolhead17 has quit IRC | 10:52 | |
| *** Masahiro has joined #openstack-infra | 10:54 | |
| *** yfried|afk is now known as yfried_ | 10:57 | |
| *** _nadya_ has joined #openstack-infra | 10:58 | |
| *** Masahiro has quit IRC | 10:58 | |
| *** Masahiro has joined #openstack-infra | 11:00 | |
| *** Abhishek has joined #openstack-infra | 11:01 | |
| *** dhritishikhar has joined #openstack-infra | 11:04 | |
| *** unicell has quit IRC | 11:04 | |
| *** pelix has joined #openstack-infra | 11:04 | |
| *** e0ne is now known as e0ne_ | 11:09 | |
| *** mase_x200 has joined #openstack-infra | 11:09 | |
| *** yfried_ is now known as yfried|afk | 11:10 | |
| *** unicell has joined #openstack-infra | 11:15 | |
| *** e0ne_ has quit IRC | 11:20 | |
| *** avozza is now known as zz_avozza | 11:21 | |
| *** mase_x200 has quit IRC | 11:22 | |
| *** yamahata has quit IRC | 11:23 | |
| *** naohirot has joined #openstack-infra | 11:27 | |
| *** bradjones has quit IRC | 11:28 | |
| *** Guest21046 has quit IRC | 11:29 | |
| *** bradjones has joined #openstack-infra | 11:30 | |
| *** e0ne has joined #openstack-infra | 11:30 | |
| *** marcusvrn1 has joined #openstack-infra | 11:30 | |
| *** marcusvrn has quit IRC | 11:31 | |
| *** andreykurilin has quit IRC | 11:31 | |
| *** rfolco has joined #openstack-infra | 11:34 | |
| *** yfried|afk is now known as yfried_ | 11:39 | |
| *** rfolco has quit IRC | 11:39 | |
| *** cdent has joined #openstack-infra | 11:42 | |
| *** rfolco has joined #openstack-infra | 11:43 | |
| *** adalbas has joined #openstack-infra | 11:43 | |
| tchaypo | oh, the issue with nagios is just that 145038 introduced the name “nagios-plugins-all”, when I think they meant “nagios-plugins-common” or something. I don’t know how to fix the patch because I don’t know what they intended, but it’s not going to break any other changes | 11:44 |
| *** Abhishek has quit IRC | 11:44 | |
| tchaypo | it just gave the appearance of a lot of failures because it’s been retried a few times | 11:44 |
| *** dims__ has joined #openstack-infra | 11:44 | |
| *** arxcruz has quit IRC | 11:47 | |
| *** rfolco has quit IRC | 11:48 | |
| *** Masahiro has quit IRC | 11:48 | |
| *** rfolco has joined #openstack-infra | 11:49 | |
| *** dims__ has quit IRC | 11:49 | |
| *** jaypipes has joined #openstack-infra | 11:51 | |
| *** Guest283 is now known as Vivek | 11:53 | |
| *** Vivek has joined #openstack-infra | 11:53 | |
| *** rfolco has quit IRC | 11:54 | |
| *** rfolco has joined #openstack-infra | 11:55 | |
| *** fandi has quit IRC | 11:56 | |
| *** teran has joined #openstack-infra | 11:58 | |
| *** rfolco has quit IRC | 11:59 | |
| *** dims__ has joined #openstack-infra | 11:59 | |
| *** teran_ has quit IRC | 12:01 | |
| *** rfolco has joined #openstack-infra | 12:02 | |
| *** zz_avozza is now known as avozza | 12:04 | |
| *** rfolco has quit IRC | 12:06 | |
| *** otter768 has joined #openstack-infra | 12:08 | |
| *** pc_m has joined #openstack-infra | 12:09 | |
| *** arxcruz has joined #openstack-infra | 12:11 | |
| *** otter768 has quit IRC | 12:13 | |
| *** ihrachyshka has joined #openstack-infra | 12:17 | |
| *** rfolco has joined #openstack-infra | 12:19 | |
| *** baoli has joined #openstack-infra | 12:19 | |
| *** rfolco has quit IRC | 12:24 | |
| *** baoli has quit IRC | 12:24 | |
| *** baoli has joined #openstack-infra | 12:24 | |
| *** dims__ has quit IRC | 12:25 | |
| *** hashar has quit IRC | 12:25 | |
| *** zz_dimtruck is now known as dimtruck | 12:29 | |
| *** koolhead17 has joined #openstack-infra | 12:29 | |
| *** koolhead17 has joined #openstack-infra | 12:29 | |
| *** doug-fish has joined #openstack-infra | 12:30 | |
| *** aysyd has joined #openstack-infra | 12:30 | |
| *** rfolco has joined #openstack-infra | 12:36 | |
| *** daya_k has quit IRC | 12:38 | |
| *** dimtruck is now known as zz_dimtruck | 12:38 | |
| *** _nadya_ has quit IRC | 12:41 | |
| *** marcusvrn1 has quit IRC | 12:42 | |
| *** zul has joined #openstack-infra | 12:42 | |
| *** marcusvrn has joined #openstack-infra | 12:43 | |
| *** avozza is now known as zz_avozza | 12:43 | |
| *** mpaolino has quit IRC | 12:43 | |
| *** ryanpetrello has joined #openstack-infra | 12:46 | |
| *** mwagner_lap has quit IRC | 12:48 | |
| *** Masahiro has joined #openstack-infra | 12:49 | |
| *** vigneshvar has quit IRC | 12:51 | |
| *** dprince has joined #openstack-infra | 12:51 | |
| *** rfolco has quit IRC | 12:52 | |
| *** rfolco has joined #openstack-infra | 12:52 | |
| *** Masahiro has quit IRC | 12:54 | |
| *** rfolco has quit IRC | 12:57 | |
| *** dims__ has joined #openstack-infra | 12:57 | |
| *** rfolco has joined #openstack-infra | 12:58 | |
| *** acruz_ has joined #openstack-infra | 12:58 | |
| *** arxcruz has quit IRC | 13:00 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add to rally non voting unstable performance job https://review.openstack.org/145098 | 13:00 |
| *** marcusvrn has quit IRC | 13:00 | |
| *** marcusvrn has joined #openstack-infra | 13:01 | |
| *** acruz_ is now known as arxcruz | 13:02 | |
| *** rfolco has quit IRC | 13:02 | |
| openstackgerrit | Chris Dent proposed openstack-infra/project-config: Update tooz project config for sentinel support https://review.openstack.org/145220 | 13:04 |
| *** mpaolino has joined #openstack-infra | 13:04 | |
| *** mfink has quit IRC | 13:05 | |
| *** Hal has joined #openstack-infra | 13:07 | |
| *** Hal is now known as Guest10439 | 13:08 | |
| *** fandi has joined #openstack-infra | 13:11 | |
| dhellmann | anteaya, morganfainberg : oslo.db 1.1 is on pypi, it's just hidden from the default view because 1.0.3 was released later. I've updated the settings so 1.3.0 shows as the latest version | 13:11 |
| *** rfolco has joined #openstack-infra | 13:15 | |
| *** rlandy has joined #openstack-infra | 13:16 | |
| *** ayoung has quit IRC | 13:17 | |
| *** rfolco has quit IRC | 13:19 | |
| *** hashar has joined #openstack-infra | 13:20 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add to rally non voting unstable performance job https://review.openstack.org/145098 | 13:20 |
| *** Guest10439 has quit IRC | 13:21 | |
| mpaolino | sdague, can we talk about https://review.openstack.org/#/c/108714/ ? | 13:23 |
| *** yfried_ has quit IRC | 13:24 | |
| sdague | mpaolino: sure | 13:24 |
| *** e0ne is now known as e0ne_ | 13:25 | |
| mpaolino | I have locally all the changes we discussed | 13:25 |
| mpaolino | so basically now we are installing both libvirt and qemu in the default folders | 13:25 |
| *** mbacchi has joined #openstack-infra | 13:26 | |
| mpaolino | I tested it using an Ubuntu Trusty VM... but this doesn't work because libguestfs installs libvirt on top of our compiled version | 13:26 |
| mpaolino | (libvirt is a dependency of libguestfs-dev) | 13:27 |
| mpaolino | do you have any suggestion about how to solve this? | 13:27 |
| *** mfink has joined #openstack-infra | 13:28 | |
| mpaolino | I can manually install libuestfs-dev without its dependencies before the installation of libvirt, but this would create problems with the other dependencies (libc6, libfuse, etc.) | 13:29 |
| *** Guest10439 has joined #openstack-infra | 13:29 | |
| *** mfink__ has joined #openstack-infra | 13:30 | |
| *** rfolco has joined #openstack-infra | 13:31 | |
| sdague | I just assumed you'd leave the packages in place and install over them | 13:31 |
| sdague | because libvirt is so embedded in distro dependency chains you can't actually remove those packages | 13:32 |
| *** mfink has quit IRC | 13:33 | |
| *** Guest10439 has quit IRC | 13:34 | |
| *** bswartz has quit IRC | 13:34 | |
| *** e0ne_ has quit IRC | 13:35 | |
| mpaolino | ah ok, so what you suggest is to install libvirt beforehand, and later our compiled version... correct? | 13:35 |
| *** rfolco has quit IRC | 13:35 | |
| sdague | so... honestly, lets step back a minute. | 13:36 |
| sdague | assume you could only do this via a devstack plugin - https://review.openstack.org/#/c/142805/5/doc/source/plugins.rst,cm, is that possible? | 13:36 |
| sdague | so let the normal libvirt install proceed, have you plugin replace libvirt | 13:37 |
| sdague | and all the devstack code runs as normal | 13:37 |
| mpaolino | I don't know, I have to check the plugin mechanism to answer | 13:38 |
| mpaolino | I'll let you know | 13:38 |
| mpaolino | thanks for the suggestion | 13:39 |
| *** mwagner_lap has joined #openstack-infra | 13:40 | |
| *** marcusvrn has quit IRC | 13:40 | |
| *** Ala has quit IRC | 13:48 | |
| *** kgiusti has quit IRC | 13:55 | |
| *** kgiusti has joined #openstack-infra | 13:56 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:57 | |
| *** mrunge is now known as mrunge_bbl | 13:57 | |
| *** Ala has joined #openstack-infra | 13:58 | |
| *** e0ne has joined #openstack-infra | 13:58 | |
| *** patrickeast has joined #openstack-infra | 13:59 | |
| *** Masahiro has joined #openstack-infra | 14:00 | |
| openstackgerrit | Jukka Lehtniemi proposed openstack-infra/zuul: Add support for Gerrit replication events https://review.openstack.org/135332 | 14:01 |
| kashyap | mpaolino, Reading the scroll, libvirt is a dep of libviguestfs, since it provides sVirt security mechanism for the temporary guests libguestfs provide | 14:02 |
| *** Longgeek has quit IRC | 14:04 | |
| *** nosnos has quit IRC | 14:04 | |
| *** Longgeek has joined #openstack-infra | 14:05 | |
| *** Masahiro has quit IRC | 14:05 | |
| *** Longgeek has quit IRC | 14:06 | |
| *** dkranz has joined #openstack-infra | 14:07 | |
| *** Longgeek has joined #openstack-infra | 14:07 | |
| kashyap | sdague, IIUC, you're suggesting to replace system installed libvirt? Don't you think it's too invasive? | 14:08 |
| *** zz_dimtruck is now known as dimtruck | 14:08 | |
| *** signed8bit has joined #openstack-infra | 14:08 | |
| *** otter768 has joined #openstack-infra | 14:09 | |
| *** Longgeek has quit IRC | 14:09 | |
| *** Longgeek_ has joined #openstack-infra | 14:09 | |
| sdague | kashyap: it's way less invasive than trying to manage multiples at the same time | 14:10 |
| sdague | and less error prone | 14:10 |
| sigmavirus24 | So https://review.openstack.org/#/c/144799/ seems to be playing Random Failure Roulette =P | 14:11 |
| sdague | if the point of this is to test with a new libvirt, you kind of need to replace the system libvirt with a new one | 14:11 |
| *** naohirot has quit IRC | 14:11 | |
| *** Longgeek_ has quit IRC | 14:11 | |
| *** dustins has joined #openstack-infra | 14:11 | |
| *** bswartz has joined #openstack-infra | 14:11 | |
| kashyap | sdague, Well, seems like you're assuming there's no other libvirt guests that are being run outside of DevStack | 14:11 |
| sdague | yep | 14:11 |
| sdague | we already assume that | 14:11 |
| sdague | devstack is super invasive | 14:12 |
| *** Longgeek has joined #openstack-infra | 14:12 | |
| sdague | if you are trying to dual purpose your devstack box with anything else... stop | 14:12 |
| kashyap | mpaolino, sdague, Yes, of course it is invasive, but still I'm not convinced messing with system installed libvirt. | 14:13 |
| *** ddieterly has joined #openstack-infra | 14:13 | |
| kashyap | sdague, And, I hear you - All my test env involves variants of DevStack | 14:13 |
| *** Longgeek has quit IRC | 14:13 | |
| *** otter768 has quit IRC | 14:14 | |
| *** Longgeek has joined #openstack-infra | 14:14 | |
| mpaolino | kashyap, the previous versions of this patch weren't accepted because of the that feature | 14:15 |
| *** mase_x200 has joined #openstack-infra | 14:15 | |
| kashyap | mpaolino, Okay, let's see the new iteration, I can be your test monkey once you post a new version | 14:15 |
| *** markmcclain has joined #openstack-infra | 14:15 | |
| *** marun has quit IRC | 14:17 | |
| kashyap | Would be nice to get this to completion. | 14:17 |
| mpaolino | yes, I agree ;) | 14:17 |
| *** dimtruck is now known as zz_dimtruck | 14:18 | |
| *** cnesa1 has quit IRC | 14:18 | |
| *** yfried_ has joined #openstack-infra | 14:19 | |
| sdague | so, honestly, I'd really like to see this doable via a devstack external plugin, so it's something that the devstack core code doesn't need to care about | 14:20 |
| sdague | because the devstack core code shouldn't care, all it should care is what libvirtd is named | 14:20 |
| kashyap | Is the plugins mechanism ready to use? Are there existing examples? | 14:20 |
| *** cnesa has quit IRC | 14:21 | |
| *** cnesa1 has joined #openstack-infra | 14:21 | |
| *** kragniz has quit IRC | 14:22 | |
| mpaolino | would it be possible to accept for the time being a non-plugin version and convert the actual implementation when the plugin mechanism will be merged? | 14:22 |
| *** kragniz has joined #openstack-infra | 14:22 | |
| *** unicell has quit IRC | 14:23 | |
| *** unicell has joined #openstack-infra | 14:23 | |
| *** patrickeast has quit IRC | 14:25 | |
| *** amuller has quit IRC | 14:26 | |
| *** mriedem has joined #openstack-infra | 14:28 | |
| *** cnesa1 has quit IRC | 14:28 | |
| kashyap | mpaolino, FWIW, that sounds reasonable to me, as this patch has been in works for over 6 months and had quite a bit of review - Ian also seems to agree as per the last comment. | 14:28 |
| *** sdake has joined #openstack-infra | 14:30 | |
| *** sdake has quit IRC | 14:30 | |
| *** sdake has joined #openstack-infra | 14:30 | |
| kashyap | You can propose that in the patch, and see what the comments will be. | 14:30 |
| *** dhritishikhar has quit IRC | 14:31 | |
| *** ayoung has joined #openstack-infra | 14:31 | |
| *** markmcclain has quit IRC | 14:31 | |
| *** markmcclain has joined #openstack-infra | 14:32 | |
| *** dkliban_afk is now known as dkliban | 14:33 | |
| *** markmcclain has quit IRC | 14:36 | |
| *** Longgeek has quit IRC | 14:36 | |
| *** weshay has joined #openstack-infra | 14:37 | |
| *** mase_x200 has quit IRC | 14:37 | |
| *** markmcclain has joined #openstack-infra | 14:38 | |
| *** markmcclain has quit IRC | 14:39 | |
| mordred | sdague, kashyap, mpaolino: luckily, libvirt is a C library | 14:39 |
| mordred | which means that it's TOTALLY possible to install a new version of it on the system without breaking the distro installed one | 14:40 |
| mordred | so I don't think you need to think of it in terms of "overwriting" the distro libvirt | 14:40 |
| kashyap | mordred, Yes, also see the Daniel Berrange's comment in Patch Set 7: Sep 29 5:33 PM | 14:40 |
| * mordred goes to read | 14:40 | |
| kashyap | I very much agree with that, as I'm one of those hypothetical users that Dan mentions. | 14:40 |
| sdague | mordred: only if you then go an massively muck with devstack with lots of extra paths and invocations | 14:41 |
| kashyap | But yea - I can appreciate DevStack developers perspective. Just wondering if there's a reasonable balance | 14:41 |
| mordred | sdague: you should totally be able to export 2 env vars and be done with it | 14:41 |
| sdague | nope | 14:41 |
| mordred | all you need is PATH and LD_LIBRARY_PATH, no? | 14:41 |
| mordred | and then possibly CFLAGS/CPPFLAGS/LDFLAGS if something is going to build a new copy of python-libvirt on top of that libvirt | 14:42 |
| mordred | of course, you could skip that madness for the purposes of this and add /usr/local/mylivirt/lib to /etc/ld.so.conf and run ldconfig so that LD_LIBRARY_PATH is not needed | 14:42 |
| anteaya | dhellmann: thank you | 14:43 |
| kashyap | mordred, This was the comment I was referring to: | 14:43 |
| *** marun has joined #openstack-infra | 14:43 | |
| mordred | (keep in mind I have not read this patch - I'm just saying taht multiple versions of a C library is way less kludgy than with python) | 14:43 |
| kashyap | "If we use /usr/local then you can only easily install one version of libvirt/QEMU at a time per machine. When people are testing with non-distro versions, it is reasonable that they might want to have different versions on the same machine, eg to try to debug & identify behavioural differences between 2 versions of libvirt or QEMU." | 14:43 |
| anteaya | tchaypo: I tend to think of our system as infra or infrastructure and external testing systems as CI | 14:44 |
| anteaya | tchaypo: it confuses me when folks use CI to refer to what the infra team runs | 14:45 |
| sdague | yeh, so please read the patch. Then also remember what happens when you are outside of the stack.sh script and actually trying to modify and start things (or upgrade things) without the env tampered with | 14:46 |
| sdague | most people completely forget - ok, so later that day, when you log in from a different terminal and all that export logic in stack.sh isn't there, what happens | 14:46 |
| * kashyap paying attention | 14:47 | |
| kashyap | sdague, Agreed - about env variables. | 14:47 |
| *** cnesa has joined #openstack-infra | 14:49 | |
| *** cnesa2 has joined #openstack-infra | 14:49 | |
| pelix | jeblair: you about? | 14:49 |
| tchaypo | anteaya: that makes sense | 14:50 |
| anteaya | tchaypo: thanks | 14:50 |
| sdague | I'm rapidly getting to the point where I don't think this belongs in devstack at all, and it should just be done as an external plugin. | 14:50 |
| mpaolino | sdague, in the case of the above patch, the PATHs variables are written in .bashrc | 14:51 |
| mordred | sdague: nod | 14:51 |
| mpaolino | so if you log in from somewhere else, you whould not have any problem | 14:51 |
| *** vipul has quit IRC | 14:52 | |
| sdague | mpaolino: unless the user uses zsh | 14:52 |
| sdague | or some other shell | 14:52 |
| *** vipul has joined #openstack-infra | 14:52 | |
| *** radez_g0n3 is now known as radez | 14:52 | |
| mpaolino | sdague, maybe it's unlikely but you're right :) | 14:53 |
| radez | mrmartin: ping, hello I wanted to follow up re: oauth2 via openstackid.org | 14:53 |
| mrmartin | hi radez | 14:53 |
| sdague | there are tons of zsh users in our community | 14:54 |
| tchaypo | Yes, many of them nah me regularly about switching | 14:54 |
| radez | mrmartin: I'm not 100% sure that keystone has oauth2 support, but that seems like a good direction to try and go | 14:55 |
| *** garyh has joined #openstack-infra | 14:55 | |
| radez | I believe it was suggested that there is support or that it's in process for keystone? | 14:55 |
| openstackgerrit | Thierry Carrez proposed openstack-infra/project-config: Publish yaml2ical to PyPI https://review.openstack.org/145243 | 14:56 |
| *** jistr has quit IRC | 14:56 | |
| mrmartin | I don't know the keystone status, but I've implemented the oauth2 provider for drupal, and it was not a rocket science | 14:56 |
| *** Longgeek has joined #openstack-infra | 14:56 | |
| mrmartin | but I'm using this from a website, and the login page is redirected to openstackid.org | 14:57 |
| mordred | radez, mrmartin: I believe that keystone has oauth2 support - but pinging morganfainberg because he's smarter | 14:57 |
| radez | thx mordred | 14:57 |
| *** jistr has joined #openstack-infra | 14:57 | |
| *** zz_dimtruck is now known as dimtruck | 14:57 | |
| radez | mrmartin: I've not implimented oauth2 beofre, I'll do some research to understand better what I'm attemting to do and be back in touch if I have some questions. | 14:58 |
| radez | thx for the reply mrmartin | 14:58 |
| mrmartin | ok, if this is the case, you just need to do some integration tests with keystone | 14:58 |
| *** mattt has quit IRC | 14:58 | |
| mrmartin | radez, if you like to do some testing, I have some vagrant scripts for dev deployment of openstackid, and it is better, because you can directly check the logs | 14:59 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 14:59 | |
| radez | mrmartin: oh that woudl be great | 14:59 |
| mordred | mrmartin, radez: I missed the context - what's the idea behind openstackid and keystone? (I may have had a similar idea so I wanted to see if you're thinking the same thing) | 15:00 |
| radez | mrmartin: is there a process for getting an app-id and app secret assigned? | 15:00 |
| mrmartin | we don't have process for this yet. | 15:02 |
| *** jklare has quit IRC | 15:02 | |
| *** cpallares has joined #openstack-infra | 15:02 | |
| *** nelsnelson has joined #openstack-infra | 15:03 | |
| mrmartin | but I suggest to do tests locally, and I can give you an app account for the dev version. For the productive one, I guess, we need to ask an approval from infra team | 15:03 |
| *** mattt has joined #openstack-infra | 15:03 | |
| *** nelsnelson has quit IRC | 15:03 | |
| *** krtaylor has quit IRC | 15:03 | |
| *** nelsnelson has joined #openstack-infra | 15:04 | |
| radez | mordred: TryStack.org is a community openstack cluster | 15:04 |
| radez | Some of the foundation leandership would like to see us move off of facebook auth to using openstackidorg | 15:04 |
| *** zz_avozza is now known as avozza | 15:05 | |
| radez | mrmartin: sounds great, how about we start with a dev account and the vagrant scripts and I'll circle back when I'm ready to move forward from there? | 15:05 |
| *** zz_jgrimm is now known as jgrimm | 15:05 | |
| mrmartin | I suggest to start with the vagrant, because you can access the logs there, and it is a great help during development. And openstackid is a bit restrictive so it is very easy to get banned by ip. | 15:06 |
| radez | mrmartin: sure, woudl you be able to reply to my email with a link to them? | 15:07 |
| mrmartin | yeap, I need to upload it somewhere, and clean-up a bit. | 15:08 |
