| jroll | mordred: lol | 00:00 |
|---|---|---|
| *** Ryan_Lane2 has joined #openstack-infra | 00:01 | |
| jroll | mordred: 122 downloads already? :o | 00:01 |
| mordred | jhesketh, jeblair: I'm too jetlagged to review those changes for accuracy - but consider me a +2 on whatever you want to do | 00:01 |
| ianw | clarkb: heh, i see your next review addresses some of what i just commented on. do think that "has_*" sounds more like a function name | 00:02 |
| jeblair | jhesketh: i do not have time to test it or learn enough about the rewrite to review it. i figured since you, fungi, and SergeyLukjanov are familiar with it it would be easiest for you to do so. so i hope that one of you will approve one of those changes. | 00:02 |
| mordred | jroll: right? | 00:02 |
| jhesketh | jeblair: fair call, I'm not set up to test it either so I only reviewed it by eye. I also noticed it depends on another large change, so maybe we do do the revert? | 00:03 |
| *** wenlock has quit IRC | 00:03 | |
| clarkb | ianw: is there a suggseted alternative name? sorry I am trying to debug why the test fails now that it shoul work properly :) | 00:03 |
| *** unicell has joined #openstack-infra | 00:04 | |
| mordred | wow. ok - I know I'm extra happy about a stupid tool - but I just ran it on a big-ish yaml file I had, and the resulting directories are actually totally understandable | 00:05 |
| ianw | clarkb: my suggestion in the comments is to keep a list called disk_image_providers[] or something, and have has_diskimage() as a fn that checks that | 00:05 |
| clarkb | ianw: that seems so very unpythonic :) | 00:05 |
| clarkb | ianw: the truthiness of [] means most of the old code just continues to work | 00:05 |
| mordred | jroll: http://paste.openstack.org/show/126212 is the yaml file I just tried - http://paste.openstack.org/show/126213 is the dir structure | 00:06 |
| clarkb | ianw: we have to iterate the list each time we parse the config so I think we should just create the list then | 00:06 |
| clarkb | ianw: but if the name is bad I don't mind using a different name | 00:07 |
| clarkb | I do not think we should recalculate that list more than we have to though | 00:07 |
| jroll | mordred: aha, I was going to ask about that | 00:07 |
| clarkb | jogo: vishy: I have a change up to test multi_host | 00:08 |
| mordred | and I have just verified that it works