| mordred | radez: yes. I think that's a great plan | 15:08 |
| radez | mordred: was this similar to plans that you had? | 15:08 |
| *** mattt has quit IRC | 15:08 | |
| *** mattt has joined #openstack-infra | 15:08 | |
| mordred | radez: so that means when you have figured that out, you'll have a story for how keystone things work with oauth things | 15:08 |
| radez | thx mrmartin I'll hold tight till you've got it ready | 15:09 |
| radez | mordred: Sounds like I'm a bit of a pioneer here? | 15:09 |
| radez | :) | 15:09 |
| mordred | radez: kinda - I have a different specific goal in mind, but it does involve keystone and oauth - so it'll be great if someone figures out the plumbing | 15:09 |
| sdague | man, so the pip mirrors got all kinds of aweful huh - http://status.openstack.org/elastic-recheck/#1343313 | 15:10 |
| mordred | radez: my (possibly stupid/possibly crazy) idea is that as we add rest apis to some of the infra services, we should just start running a keystone and register our apis with that - even though we wouldn't be running any other openstack services | 15:10 |
| *** jklare has joined #openstack-infra | 15:10 | |
| mordred | radez: beacuse it's got a nice service catalog and a nice python library and whatnot and we could just re-use that work | 15:10 |
| mordred | but there are WAY too many things I don't understand about that as of yet to know if it's a good or a bad idea | 15:10 |
| *** mattfarina has joined #openstack-infra | 15:10 | |
| radez | mordred: maybe I can get the oauth figured out enought to help you out | 15:11 |
| *** krtaylor has joined #openstack-infra | 15:11 | |
| mordred | radez: I'm excited about this | 15:11 |
| mordred | sdague: hopefully soon jeblair's new pip mirroring on AFS stuff will prevent us from getting broken mirrors | 15:12 |
| mordred | in case anyone is interested - dubstep does a great job of drowning out episodes of law and order that are playing on a nearby TV | 15:14 |
| *** changbl has quit IRC | 15:14 | |
| kaisers1 | Hi! Does anyone have a min to help me with a zuul issue? Just installed on an ubuntu 14.04, zuul runs fine when calling 'zuul-server -d' but exits without comment when running just 'zuul-server' (daemon mode) | 15:14 |
| sdague | mordred: is there an ETA on that? | 15:14 |
| sdague | that's a ton of failed jobs today around this | 15:14 |
| dims__ | sdague: y "Bad md5 hash" all around | 15:14 |
| radez | mordred: me too, I can't wait to dump facebook! | 15:14 |
| sdague | 181 fails in 24 hrs is *a lot* | 15:15 |
| *** mtanino has joined #openstack-infra | 15:15 | |
| mordred | sdague: I do not expect it will be today | 15:15 |
| mordred | sdague: but I know he made really good progress on it before the break | 15:15 |
| sdague | ok | 15:15 |
| mordred | sdague: this is one of the pieces of work: https://bitbucket.org/pypa/bandersnatch/pull-request/12/add-option-to-dir-hash-index-files/diff | 15:17 |
| mordred | sdague: in case you just happen to be interested | 15:18 |
| sdague | cool | 15:19 |
| hughsaunders | hey all, do the nodes provided by nodepool for jobs have consistent disk layout (size, partitions)? | 15:19 |
| *** marcusvrn has joined #openstack-infra | 15:19 | |
| mordred | hughsaunders: I do not believe so - we're a bit limited at the moment by the layout we get in the flavor from each cloud | 15:20 |
| *** mpaolino has quit IRC | 15:20 | |
| *** timcline has joined #openstack-infra | 15:20 | |
| hughsaunders | mordred: thanks, where can I look in the config repo to find the flavours that are used? | 15:21 |
| mordred | one sec ... | 15:21 |
| *** stevemar has joined #openstack-infra | 15:22 | |
| *** dmsimard_away is now known as dmsimard | 15:22 | |
| mordred | hughsaunders: you want to look at modules/openstack_project/templates/nodepool/nodepool.yaml.erb in the system-config repo | 15:22 |
| mordred | hughsaunders: we select flavors based on the "min-ram" parameter | 15:23 |
| mordred | hughsaunders: combined with name-filter | 15:23 |
| *** amitgandhinz has joined #openstack-infra | 15:23 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:23 | |
| hughsaunders | mordred: thanks | 15:24 |
| mordred | hughsaunders: so, http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/templates/nodepool/nodepool.yaml.erb#n229 for instance | 15:24 |
| mordred | hughsaunders: sure thing! | 15:24 |
| hughsaunders | annoying that you can't select images based on name or ID as both have a habbit of changing... | 15:25 |
| *** mrunge_bbl is now known as mrunge | 15:27 | |
| *** avozza is now known as zz_avozza | 15:28 | |
| mordred | yah | 15:28 |
| mordred | tell me about it | 15:28 |
| mordred | we have a fix for that which I'm working on | 15:28 |
| mordred | which is that we're going to stop using images that the provides provide and build our own so taht we can control the names | 15:29 |
| mordred | amongst other things | 15:29 |
| *** tonytan4ever has joined #openstack-infra | 15:30 | |
| sdake | is there some special trick to pushing a signed tag? | 15:31 |
| sdake | I am getting perm denied, but I'm certain I've done it before | 15:31 |
| anteaya | sdake: which repo? | 15:31 |
| hughsaunders | mordred: sounds good, but may increase boot times? guess node pool mitigates that though. | 15:32 |
| sdake | github.com/stackforge/kolla | 15:32 |
| sdake | i am in the milestone group | 15:32 |
| anteaya | sdake: I'll check the permissions | 15:32 |
| sdake | thanks :) | 15:32 |
| sdake | maybe I'm just doing something wrong on my end more likely | 15:32 |
| *** obondarev has joined #openstack-infra | 15:32 | |
| anteaya | let's knock down the easy thing first | 15:32 |
| *** mpaolino has joined #openstack-infra | 15:32 | |
| sdake | i should clone the SSH link right? | 15:32 |
| sdake | or a different link | 15:32 |
| mordred | sdake: you're pushign the tag to gerrit and not github, right? | 15:32 |
| sdake | oh ya pushing to github | 15:33 |
| sdake | how do you push to gerrit? | 15:33 |
| mordred | that's not going to wrok | 15:33 |
| mordred | git push gerrit $name_of_my_tag | 15:33 |
| sdake | thanks! | 15:33 |
| mordred | assuming you've run git review in that repo | 15:33 |
| sdake | right | 15:33 |
| anteaya | sdake: you can only submit tags if you are koala release: http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/stackforge/kolla.config | 15:33 |
| sdake | ya i'm in that group | 15:34 |
| mrmartin | I need a little help with openstackid.org, somebody who have an SSH access to that instance, could send me the output of this command: /opt/deploy/deploy.sh openstackid ? | 15:34 |
| sdake | the problem is the git push gerrit I think :) | 15:34 |
| mrmartin | so this: /opt/deploy/deploy.sh status openstackid - must contain the status command | 15:34 |
| sdake | thanks mordred anteaya :) | 15:34 |
| sdake | I was able to push with git push gerrit | 15:34 |
| anteaya | great | 15:35 |
| mordred | mrmartin: troot@openstackid:~# /opt/deploy/deploy.sh status openstackid | 15:35 |
| mordred | INSTALLED [02-Oct-2014 12:32] | 15:35 |
| mordred | mrmartin: and yeah, without the status command it prints help :) | 15:35 |
| mrmartin | thankyou. | 15:35 |
| clarkb | sdague mordred I think mirror awfulness is related to hpcloud networking fwiw | 15:35 |
| mordred | clarkb: awesome | 15:35 |
| clarkb | I did look at it a bit yesterday and the mirror looked fine | 15:35 |
| *** markmcclain has joined #openstack-infra | 15:36 | |
| clarkb | but connection resets and the like were happening quite a bit | 15:36 |
| jroll | yay, we're not the only broken cloud :P | 15:36 |
| anteaya | dhellmann: if you are about help me understand two things 1) how can you publish something to pypi and then hide it (and if you do can you still access the hidden thing to download it) and 2) what happened with olso.db version releases that 1.0.3 was published after 1.1.0? | 15:37 |
| sdague | clarkb: so could we solve this at the zuul layer, and treat these kinds of failures like the random jenkins fails which zuul auto restarts? | 15:37 |
| clarkb | I dont think zuul currebtly has enough info to do that | 15:38 |
| sdague | because it seems like making the underlying operations reliable has been "challenging" | 15:38 |
| sdague | clarkb: there are a set of jenkins explodes that it restarts? right? | 15:38 |
| clarkb | but possibly if we can put more info in the gearman failures | 15:38 |
| *** david-lyle has joined #openstack-infra | 15:38 | |
| jesusaurus | clarkb: yesterday we weren't able to figure out what caused the gerrit replication problem for puppet-lodgeit, do you want to just manually replicate that project? | 15:38 |
| anteaya | jesusaurus: you read my mind! | 15:38 |
| clarkb | yes because jenkins explodes is a specific gearman failure | 15:38 |
| fungi | mrmartin: radez: still skimming scrollback, but morganfainberg did mention expected working openid support in keystone master now as well, depending on whether openid or oauth works better for your use case | 15:39 |
| * anteaya dislikes split out modules that aren't deleted from the home branch | 15:39 | |
| anteaya | jesusaurus: the problem is that we need to find the issue | 15:39 |
| clarkb | jesusaurus ya we can force replication | 15:39 |
| anteaya | since there are 150 module splits to come | 15:40 |
| clarkb | sdague its just a matter of making that mapping. jenkins exploding already has a neat map | 15:40 |
| mrmartin | fungi, openid worked well for my use-case, I have only one problem, openid not passed the profile picture url, but oauth2 have support for that. | 15:40 |
| morganfainberg | fungi: master keystone should support oidc | 15:40 |
| sdague | clarkb: can you point me to where that is? | 15:40 |
| sdague | in the code? | 15:40 |
| jesusaurus | anteaya: yeah, i agree that we need to know what broke and how to prevent it from breaking again | 15:41 |
| clarkb | sdague its in the jenkins gearman plugin. jenkins exception maps to gearman job exception iirc | 15:41 |
| morganfainberg | Oauth2 is not supported we use oauth 1.1 and its a bit weird. | 15:41 |
| anteaya | jesusaurus: I'm for fixing the lose end today rather than keeping it open | 15:41 |
| anteaya | as long as we keep the find the brokeness on the radar | 15:41 |
| clarkb | sdague then in zuul the gearman driver treats gearman exception differently than gearman failure | 15:42 |
| fungi | anteaya: i tend to think of ci (continuous integration) as one of the various things our infrastructure is designed to accomplish (we just happen to also run a lot of other non-ci stuff like wiki, mailing lists, etherpad, and so on) | 15:42 |
| *** esker has joined #openstack-infra | 15:42 | |
| sigmavirus24 | https://review.openstack.org/#/c/144799/ finally passed jenkins, and I'd love some more review on it from anyone interested in preventing whitespace changes from triggering a global-requirements update on projects | 15:43 |
| *** wenlock has joined #openstack-infra | 15:43 | |
| *** jcoufal_ has joined #openstack-infra | 15:44 | |
| anteaya | fungi: fair enough | 15:44 |
| anteaya | fungi: my mind is just triggered to think of "the other folks" when I hear the acronym | 15:44 |
| anteaya | or see it | 15:44 |
| anteaya | feel free to wag a finger at me | 15:44 |
| fungi | assumptions! tsk tsk | 15:45 |
| anteaya | ha ha ha | 15:45 |
| clarkb | sdague: http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/launcher/gearman.py#n421 thats the zuul side | 15:46 |
| *** marcusvrn1 has joined #openstack-infra | 15:46 | |
| *** marcusvrn has quit IRC | 15:46 | |
| *** jcoufal has quit IRC | 15:47 | |
| *** rwsu has joined #openstack-infra | 15:48 | |
| sdague | clarkb: so setting the build.retry is the thing? | 15:48 |
| openstackgerrit | Marton Kiss proposed openstack-infra/system-config: Release upgrade of openstackid 1.0.3 https://review.openstack.org/145253 | 15:48 |
| *** Masahiro has joined #openstack-infra | 15:49 | |
| clarkb | sdague: ya | 15:50 |
| clarkb | trying to sort out now where the gearman plugin decides what the build status is exception vs failure | 15:50 |
| *** erikwilson has joined #openstack-infra | 15:51 | |
| mrmartin | fungi: this must roll out the new changes to openstackid.org | 15:51 |
| *** erikwilson is now known as erikmwilson_ | 15:51 | |
| *** erikmwilson has quit IRC | 15:52 | |
| fungi | mrmartin: yep, i saw. lgtm | 15:52 |
| *** barra204_ has joined #openstack-infra | 15:52 | |
| mrmartin | but a question, on puppetboard, for openstackid.org the "catalog compiled at" status is unknown. http://puppetboard.openstack.org/node/openstackid.org | 15:52 |
| fungi | clarkb: i've been noticing a lot of random checksum mismatch failures from pip in hpcloud over the past couple days... is there discussion of that buried in the morning's scrollback? | 15:53 |
| *** erikmwilson has joined #openstack-infra | 15:53 | |
| mrmartin | but the reports seems to be updated well. | 15:53 |
| *** barra204_ is now known as shakamunyi | 15:53 | |
| *** obondarev has quit IRC | 15:53 | |
| clarkb | sdague: http://git.openstack.org/cgit/openstack-infra/gearman-plugin/tree/src/main/java/hudson/plugins/gearman/StartJobWorker.java#n230 | 15:54 |
| clarkb | fungi: yes | 15:54 |
| *** Masahiro has quit IRC | 15:54 | |
| clarkb | fungi: but also connection resets and other similar errors | 15:54 |
| fungi | clarkb: okay, so just worse than usual network malaise in hpcloud i guess | 15:55 |
| clarkb | that was basically what I determined yesterday as the server itself seemed happy. | 15:55 |
| *** banix has joined #openstack-infra | 15:55 | |
| dhellmann | anteaya: there's an administrative interface on the pypi site you can use to control which dists associated with a project are visible on that projects' home page. New uploads hide old dists by default. All uploads are still available to pip for installation, so "hiding" it only controls what is listed on the project's main page. | 15:55 |
| clarkb | we use columes for the mirror data too so network trouble may be twice as bad on those nodes :/ | 15:55 |
| fungi | clarkb: same... i did a few download tests from it for files i saw reported broken in the logs and they were fine | 15:56 |
| anteaya | dhellmann: odd, I seem to have been left with the impression that he was getting errors for that download | 15:56 |
| anteaya | dhellmann: whereas according you you, he shouldn't be | 15:56 |
| dhellmann | anteaya: oslo.db 1.0.3 was the patch release to fix an issue with setuptools/pip in the stable/juno series from just before the holiday break; we changed the sqlalchemy dependencies | 15:56 |
| anteaya | dhellmann: I'll follow up with him | 15:56 |
| fungi | mrmartin: i'll manually inspect the server just to make sure it's up to date according to local puppet agent info | 15:57 |
| *** e0ne is now known as e0ne_ | 15:57 | |
| dhellmann | anteaya: ok, let me know if you determine there's something wrong with the file | 15:57 |
| mrmartin | fungi: ok | 15:57 |
| *** hemnafk is now known as hemna | 15:57 | |
| mrmartin | this must show the installed status: /opt/deploy/deploy.sh status openstackid | 15:57 |
| anteaya | dhellmann: ah yes I recall now | 15:58 |
| clarkb | jesusaurus: I just triggered replication for the lodgeit repo | 15:58 |
| anteaya | dhellmann: thanks for the explaination | 15:58 |
| fungi | mrmartin: puppet agent seems to think it's in sync with the data on the puppetmaster, and the deploy status says "INSTALLED [02-Oct-2014 12:32]" | 15:58 |
| *** pblaho_ has joined #openstack-infra | 15:59 | |
| *** bswartz has quit IRC | 15:59 | |
| mrmartin | fungi: and we need a /srv/openstackid/w/app/config/dev/server.php symlink | 15:59 |
| *** reed has joined #openstack-infra | 15:59 | |
| fungi | mrmartin: yep, does not exist | 15:59 |
| anteaya | clarkb: expectations for how long that might take? | 15:59 |
| mrmartin | fungi: not worry, the 1.0.3 patch is not there | 16:00 |
| fungi | anteaya: near-instant | 16:00 |
| *** andreaf is now known as andreaf_ | 16:00 | |
| jesusaurus | clarkb: thanks | 16:00 |
| mrmartin | fungi: I guess the post gating job not finished for 1.0.3 upgrade patch | 16:00 |
| anteaya | fungi: awesome | 16:01 |
| anteaya | yes I can see the lodgit repo on git 1-5 now | 16:01 |
| *** pblaho has quit IRC | 16:01 | |
| anteaya | fungi clarkb is there anything anyone without server access can do to help figure out what broke with repo replcation? | 16:02 |
| *** armax has joined #openstack-infra | 16:02 | |
| *** imcsk8 has quit IRC | 16:02 | |
| *** imcsk8 has joined #openstack-infra | 16:02 | |
| *** dustins_ has joined #openstack-infra | 16:02 | |
| *** dustins has quit IRC | 16:03 | |
| fungi | clarkb: and did anyone say what the story was behind the graphite dead air yet, or does it still need investigation? | 16:03 |
| anteaya | jesusaurus: I rechecked the patch | 16:03 |
| clarkb | fungi: there is graphite dead air? I am unaware of that and probably need to read more scrollback? | 16:04 |
| fungi | anteaya: current best theory, could be a bug in push collision avoidance. you could check gerrit bug reports to see if it's a known issue in 2.8 | 16:04 |
| clarkb | anteaya: you can try to reproduce it on a dev server maybe? my guess is a bug with the replication plugin | 16:04 |
| anteaya | fungi: I can do that | 16:04 |
| jesusaurus | anteaya: awesome | 16:05 |
| fungi | oh, right, not sure if the replication plugin release cycle follows gerrit's, nor whether they share a bug tracker | 16:05 |
| anteaya | clarkb: ah I am not sure how to go about replicating that | 16:05 |
| jesusaurus | mordred: do you happen to remember what kind of "tizzy" gerrit got itself into when the replicationDelay was 0? | 16:05 |
| anteaya | jesusaurus: would you know how to see if we can replicate on the review-dev server? | 16:05 |
| anteaya | I'll start with the replication plugin and report back | 16:05 |
| *** e0ne_ is now known as e0ne | 16:06 | |
| jesusaurus | im wondering if we want to bump that delay up from 1 to the default of 15 seconds | 16:06 |
| *** imcsk8 has quit IRC | 16:06 | |
| *** pblaho_ has quit IRC | 16:06 | |
| *** imcsk8 has joined #openstack-infra | 16:06 | |
| *** pblaho has joined #openstack-infra | 16:07 | |
| jesusaurus | anteaya: to trigger a replication you need to be in the Administrators group then you can use "ssh -p 29418 $server gerrit replicate $project" to trigger the command from the ssh api | 16:07 |
| *** dkranz has quit IRC | 16:08 | |
| kaisers1 | Hi! Can someone give me a hint where to find documentation about the relations between zuul, gearman and the Jenkins Job Builder? I have a setup running zuul, who creates jobs for gearman and i have a Jenkins server with gearman plugin. Is the JBB required to create the Jenkins jobs? Or is it deprecated? | 16:08 |
| jesusaurus | anteaya: but im not sure how to get a colliding replication in the way to get into the same code path | 16:08 |
| *** pblaho has quit IRC | 16:09 | |
| anteaya | zaro: where would I find the version of the gerrit replication plugin we are using? | 16:09 |
| sdague | clarkb: ok, I'm sort of lost in the weeds of the jenkins stack to figure out what sets this at the base level, and how we manage to get the unset state we need for the retry | 16:10 |
| anteaya | jesusaurus: yeah, it is the collision that I don't know about | 16:10 |
| sdague | so I'm going to give up that train of though | 16:10 |
| sdague | thought | 16:10 |
| *** dustins_ has quit IRC | 16:10 | |
| fungi | kaisers1: the relationship (connector) between jenkins and gearman is a jenkins plugin | 16:10 |
| anteaya | jesusaurus: as I don't entirely follow what happened with lodgeit to create a problem yesterday | 16:10 |
| *** otter768 has joined #openstack-infra | 16:10 | |
| sdague | clarkb: unrelated to that, because I know you have strong opinions on additional log levels, I'm curious what you think of this - https://review.openstack.org/#/c/145245/ | 16:10 |
| fungi | kaisers1: jenkins job builder is just a tool to render jenkins job configuration and push it into a jenkins master | 16:11 |
| dhellmann | anteaya: any time! | 16:11 |
| *** _nadya_ has joined #openstack-infra | 16:11 | |
| anteaya | dhellmann: :d | 16:11 |
| anteaya | :D | 16:11 |
| *** dustins has joined #openstack-infra | 16:11 | |
| clarkb | sdague: will that affect how TRACE is used today? or is that mostly just formalizing how it works now? | 16:11 |
| kaisers1 | fungi: Ok, i did install that. Should this be sufficient to create the jenkins jobs? My current test setup has 2 jobs in the gearman queue but so far they do not progress into jenkins, therefore i'm looking for documentation how the details of this integration work... | 16:11 |
| jesusaurus | anteaya: i was looking through the code a little bit yesterday, and it looks like gerrit didnt push to some of the git servers because it detected a push to the same uri already in progress | 16:12 |
| clarkb | sdague: current TRACE is a bit fuzzy to me since its just there as that prefix thing on tracebacks | 16:12 |
| jesusaurus | in that case it's supposed to wait $replicationDelay seconds then try again | 16:12 |
| *** bswartz has joined #openstack-infra | 16:12 | |
| *** vhoward- has left #openstack-infra | 16:12 | |
| sdague | clarkb: correct, it's not actually a log level, I think it was appended because someone didn't understand what it was | 16:12 |
| kaisers1 | fungi: This means the JBB is just a tool used by one of these components and set up by that component or does it need to be set up by me? | 16:13 |
| kaisers1 | Sorry, newbie questions, in know :) | 16:13 |
| *** david-lyle has quit IRC | 16:13 | |
| *** mikedillion has joined #openstack-infra | 16:13 | |
| *** yaguang has quit IRC | 16:14 | |
| fungi | kaisers1: you would need to configure something to put your jjb configuration on your jenkins master and run the jenkins-jobs command line (that could be a human, though we do it with puppet right now) | 16:14 |
| sdague | clarkb: there should be enough info in the change to describe what the proposed change is, hopefully, if not, I need to update that change | 16:14 |
| *** mikedillion has quit IRC | 16:14 | |
| *** erikmwilson_ has quit IRC | 16:14 | |
| anteaya | jesusaurus: hmmm, okay I'm not sure if this is helpful or not: https://gerrit.googlesource.com/plugins/replication/+/46673e4086dff8287a21796534386dd04cd0163b | 16:14 |
| *** erikwilson has joined #openstack-infra | 16:15 | |
| *** pblaho has joined #openstack-infra | 16:15 | |
| clarkb | sdague: I think what is missing for me is what the delta in behavior is. The new behavior is described but not much of the old | 16:15 |
| anteaya | I'm mostly just looking for things that might be remotely related | 16:15 |
| *** otter768 has quit IRC | 16:15 | |
| *** hashar has quit IRC | 16:15 | |
| *** erikwilson is now known as erikmwilson_ | 16:15 | |
| fungi | kaisers1: we have documentation on the individual pieces we've written (zuul, jenkins-job-builder, the gear library, the jenkins-gearman plugin), but not really much in the way of integration walkthroughs. we have puppet modules which demonstrate how we deploy them together for our use | 16:15 |
| *** jistr has quit IRC | 16:15 | |
| sdague | " | 16:15 |
| sdague | TRACE is a logging keyword that is understood by most logging | 16:15 |
| sdague | tools. OpenStack has repurposed this in the past to not be TRACE | 16:15 |
| sdague | logging but instead be used whenever a Stacktrace was dumped." | 16:15 |
| kaisers1 | fungi: aah, ok. So JJB _is_ needed to create the jenkins jobs, thanks! | 16:16 |
| clarkb | sdague: which doesn't cover how it wasn't a log level | 16:16 |
| *** jistr has joined #openstack-infra | 16:16 | |
| clarkb | sdague: or that it is always currently paired with at least one ERROR line iirc | 16:16 |
| kaisers1 | fungi: Yeah, tried the puppet stuff but it contains some stuff that's over the top for my setup. Therefore i went to write an ansible script that i'm working on right now | 16:16 |
| fungi | kaisers1: well, you could use zuul and jenkins together without jjb, for example if you manually configured jobs in the jenkins webui | 16:16 |
| sdague | clarkb: sometimes it is | 16:17 |
| *** jistr has quit IRC | 16:17 | |
| sdague | I've seen cases where it wasn't | 16:17 |
| clarkb | wait sometimes it is a proper level? | 16:17 |
| *** dhritishikhar has joined #openstack-infra | 16:17 | |
| clarkb | or the ERROR prefix line? | 16:17 |
| *** jistr has joined #openstack-infra | 16:17 | |
| sdague | no, it's an appended piece of the output stream | 16:17 |
| *** dannywilson has joined #openstack-infra | 16:17 | |
| *** denis_makogon has quit IRC | 16:17 | |
| kaisers1 | fungi: I think i want to go with automation, instead of manual. :-D | 16:18 |
| *** dannywilson has quit IRC | 16:19 | |
| *** dannywilson has joined #openstack-infra | 16:19 | |
| fungi | kaisers1: of course, just mentioning for completeness that jjb isn't technically a requirement of any of the other pieces--it's an enetirely separate (but extremely complimentary) building block | 16:19 |
| mordred | kaisers1: always the right choice | 16:19 |
| *** ildikov has quit IRC | 16:19 | |
| kaisers1 | fungi: I'm not sure i completely get how JJB and Gearman correlate. From the docs i gather that both work with zuul & Jenkins but do they have any relationship with each other? | 16:19 |
| fungi | kaisers1: jjb and gearman have no relationship whatsoever | 16:20 |
| kaisers1 | Aaaah | 16:20 |
| kaisers1 | ok | 16:20 |
| clarkb | sdague: the new state seems reasonable though. Make TRACE somehing useful rather than just an openstack behavior | 16:20 |
| kaisers1 | fungi: that explains my confusion | 16:20 |
| fungi | kaisers1: gearman is there primarily to provide a scalable interconnect between zuul and workers (internal zuul task workers, jenkins masters, turbo-hipster runners, whatever) | 16:21 |
| jesusaurus | anteaya: no, not quite, we don't quite make it that far through the function. we hit the "Rescheduling replication" log message yesterday | 16:21 |
| *** erikmwilson_ has quit IRC | 16:21 | |
| sdague | clarkb: https://github.com/openstack/neutron/blob/master/neutron/openstack/common/log.py#L144-L148 | 16:21 |
| *** erikwilson has joined #openstack-infra | 16:21 | |
| clarkb | sdague: that is awesome | 16:21 |
| *** Ala has quit IRC | 16:21 | |
| kaisers1 | fungi: i just found out i had a bad misconception of what exactly a Jenkins job is =-O | 16:22 |
| jesusaurus | oh actually, am i reading the right code? | 16:22 |
| jesusaurus | what version of the plugin do we have installed? | 16:22 |
| sdague | so it actually is bypassing the entire python logging mechanism | 16:22 |
| *** erikmwilson has quit IRC | 16:23 | |
| anteaya | jesusaurus: yeah, that is what I | 16:23 |
| *** mkoderer_cloud has quit IRC | 16:23 | |
| anteaya | m wondering myself | 16:23 |
| *** dkranz has joined #openstack-infra | 16:23 | |
| clarkb | jesusaurus: https://review.openstack.org/#/admin/plugins/ can you see that page? | 16:23 |
| sdague | clarkb: ok, once the base logging spec goes in, I'll try to drum up more on this. | 16:23 |
| *** wuhg has quit IRC | 16:24 | |
| anteaya | clarkb: no you have to be admin | 16:24 |
| *** thedodd has joined #openstack-infra | 16:24 | |
| clarkb | anteaya: jesusaurus v2.8.4-15-g6dc8444 replication is bundled with the server iirc | 16:24 |
| clarkb | so that version matches the server too | 16:24 |
| kaisers1 | fungi: I do know that gearman schedules builds to e.g. multiple jenkins masters (in my setup there currently is only one). JJB does set up the Jenkins Jobs that are used to run these builds. Is that correct? | 16:24 |
| *** erikmwilson has joined #openstack-infra | 16:24 | |
| anteaya | clarkb: thanks | 16:24 |
| jesusaurus | clarkb: no i cant open that page | 16:25 |
| *** erikwilson is now known as erikmwilson_ | 16:25 | |
| fungi | kaisers1: that's basically right. gearman is more just a work queue... zuul's scheduler is what's adding the tasks to the gearman queue | 16:25 |
| jesusaurus | clarkb: and im not seeing the replication code in openstack-infra/gerrit | 16:26 |
| *** erikmwilson_ has quit IRC | 16:26 | |
| clarkb | fungi: http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=441&rra_id=all it looks like graphite may have just died? | 16:26 |
| fungi | kaisers1: and if you're using jenkins with the jenkins-gearman plugin, it registers any jobs it's able to run with gearman, so that zuul knows whether there's any reason to solicit that work | 16:26 |
| *** erikwilson has joined #openstack-infra | 16:27 | |
| *** erikmwilson has quit IRC | 16:27 | |
| *** marun has quit IRC | 16:27 | |
| *** yolanda has quit IRC | 16:27 | |
| fungi | clarkb: yes, that's what it's looking like to me too | 16:27 |
| *** sabeen has joined #openstack-infra | 16:27 | |
| *** erikwilson is now known as erikmwilson | 16:27 | |
| clarkb | jesusaurus: http://git.openstack.org/cgit/openstack-infra/gerrit/tree/plugins?h=openstack/2.8.4 apparently it is a submodule? | 16:27 |
| jesusaurus | ah, okay | 16:27 |
| fungi | yeah, gerrit plugins are git submodules | 16:28 |
| *** erikwilson has joined #openstack-infra | 16:28 | |
| kaisers1 | fungi: That does help me for progressing, thanks a lot! :) | 16:28 |
| *** dimtruck is now known as zz_dimtruck | 16:29 | |
| *** _nadya_ has quit IRC | 16:29 | |
| fungi | clarkb: [Tue Jan 6 14:42:12 2015] Out of memory: Kill process 1281 (carbon-cache.py) score 756 or sacrifice child | 16:31 |
| fungi | that 'splains it | 16:31 |
| *** koolhead17 has quit IRC | 16:31 | |
| fungi | http://cacti.openstack.org/cacti/graph.php?action=view&local_graph_id=443&rra_id=all | 16:31 |
| *** e0ne is now known as e0ne_ | 16:31 | |
| *** yolanda has joined #openstack-infra | 16:31 | |
| fungi | may be leaky | 16:31 |
| *** mpaolino has quit IRC | 16:31 | |
| fungi | i've started carbon-cache again | 16:32 |
| jeblair | clarkb, mordred: we still need to move onto more hpcloud networks, right? | 16:33 |
| mordred | jeblair: yes | 16:33 |
| mordred | jeblair: which is currently pending the dib-nodepool work - which I'm ACTUALLY hacking on today | 16:33 |
| jeblair | i think we were hoping that we could wait until we were using dib, so we weren't making 9x images | 16:33 |
| jeblair | yeah | 16:33 |
| * mordred is making good progress today for the first time in a while | 16:34 | |
| *** zz_dimtruck is now known as dimtruck | 16:34 | |
| *** teran has quit IRC | 16:34 | |
| *** bcourt has quit IRC | 16:34 | |
| jeblair | do we think the network issues might be caused by that and might be bad enough to just take the hit and go ahead and move now? | 16:34 |
| mordred | it's entirely possible - we _are_ slamming those virtual routers pretty hard all the time | 16:34 |
| pelix | jeblair: can you comment further on https://review.openstack.org/#/c/133465/ | 16:35 |
| clarkb | my concern is that it is the mirror side network not the client side | 16:35 |
| mordred | that's a good point | 16:35 |
| anteaya | jesusaurus: I'm not sure what git command to use to get the exact v2.8.4-15-g6dc8444 code, so far just doing git checkout v2.8.4-15-g6dc8444 doesn't work | 16:35 |
| clarkb | especially with md5sum errors that seems more likely to be a result of mirror <-> volume communication | 16:35 |
| jeblair | well, if by 'slamming those virtual routers' you mean 'using the network at moderate bandwidth', yeah. i think the actual problem was that the virtual routers are on shared real hardware and therefore noisy neighbors | 16:35 |
| fungi | anteaya: git checkout 6dc8444 | 16:35 |
| anteaya | fungi: thanks | 16:36 |
| jeblair | clarkb: are we getting consistent md5sum errors? | 16:36 |
| clarkb | though in theory that is all over tcp too so shouldn't really be any more or less vulnerable. | 16:36 |
| clarkb | jeblair: ya that is the error that e-r is tracking iirc | 16:36 |
| anteaya | fungi: where did you get the 6dc8444 from? | 16:36 |
| jeblair | clarkb: have a link handy? | 16:36 |
| mordred | clarkb: unless the virtual nics are doing checksum offloading wrong | 16:36 |
| anteaya | fungi: ah just dropped the g | 16:36 |
| anteaya | thanks | 16:36 |
| jeblair | clarkb: and how did you determine that it's mirror side, or is that just a suspicion? | 16:36 |
| mordred | clarkb: we actually used to run in to that when doing drbd mysql gigs | 16:36 |
| jeblair | mordred: wow | 16:36 |
| clarkb | jeblair: http://status.openstack.org/elastic-recheck/ first bug there | 16:37 |
| mordred | clarkb: and had to get the DRBD folks to add an option to DRBD to not trust the nic checksumming | 16:37 |
| mordred | like, it's WAY more common than you'd expect, sadly, even with real nics | 16:37 |
| jeblair | clarkb: okay, so what's the evidence that it's mirror-side? | 16:37 |
| *** banix has quit IRC | 16:37 | |
| jesusaurus | anteaya: based on http://git.openstack.org/cgit/openstack-infra/gerrit/tree/plugins?h=openstack/2.8.4 that clark linked, I checked out 24438f6 | 16:38 |
| fungi | yeah, hardware network offload from virtual machines is tricky business, and depends a lot on the driver vendor's ability to implement it. in a past life we used to end up turning it off in a lot of places where it was exceedingly buggy | 16:38 |
| anteaya | fungi: not yet: http://paste.openstack.org/show/155763/ | 16:38 |
| clarkb | jeblair: well after I thought about it a bit more I don't think it necessarily is :) | 16:38 |
| clarkb | jeblair: if it was real hardware I would be start looking at the disk much more closely because tcp | 16:38 |
| clarkb | but in this case its tcp the whole way down | 16:38 |
| fungi | anteaya: you simply don't have a copy of whatever commit that is | 16:38 |
| fungi | anteaya: it may be a local revision, or a patchset in a gerrit review or something | 16:39 |
| anteaya | fungi: ah would make it hard for me to see then | 16:39 |
| anteaya | fungi: thanks though | 16:39 |
| anteaya | jesusaurus: how did you come up with 24438f6, trying to follow along | 16:39 |
| fungi | anteaya: aha, i was missing context, so this is the version of gerrit we're running in production | 16:40 |
| *** pcrews has joined #openstack-infra | 16:40 | |
| jeblair | d41d8cd98f00b204e9800998ecf8427e is the md5sum of the empty file | 16:40 |
| *** pblaho_ has joined #openstack-infra | 16:40 | |
| *** pblaho has quit IRC | 16:40 | |
| anteaya | and the system-config patch for the lodgeit module split passed jenkins: https://review.openstack.org/#/c/142918/ | 16:40 |
| anteaya | yay who wants to approve it? jeblair? | 16:40 |
| *** gyee has joined #openstack-infra | 16:40 | |
| jesusaurus | anteaya: on that page it says "plugins/replication @ 24438f6" so i went to my gerrit-replication clone and checked out 24438f6 | 16:41 |
| *** e0ne_ has quit IRC | 16:41 | |
| anteaya | fungi: seems to be yes, trying to look at the code for the plugin | 16:41 |
| fungi | anteaya: if i clone openstack-infra/gerrit i'm able to checkout 6dc8444 with no problem | 16:41 |
| anteaya | jesusaurus: ah | 16:41 |
| *** punith has quit IRC | 16:41 | |
| anteaya | ah I had cloned the google replication plugin | 16:42 |
| * anteaya tries again | 16:42 | |
| jesusaurus | fungi: but we want to look at the replication plugin's code | 16:42 |
| clarkb | though if we assume tcp will do its tcp thing then perhaps the backing volume hardware is at fault? but I have no evidence to suggest that is the problem | 16:42 |
| fungi | jesusaurus: yep, simply explaining to anteaya where commit 6dc8444 is | 16:42 |
| anteaya | fungi: thanks | 16:42 |
| jesusaurus | fungi: ah | 16:42 |
| fungi | clarkb: if the backing volume hardware is at fault, i would expect filesystem errors in dmesg | 16:43 |
| *** cnesa has quit IRC | 16:43 | |
| clarkb | fungi: ya and that is clean | 16:43 |
| fungi | it would be a strange situation indeed where your block storage could cause your filesystem to silently return empty files | 16:44 |
| clarkb | bandersnatch isn't complaining about sum mismatches when syncing. Though the sync volume vs client download volumes are nowhere near the same magnitude of data transfer | 16:44 |
| *** cnesa2 has quit IRC | 16:44 | |
| *** mpaolino has joined #openstack-infra | 16:44 | |
| jeblair | clarkb, fungi, mordred: over the past 4 hours, about 1/3 of the instances had the hash d41d8cd98f00b204e9800998ecf8427e which is the hash of the empty file | 16:44 |
| *** banix has joined #openstack-infra | 16:45 | |
| fungi | it it possible there's a transparent web proxy handling traffic between those systems? | 16:45 |
| *** jlibosva has quit IRC | 16:45 | |
| mrmartin | fungi: don't we need an additional +2 here: https://review.openstack.org/#/c/145253/ ? | 16:45 |
| * mordred starts to cry | 16:45 | |
| anteaya | fungi: thanks, I confirm | 16:45 |
| fungi | mrmartin: yes, getting a second +2 would be preferable | 16:46 |
| anteaya | so now what repo should I be looking at to view the replcation plugin we are running in production? | 16:46 |
| *** packet has joined #openstack-infra | 16:46 | |
| anteaya | no sense me looking at the wrong code | 16:46 |
| anteaya | I'm currently on openstack-infra/gerrit HEAD detached at 6dc8444 | 16:46 |
| *** bcourt has joined #openstack-infra | 16:47 | |
| anteaya | I'm in gerrit/plugins/replication and I see no code, should their be code? | 16:48 |
| clarkb | there is a bit of 'File does not exist: /srv/static/mirror/web/simple/pep8/1.4.5' going on in the logs | 16:48 |
| clarkb | er apache logs. and not just for that package | 16:48 |
| *** pradk has joined #openstack-infra | 16:48 | |
| jesusaurus | its not clear to me how to get the plugins in the gerrit project | 16:48 |
| *** pblaho_ has quit IRC | 16:48 | |
| *** pblaho_ has joined #openstack-infra | 16:48 | |
| anteaya | jesusaurus: glad it isn't just me | 16:48 |
| anteaya | jesusaurus: looks like zaro is who we are waiting on, he should be here soon | 16:49 |
| clarkb | that file doesn't actually exist though so I don't think that is the problem (file is pep8-1.4.5) | 16:49 |
| fungi | clarkb: yeah, the way pip works, expect a lot of 404 errors | 16:50 |
| jesusaurus | i just have a separate clone of the replication plugin and then ran "git checkout 24438f6" under the assumption that the link clarkb posted is what we are running in production | 16:50 |
| jesusaurus | which is a pretty large assumption | 16:50 |
| clarkb | jesusaurus: I think that is the proper assumption | 16:50 |
| clarkb | jesusaurus: when you build gerrit you get the "in tree" submodule plugins for free | 16:50 |
| clarkb | and since the submodule is at that ref that should be what we get | 16:50 |
| jesusaurus | ah, okay | 16:50 |
| jeblair | pelix: oh you're darragh. you changed your name. hi. | 16:51 |
| *** jerryz has quit IRC | 16:51 | |
| jeblair | pelix: hacking is full of checks that are not widely agreed upon, are highly contentious, often the opinion of one person, and change all the time. enforcing hacking turns into a never ending argument about new rules. | 16:52 |
| jeblair | pelix: if reviewers are nitping people about stuff covered by hacking on jjb, then i would ask them to stop | 16:52 |
| jeblair | pelix: none of those things are important and they are wasting the time of contributors | 16:52 |
| *** pblaho_ has quit IRC | 16:54 | |
| anteaya | jesusaurus: okay I see the same: Fix JAR manifest claiming it is Commons IO project | 16:54 |
| mrmartin | clarkb, jeblair: may I ask you to approve this patch: https://review.openstack.org/#/c/145253/ ? it is required for oauth2 configuration on openstackid.org | 16:56 |
| clarkb | jeblair: fungi: for http://logs.openstack.org/75/145175/2/gate/gate-heat-python27/6ce3572/console.html#_2015-01-06_15_03_46_035 the apache log is: http://paste.openstack.org/show/155764/ | 16:57 |
| anteaya | jesusaurus: could they nest that a bit deeper | 16:57 |
| *** garyk has quit IRC | 16:57 | |
| *** packet has quit IRC | 16:57 | |
| jeblair | jesusaurus: 142918 aprvd | 16:58 |
| jeblair | mrmartin: done | 16:58 |
| mrmartin | jeblair: thanks! | 16:58 |
| clarkb | jeblair: fungi: that whl is 70021 bytes according to ls on a local download with matching md5sums | 16:58 |
| *** dkranz has quit IRC | 16:58 | |
| jeblair | clarkb: i'm not quite sure what to make of "200 31856" | 16:59 |
| clarkb | GET /packages/py2.py3/p/pbr/pbr-0.10.7-py2.py3-none-any.whl HTTP/1.1" 200 31856 implies we didn't transfer enough bytes right? but | 16:59 |
| *** yamahata has joined #openstack-infra | 16:59 | |
| *** MaxV has quit IRC | 16:59 | |
| clarkb | er but 31856 is > empty file | 16:59 |
| *** apmelton has quit IRC | 16:59 | |
| jeblair | yeah, i guess that still counts as a 200 since it started sending it | 16:59 |
| jesusaurus | anteaya: im not sure what you are looking at, what is Fix JAR manifest claiming it is Commons IO project? | 16:59 |
| *** apmelton has joined #openstack-infra | 17:00 | |
| clarkb | jeblair: ya I think 200 is fine. But we clearly didn't ship enough bytes | 17:00 |
| jeblair | so we're getting truncated files from mirror <-> slave, sometimes 0 bytes. | 17:00 |
| *** jlibosva has joined #openstack-infra | 17:01 | |
| *** thedodd has quit IRC | 17:02 | |
| *** armax has quit IRC | 17:02 | |
| *** bswartz has quit IRC | 17:02 | |
| *** timcline has quit IRC | 17:02 | |
| clarkb | in this case 31k bytes but hashed to zero byte file | 17:03 |
| *** sarob has joined #openstack-infra | 17:03 | |
| anteaya | jesusaurus: that is the commit message for the top commit when I run git log | 17:03 |
| *** shashankhegde has joined #openstack-infra | 17:03 | |
| clarkb | I wonder if pip is detecting a failure, truncating the result file but hashing it anyways? | 17:03 |
| anteaya | jesusaurus: by way of clarifying where I am, and that we are looking at teh same code - I doubt myself here, not you | 17:03 |
| *** marun has joined #openstack-infra | 17:04 | |
| jeblair | clarkb: oh, i didn't notice it was a d41 hash... maybe the server sent 31k but pip didn't end up receiving it | 17:04 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for heat bug 1344150 https://review.openstack.org/145267 | 17:04 |
| uvirtbot | Launchpad bug 1344150 in heat "StacksTestJSON.test_suspend_resume_stack failure" [High,Triaged] https://launchpad.net/bugs/1344150 | 17:04 |
| *** thedodd has joined #openstack-infra | 17:04 | |
| clarkb | jeblair: or that | 17:04 |
| jesusaurus | anteaya: ah, yeah, that's the top commit in my log as well | 17:04 |
| anteaya | jesusaurus: I don't know enough about java to know what void means in a function name | 17:04 |
| *** jpich has quit IRC | 17:04 | |
| anteaya | jesusaurus: yay, looking at the same code, it is a start | 17:04 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/elastic-recheck: Add query for heat bug 1344150 https://review.openstack.org/145267 | 17:05 |
| fungi | clarkb: jeblair: i don't suppose that length turns out to be commonly recurring in the apache log? looking now | 17:05 |
| *** dustins has quit IRC | 17:05 | |
| clarkb | I would half expect a different better error message from pip in that case. Maybe its time to go spelunking | 17:05 |
| jesusaurus | anteaya: if you run 'git grep requestRunway' you will see the two bits of code I was looking at yesterday | 17:05 |
| zaro | morning. | 17:06 |
| anteaya | zaro: morning | 17:06 |
| anteaya | after you have had coffee jesusaurus and I would like to monopolize you for a bit | 17:06 |
| jesusaurus | anteaya: that void means that there is no return value for the function | 17:06 |
| *** e0ne has joined #openstack-infra | 17:07 | |
| *** dustins has joined #openstack-infra | 17:07 | |
| anteaya | jesusaurus: ah cool thanks I had found this: http://docs.oracle.com/javase/tutorial/java/javaOO/accesscontrol.html for understanding scope | 17:07 |
| fungi | hrm, only 7 hits for ' 200 31856 ' in the access log from today, all from different files though | 17:07 |
| fungi | actually, a lot of those are for the pbr 0.10.7 wheel | 17:08 |
| jesusaurus | fungi: maybe grep for the file and not the correct byte count? | 17:08 |
| openstackgerrit | Merged openstack-infra/system-config: Release upgrade of openstackid 1.0.3 https://review.openstack.org/145253 | 17:09 |
| fungi | jesusaurus: yeah, i guess i can just pick one at random | 17:09 |
| clarkb | jeblair: fungi: reading https://github.com/pypa/pip/blob/6.0.6/pip/download.py#L615-L622 it appears that the client is getting 0 bytes | 17:09 |
| fungi | that pbr wheel ought to be a good choice | 17:09 |
| *** daya_k has joined #openstack-infra | 17:10 | |
| anteaya | zaro: specifically we are trying to understand the replicaton plugin and how to debug a requestRunway error | 17:10 |
| *** erikmwilson is now known as erikwilson_ | 17:11 | |
| *** packet has joined #openstack-infra | 17:11 | |
| openstackgerrit | Merged openstack-infra/system-config: Split out lodgeit module https://review.openstack.org/142918 | 17:11 |
| *** dkranz has joined #openstack-infra | 17:12 | |
| *** erikwilson_ is now known as erikmwilson | 17:12 | |
| *** mikedillion has joined #openstack-infra | 17:13 | |
| *** sarob has quit IRC | 17:13 | |
| clarkb | the 70kb at http://logs.openstack.org/75/145175/2/gate/gate-heat-python27/6ce3572/console.html#_2015-01-06_15_03_46_034 indicates pip received a content-length header of 70kb | 17:13 |
| *** erikmwilson is now known as erikwilson_ | 17:14 | |
| *** erikwilson_ has quit IRC | 17:14 | |
| clarkb | that means the mirror side saw a 70kb file locally and set its headers properly then only transfered 31kb. On the pip side it appears to have read zero bytes based on the hash | 17:14 |
| clarkb | oh I see now pip could be masking a more specific error because the hash check happens in a finally | 17:14 |
| *** erikmwilson has joined #openstack-infra | 17:14 | |
| clarkb | and the hash mismatch happens and you raise a new exception | 17:15 |
| clarkb | dstufft: ^ https://github.com/pypa/pip/blob/6.0.6/pip/download.py#L615-L622 is that a bug? | 17:15 |
| *** mpaolino has quit IRC | 17:15 | |
| fungi | analysis of 200 response lengths for that pbr wheel in the current apache access log: http://paste.openstack.org/show/155767/ | 17:15 |
| *** erikmwilson is now known as erikwilson_ | 17:15 | |
| *** erikwilson_ is now known as erikmwilson | 17:15 | |
| *** sarob has joined #openstack-infra | 17:15 | |
| annegentle_ | jeblair: I'd like to put in my formal attic request for the <service>-api repos. How can I write it up with the details? Or make the patch myself to move those repos? My brain is fuzzy after the last few weeks of low activity. | 17:15 |
| dstufft | clarkb: probably it doesn't make sense to use a finally there | 17:15 |
| *** dhritishikhar has quit IRC | 17:15 | |
| *** timcline has joined #openstack-infra | 17:15 | |
| dstufft | since if an exception happened we should likely just raise that instead of comparing the hash | 17:16 |
| clarkb | dstufft: that is what I am thinking. Would probably help us debug this issue too | 17:16 |
| clarkb | I will write a blurb on the bug tracker there | 17:16 |
| *** e0ne has quit IRC | 17:17 | |
| *** e0ne has joined #openstack-infra | 17:17 | |
| pelix | jeblair: that seems like an equally arbitrary decision not to have them enabled. If they are that contentious why are they enabled on some infra projects and disabled on others. | 17:17 |
| fungi | pelix: the cult of cargo | 17:17 |
| *** e0ne has quit IRC | 17:17 | |
| *** timcline_ has joined #openstack-infra | 17:18 | |
| jeblair | annegentle_: if you want to write the patch, it would be against the 'project-config' repo; would pretty much just be string substitution of openstack -> openstack-attic for those repos | 17:18 |
| *** jcoufal_ has quit IRC | 17:18 | |
| *** daya_k has quit IRC | 17:18 | |
| *** Longgeek has quit IRC | 17:19 | |
| jesusaurus | the last bit of the lodgeit module split-out passed tests: https://review.openstack.org/#/c/145031/ | 17:20 |
| fungi | clarkb: jeblair: so those top three response lengths look like they're probably legit. i also find 70291, 70236 and 70290 returned for the same file on the other three pypi mirrors. the 70292 may also be normal though still less frequent... but the 31856, 30408 and 21720 lengths are definitely indicating a problem | 17:20 |
| *** timcline has quit IRC | 17:21 | |
| *** bknudson has quit IRC | 17:21 | |
| fungi | apparently we returned short copies of that file for 0.04% of the requests made | 17:22 |
| *** shashankhegde has quit IRC | 17:22 | |
| *** jistr has quit IRC | 17:22 | |
| *** bcourt has quit IRC | 17:22 | |
| clarkb | dstufft: https://github.com/pypa/pip/issues/2332 | 17:23 |
| *** marcusvrn1 has quit IRC | 17:23 | |
| jeblair | jesusaurus: +2. also i subscribed to the new module in gertty | 17:24 |
| clarkb | jeblair: jesusaurus and approved | 17:24 |
| dstufft | clarkb: looks good to me | 17:25 |
| nibalizer | woo split outs! | 17:25 |
| *** Longgeek has joined #openstack-infra | 17:25 | |
| fungi | i'm going to wager that pip ran into a connection reset partway through the transfer, and so did not retrieve the entirely of the file (based on the stray connection resets we're seeing impact other network traffic in hpcloud) | 17:26 |
| anteaya | annegentle_: I'll get you an example | 17:26 |
| *** markmcclain has quit IRC | 17:27 | |
| zaro | anteaya: you get list of plugins from https://review.openstack.org/#/admin/plugins/ | 17:27 |
| dstufft | clarkb: if you make a PR I Can review it :D if not I'll try to get around to it | 17:27 |
| zaro | but you don't have permission | 17:27 |
| clarkb | dstufft: I was just going to figure that out | 17:27 |
| zaro | it's basically the core plugins+javamelody | 17:27 |
| zaro | anteaya: replication is a core plugin | 17:27 |
| clarkb | dstufft: I have to fork, then clone, make change, edit push, pull request its a slightly different workflow :) also do I use the develop branch or master? | 17:27 |
| dstufft | clarkb: develop branch | 17:28 |
| zaro | which means it's the same version as gerrit version | 17:28 |
| dstufft | I'm probably going to delete the develop branch and just have the master branch, but there's some crap that needs fixed that assumes master == stable first | 17:28 |
| dstufft | because having two branches confuses people | 17:28 |
| *** dhritishikhar has joined #openstack-infra | 17:28 | |
| anteaya | annegentle_: here is an example of the patch to rename a repo: https://review.openstack.org/#/c/131558/ | 17:28 |
| *** mikedillion has quit IRC | 17:29 | |
| jeblair | nibalizer: on 136591, why does it depend on puppetlabs/apache? do we depend on that, or our own apache? | 17:29 |
| anteaya | annegentle_: offer a patch, we'll get it into shape and add the rename request to the infra meeting wikipage: https://wiki.openstack.org/wiki/Meetings/InfraTeamMeeting#Upcoming_Project_Renames | 17:29 |
| jesusaurus | zaro: yesterday there was a problem with the replication plugin where it successfully pushed a new project to two git servers, but three git servers had empty projects | 17:29 |
| anteaya | zaro: you have to be admin to see the review.o.o/admin page, jesusaurus and I already tried and failed to view that page | 17:30 |
| jesusaurus | zaro: the replication plugin logged a "Rescheduling replication..." message | 17:30 |
| dstufft | clarkb: sometimes i think about trying to push to move us to something like phabricator which has a more gerrit like workflow, but some folks are against it | 17:30 |
| dstufft | I don't really see a lot of difference between the two workflows | 17:31 |
| *** thingee has joined #openstack-infra | 17:31 | |
| zaro | jesusaurus: do you know if we are replicating review-dev.o.o? maybe we can use that for testing? | 17:31 |
| anteaya | zaro: clarkb gave us v2.8.4-15-g6dc8444 replication is bundled with the server iirc for plugin version | 17:31 |
| jesusaurus | zaro: i dont know if we are replicating review-dev | 17:32 |
| zaro | yes, replication is a core plugin, so it will have the same version | 17:32 |
| nibalizer | jeblair: it was a direct port of the modulefile | 17:33 |
| nibalizer | i can update it | 17:33 |
| nibalizer | though if we're gonna write our modules into the metadata.json file as dependencies, we should cut releases and upload them to the forge | 17:33 |
| anteaya | zaro: so the big picture is when we do the puppet module split outs (we have about 150 to do) we need the repo replcation to work | 17:34 |
| zaro | jesusaurus: do you know where the configurations are for the replication plugin? | 17:34 |
| anteaya | zaro: and yesterday it didnt' on 3 out of 5 git repos | 17:34 |
| anteaya | and jesusaurus and I are doing our best to figure out how to fix it | 17:34 |
| clarkb | dstufft: the biggest difference (though I think github fixed this recently) is with github you have had to go push buttons in a webbrowser | 17:34 |
| clarkb | dstufft: its nice to be able to just push code to where it needs to go | 17:34 |
| *** arxcruz has quit IRC | 17:34 | |
| zaro | anteaya: ok. we work on it today. | 17:34 |
| *** teran has joined #openstack-infra | 17:34 | |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-storyboard: Add metadata.json to puppet-storyboard https://review.openstack.org/136591 | 17:34 |
| clarkb | (biggest difference as the person pushing code, there are lots of differences in the review process) | 17:35 |
| anteaya | zaro: great thanks | 17:35 |
| jesusaurus | zaro: modules/openstack_project/manifests/review.pp in system-config has blocks that set options like replicationDelay | 17:35 |
| dstufft | clarkb: I used to have my ``git`` actually be an alias to ``hub`` which meant ``git pull-request`` was a thing that worked for me | 17:35 |
| dstufft | but I got lazy during one of my OS reinstalls and didn't reinstall it | 17:35 |
| dstufft | clarkb: I am possibly wierd because I don't find much value in the ability to diff between different versions of the patch I always review the whole diff every time | 17:36 |
| openstackgerrit | Merged openstack-infra/puppet-lodgeit: Add needed files to the new puppet module https://review.openstack.org/145031 | 17:37 |
| zaro | jesusaurus: looking | 17:37 |
| openstackgerrit | Merged openstack-infra/publications: Prettified all HTML files https://review.openstack.org/98654 | 17:37 |
| *** bcourt has joined #openstack-infra | 17:37 | |
| openstackgerrit | Merged openstack-infra/puppet-github: Adding initial files per spec https://review.openstack.org/137757 | 17:38 |
| *** Masahiro has joined #openstack-infra | 17:38 | |
| clarkb | dstufft: I am running tests now. Is it expected that some will fail? | 17:38 |
| dstufft | clarkb: if your environment isn't correctly setup yea | 17:38 |
| dstufft | tox -e py34 -- -m unit will give you ones that shouldn't fail and are a lot faster | 17:39 |
| dstufft | I normally run unit tests locally and let travis do the rest | 17:39 |
| clarkb | kk /me kills current run | 17:39 |
| dstufft | I think our test suite requires installing every scm we use | 17:39 |
| dstufft | s/use/support/ | 17:40 |
| *** teran has quit IRC | 17:40 | |
| openstackgerrit | Merged openstack-infra/publications: New CI Automation doc https://review.openstack.org/118439 | 17:40 |
| openstackgerrit | Merged openstack-infra/publications: Update tools and review purposes. https://review.openstack.org/128722 | 17:41 |
| openstackgerrit | Merged openstack-infra/publications: Add README.rst to using-your-own slide deck https://review.openstack.org/143195 | 17:41 |
| openstackgerrit | Merged openstack-infra/publications: Add info about adding new branch to gerritbot https://review.openstack.org/143198 | 17:41 |
| zaro | anteaya, jesusaurus : so which server is it not replicating to? | 17:42 |
| anteaya | well in this instance git1, git2 and git5 were empty | 17:42 |
| jesusaurus | zaro: clarkb manually triggered a replication | 17:42 |
| anteaya | the repo went to git3 and git4 | 17:42 |
| *** sarob has quit IRC | 17:42 | |
| *** Masahiro has quit IRC | 17:43 | |
| anteaya | zaro: we needed to move ahead and finish the split | 17:43 |
| anteaya | zaro: didn't want to leave things open | 17:43 |
| *** sarob has joined #openstack-infra | 17:43 | |
| zaro | anteaya: so everything is in a good state after clarkb triggered the replication? | 17:44 |
| anteaya | yes | 17:44 |
| anteaya | however we need to find out why it broke and fix it | 17:44 |
| anteaya | as we have 150 more split outs to do | 17:44 |
| anteaya | so manually fixing isn't something we can incorporate in the default workflow | 17:44 |
| openstackgerrit | Merged openstack-infra/puppet-jenkins: Update to direct storyboard link https://review.openstack.org/137249 | 17:44 |
| *** SumitNaiksatam has quit IRC | 17:45 | |
| clarkb | dstufft: https://github.com/pypa/pip/pull/2333 | 17:45 |
| *** changbl has joined #openstack-infra | 17:45 | |
| clarkb | not sure I did that right, let me know if I need to make other changes | 17:45 |
| zaro | anteaya: understood | 17:45 |
| zaro | jesusaurus: is there anymore info from gerrit log besides the "Rescheduling replication..." message? | 17:46 |
| dstufft | clarkb: looks right to me | 17:46 |
| jesusaurus | zaro: im not sure, i dont have access to the logs | 17:48 |
| jesusaurus | clarkb: did you see other relavent log messages yesterday? | 17:48 |
| dstufft | clarkb: now it's just wait for travis to finish running the tests, which takes idk 15 minutes or so I think | 17:48 |
| dstufft | I wish it was faster, working on the new virtualenv with it's tests are nice | 17:48 |
| annegentle_ | perfect anteaya thanks (had a phone call and just circled back) :) | 17:48 |
| dstufft | tests in under a minute <3 | 17:48 |
| *** bknudson has joined #openstack-infra | 17:48 | |
| anteaya | annegentle_: sure, offer something up and me, andreas and pleia2 will get in shape | 17:49 |
| clarkb | zaro: jesusaurus: no that is it. It says rescheduling then nothing | 17:50 |
| clarkb | zaro: jesusaurus and it never actually replicates | 17:50 |
| annegentle_ | anteaya: I think I'll do one repo per patch, sound good? | 17:50 |
| *** dhritishikhar has quit IRC | 17:50 | |
| jesusaurus | clarkb: did it give that message once or more than once? | 17:50 |
| anteaya | annegentle_: well there is no harm in doing multiple repos in one patch | 17:50 |
| anteaya | annegentle_: unless you have a reason to separate them | 17:51 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Added manifest for storyboard-dev.openstack.org https://review.openstack.org/140466 | 17:51 |
| annegentle_ | anteaya: seven patches for seven repos :) | 17:51 |
| *** markmcclain has joined #openstack-infra | 17:51 | |
| anteaya | fungi clarkb jeblair any reason why annegentle_ can't put all the repos she wants in attic in one patch? | 17:51 |
| anteaya | I can't think of any | 17:51 |
| clarkb | jesusaurus: just once | 17:51 |
| jeblair | annegentle_: actually i think one patch would be preferred in this case | 17:52 |
| clarkb | jesusaurus: once per target so three times total | 17:52 |
| annegentle_ | anteaya: okay, yeah it's very similar work | 17:52 |
| *** shashankhegde has joined #openstack-infra | 17:52 | |
| jesusaurus | clarkb: gotcha | 17:52 |
| annegentle_ | jeblair: what no seven brides for seven brothers counter argument? :) | 17:52 |
| *** MaxV has joined #openstack-infra | 17:52 | |
| fungi | anteaya: annegentle_: seems like a fine idea to me. less review effort | 17:52 |
| jeblair | annegentle_, anteaya: we have to merge them all at once, and gerrit can only handle merging about 4 at once before you have to go manually rebasing them, so it will be easier on the day we do the move | 17:52 |
| cody-somerville | fungi: What's the sitrep on the hashsum issue? | 17:52 |
| jesusaurus | zaro: so maybe pool.reschedule() isnt actually causing it to try again? | 17:53 |
| fungi | cody-somerville: know anybody who works at hpcloud? ;) | 17:53 |
| *** sarob_ has joined #openstack-infra | 17:53 | |
| annegentle_ | jeblair: thanks | 17:53 |
| *** melwitt has joined #openstack-infra | 17:53 | |
| cody-somerville | fungi: I do. | 17:53 |
| fungi | cody-somerville: best guess is stray connection resets between the neutron nets for our workers in hpcloud and our pypi mirror in hpcloud are causing truncated gets, but pip misrepresents the failure as a hash mismatch | 17:54 |
| fungi | cody-somerville: clarkb has a pull request proposed to improve pip's error handling there so we can more easily confirm | 17:54 |
| cody-somerville | fungi: I also see the issue happening when downloading from pypi.python.org. | 17:55 |
| *** sarob has quit IRC | 17:55 | |
| jeblair | clarkb, fungi it's all hpcloud-b3 | 17:55 |
| *** sarob_ is now known as sarob | 17:55 | |
| jeblair | cody-somerville: ^ | 17:55 |
| *** marcusvrn has joined #openstack-infra | 17:55 | |
| fungi | jeblair: oh! i didn't think to check... do we pin that router to a single az? | 17:55 |
| jeblair | fungi: yeah, we are at one router per pseudo-region | 17:55 |
| *** markmcclain has quit IRC | 17:56 | |
| *** markmcclain1 has joined #openstack-infra | 17:56 | |
| jeblair | so b3 is exactly one neutron virt router | 17:56 |
| *** shashankhegde_ has joined #openstack-infra | 17:56 | |
| pelix | zaro: any thoughts on config file settings naming in https://review.openstack.org/#/c/68895/12/doc/source/installation.rst? | 17:56 |
| anteaya | jesusaurus zaro I'm going to get a walk in before meeting season starts, will read backscroll when I return (I don't think I am holding onto any knowledge you don't already have) | 17:56 |
| *** Longgeek has quit IRC | 17:56 | |
| fungi | yep, just confirmed all our b3 nodes are in az3 | 17:56 |
| *** dimtruck is now known as zz_dimtruck | 17:56 | |
| *** shashankhegde has quit IRC | 17:56 | |
| *** shashankhegde_ is now known as shashankhegde | 17:56 | |
| jeblair | fungi: oh, i don't think the az has much to do with it actually | 17:56 |
| *** Longgeek has joined #openstack-infra | 17:57 | |
| fungi | jeblair: i was checking to see whether there was a correlation to az as well | 17:57 |
| jeblair | i mean, it could, but based on discussions with the networking folks last time, the virtual network layout crosses azs | 17:57 |
| jeblair | fungi: oh ok | 17:57 |
| jeblair | fungi: but we probably have other pseudo regions using az3 | 17:57 |
| *** che-arne has quit IRC | 17:57 | |
| fungi | jeblair: but yeah, b1 through b5 all have nodes in az3, so no az correlation just router correlation | 17:58 |
| *** sarob has quit IRC | 17:59 | |
| fungi | and node count in b3 is not especially higher than any of the others, so perhaps the router for b3 is sharing a host with someone who is monopolizing resources on its interfaces? | 17:59 |
| zaro | clarkb: maybe try show-queue command to see if there's a bunch of replication jobs backed up? https://review.openstack.org/Documentation/cmd-show-queue.html | 17:59 |
| clarkb | zaro: I did, there werent | 17:59 |
| clarkb | zaro: there were no replication jobs for that repo | 18:00 |
| clarkb | zaro: and just a handful of recent jobs for updates to neutron and nova | 18:00 |
| *** zz_dimtruck is now known as dimtruck | 18:01 | |
| *** jamielennox|away is now known as jamielennox | 18:01 | |
| clarkb | note I am currently cleaning up some old ready nodes on nodepool. Jenkins doesn't seem to agree that those nodes are indeed available for use | 18:01 |
| *** afazekas has quit IRC | 18:01 | |
| clarkb | and it impacts nodepool demand calculations | 18:01 |
| jeblair | fungi, clarkb, cody-somerville: i'm digging up our router configuration | 18:02 |
| zaro | clarkb, jesusaurus : this might be the issue? https://gerrit-review.googlesource.com/#/c/60414/ | 18:03 |
| *** derekh has quit IRC | 18:04 | |
| clarkb | zaro: maybe? replication did work when I triggered it again thouhg. I think it just never got rescheduled | 18:04 |
| zaro | i believe jgit was updated in gerrit ver 2.9.4 as well. | 18:04 |
| zaro | clarkb: Luca reported exact same problem. maybe we should apply same patch? https://gerrit-review.googlesource.com/#/c/58502/ | 18:07 |
| jesusaurus | zaro: looks very similar, yeah | 18:08 |
| *** e0ne has joined #openstack-infra | 18:08 | |
| jeblair | zaro, clarkb: should we have manage-projects trigger replication again after a delay or something? | 18:09 |
| *** armax has joined #openstack-infra | 18:09 | |
| clarkb | jeblair: maybe an end run catch all? | 18:10 |
| jeblair | ya | 18:10 |
| jeblair | clarkb: i'm assuming openstackjenkins2-network3 uses openstackjenkins2-router3. is that correct? | 18:11 |
| clarkb | jeblair: ya they are all mapped like that | 18:11 |
| *** otter768 has joined #openstack-infra | 18:11 | |
| *** amitgandhinz has quit IRC | 18:11 | |
| clarkb | its one subnet, one network, and one router all 1:1:1 like that | 18:11 |
| mordred | ok. during my last phone call, I got the swift related bits of dib-nodepool done - maybe during the next meeting I'll get the glance bits done | 18:12 |
| *** SumitNaiksatam has joined #openstack-infra | 18:13 | |
| *** patrickeast has joined #openstack-infra | 18:13 | |
| mrmartin | fungi: may I ask you to check the status at openstackid.org? /opt/deploy/deploy.sh status openstackid | 18:14 |
| jeblair | clarkb, fungi, cody-somerville: i think the hpcloud network issues stopped about 3 hours ago | 18:15 |
| *** thedodd has quit IRC | 18:15 | |
| clarkb | jeblair: is that based on logstash info? | 18:15 |
| fungi | mrmartin: sure, just a moment | 18:15 |
| jeblair | clarkb: yeah | 18:15 |
| jeblair | clarkb: and the logstash queue seems small enough for that to be accurate | 18:15 |
| clarkb | jeblair: yup | 18:15 |
| fungi | mrmartin: INSTALLED [06-Jan-2015 07:03] | 18:16 |
| mrmartin | ok it is better now | 18:16 |
| *** otter768 has quit IRC | 18:16 | |
| fungi | mrmartin: also /srv/openstackid/w/app/config/dev/server.php is now a symlink to /etc/openstackid/server.php as expected | 18:16 |
| mrmartin | and could you paste the content of server.php here? | 18:16 |
| mrmartin | I've logged in with my account, but cannot see the Oauth2 menu. | 18:18 |
| fungi | mrmartin: http://paste.openstack.org/show/155775/ | 18:18 |
| *** yamahata has quit IRC | 18:18 | |
| mrmartin | this works well both on openstackid-dev and on my local dev environment | 18:18 |
| mrmartin | the server.php content seems to be ok. | 18:18 |
| *** afazekas has joined #openstack-infra | 18:18 | |
| *** dhritishikhar has joined #openstack-infra | 18:19 | |
| *** garyk has joined #openstack-infra | 18:20 | |
| *** thingee has quit IRC | 18:21 | |
| *** amitgandhinz has joined #openstack-infra | 18:21 | |
| *** e0ne has quit IRC | 18:21 | |
| *** ssam2 has quit IRC | 18:21 | |
| *** doug-fish has quit IRC | 18:21 | |
| *** jp_at_hp has quit IRC | 18:22 | |
| armax | anyone willing to nudge this one in? https://review.openstack.org/#/c/142603/ tnx and happy new year :) | 18:22 |
| mrmartin | fungi: I have an idea, may I ask you to clear the keys cached by redis? | 18:23 |
| *** patrickeast has quit IRC | 18:23 | |
| *** patrickeast_ has joined #openstack-infra | 18:23 | |
| *** patrickeast_ is now known as patrickeast | 18:23 | |
| *** asselin has quit IRC | 18:23 | |
| *** primeministerp has quit IRC | 18:24 | |
| fungi | mrmartin: sure, just need to let me know how ;) | 18:25 |
| *** bswartz has joined #openstack-infra | 18:25 | |
| zaro | clarkb: this is probably the replication retry fix (fixed in 2.9.x) https://gerrit-review.googlesource.com/#/c/41320/ | 18:26 |
| mrmartin | fungi: redis -p 6378 -a <redis-password>, then you can query the entries with KEYS * and remove all of them with FLUSHALL | 18:26 |
| *** dangers_away is now known as dangers | 18:26 | |
| mrmartin | you can lookup redis pwd at /etc/redis/redis.conf | 18:26 |
| fungi | mrmartin: doing now | 18:26 |
| *** Longgeek_ has joined #openstack-infra | 18:27 | |
| *** asselin has joined #openstack-infra | 18:28 | |
| *** mpaolino has joined #openstack-infra | 18:28 | |
| fungi | mrmartin: what package provides the 'redis' command-line tool? | 18:28 |
| *** zz_avozza is now known as avozza | 18:28 | |
| mrmartin | redis-cli | 18:28 |
| annegentle_ | anteaya: do you know if the gerrit/acls/openstack/*.config files remain for projects that go to the attic? | 18:28 |
| mrmartin | sry | 18:28 |
| annegentle_ | anteaya: what I mean is, we don't intend to review those repos any more, so are the the review teams meaningful? | 18:29 |
| annegentle_ | trying to decide whether to rename or delete the file | 18:29 |
| *** mpaolino has quit IRC | 18:29 | |
| fungi | mrmartin: from where do you obtain the redis-cli package? it doesn't seem to be in ubuntu | 18:30 |
| *** jlibosva has quit IRC | 18:30 | |
| *** Longgeek has quit IRC | 18:30 | |
| mrmartin | must be there, with redis | 18:30 |
| *** jistr has joined #openstack-infra | 18:30 | |
| mrmartin | dpkg -l | grep redis | 18:31 |
| mrmartin | ii redis-server 2:2.2.12-1build1 Persistent key-value database with network interface | 18:31 |
| fungi | mrmartin: oh, redis-cli is the command, and it's in the redis-server package | 18:31 |
| fungi | mrmartin: got it | 18:31 |
| *** dtantsur is now known as dtantsur|afk | 18:31 | |
| *** jistr has quit IRC | 18:31 | |
| armax | fungi: tnx | 18:31 |
| *** jistr has joined #openstack-infra | 18:32 | |
| fungi | mrmartin: there were 58 cached keys, and now they've all been flushed | 18:32 |
| *** Ryan_Lane has joined #openstack-infra | 18:32 | |
| *** jlibosva has joined #openstack-infra | 18:32 | |
| openstackgerrit | Merged openstack-infra/project-config: move neutron-api job to use gate hooks https://review.openstack.org/142603 | 18:33 |
| mrmartin | fungi: and yes! I've access to oauth2 menu. Need to add redis cache clear to upgrade process somehow. | 18:33 |
| mrmartin | awesome | 18:34 |
| fungi | mrmartin: agreed, sounds like it should be part of the deployment update's exec | 18:34 |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 18:34 |
| *** hdd has joined #openstack-infra | 18:35 | |
| *** bswartz has quit IRC | 18:35 | |
| jogo | sigmavirus24: ping https://github.com/jcrocholl/pep8/issues/362 | 18:36 |
| *** yamahata has joined #openstack-infra | 18:37 | |
| jogo | sigmavirus24: in openstack-infra we use jenkins and don't currently redirect tox output | 18:37 |
| *** avozza is now known as zz_avozza | 18:38 | |
| jogo | sigmavirus24: in part because we run pip freeze as well https://github.com/openstack-infra/project-config/blob/master/jenkins/scripts/run-pep8.sh | 18:38 |
| *** zz_gondoi is now known as gondoi | 18:40 | |
| *** doug-fish has joined #openstack-infra | 18:41 | |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:41 | |
| mrmartin | anyway, as we are starting to upload prod data to the groups portal, do we have backups for the mysql databases at rackspace side, or we need to implement something? | 18:42 |
| *** AJaeger has joined #openstack-infra | 18:42 | |
| jeblair | mrmartin: there is a puppet module that can be used for database backups, and then we should enable off-site system backups for the host | 18:43 |
| mrmartin | ok. | 18:43 |
| jeblair | mrmartin: you can make the change for the database backups (it's easy, see the gerrit or storyboard hosts) | 18:44 |
| jeblair | mrmartin: then when that's done, ask an infra-root to enable host backups | 18:44 |
| mrmartin | and how can I backup the files uploaded by end-users? | 18:44 |
| mrmartin | it is sitting in the filesystem. | 18:44 |
| jeblair | mrmartin: should be handled by the system backup | 18:45 |
| mrmartin | ok, I'll check that. | 18:45 |
| *** yfried_ is now known as yfried|afk | 18:45 | |
| jeblair | mrmartin: the puppet is something like "mysql::backup" for the db part, and then i think there's a "bup" module for the client part of the system backup, then the server part and key exchange needs to be done manually | 18:46 |
| jeblair | mrmartin: so those are the bits to search for on the gerrit and storyboard hosts | 18:46 |
| mrmartin | jeblair: thanks | 18:46 |
| *** pelix has quit IRC | 18:47 | |
| *** bdpayne has joined #openstack-infra | 18:48 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Consolidated OAuth Configuration, added enable flag to oauth cleaner. https://review.openstack.org/144355 | 18:48 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Added cron plugin to clean old access tokens. https://review.openstack.org/140536 | 18:48 |
| *** packet has quit IRC | 18:49 | |
| *** reed has quit IRC | 18:49 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Consolidated OAuth Configuration, added enable flag to oauth cleaner. https://review.openstack.org/144355 | 18:50 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/puppet-storyboard: Added Cron and Token Cleanup to configuration. https://review.openstack.org/144356 | 18:53 |
| *** signed8bit is now known as signed8bit_ZZZzz | 18:53 | |
| *** che-arne has joined #openstack-infra | 18:55 | |
| openstackgerrit | Mauro Rodrigues proposed openstack-infra/subunit2sql: Ensure that transactions are made at each migration https://review.openstack.org/145297 | 18:55 |
| *** banix has quit IRC | 18:56 | |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/storyboard: Consolidated OAuth Configuration, added enable flag to oauth cleaner. https://review.openstack.org/144355 | 18:57 |
| anteaya | annegentle_: rename for now | 18:58 |
| anteaya | annegentle_: we need to change the contents to read only | 18:58 |
| anteaya | annegentle_: example: http://git.openstack.org/cgit/openstack-infra/project-config/tree/gerrit/acls/openstack-attic/melange.config | 18:59 |
| anteaya | but the file needs to be there for each repo | 18:59 |
| pleia2 | meeting time | 19:00 |
| *** signed8bit_ZZZzz is now known as signed8bit | 19:00 | |
| *** Longgeek_ has quit IRC | 19:01 | |
| *** yfried|afk is now known as yfried_ | 19:02 | |
| jhesketh | Morning | 19:02 |
| pleia2 | morning jhesketh | 19:02 |
| anteaya | morning jhesketh | 19:02 |
| *** banix has joined #openstack-infra | 19:02 | |
| *** jlibosva has quit IRC | 19:03 | |
| *** Longgeek has joined #openstack-infra | 19:04 | |
| annegentle_ | anteaya: ah ok | 19:04 |
| anteaya | and you got some more help too | 19:04 |
| anteaya | we are discussing the scheduing in -meeting if you are in there | 19:05 |
| *** julim has joined #openstack-infra | 19:05 | |
| *** bknudson has quit IRC | 19:06 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add tool for generating line graphs of test run times https://review.openstack.org/142649 | 19:07 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Improve query efficiency on populate_run_time migration https://review.openstack.org/143237 | 19:07 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add db_api method for updating an existing test_run https://review.openstack.org/143238 | 19:07 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add migration to add test attrs column to test_runs table https://review.openstack.org/143239 | 19:07 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Switch subunit2sql to set attrs in test_runs table https://review.openstack.org/143240 | 19:07 |
| *** shakamunyi has quit IRC | 19:07 | |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for heat bug 1344150 https://review.openstack.org/145267 | 19:08 |
| uvirtbot | Launchpad bug 1344150 in heat "StacksTestJSON.test_suspend_resume_stack failure" [High,Triaged] https://launchpad.net/bugs/1344150 | 19:08 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add db_api method for updating an existing test_run https://review.openstack.org/143238 | 19:09 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add migration to add test attrs column to test_runs table https://review.openstack.org/143239 | 19:09 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Switch subunit2sql to set attrs in test_runs table https://review.openstack.org/143240 | 19:10 |
| *** dkingshott has joined #openstack-infra | 19:11 | |
| openstackgerrit | Travis Tripp proposed openstack/requirements: Remove old deprecated xstatic angular packages https://review.openstack.org/145096 | 19:14 |
| *** yamahata has quit IRC | 19:14 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/devstack-gate: Handle more than one stream stored testrepository https://review.openstack.org/143727 | 19:15 |
| *** reed has joined #openstack-infra | 19:17 | |
| *** hdd has quit IRC | 19:18 | |
| *** rickerc has joined #openstack-infra | 19:18 | |
| *** markmcclain1 has quit IRC | 19:18 | |
| *** e0ne has joined #openstack-infra | 19:18 | |
| *** e0ne is now known as e0ne_ | 19:19 | |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 19:21 |
| *** yamahata has joined #openstack-infra | 19:21 | |
| *** ZZelle_ has joined #openstack-infra | 19:23 | |
| *** markmcclain has joined #openstack-infra | 19:24 | |
| *** ryanpetrello_ has joined #openstack-infra | 19:25 | |
| *** vponomaryov has quit IRC | 19:26 | |
| *** yamahata has quit IRC | 19:26 | |
| *** Masahiro has joined #openstack-infra | 19:27 | |
| *** dhritishikhar has quit IRC | 19:27 | |
| *** ryanpetrello has quit IRC | 19:27 | |
| *** ryanpetrello_ is now known as ryanpetrello | 19:27 | |
| *** bswartz has joined #openstack-infra | 19:28 | |
| *** e0ne_ has quit IRC | 19:28 | |
| openstackgerrit | Joe Gordon proposed openstack-infra/elastic-recheck: Remove resolved bugs https://review.openstack.org/145310 | 19:29 |
| *** packet has joined #openstack-infra | 19:30 | |
| *** che-arne has quit IRC | 19:31 | |
| *** yfried_ is now known as yfried|afk | 19:31 | |
| *** Masahiro has quit IRC | 19:31 | |
| *** cdent has quit IRC | 19:32 | |
| *** e0ne has joined #openstack-infra | 19:34 | |
| *** gondoi is now known as zz_gondoi | 19:35 | |
| *** ihrachyshka has quit IRC | 19:35 | |
| *** thedodd has joined #openstack-infra | 19:37 | |
| *** david-lyle_afk is now known as david-lyle | 19:39 | |
| *** mfink__ has quit IRC | 19:39 | |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 19:42 |
| *** mfink has joined #openstack-infra | 19:42 | |
| *** baoli has quit IRC | 19:43 | |
| *** dims__ has quit IRC | 19:45 | |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 19:46 |
| *** dims__ has joined #openstack-infra | 19:46 | |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 19:48 |
| *** dims__ has quit IRC | 19:49 | |
| *** sarob has joined #openstack-infra | 19:49 | |
| *** dims__ has joined #openstack-infra | 19:49 | |
| *** Rockyg has joined #openstack-infra | 19:51 | |
| *** hashar has joined #openstack-infra | 19:51 | |
| *** shakamunyi has joined #openstack-infra | 19:51 | |
| *** barra204_ has joined #openstack-infra | 19:55 | |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 19:57 |
| *** yfried|afk is now known as yfried_ | 19:57 | |
| clarkb | mordred: over here to not interfere with last minutes of meeting | 19:58 |
| clarkb | mordred: but we know it does not work on ubuntu | 19:58 |
| mordred | right | 19:58 |
| clarkb | mordred: that is the value in this. If someone asked I can honestly say don't use py34 with openstack on ubuntu | 19:58 |
| clarkb | actually don't use py34 on ubuntu at all | 19:58 |
| *** shakamunyi has quit IRC | 19:58 | |
| mordred | yes - those are useful things | 19:58 |
| mordred | let me clarify | 19:59 |
| mordred | I agree that we learn things by using non-boutique distros | 19:59 |
| clarkb | so the value is in knowing either way and not necessarily make it owrk in all these iterations on platform X | 19:59 |
| mordred | BUT | 19:59 |
| mordred | I don't necessarily agree that just using a non-boutique distro achieves our goals of testing on python 3.4 since python 3.4 apparently doesn't mean anything | 19:59 |
| sdague | question time, it seems that in devstack-gate we've grown a ton of DEVSTACK_GATE_SOME_LONG_VARIABLE=1 values that are basically just there to set a single devstack config option | 20:00 |
| sdague | instead of doing that, should we just provide the ability for a pass through block? | 20:00 |
| mordred | sdague: ++ | 20:00 |
| *** CIBuilds has joined #openstack-infra | 20:00 | |
| anteaya | sdague: I see the same | 20:01 |
| anteaya | sdague: it looks like something is needed | 20:01 |
| *** MaxV has quit IRC | 20:01 | |
| *** hyakuhei has quit IRC | 20:03 | |
| clarkb | in the psat we intentionally did not pass anything through | 20:04 |
| clarkb | so that the localrc can be copy pasta'd | 20:04 |
| clarkb | among other things | 20:04 |
| sdague | it still will be | 20:04 |
| clarkb | I guess I don't understand the suggestion then | 20:04 |
| *** ayoung has quit IRC | 20:04 | |
| openstackgerrit | Sean Dague proposed openstack-infra/devstack-gate: create the ability to pass through devstack config https://review.openstack.org/145321 | 20:04 |
| *** asselin has quit IRC | 20:05 | |
| sdague | that's basically the suggestion | 20:05 |
| *** KurtMartin has quit IRC | 20:05 | |
| *** asselin has joined #openstack-infra | 20:05 | |
| *** KurtMartin has joined #openstack-infra | 20:05 | |
| clarkb | I see pass through into the config not preserver your env to devstack | 20:05 |
| sdague | right | 20:05 |
| clarkb | sdague: fwiw setting multiline bash variables sucks | 20:05 |
| *** ayoung has joined #openstack-infra | 20:06 | |
| clarkb | might be easier with file to append? | 20:06 |
| sdague | here doc isn't bad | 20:06 |
| sdague | that's what I'd expect people to do honestly | 20:06 |
| *** teran has joined #openstack-infra | 20:06 | |
| clarkb | except you have to do that with read iirc and when it hits EOF it returns not 0 so set -e fails and you have to ignore errors | 20:07 |
| sdague | ? | 20:07 |
| clarkb | the heardoc trick | 20:07 |
| clarkb | you read into a var with a heredoc | 20:08 |
| *** KurtMartin has quit IRC | 20:08 | |
| clarkb | heredocs end with an EOF which makes read return an error | 20:08 |
| *** asselin has quit IRC | 20:08 | |
| *** KurtMartin has joined #openstack-infra | 20:08 | |
| clarkb | its just one of those things that I get annoyed with in bash | 20:08 |
| pleia2 | jeblair: thanks for the publications reviews, I really appreciate it | 20:08 |
| anteaya | pc_m: do ask | 20:08 |
| pc_m | Hi! | 20:09 |
| anteaya | hi | 20:09 |
| pc_m | I'm trying to get DevStack to run with two routers, so that I can test reference VPN under a single VM (versus two nodes and bare metal). | 20:09 |
| *** dprince has quit IRC | 20:09 | |
| pc_m | Been struggling for weeks. | 20:09 |
| clarkb | pc_m: see the stack at https://review.openstack.org/#/c/131604/ | 20:10 |
| pc_m | I *think* I finally have it working, but there are some odd things. Wondering if anyone knows | 20:10 |
| *** bknudson has joined #openstack-infra | 20:10 | |
| pc_m | clarkb: So I'm trying to do one node, single devstack, with two private nets, two routers, and VPN across the shared public net. | 20:11 |
| *** teran has quit IRC | 20:11 | |
| anteaya | jeblair: might I be so bold as to request a copy of your gertty.yaml (password redacted)? | 20:11 |
| clarkb | oh thats different than the stack at 131604 | 20:11 |
| pc_m | I added the second net, subnet, and router. | 20:11 |
| jeblair | anteaya: it's in the gertty repo | 20:11 |
| jeblair | anteaya: it's the openstack sample | 20:11 |
| anteaya | jeblair: ah okay thanks | 20:11 |
| pc_m | The odd thing I see is that, from the router2 namespace or a VM in router2 namespace, I can ping router1! | 20:11 |
| pc_m | This does not work the other way. | 20:11 |
| *** markmcclain has quit IRC | 20:12 | |
| pc_m | I cannot ping VM to VM (one in each private net). | 20:12 |
| *** otter768 has joined #openstack-infra | 20:12 | |
| pc_m | I did get VPN to work, and packets to pass encrypted from VM to VM. | 20:12 |
| pc_m | Just puzzled right now, as to why I can ping from 10.2.0.2 to 10.1.0.1 (router) | 20:12 |
| openstackgerrit | Anne Gentle proposed openstack-infra/project-config: Moves the <service>-api repos to the attic https://review.openstack.org/145289 | 20:12 |
| pc_m | Anyone have any ideas? | 20:13 |
| clarkb | pc_m: you probably have a route from one to the other | 20:13 |
| clarkb | maybe an ip assigned to an interface adds an implicit route somewhere | 20:13 |
| anteaya | annegentle_: what is up with that long long line at teh bottom of the commit message? | 20:13 |
| annegentle_ | anteaya: it's a link, can I wrap it? | 20:14 |
| sdague | clarkb: ok, let me figure out if there is a sane way to do this | 20:14 |
| anteaya | annegentle_: I dont' know | 20:14 |
| anteaya | it does stick out as a bit of a sore thumb | 20:14 |
| annegentle_ | anteaya: heh. yeah. it's to answer Joe Gordon's question about background. | 20:14 |
| *** markmcclain has joined #openstack-infra | 20:15 | |
| annegentle_ | anteaya: I could use a link shortener but that seems sketchy | 20:15 |
| clarkb | pc_m: I would expect routers to be able to pass traffic between them if they can "see" each other | 20:15 |
| *** ildikov has joined #openstack-infra | 20:15 | |
| pc_m | clarkb: Hmmm. I do see a route on the host for 10.1.0.0 that has a public IP as a GW to br-ex | 20:15 |
| pc_m | clarkb: Wonder if that gets added as part of the original stacking. | 20:16 |
| anteaya | annegentle_: ah okay well it lives then | 20:17 |
| clarkb | sdague: its probably ok to do what you have if nothing else is cleaner. Bash is just bad at this particular thing | 20:17 |
| anteaya | jeblair: we didn't get to the procedural -2 in the meeting | 20:17 |
| *** otter768 has quit IRC | 20:17 | |
| anteaya | so to prevent annegentle_'s patch merging before the gerrit downtime, who should -2 it | 20:17 |
| anteaya | it looks ready to me | 20:17 |
| jeblair | anteaya: you can -2 or wip it if you want. the folks doing the rename can remove -2s anyway. | 20:18 |
| *** shashankhegde has quit IRC | 20:18 | |
| anteaya | okay I will -2 then, to prevent the merge | 20:19 |
| *** AJaeger has quit IRC | 20:19 | |
| *** HeOS has joined #openstack-infra | 20:20 | |
| *** MaxV has joined #openstack-infra | 20:20 | |
| sdague | clarkb: so this seems to work fine http://paste.openstack.org/show/155793/ | 20:23 |
| openstackgerrit | Adam Gandelman proposed openstack-infra/devstack-gate: Set ironic callback timeout < build timeout https://review.openstack.org/145327 | 20:24 |
| *** jistr has quit IRC | 20:25 | |
| clarkb | sdague: huh if I do interactively it does not work | 20:25 |
| clarkb | and I am using bash | 20:25 |
| clarkb | but does work when run as a script. so where is my derp | 20:25 |
| clarkb | lol quotes. Now I feel silly | 20:26 |
| clarkb | echo $bob != echo "$bob" | 20:26 |
| pc_m | clarkb: Looks like devstack creates the route, if L3 service is enabled. | 20:27 |
| sdague | yes, you need the quotes | 20:28 |
| sdague | otherwise it strips the new lines | 20:28 |
| sdague | that took me about 20 minutes to figure out myself :) | 20:28 |
| *** ociuhandu has quit IRC | 20:29 | |
| *** _nadya_ has joined #openstack-infra | 20:30 | |
| *** mrunge has quit IRC | 20:30 | |
| fungi | right, without quotes it assumes you're wrapping for readability | 20:35 |
| *** mfink has quit IRC | 20:35 | |
| fungi | and helpfully replaces them with spaces while parsing | 20:35 |
| *** dimtruck is now known as zz_dimtruck | 20:38 | |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add tool for generating line graphs of test run times https://review.openstack.org/142649 | 20:39 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Improve query efficiency on populate_run_time migration https://review.openstack.org/143237 | 20:39 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add db_api method for updating an existing test_run https://review.openstack.org/143238 | 20:39 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Add migration to add test attrs column to test_runs table https://review.openstack.org/143239 | 20:39 |
| openstackgerrit | Matthew Treinish proposed openstack-infra/subunit2sql: Switch subunit2sql to set attrs in test_runs table https://review.openstack.org/143240 | 20:39 |
| *** andreykurilin has joined #openstack-infra | 20:43 | |
| anteaya | jeblair: I'm finding all sorts of fun things to make gertty unhappy, my latest accomplishment: http://paste.openstack.org/show/155795/ | 20:43 |
| *** hdd has joined #openstack-infra | 20:44 | |
| anteaya | to replicate I blow away gertty.db, resync with project-config, press f2 for the my changes dashboard, press f3 for the unreviewed project-config and gertty quits | 20:44 |
| *** pcaruana is now known as pcaruana|afk| | 20:45 | |
| openstackgerrit | K Jonathan Harker proposed openstack-infra/jeepyb: Make the gerrit review expiry age tunable https://review.openstack.org/145330 | 20:45 |
| Rockyg | anteaya: have you ever considered a career in QA;-) | 20:50 |
| *** yfried_ is now known as yfried|afk | 20:51 | |
| anteaya | it breaks on label:Code-Review=0,self | 20:51 |
| anteaya | well I'm better with no labels | 20:51 |
| anteaya | I need to move around | 20:51 |
| anteaya | but there are few things I can't figure out how to break | 20:53 |
| anteaya | it is the ,self | 20:59 |
| anteaya | it needs ,user=self | 20:59 |
| *** harlowja_at_home has joined #openstack-infra | 20:59 | |
| *** zz_avozza is now known as avozza | 21:00 | |
| anteaya | jeblair: do you consider this a bug or a hole in the documentation? | 21:00 |
| *** gyee has quit IRC | 21:00 | |
| *** esker has quit IRC | 21:01 | |
| *** baoli has joined #openstack-infra | 21:03 | |
| *** shashankhegde has joined #openstack-infra | 21:04 | |
| *** unicell has quit IRC | 21:06 | |
| *** thedodd has quit IRC | 21:06 | |
| *** teran has joined #openstack-infra | 21:07 | |
| *** SumitNaiksatam has quit IRC | 21:07 | |
| *** thedodd has joined #openstack-infra | 21:07 | |
| *** zz_dimtruck is now known as dimtruck | 21:08 | |
| *** mrmartin has quit IRC | 21:08 | |
| *** teran has quit IRC | 21:12 | |
| *** asselin has joined #openstack-infra | 21:12 | |
| pleia2 | yolanda: your "Use of OpenStack CI for your own projects" talk is now up on our publications site with the correct title! http://docs.openstack.org/infra/publications/ | 21:13 |
| *** mfink has joined #openstack-infra | 21:13 | |
| anteaya | but ,user=self doesn't work, it returns patches I have reviewed already | 21:14 |
| zaro | fungi: need your help. | 21:15 |
| fungi | anteaya: i think there's a gerrit bug/feature request open for having those treated as qualifiers like the gerrit query docs imply they should, but right now they're cumulative | 21:15 |
| zaro | fungi: error when i ran gerrit upgrade: http://paste.openstack.org/show/155798/ | 21:15 |
| fungi | zaro: looking | 21:15 |
| *** Masahiro has joined #openstack-infra | 21:16 | |
| anteaya | fungi: they are cumulative, I know the meaning of that word but its use in this context loses me | 21:16 |
| zaro | fungi: i think user doesn't have permission to drop indexes? | 21:16 |
| fungi | zaro: "Can't DROP 'patch_set_approvals_closedByU'; check that column/key exists" | 21:16 |
| zaro | is that right? | 21:16 |
| *** banix has quit IRC | 21:16 | |
| zaro | column definately exists | 21:16 |
| fungi | zaro: possibly. i'm going digging in the db now | 21:16 |
| sigmavirus24 | dhellmann: do you have a second to look at https://review.openstack.org/#/c/144799/ ? | 21:16 |
| zaro | fungi: http://paste.openstack.org/show/155799/ | 21:16 |
| mattoliverau | Morning | 21:17 |
| anteaya | morning mattoliverau | 21:18 |
| zaro | morning mattoliverau | 21:18 |
| openstackgerrit | Joe Gordon proposed openstack-infra/system-config: Add screen-n-cpu-1 to logstash https://review.openstack.org/145343 | 21:18 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: Prepare project-config for puppet module split https://review.openstack.org/140523 | 21:18 |
| *** Longgeek has quit IRC | 21:18 | |
| fungi | anteaya: cumulative as in label:Code-Review=0,user=self right now gets interpreted as "changes with at least one code review vote of 0 and also changes which you have reviewed, added together" | 21:19 |
| anteaya | oh | 21:19 |
| fungi | anteaya: i think sdague ran into this and may have opened the bug with gerrit | 21:19 |
| anteaya | yeah, not what I am after | 21:19 |
| asselin | jeblair, mordred, krtaylor, sweston, nibalizer, fungi, AJaeger, pleia2, jesusaurus Puppet module split patch rebased ^^ | 21:20 |
| *** Masahiro has quit IRC | 21:20 | |
| sdague | fungi: I never filed a bug for that, I actually sort of use that behavior | 21:20 |
| anteaya | label:Code-Review=0,user=self seems to return the expected result in the gerrit gui | 21:21 |
| sdague | anteaya: what are you trying to get out of gerrit? | 21:21 |
| anteaya | gertty | 21:21 |
| pleia2 | thanks asselin, having a look | 21:21 |
| anteaya | I am trying get gertty to display my custom query as a dashboard | 21:21 |
| anteaya | NOT owner:self status:open label:Code-Review=0,self label:Workflow=0 label:Verified=+1,jenkins project:openstack-infra/project-config | 21:21 |
| fungi | zaro: and the schema migrations were being applied with the gerrit2 database account? | 21:21 |
| anteaya | it crashes on the ,self | 21:21 |
| nibalizer | anteaya: i have also seen it crash on complex queries in dashboards | 21:22 |
| nibalizer | mine was a bit different | 21:22 |
| anteaya | and ,user=self doesn't return the expected values, as in patches I have already reviewed | 21:22 |
| *** thedodd has quit IRC | 21:22 | |
| anteaya | though ,user=self works in the gerrit gui | 21:22 |
| zaro | fungi: yes, i ran as gerrit2 | 21:22 |
| anteaya | nibalizer: did you file any bugs? I can add to yours | 21:22 |
| *** banix has joined #openstack-infra | 21:23 | |
| fungi | zaro: looks like no drop_priv for gerrit2 http://paste.openstack.org/show/155805/ | 21:23 |
| nibalizer | anteaya: no i assumed pebkac | 21:23 |
| *** thedodd has joined #openstack-infra | 21:23 | |
| jeblair | anteaya: the short version is that if it works in gerrit, the same syntax should work in gertty. that code is poorly exercised and tested, so it is not surprising that there would be bugs | 21:24 |
| zaro | fungi: ahh. that will need to change for upgrade. thanks. | 21:24 |
| anteaya | nibalizer: let's file a bug, shall we? | 21:24 |
| anteaya | jeblair: great | 21:24 |
| fungi | zaro: oh, though that's the global privilege. i should probably be looking at the database-specific grants | 21:24 |
| fungi | zaro: pardon me while i fumble around in mysql for a few more minutes | 21:24 |
| jeblair | anteaya: so the best bug reports would be: if the behavior doesn't match, describe the difference in behavior in both gerrit and gertty | 21:24 |
| anteaya | jeblair: I will aim to do so | 21:25 |
| *** erikmwilson has quit IRC | 21:25 | |
| jeblair | anteaya: and if gertty crashes, that's enough there -- any time it crashes that's a bug | 21:25 |
| anteaya | nibalizer: do you want to kick it off, since you saw it first? | 21:25 |
| anteaya | nibalizer: and I will add | 21:25 |
| anteaya | jeblair: okay thanks | 21:25 |
| zaro | fungi: np, take your time. | 21:27 |
| *** russellb has joined #openstack-infra | 21:27 | |
| nibalizer | anteaya: im a bit busy right now | 21:27 |
| anteaya | nibalizer: I will do it then, didn't mean to overload | 21:27 |
| *** bhuvan has joined #openstack-infra | 21:27 | |
| nibalizer | np :) | 21:29 |
| nibalizer | thanks for getting it reported | 21:29 |
| sweston | asselin: cool, will check it out in a bit | 21:31 |
| fungi | zaro: i think http://paste.openstack.org/show/155806/ means it shouldn't be a permissions issue | 21:32 |
| sigmavirus24 | jogo: flake8 already supports --output-file | 21:33 |
| fungi | zaro: specifically, gerrit2 has alter_priv on reviewdb, which means it should be able to use alter table drop column on any of the tables in that db | 21:34 |
| sigmavirus24 | 2.3.0 now has that in | 21:34 |
| *** dangers is now known as dangers_away | 21:34 | |
| asselin | sweston, pleia2 jenkions failed on that patch, and I can't seem to access the error log. let me know if you can see it. | 21:34 |
| fungi | zaro: but if you just `select * from mysql.db where db='reviewdb';` you should see it is granted all permissions on that database anyway, not just those | 21:35 |
| sdague | jeblair: I don't know if gertty would be impacted by it, however I ended up having to uri encode commas otherwise there was some terrible brokeness in building gerrit dashboards | 21:36 |
| zaro | fungi: i didn't see this problem when i ran against my test mysql db. the difference was that i ran with a mysql admin account | 21:36 |
| zaro | let me check my test db for the tables. | 21:36 |
| *** _nadya_ has quit IRC | 21:36 | |
| anteaya | nibalizer: when you feel like adding your observations: https://storyboard.openstack.org/#!/story/2000100 | 21:37 |
| *** SumitNaiksatam has joined #openstack-infra | 21:37 | |
| jeblair | sdague: ah, thanks. yeah, i don't think it should be since gertty's searches are entirely local (needed to support full offline operation), so i wrote a parser for the query language. it's just possible i may have not gotten that perfect on the first try. :) | 21:38 |
| fungi | zaro: https://groups.google.com/forum/#!topic/repo-discuss/0qP1xZoOoDY | 21:38 |
| anteaya | annegentle_: oh yes please suggest someone for helping with docs for the nova-net to neutron migration | 21:38 |
| pleia2 | asselin: File not found is indeed unfortunate! not sure what happened there | 21:38 |
| asselin | pleia2, ok thanks, I try recheck | 21:39 |
| fungi | zaro: looks like maybe this is an artifact of having started from a much older gerrit release and incrementally upgraded to the current state... | 21:39 |
| clarkb | look for the job on jenkins nodes | 21:39 |
| *** barra204_ has quit IRC | 21:39 | |
| jogo | sigmavirus24: yeah I saw in your comments, but I agree that it makes more sense in pep8 | 21:39 |
| clarkb | and access the log there. likely swift upload is broken somehow | 21:39 |
| sigmavirus24 | jogo: yeah. it's a trivial change in flake8 so I'll maintain it for as long as it takes to convince Florent | 21:39 |
| sigmavirus24 | jogo: also ##python-code-quality | 21:39 |
| *** barra204_ has joined #openstack-infra | 21:39 | |
| *** mestery has quit IRC | 21:40 | |
| zaro | fungi: i actually was using the work around mentioned by Bassem (6th reply). | 21:41 |
| *** jedimike has quit IRC | 21:41 | |
| zaro | fungi: i ran into this same issue when i tried upgrade with my own mysql db, it worked after adding the indexes. | 21:42 |
| fungi | zaro: oh. not really sure what's going on in that case | 21:42 |
| zaro | fungi: but now it's failing on review-dev, that's why i thought it was permission issue. | 21:43 |
| anteaya | oh I guess I should let folks know I am away most of tomorrow running errands and then I fly on Thursday | 21:45 |
| clarkb | pleia2: asselin which log is a 404? | 21:45 |
| fungi | zaro: i can't say for sure it's not... i have at best a tenuous grasp of the mysql access control implementation, so someone else may be able to point out the actual issue i'm missing | 21:45 |
| asselin | clarkb, https://review.openstack.org/#/c/140523/ | 21:45 |
| asselin | http://logs.openstack.org/23/140523/11/check/gate-project-config-projects-yaml/31a9f02/ | 21:45 |
| pleia2 | I thought it ran on jenkins05, but couldn't find it there | 21:46 |
| zaro | fungi: is there a way to run a manual drop of the index from gerrit2 account? | 21:46 |
| fungi | zaro: alter table <tablename> drop column <columnname>; | 21:47 |
| fungi | i think | 21:47 |
| pleia2 | nope, 07 | 21:47 |
| clarkb | pleia2: did you find it ? I was looking would be good to track this down if it is a swift upload problem | 21:47 |
| fungi | pleia2: check the console log and hopefully you'll see an error near the end | 21:48 |
| pleia2 | clarkb: yeah, it's on 07 | 21:48 |
| clarkb | pleia2: can haz link? | 21:48 |
| fungi | pleia2: which might tell us why it failed to upload | 21:48 |
| dhellmann | sigmavirus24: +2a | 21:48 |
| sigmavirus24 | dhellmann: thanks! | 21:48 |
| pleia2 | https://jenkins05.openstack.org/job/project-config-compare-xml/195/console | 21:49 |
| dhellmann | sigmavirus24: nice improvement | 21:49 |
| *** erikwilson is now known as erikmwilson | 21:49 | |
| clarkb | pleia2: thats 05 | 21:49 |
| pleia2 | arg, sec,w rong link | 21:49 |
| pleia2 | https://jenkins05.openstack.org/job/project-config-compare-xml/195/console | 21:49 |
| pleia2 | darn it | 21:49 |
| pleia2 | https://jenkins07.openstack.org/job/gate-project-config-projects-yaml/lastUnsuccessfulBuild/ | 21:50 |
| clarkb | silly paste buffers | 21:50 |
| *** Katherine has joined #openstack-infra | 21:50 | |
| zaro | fungi: yes, but how do you run it as the gerrit2 user? | 21:50 |
| pleia2 | but that seems the wrong change too | 21:50 |
| clarkb | pleia2: ya the link to the logs is different | 21:50 |
| pleia2 | https://jenkins07.openstack.org/job/gate-project-config-projects-yaml/ | 21:50 |
| pleia2 | so that shows the last successful build because of the recheck | 21:50 |
| zaro | fungi: i can only run mysql using sudo | 21:51 |
| pleia2 | my issue was that I was clicking on last unsuccessful, which was a different change | 21:51 |
| pleia2 | apparently Jenkins' interface confounds me | 21:51 |
| openstackgerrit | Joe Gordon proposed openstack-infra/devstack-gate: Add large-ops mode to feature matrix https://review.openstack.org/145352 | 21:51 |
| mordred | pleia2: it's a confounding interface | 21:51 |
| fungi | zaro: `mysql -u gerrit2 -p reviewdb` and then when prompted for a password give it the one from ~gerrit2/review_site/etc/secure.config | 21:51 |
| asselin | pleia2, clarkb https://jenkins04.openstack.org/job/gate-project-config-projects-yaml/ | 21:52 |
| clarkb | https://jenkins04.openstack.org/job/gate-project-config-projects-yaml/329/console | 21:52 |
| clarkb | I think thats it | 21:52 |
| pleia2 | yep, that's the one | 21:53 |
| openstackgerrit | James E. Blair proposed stackforge/gertty: Document Debian/Fedora installation https://review.openstack.org/145353 | 21:53 |
| asselin | yes | 21:53 |
| clarkb | and it failed because fetching the console log resulted in a 502 proxy error | 21:54 |
| fungi | clarkb: pleia2: so i think that means that the swift upload script ended up getting a proxy error from the jenkins master? | 21:54 |
| clarkb | jhesketh: ^ | 21:54 |
| fungi | yeah, that | 21:54 |
| clarkb | fungi: yup | 21:54 |
| pleia2 | yep | 21:54 |
| fungi | not entirely uncommon for these servers either | 21:54 |
| fungi | may make sense for it to retry a limited number of times before giving up | 21:55 |
| *** ryanpetrello_ has joined #openstack-infra | 21:55 | |
| clarkb | and upload a blurb about the error if that happens | 21:55 |
| zaro | fungi: i get this error.. ERROR 1045 (28000): Access denied for user 'gerrit2'@'localhost' (using password: YES) | 21:55 |
| clarkb | "Could not retrieve console.html" | 21:55 |
| clarkb | or something | 21:55 |
| asselin | clarkb, yes, but that may also get a proxy error | 21:56 |
| *** ryanpetrello has quit IRC | 21:56 | |
| *** ryanpetrello_ is now known as ryanpetrello | 21:56 | |
| clarkb | asselin: no because the proxy error is talking to jenkins | 21:57 |
| clarkb | asselin: so we should be able to reliable upload an error message instead to swift | 21:57 |
| *** mestery has joined #openstack-infra | 21:57 | |
| Katherine | happy 2015, all. i'm having some issues with my zuul gerrit trigger script so i am unable to do merges for dependent pipeline testing | 21:57 |
| Katherine | in zuul/trigger/gerrit.py, in _getInfoRefs, the url is being set in order to pull the refs from the gerrit server | 21:58 |
| asselin | ok I see | 21:58 |
| clarkb | sdague: I am not sure I grok the -1 on https://review.openstack.org/#/c/142327/3 are you saying you want that venv to be executed out of $PATH via a script in /usr/local/bin? | 21:58 |
| fungi | zaro: me too... trying to figure out why authentication isn't working for that account | 21:58 |
| Katherine | in hp, this url is not the same as what the gerrit trigger script is expecting so the data i'm getting back from the urlopen.read is not the correct page | 21:58 |
| Katherine | i'm hoping someone may be able to tell me how to figure out what my url should be | 21:59 |
| jhesketh | clarkb: interesting... not sure there is much we can do about that | 21:59 |
| jhesketh | maybe multiple retries | 21:59 |
| Katherine | i should say, for my project here at hp, the url is not the same as what openstack is expecting | 21:59 |
| clarkb | jhesketh: ya I think retries (a sanish number) and upload a replacement console.html if that fails | 21:59 |
| clarkb | jhesketh: to avoid 404ing without any useful info | 21:59 |
| jhesketh | yep | 22:00 |
| clarkb | Katherine: it should be a standard path for gerrit servers | 22:00 |
| sdague | clarkb: that's what I was thinking. So that there would be a script in the path that would do the right thign | 22:00 |
| clarkb | something like https://host/p/repo/?inforefs or whatever | 22:00 |
| sdague | however, if no one else thinks so, feel free to override | 22:00 |
| clarkb | sdague: I think since its hidden away behind the jenkins macros its not a big deal | 22:01 |
| clarkb | er jjb macros | 22:01 |
| sdague | ok | 22:01 |
| sdague | I'll zero my review | 22:01 |
| Katherine | clarkb: does openstack specify a gerrit baseurl in their zuul.conf? | 22:02 |
| clarkb | sdague: thanks | 22:02 |
| clarkb | Katherine: https://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/zuul/templates/zuul.conf.erb is what we use | 22:02 |
| *** e0ne has quit IRC | 22:02 | |
| zaro | nibalizer: how do you stop puppet agent? | 22:03 |
| clarkb | Katherine: looking at https://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/zuul_prod.pp#n40 we don't seem to override the baseurl value | 22:03 |
| clarkb | zaro: `puppet agent --disable` | 22:03 |
| zaro | thanks | 22:04 |
| Katherine | clarkb: yes, that's what i saw | 22:04 |
| *** asettle has joined #openstack-infra | 22:05 | |
| clarkb | jhesketh: https://review.openstack.org/#/c/142327/3/nodepool/scripts/install_devstack_dependencies.sh I think that is the wrong file to install the venv into | 22:05 |
| *** dustins has quit IRC | 22:05 | |
| *** hdd has quit IRC | 22:05 | |
| clarkb | jhesketh: it will only do that on devstack nodes iirc. You probably want to do it closer to where the zuul venv is installed so that it happens on all nodes | 22:05 |
| jhesketh | clarkb: where would you suggest? | 22:05 |
| *** bhuvan_ has joined #openstack-infra | 22:05 | |
| clarkb | jhesketh: let me find a specific spot | 22:05 |
| *** kmartin has joined #openstack-infra | 22:06 | |
| *** bhuvan_ has quit IRC | 22:06 | |
| jhesketh | clarkb: that's why that change also updates prepare_node.sh | 22:06 |
| openstackgerrit | Richard Jones proposed openstack/requirements: Add new angular deps to global-requirements.txt for Horizon https://review.openstack.org/139298 | 22:06 |
| Katherine | clarkb: when i don't specify a baseurl in my conf file, the url used is https://cvp-test-gerrit.vse.adapps.hp.com/fusion/refs?service=git-upload-pack, which returns a 404 | 22:06 |
| clarkb | jhesketh: is that not sufficient then? | 22:06 |
| *** MarkAtwood has joined #openstack-infra | 22:06 | |
| jhesketh | I don't think that is ran on devstack nodes? | 22:06 |
| clarkb | let me double check. for some reason I thought prepare_node.sh was part of the dib stuff (which we should update too since we are trying to get moving in that space again) | 22:07 |
| *** asettle has quit IRC | 22:07 | |
| *** Alexandra_ is now known as asettle | 22:07 | |
| clarkb | jhesketh: prepare_node_devstack.sh calls prepare_node.sh so prepare_node.sh should be enough | 22:08 |
| *** bhuvan has quit IRC | 22:08 | |
| clarkb | jhesketh: we should however update the dib scripts too will show you where that should go | 22:08 |
| *** teran has joined #openstack-infra | 22:08 | |
| *** KurtMartin has quit IRC | 22:08 | |
| *** MaxV has quit IRC | 22:08 | |
| *** bswartz has quit IRC | 22:09 | |
| anteaya | sdague: can we get a comment with your changed vote? | 22:09 |
| jhesketh | clarkb: oh... hmm.. then I wonder why installing the dependencies via puppet didn't work for devstack nodes | 22:09 |
| clarkb | jhesketh: https://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/nodepool-base/install.d you can add another 99- scrupt for this venv. Also make sure you check it in as +x | 22:09 |
| clarkb | jhesketh: they do run different puppet so maybe it was a branch within puppet? | 22:09 |
| jhesketh | clarkb: https://review.openstack.org/#/c/142330/1 right, that's likely | 22:10 |
| jhesketh | clarkb: in patchset 1 there I moved it into the common slave, would that be a better place to set up a venv rather than these scripts? | 22:11 |
| *** markmcclain has quit IRC | 22:11 | |
| * zaro needs to afk for next 30mins | 22:11 | |
| clarkb | jhesketh: no I think zuul venv sort of set the precedent to do it outside of puppet | 22:11 |
| *** markmcclain has joined #openstack-infra | 22:12 | |
| clarkb | though maybe jeblair et al would like to move that venv too? | 22:12 |
| fungi | zaro: i don't suppose there's any chance the password in ~gerrit2/review_site/etc/secure.config isn't the correct database password for that account? | 22:12 |
| jhesketh | yeah I'm not sure that's a good precedent... we're setting up puppet so we can install everything via puppet, so shouldn't we put what we need there? | 22:12 |
| *** barra204_ has quit IRC | 22:12 | |
| fungi | zaro: the filesystem says it was modified a little less than an hour ago | 22:13 |
| *** teran has quit IRC | 22:13 | |
| Katherine | clarkb: actually, no, if i don't specify a baseurl, my url is https://cvp-test-gerrit.vse.adapps.hp.com/p/fusion/info/refs?service=git-upload-pack which returns 401 | 22:13 |
| *** otter768 has joined #openstack-infra | 22:13 | |
| Katherine | clarkb: hmm... perhaps this is my problem | 22:13 |
| *** barra204 has joined #openstack-infra | 22:13 | |
| *** dkliban is now known as dkliban_afk | 22:13 | |
| *** MaxV has joined #openstack-infra | 22:13 | |
| clarkb | jhesketh: I could go either way really. Installing a venv is sort of clunky in puppet though which is probably why we havne't done it that way | 22:14 |
| *** gyee has joined #openstack-infra | 22:14 | |
| *** mwagner_lap has quit IRC | 22:15 | |
| *** dmsimard is now known as dmsimard_away | 22:15 | |
| fungi | zaro: nevermind, that password is the same as we have in hiera, and that hasn't been altered for a long time | 22:15 |
| *** shashankhegde has quit IRC | 22:16 | |
| fungi | clarkb: yeah when i originally suggested that alternative i just assumed the publisher macro would create the virtualenv and install the right stuff into it, then call that script from it | 22:16 |
| fungi | minimal footprint on the worker that way, presumably for a very minimal increase in runtime | 22:17 |
| jhesketh | jeblair: any opinion on setting up a venv for zuul_swift_upload in puppet vs ready scripts? | 22:17 |
| *** otter768 has quit IRC | 22:17 | |
| *** hashar has quit IRC | 22:18 | |
| fungi | jhesketh: i'm still curious why not in the publisher macro itself | 22:18 |
| fungi | jhesketh: that doesn't need any modifications to worker images at all | 22:19 |
| fungi | jhesketh: one argument against the ready script specifically though is that it assumes the only place we'll use it is on nodepool nodes, but we have other jenkins slaves which aren't built by nodepool and will want to upload jobs somewhere | 22:20 |
| fungi | er, upload logs somewhere | 22:21 |
| nibalizer | zaro: service puppet stop should do it | 22:21 |
| jhesketh | fungi: so you're in line with what sdague was suggesting | 22:21 |
| nibalizer | if puppet is wedged | 22:21 |
| jhesketh | (I think) | 22:21 |
| nibalizer | you might need to kill it manually | 22:21 |
| jhesketh | so have a shell script that checks for a venv, creates it if it doesn't exist, and then runs the upload stuf | 22:22 |
| jhesketh | *stuff | 22:22 |
| fungi | jhesketh: not sure--i thought sdague was suggesting to put it in a script somewhere, yeah? i'm suggesting to just build the virtualenv from code directly included in the publisher macro in the jjb config | 22:22 |
| fungi | oh, though maybe the somewhere he was suggesting wasn't on the worker filesystems | 22:22 |
| fungi | but just ephemerally in the publisher macro | 22:23 |
| jhesketh | fungi: right, same difference I guess | 22:23 |
| fungi | jhesketh: well, if it has to go on the filesystem, that's when nodepool cached images come into play. if it's just embedded in the job config then it will work as soon as the jobs get updated on the jenkins masters with no need to split into staged changes | 22:24 |
| jhesketh | staged changes don't matter in the long run | 22:24 |
| fungi | nibalizer: stopping the puppet agent doesn't help if you're not running it as a daemon, and we're not | 22:24 |
| jhesketh | adding it to the macro has the (small) overhead of fetching the deps from pip every time | 22:24 |
| *** kgiusti has quit IRC | 22:25 | |
| fungi | jhesketh: they do if you find a bug and need to iterate quickly to fix it... don't want to have to scramble to rebuild images to address a bug in that script if you can avoid it | 22:25 |
| nibalizer | fungi oh right | 22:25 |
| nibalizer | uh you'll have to touch the file that puppetdlockfile is set to in the ansible scripts then | 22:25 |
| *** MaxV_ has joined #openstack-infra | 22:25 | |
| jhesketh | fungi: fair call | 22:25 |
| fungi | nibalizer: puppet agent --disable, like clarkb suggested, has that effect | 22:26 |
| *** pc_m has quit IRC | 22:27 | |
| nibalizer | okay sweet | 22:27 |
| nibalizer | i thought we got fancy and set a lockfile manualy in the ansible scripts | 22:27 |
| *** garyk1 has joined #openstack-infra | 22:28 | |
| *** garyk has quit IRC | 22:28 | |
| *** MaxV has quit IRC | 22:28 | |
| *** dims_ has joined #openstack-infra | 22:28 | |
| fungi | jhesketh: though a good counterargument is that we're currently adding /usr/zuul-env via nodepool | 22:29 |
| fungi | jhesketh: and consistency is a pretty big trump card | 22:29 |
| *** shashankhegde has joined #openstack-infra | 22:29 | |
| fungi | jhesketh: via nodepool/elements/nodepool-base/install.d/99-install-zuul and nodepool/scripts/prepare_node.sh depending on the image creation mechanism | 22:30 |
| fungi | jhesketh: so i'm not terribly keen on the ready script, but between embedding it in the publisher macro config or copying what we did for zuul-env i could go either way | 22:32 |
| *** dims__ has quit IRC | 22:32 | |
| *** sarob has quit IRC | 22:34 | |
| *** emagana has joined #openstack-infra | 22:38 | |
| *** sputnik13 has joined #openstack-infra | 22:38 | |
| ianw | clarkb / fungi: re that pbr problem yesterday : https://review.openstack.org/#/c/145119/ (Make tox use develop mode) ; i'm not fully across namespace packages, but i guess the only way to do it is to have pbr installed on the system | 22:40 |
| *** dkranz has quit IRC | 22:40 | |
| ianw | also, my f21 topic got dropped from the meeting minutes by an edit (i'd assumed it was dropped for time) | 22:41 |
| ianw | i'm thinking, giving the ongoing issues with d-i-b and the clear usage of f21 given the incoming devstack reviews we can get at least some testing on rax nodes which have an image | 22:41 |
| ianw | just to keep basic sanity | 22:42 |
| patrickeast | Hi all, so i’ve gotten myself into a bad state with nodepool, it has created an instance, installed everything, and created an image from it, but then nova is stuck deleting the instance.. if I look in the nodepool db it has the image id so that looks ok. I tried to stop nodepool but it is hanging (probably on the nova delete)… whats a safe way to fix this? Can I kill nodepool, manually remove the instance in | 22:42 |
| patrickeast | nova (leaving the image behind), and restart nodepool? Or will I need to manually clear out the entry in the snapshot_image table and let nodepool build an image again? | 22:42 |
| openstackgerrit | Merged openstack/requirements: Update only if version requirement has changed https://review.openstack.org/144799 | 22:43 |
| zaro | fungi: turn up anything yet? | 22:45 |
| openstackgerrit | Joe Gordon proposed openstack-infra/project-config: Use large-ops-testing to test reducing the number of services https://review.openstack.org/145367 | 22:46 |
| fungi | zaro: nothing. i need someone else with more confidence in their mysql skills to tell me why the password in ~gerrit2/review_site/etc/secure.config on review-dev doesn't work with `mysql -u gerrit2 -p reviewdb` there | 22:47 |
| zaro | fungi: i don't think the db is local is it? ~gerrit2/review_site/etc/secure.config points to a db @ rackspace? | 22:47 |
| *** jedimike has joined #openstack-infra | 22:47 | |
| zaro | fungi: opps meant ~gerrit2/review_site/etc/gerrit.config points | 22:47 |
| fungi | zaro: oh! did we move review-dev to trove? looks like we did | 22:48 |
| * fungi sighs | 22:48 | |
| fungi | zaro: so add -h followed by the hostname mentioned in that file | 22:49 |
| zaro | we should probably stop mysql locally to prevent confusion again. | 22:49 |
| jogo | top bug: http://status.openstack.org/elastic-recheck/index.html#1343313 | 22:49 |
| jogo | Bad md5 hash for package | 22:49 |
| jogo | fungi: ^ | 22:49 |
| fungi | zaro: good call. with -h pointing at the trove db hostname i can authenticate. i blame the flu | 22:50 |
| jeblair | fungi, jhesketh: back... | 22:51 |
| fungi | jogo: not surprising. hpcloud neutron has been pretty broken for us for a few days now, at least one of our routers there anyway | 22:51 |
| jeblair | fungi, jhesketh: i think puppet or image build is the place to put it -- | 22:51 |
| *** dimtruck is now known as zz_dimtruck | 22:52 | |
| jeblair | venv creation is _highly_ likely to fail, so i don't think we should do it within jobs at all | 22:52 |
| jhesketh | jeblair: why's that? | 22:52 |
| jhesketh | jeblair: doesn't that risk us creating bad images? | 22:52 |
| jogo | fungi: yeah looks like the failures are all in hp | 22:52 |
| jeblair | (i mean, if we were doing that, we would have no record of the md5sum errors we've been looking at right now because we would not be able to download the packages to create the venv to upload the log files) | 22:52 |
| jeblair | jogo: hpcloud-b3 specifically -- it's that particular neutron router that was dropping packets | 22:53 |
| jeblair | jhesketh: image creation should fail if any of the steps therein fail -- so if we fail to create the venv, the image itself should fail to be built, so we just stick with the one we have | 22:53 |
| jogo | jeblair: just noticed that via logstash (build_node trends) | 22:53 |
| jogo | mestery: neutron fail ^ | 22:54 |
| fungi | jeblair: fair enough, my vote is to copy what we're doing for /usr/zuul-env creation in that case | 22:54 |
| fungi | jhesketh: ^ | 22:54 |
| jeblair | jogo: yep. it stopped at 1500 utc today; if it has started again, we can alert the hpcloud noc | 22:54 |
| jeblair | fungi: yeah, i like that, though you are correct in that it doesn't get us on the long lived nodes though | 22:54 |
| jogo | jeblair: see only one failure after 1500 utc | 22:54 |
| jeblair | fungi: we could either do something in puppet to bridge that, or figure that removing the last of the long-lived nodes is a requirement to finalizing this | 22:55 |
| fungi | jeblair: jhesketh: or do something better than what we're doing to create /usr/zuul-env and then change zuul-env to get built the same way ;) | 22:55 |
| jogo | jeblair fungi: thanks, glad to see it has been taken care of | 22:55 |
| jhesketh | jeblair: sure, but it'd be better not to attempt building bad images and failing | 22:55 |
| jhesketh | maybe we can harden creating venvs somehow (retries for example) | 22:55 |
| jhesketh | jeblair: if we assume we can make the venv creation stable, are there any other objections? | 22:56 |
| fungi | pre-caching a wheelhouse as clarkb suggested | 22:56 |
| jeblair | jhesketh: sure, i'm not sure it's that much of a problem at the scale we're talking about | 22:56 |
| *** ryanpetrello has quit IRC | 22:56 | |
| jeblair | jhesketh: wait | 22:56 |
| jhesketh | that's fair | 22:56 |
| fungi | zaro: so anyway, i'll have to go hunting in trove to confirm what database privs the gerrit2 account actually has, but it's entirely likely it's granted a limited subset of privileges | 22:56 |
| *** timcline_ has quit IRC | 22:56 | |
| jeblair | jhesketh: are you suggesting that we make creating a venv _within_ the job never fail? | 22:56 |
| *** jedimike has quit IRC | 22:56 | |
| fungi | zaro: i'll have to look tomorrow though. i'm being commanded to get some bed rest now | 22:56 |
| fungi | night all | 22:56 |
| jhesketh | jeblair: for the purpose of argument, yes | 22:57 |
| jeblair | fungi: good night | 22:57 |
| jhesketh | but I realise that's not the case | 22:57 |
| jeblair | jhesketh: okay, i don't understand why that's under discussion -- the entire point of creating images is to have things ready to run immediately and reliably within the jobs themselves | 22:57 |
| jeblair | i'm not sure why we would try to move really basic pre-reqs like that out of image building | 22:58 |
| jeblair | (it's literally a thing that every single job would need to do) | 22:58 |
| jhesketh | yep, good point | 22:58 |
| jhesketh | okay, I'll fix up where the venv is being created and reroll | 22:58 |
| zaro | fungi: you should rest. i'll look into it. i don't think it's a permissions problem. it's because I applied the workaround (from group discussion thread) to the local db instead of the trove one. | 22:58 |
| *** hdd has joined #openstack-infra | 22:58 | |
| clarkb | jhesketh: sounds good | 23:00 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:01 | |
| *** bknudson has quit IRC | 23:02 | |
| *** Rockyg has quit IRC | 23:02 | |
| jeblair | oops, i typod a gertty open command and now i'm cloning fuel-docs :/ | 23:02 |
| *** Rockyg has joined #openstack-infra | 23:03 | |
| jeblair | (we really need an esc from that) | 23:03 |
| *** Rockyg has quit IRC | 23:03 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/pbr: Add pointers to various resources https://review.openstack.org/144838 | 23:03 |
| *** Masahiro has joined #openstack-infra | 23:04 | |
| clarkb | jeblair: do you want to review https://review.openstack.org/#/c/140106/ before I aprove it? it is one of the nodepool changes to fix dib bugs. has two +2's but no +A | 23:05 |
| jeblair | clarkb: i'm actually going through the list of changes from the meeting right now, so i'll probably hit it in 5 mins | 23:06 |
| clarkb | cool I will wait then | 23:06 |
| *** HeOS has quit IRC | 23:06 | |
| clarkb | and ya thats what I have sort of been doing. (Now waiting on new patche sfrom jhesketh) | 23:06 |
| *** asettle has quit IRC | 23:06 | |
| *** garyk1 has quit IRC | 23:06 | |
| *** KurtMartin has joined #openstack-infra | 23:07 | |
| zaro | fungi: yep, upgrade ran without error. | 23:07 |
| *** weshay has quit IRC | 23:08 | |
| *** KurtMartin has quit IRC | 23:09 | |
| *** teran has joined #openstack-infra | 23:09 | |
| *** Masahiro has quit IRC | 23:09 | |
| jeblair | asselin: couple of extra modules in your list in 140523 | 23:10 |
| *** bhuvan has joined #openstack-infra | 23:10 | |
| asselin | jeblair, really? which ones? | 23:10 |
| *** alex7376 has joined #openstack-infra | 23:10 | |
| jeblair | clarkb: wow, exactly 5 mins, as it turns out :) | 23:10 |
| jeblair | asselin: commented | 23:10 |
| *** kmartin has quit IRC | 23:10 | |
| *** alex7376 is now known as asettle | 23:11 | |
| jeblair | clarkb: in 140106 is there any cleanup we can/should do? | 23:12 |
| jeblair | clarkb: like deleting whatever dib might have written out to disk to reclaim space? or will it do that itself on the next run? | 23:12 |
| *** alexpilotti has quit IRC | 23:12 | |
| clarkb | jeblair: dib should clean up after itself during failures | 23:12 |
| asselin | jeblair, ok I see, b/c it was at the bottom of the list of repositories here: http://specs.openstack.org/openstack-infra/infra-specs/specs/puppet-modules.html | 23:13 |
| clarkb | jeblair: and nodepool should cycle out the old images that "succeeded" over time | 23:13 |
| asselin | jeblair, will take them out | 23:13 |
| *** murkk has joined #openstack-infra | 23:13 | |
| murkk | I am trying to send up a commit and I get an error: fatal: ICLA contributor agreement requires current contact information. | 23:14 |
| *** teran has quit IRC | 23:14 | |
| marun | so, would anyone hazard a guess why 'check experimental' on a neutron change ran experimental-tripelo? | 23:14 |
| marun | https://review.openstack.org/#/c/72585/ | 23:14 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack-dev/pbr: Add pointers to various resources https://review.openstack.org/144838 | 23:14 |
| clarkb | marun: because experimental tripleo jobs are triggered by the same comment | 23:14 |
| clarkb | murkk: have you attempted to update your contact info? | 23:15 |
| murkk | I try to update at the url: https://review.openstack.org/#/settings/contact and I get an error: code review error server error cannot store contact information | 23:15 |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Bring up Fedora 21 nodes https://review.openstack.org/140857 | 23:15 |
| clarkb | murkk: you will also need to be a member of the openstak foundation if you haven't done that yet | 23:15 |
| marun | clarkb: that seems like something of a waste | 23:15 |
| clarkb | murkk: that error typically results from not being a member of the openstack foundation | 23:15 |
| murkk | Maybe I am not a member | 23:15 |
| marun | clarkb: why not 'check experimental-tripleo'? | 23:15 |
| clarkb | marun: simplicity. We don't want a giant grammar | 23:15 |
| *** x1b2j has joined #openstack-infra | 23:16 | |
| clarkb | murkk: one second, I can link the wiki doc that walk you through this | 23:16 |
| openstackgerrit | Merged openstack-infra/nodepool: Fix error handling when dib image builds fail https://review.openstack.org/140106 | 23:16 |
| *** bhuvan_ has joined #openstack-infra | 23:16 | |
| murkk | oh I have been looking at it | 23:16 |
| mordred | clarkb, jeblair: you're the two who are the most likely to have strong opinions on this ... | 23:16 |
| murkk | I think I submitted for membership today, does it need to be reviewed? | 23:17 |
| clarkb | murkk: http://docs.openstack.org/infra/manual/developers.html#account-setup | 23:17 |
| marun | clarkb: ok. I guess wasting 3+ vm hours if I only wanted to run the neutron-specific queue is a lesser issue then. | 23:17 |
| clarkb | murkk: no it doesn't need review | 23:17 |
| *** packet has quit IRC | 23:17 | |
| *** jgrimm is now known as zz_jgrimm | 23:18 | |
| mordred | clarkb, jeblair: for uploading images to rackspace, I need to use the modified vhd-util that I made a docker for | 23:18 |
| jeblair | marun: yeah, if it's cost savings you're after, we've already spent more time discussing it than the project spent in CPU cost. ;) (clouds!) | 23:18 |
| marun | jeblair: it's my time, too. | 23:18 |
| jeblair | marun: i'm including that, i assume you're expensive. :) | 23:19 |
| marun | jeblair: the neutron queue is apparently after the triplo queue | 23:19 |
| mordred | marun: we have a marun as a service in the works so we can cloud you too ... | 23:19 |
| *** bhuvan has quit IRC | 23:19 | |
| murkk | Just tried to join and got error: We already have a user with that name. If your email is c***************t@r*******e.com log in and update your contact info. | 23:19 |
| clarkb | marun: I am not sure I understand | 23:19 |
| *** MaxV has joined #openstack-infra | 23:19 | |
| *** baoli has quit IRC | 23:19 | |
| murkk | I will review the link you sent me | 23:19 |
| jeblair | marun: nope, they are independent and running those jobs does not subtract from the normal experimental queue resources | 23:19 |
| mordred | jeblair, clarkb: I don't have to do anything about it this instant ... but I'll need to do something about it eventually | 23:19 |
| *** baoli has joined #openstack-infra | 23:20 | |
| marun | clarkb: status.openstack.org/zuul - 72585,33 | 23:20 |
| mordred | jeblair, clarkb: do you have strong opinions on acceptable and unacceptable ways to deal with it? | 23:20 |
| clarkb | mordred: and you did test qemu img right? | 23:20 |
| marun | clarkb: the tripleo jobs already ran - despite my not wanting to run them - and the neutron experimental jobs are still queued | 23:20 |
| clarkb | marun: they are completely independent as jeblair said | 23:20 |
| *** MaxV has quit IRC | 23:20 | |
| *** emagana has quit IRC | 23:20 | |
| *** shashankhegde has quit IRC | 23:20 | |
| *** mestery has quit IRC | 23:20 | |
| *** julim has quit IRC | 23:20 | |
| *** marcusvrn has quit IRC | 23:20 | |
| *** changbl has quit IRC | 23:20 | |
| *** stevemar has quit IRC | 23:20 | |
| *** kaisers1 has quit IRC | 23:20 | |
| *** ChuckC has quit IRC | 23:20 | |
| *** asadoughi has quit IRC | 23:20 | |
| *** miqui has quit IRC | 23:20 | |
| *** dkliban_afk has quit IRC | 23:21 | |
| *** radez has quit IRC | 23:21 | |
| *** tobiash has quit IRC | 23:21 | |
| *** notmyname has quit IRC | 23:21 | |
| *** shardy has quit IRC | 23:21 | |
| *** fabo has quit IRC | 23:21 | |
| *** esmute has quit IRC | 23:21 | |
| *** hughsaunders has quit IRC | 23:21 | |
| *** isq has quit IRC | 23:21 | |
| *** mwhudson has quit IRC | 23:21 | |
| *** SotK has quit IRC | 23:21 | |
| *** pfallenop has quit IRC | 23:21 | |
| *** tsufiev has quit IRC | 23:21 | |
| *** waynr has quit IRC | 23:21 | |
| *** NobodyCam has quit IRC | 23:21 | |
| *** wolsen has quit IRC | 23:21 | |
| *** dstanek has quit IRC | 23:21 | |
| *** dosaboy has quit IRC | 23:21 | |
| *** bodepd has quit IRC | 23:21 | |
| *** TheJulia has quit IRC | 23:21 | |
| *** bradm has quit IRC | 23:21 | |
| *** gaelL has quit IRC | 23:21 | |
| *** mkoderer has quit IRC | 23:21 | |
| *** markmcclain has quit IRC | 23:21 | |
| *** rickerc has quit IRC | 23:21 | |
| *** reed has quit IRC | 23:21 | |
| *** patrickeast has quit IRC | 23:21 | |
| *** afazekas has quit IRC | 23:21 | |
| *** armax has quit IRC | 23:21 | |
| *** mtanino has quit IRC | 23:21 | |
| *** med_ has quit IRC | 23:21 | |
| *** jpeeler has quit IRC | 23:21 | |
| *** nijaba has quit IRC | 23:21 | |
| *** bnemec has quit IRC | 23:21 | |
| *** cody-somerville has quit IRC | 23:21 | |
| *** krotscheck has quit IRC | 23:21 | |
| *** gema has quit IRC | 23:21 | |
| *** gpocentek has quit IRC | 23:21 | |
| *** redrobot has quit IRC | 23:21 | |
| *** jkt has quit IRC | 23:21 | |
| *** SergeyLukjanov has quit IRC | 23:21 | |
| *** Vivek has quit IRC | 23:21 | |
| *** igorbelikov has quit IRC | 23:21 | |
| *** praveenkumar has quit IRC | 23:21 | |
| *** therve has quit IRC | 23:21 | |
| *** ekarlso- has quit IRC | 23:21 | |
| *** cburgess_ has quit IRC | 23:21 | |
| *** DinaBelova has quit IRC | 23:21 | |
| *** nithyag_ has quit IRC | 23:21 | |
| *** jcooley has quit IRC | 23:21 | |
| *** leifmadsen has quit IRC | 23:21 | |
| *** timfreund has quit IRC | 23:21 | |
| *** jogo has quit IRC | 23:21 | |
| *** grantbow has quit IRC | 23:21 | |
| *** Clint has quit IRC | 23:21 | |
| *** niedbalski has quit IRC | 23:21 | |
| clarkb | marun: one does not affect the other in the way that is inferred | 23:21 |
| *** bodepd has joined #openstack-infra | 23:21 | |
| clarkb | woo netsplit | 23:21 |
| *** pfallenop has joined #openstack-infra | 23:21 | |
| *** esmute has joined #openstack-infra | 23:21 | |
| *** TheJulia has joined #openstack-infra | 23:21 | |
| *** waynr has joined #openstack-infra | 23:21 | |
| marun | clarkb: all I know is, my jobs aren't running and triplo jobs are. | 23:21 |
| *** fabo has joined #openstack-infra | 23:21 | |
| *** fabo has joined #openstack-infra | 23:21 | |
| *** mwhudson has joined #openstack-infra | 23:21 | |
| *** NobodyCam has joined #openstack-infra | 23:21 | |
| *** shashankhegde has joined #openstack-infra | 23:21 | |
| *** kaisers has joined #openstack-infra | 23:21 | |
| *** dkliban_afk has joined #openstack-infra | 23:21 | |
| *** changbl has joined #openstack-infra | 23:21 | |
| *** hughsaunders has joined #openstack-infra | 23:21 | |
| *** julim has joined #openstack-infra | 23:21 | |
| clarkb | marun: because we had tripleo nodes available so they ran | 23:21 |
| *** emagana has joined #openstack-infra | 23:21 | |
| marun | clarkb: ah, ok. | 23:21 |
| *** wolsen has joined #openstack-infra | 23:21 | |
| *** mestery has joined #openstack-infra | 23:21 | |
| *** stevemar has joined #openstack-infra | 23:21 | |
| *** MaxV has joined #openstack-infra | 23:21 | |
| *** tsufiev has joined #openstack-infra | 23:21 | |
| *** marcusvrn has joined #openstack-infra | 23:22 | |
| mordred | clarkb: yes. it does not work | 23:22 |
| *** tobiash has joined #openstack-infra | 23:22 | |
| *** bradm has joined #openstack-infra | 23:22 | |
| *** radez has joined #openstack-infra | 23:22 | |
| *** mriedem has quit IRC | 23:22 | |
| mordred | clarkb: vhd-util with a patch applied is required | 23:22 |
| *** ChuckC has joined #openstack-infra | 23:22 | |
| *** notmyname has joined #openstack-infra | 23:22 | |
| *** asadoughi has joined #openstack-infra | 23:22 | |
| *** mkoderer has joined #openstack-infra | 23:22 | |
| *** patrickeast has joined #openstack-infra | 23:22 | |
| *** MaxV_ has quit IRC | 23:22 | |
| *** markmcclain has joined #openstack-infra | 23:22 | |
| *** rickerc has joined #openstack-infra | 23:22 | |
| *** reed has joined #openstack-infra | 23:22 | |
| *** afazekas has joined #openstack-infra | 23:22 | |
| *** armax has joined #openstack-infra | 23:22 | |
| *** mtanino has joined #openstack-infra | 23:22 | |
| *** med_ has joined #openstack-infra | 23:22 | |
| *** jpeeler has joined #openstack-infra | 23:22 | |
| *** nijaba has joined #openstack-infra | 23:22 | |
| *** praveenkumar has joined #openstack-infra | 23:22 | |
| *** bnemec has joined #openstack-infra | 23:22 | |
| *** cody-somerville has joined #openstack-infra | 23:22 | |
| *** krotscheck has joined #openstack-infra | 23:22 | |
| *** gema has joined #openstack-infra | 23:22 | |
| *** gpocentek has joined #openstack-infra | 23:22 | |
| *** redrobot has joined #openstack-infra | 23:22 | |
| *** jkt has joined #openstack-infra | 23:22 | |
| *** SergeyLukjanov has joined #openstack-infra | 23:22 | |
| *** Vivek has joined #openstack-infra | 23:22 | |
| *** igorbelikov has joined #openstack-infra | 23:22 | |
| *** cburgess_ has joined #openstack-infra | 23:22 | |
| *** therve has joined #openstack-infra | 23:22 | |
| *** ekarlso- has joined #openstack-infra | 23:22 | |
| *** DinaBelova has joined #openstack-infra | 23:22 | |
| *** Clint has joined #openstack-infra | 23:22 | |
| *** nithyag_ has joined #openstack-infra | 23:22 | |
| *** jcooley has joined #openstack-infra | 23:22 | |
| *** leifmadsen has joined #openstack-infra | 23:22 | |
| *** timfreund has joined #openstack-infra | 23:22 | |
| *** jogo has joined #openstack-infra | 23:22 | |
| *** grantbow has joined #openstack-infra | 23:22 | |
| *** niedbalski has joined #openstack-infra | 23:22 | |
| clarkb | mordred: ugh, is that patch not going to go into upstream? | 23:23 |
| *** SotK has joined #openstack-infra | 23:23 | |
| mordred | clarkb: I'm _personally_ fine with the command-in-docker-container - but I know you're not hugely crazy about that | 23:23 |
| *** gaelL has joined #openstack-infra | 23:23 | |
| mordred | clarkb: AIUI it's been in queue for $time | 23:23 |
| clarkb | mordred: I don't see why it gains us anything | 23:23 |
| mordred | where time is over a year | 23:23 |
| clarkb | mordred: that just adds a dependency of docker to nodepool for little benefit imo | 23:24 |
| zaro | can you guys help me with signer info of a class? | 23:24 |
| mordred | clarkb: well, the alternative is that we add a billion build depends to nodepool | 23:24 |
| zaro | do you know what this means? http://paste.openstack.org/show/155817/ | 23:24 |
| mordred | clarkb: and I do mean a billion - it's not a small build | 23:24 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/project-config: Prepare project-config for puppet module split https://review.openstack.org/140523 | 23:24 |
| *** boris-42 has quit IRC | 23:24 | |
| mordred | clarkb: but we can do that for sure | 23:24 |
| jeblair | i'm not sure we can do either | 23:24 |
| mordred | clarkb: we can also create a debian package of it | 23:24 |
| anteaya | zaro: I have never seen that error before | 23:24 |
| mordred | we don't really have the infrastructure for that at the moment in a systemic way, but I've been wanting us to have it for a while | 23:24 |
| zaro | i think it means i need i need to find the right jar file. but no idea. | 23:25 |
| clarkb | zaro: double check the bouncycastle install? | 23:25 |
| mordred | thus why I'm asking - I think there are pros and cons to pretty much all of the available choices I can think of | 23:25 |
| *** miqui has joined #openstack-infra | 23:25 | |
| clarkb | are there any other tools that work? | 23:25 |
| zaro | gerrit downloaded bouncycastle jar file on upgrade. i guess it's needed in 2.9 wasn't there in 2.8 | 23:25 |
| mordred | clarkb: no | 23:25 |
| jeblair | mordred: i don't think nodepool should be involved in this at all, which makes deb package seem like the least insane way | 23:25 |
| mordred | clarkb: well, that's not true - we can install xenserver | 23:25 |
| anteaya | zaro: oh this is from the 2.9 work? | 23:26 |
| *** boris-42 has joined #openstack-infra | 23:26 | |
| *** MaxV has quit IRC | 23:26 | |
| zaro | clarkb: i'm guessing that it's not correctly signed? | 23:26 |
| *** dosaboy has joined #openstack-infra | 23:26 | |
| clarkb | zaro: ya that is what the error implies | 23:26 |
| murkk | well I am confused, I am a member of the openstack foundation but I still get error when trying to save my contact information | 23:26 |
| mordred | jeblair: ok. how much do we care about making a full end to end system for building and publishing deb packages from source? | 23:26 |
| *** zz_jgrimm has quit IRC | 23:26 | |
| anteaya | murkk: an individual member? | 23:26 |
| *** mfink has quit IRC | 23:26 | |
| clarkb | murkk: the other thing to check is that the primary email address in gerrit matches the primary email address in your foundation profile | 23:26 |
| anteaya | murkk: if memory serves if you aren't an individual member that can cause issues | 23:26 |
| zaro | i have no idea where i can get a properly signed one. | 23:27 |
| clarkb | and yes that you are signed up as an individual member. There should be a button in your openstack account to become and individual member if you are not one already | 23:27 |
| zaro | guess need to ask gerrit list on this one. | 23:27 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Start using the zuul-swift-upload venv https://review.openstack.org/143340 | 23:27 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Set up venv for zuul-swift-logs https://review.openstack.org/142327 | 23:27 |
| *** ayoung has quit IRC | 23:27 | |
| murkk | let me look at my profile | 23:27 |
| jhesketh | clarkb, fungi, jeblair: ^ | 23:27 |
| clarkb | mordred: it seems weird that a common disk image format doens't have any working tooling | 23:27 |
| jeblair | mordred: i don't feel terribly strong about that. i could probably be convinced to use a ppa. but only because this entire situation is just completely insane. | 23:28 |
| *** jedimike has joined #openstack-infra | 23:28 | |
| JayF | mordred: you guys talking about VHD? | 23:28 |
| * JayF might have a gift for you | 23:28 | |
| clarkb | JayF: yes it turns out that Xens disk image format is bong hits | 23:28 |
| clarkb | JayF: because there are no working tools apparently | 23:28 |
| JayF | I'm well aware | 23:28 |
| JayF | we were trying to fix that | 23:28 |
| *** zz_jgrimm has joined #openstack-infra | 23:29 | |
| JayF | we have a repo somewhere with the vhdutil broken out of xenserver | 23:29 |
| JayF | I'm going to find it and if it's open sourced, link it | 23:29 |
| *** bdpayne has quit IRC | 23:29 | |
| zaro | clarkb: ohh is it because the bouncy castle jar is coming from google and the gerrit we deployed is our own build and signed by us? | 23:29 |
| jhesketh | sdague: re https://review.openstack.org/#/c/141286/3 I agree a template engine would be nice, but for the interim won't this do? | 23:29 |
| clarkb | zaro: maybe? that would be odd though | 23:29 |
| JayF | clarkb: mordred: https://github.com/comstud/libvhd-builder might be useful | 23:29 |
| clarkb | zaro: since that is how we have done this since forever | 23:29 |
| jhesketh | sdague: since long term we probably want to move os-log-analyze into its own repository | 23:29 |
| JayF | clarkb: mordred: If that was packaged; ironic-python-agent would consume it. This is also a problem that if you wait long enough I (or someone I work with) will fix it :) | 23:30 |
| jeblair | jhesketh: i think os-loganalyze is in its own repo? | 23:30 |
| jhesketh | sdague: also if you get a chance to see my comments here: https://review.openstack.org/#/c/141277/ | 23:31 |
| *** baoli has quit IRC | 23:31 | |
| jhesketh | jeblair, sdague: sorry; s/os-log-analyze/zuul-swift-logs | 23:32 |
| clarkb | I wonder how hard it would be to make qemu-img work | 23:32 |
| *** baoli has joined #openstack-infra | 23:32 | |
| clarkb | then get them to merge the fix since in theory qemu is an active community | 23:32 |
| openstackgerrit | patrick-crews proposed openstack-infra/elastic-recheck: Adding namespaces to elastic-recheck https://review.openstack.org/101958 | 23:33 |
| mordred | JayF: well, I believe I'm actually using that or something similar for the docker image ... | 23:33 |
| *** Mmike has quit IRC | 23:33 | |
| *** melwitt has quit IRC | 23:34 | |
| *** loquacities has quit IRC | 23:34 | |
| mordred | JayF: main problem is that "I ran a script to grab some stuff from git and compile it" is not the usual way we want to make software that we install on our production servers | 23:34 |
| clarkb | jhesketh: oh I just realized your sudo pip install's will need to be sudo -H to accomodate new pip's caching | 23:34 |
| clarkb | jhesketh: sorry I didn't catch that until just now | 23:34 |
| *** Mmike has joined #openstack-infra | 23:34 | |
| JayF | mordred: I agree; that is not the final product for that. That's research work that someone who still works with me has to change into Real Software Packaging(tm) | 23:35 |
| mordred | yah | 23:35 |
| JayF | mordred: I just didn't know if it'd be helpful or not; so I wanted to share | 23:35 |
| mordred | JayF: thanks! | 23:35 |
| *** loquacities has joined #openstack-infra | 23:35 | |
| mordred | JayF: _everything_ people have done related to this is helpful | 23:35 |
| *** jedimike has quit IRC | 23:35 | |
| JayF | mordred: We want to add VHD support to IPA, so that'll be done "eventually" | 23:35 |
| mordred | :) | 23:36 |
| cody-somerville | This new pip release is kinda painful. Maybe the powers that be can try small and incremental changes in teh futures? :) | 23:36 |
| mordred | JayF: I really shouldn't go down the rathole of "how has this not been fixed yet in general" should I? | 23:36 |
| mordred | cody-somerville: which one? | 23:36 |
| mordred | cody-somerville: we spent most of the last 4 weeks dealing with pip and setuptols releases | 23:37 |
| cody-somerville | which ever one is causing all my things to break. | 23:37 |
| *** amitgandhinz has quit IRC | 23:37 | |
| JayF | mordred: I mean, it's your stress :P | 23:37 |
| *** pradk has quit IRC | 23:38 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Start using the zuul-swift-upload venv https://review.openstack.org/143340 | 23:39 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Set up venv for zuul-swift-logs https://review.openstack.org/142327 | 23:39 |
| jhesketh | clarkb: done ^ | 23:39 |
| *** murkk has quit IRC | 23:39 | |
| jhesketh | good catch :-) | 23:39 |
| clarkb | I don't think virtualenv needs it but that shouldn't matter +2 | 23:40 |
| *** baoli has quit IRC | 23:40 | |
| jhesketh | yeah I figured safe than sorry | 23:40 |
| dstufft | cody-somerville: which things | 23:40 |
| dstufft | cody-somerville: breaking things is so fun though | 23:40 |
| dstufft | it means I can get yelled at for extended periods of time by complete strangers | 23:41 |
| zaro | jeblair, clarkb : review-dev is in a broken state. this is the problem: https://groups.google.com/d/msg/repo-discuss/j9x0BcnI08c/l7PScWF1z_AJ | 23:42 |
| *** thedodd has quit IRC | 23:42 | |
| zaro | would you have any issues if i left it in bad state for today? | 23:42 |
| harlowja_at_home | anyway i can get a +2 power on a requirements change so i can start adjusting cinder code? https://review.openstack.org/#/c/143186/ (thx much) | 23:42 |
| clarkb | zaro: thta should be fine | 23:43 |
| *** barra204 has quit IRC | 23:43 | |
| clarkb | harlowja_at_home: you can start adjusting cinder now... | 23:43 |
| zaro | good because i gotta go, bb much later | 23:43 |
| clarkb | because there is no upper bound that shouldn't be a blocker to your work | 23:43 |
| clarkb | just something to get in concurrently | 23:44 |
| *** banix has quit IRC | 23:45 | |
| clarkb | jeblair: if you review 142327 I can work on rebuilding images once that merges | 23:45 |
| dstufft | clarkb: fwiw the newest pip won't break if you forget -H | 23:46 |
| dstufft | clarkb: you just won't get caching | 23:46 |
| clarkb | dstufft: oh thats good | 23:46 |
| dstufft | and it'll whine at you | 23:46 |
| dstufft | something something this directory isn't owned by the same user | 23:46 |
| dstufft | a message like that | 23:46 |
| dstufft | 6.0.6 | 23:47 |
| dstufft | techincally 6.0.5 and 6.0.4 too! but 4 and 5 were broken on Windows ;( | 23:47 |
| dstufft | stupid windows | 23:47 |
| harlowja_at_home | clarkb, ya, i suppose i should, although without that requirements adjustment any change i make shouldn't merge | 23:47 |
| jeblair | clarkb: see my comments on 130878 and let me know if you want to respin | 23:48 |
| *** dims_ has quit IRC | 23:48 | |
| openstackgerrit | patrick-crews proposed openstack-infra/elastic-recheck: Adding namespaces to elastic-recheck https://review.openstack.org/101958 | 23:48 |
| mordred | clarkb, jeblair: ok, I'm going to tread water on docker/packages/vhd-util for now beause it's not blocking the other things I need to do - but I'll either | 23:48 |
| clarkb | jeblair: so 'for provider_name in label disimage_providers:' ? | 23:49 |
| clarkb | jeblair: I can update | 23:49 |
| jeblair | clarkb: yeah. usually a variable called "provider" is a Provider() instance | 23:51 |
| clarkb | harlowja_at_home: or make your change forward and backward compatibile | 23:51 |
| *** bdpayne has joined #openstack-infra | 23:51 | |
| harlowja_at_home | clarkb, hmmm, possible | 23:51 |
| *** nelsnelson has quit IRC | 23:51 | |
| jeblair | clarkb: actually, i'm a little confused now | 23:51 |
| grue_pm | mordred: did you get truncated then or just word funny? You finished with "..but I'll either" | 23:51 |
| mordred | grue_pm: I did hit enter early - but then I realied I didn't know how to finish the sentence | 23:51 |
| mordred | :) | 23:51 |
| grue_pm | *grin* | 23:51 |
| jeblair | clarkb: why did you decide to make those names and not LabelProviders? | 23:51 |
| *** nelsnelson has joined #openstack-infra | 23:51 | |
| *** kumartin has joined #openstack-infra | 23:52 | |
| *** kumartin is now known as kmartin | 23:52 | |
| clarkb | jeblair: because the old code there used the name | 23:52 |
| jeblair | clarkb: since l.providers is a dict of LabelProvider()s, but l.diskimage_providers is a list of names | 23:52 |
| jeblair | clarkb: does l.providers get used anywhere anymore? | 23:53 |
| *** dims__ has joined #openstack-infra | 23:53 | |
| clarkb | jeblair: ya in the allocation calculation iirc | 23:54 |
| jeblair | yep | 23:54 |
| *** markmcclain has quit IRC | 23:54 | |
| jeblair | clarkb: do you think both of those could also be dicts in the same format as providers? | 23:54 |
| openstackgerrit | patrick-crews proposed openstack-infra/elastic-recheck: Adding namespaces to elastic-recheck https://review.openstack.org/101958 | 23:55 |
| clarkb | jeblair: yes could update diskimage_providers and snapshot_providers to refer to the Provider() objcets then dig the name out of them when we need it | 23:55 |
| *** dstanek has joined #openstack-infra | 23:55 | |
| jeblair | (and really providers could become an accessor method that just merges the dicts) | 23:55 |
| clarkb | in fact I think this change may already use it both ways. Not sure why the tests didn't catch that proably got faked out | 23:56 |
| jeblair | clarkb: sorry this is kinda nitpicky/bikesheddy, but it's a mess in there and i think this sort of thing help maintainability (slightly) | 23:56 |
| clarkb | yup its fine | 23:56 |
| clarkb | I have other nitpicky changes like the path one | 23:56 |
| clarkb | simply because reading the code is much easier that way. Ok let me stab at making those proper lists of Providers | 23:57 |
| *** shardy has joined #openstack-infra | 23:57 | |
| *** mattfarina has quit IRC | 23:57 | |
| jeblair | clarkb: or dicts of providers? :) | 23:57 |
| clarkb | oh the other providers is a dict so ya | 23:57 |
| clarkb | will make merging easier | 23:57 |
| *** ryanpetrello has joined #openstack-infra | 23:57 | |
| *** dkingshott has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!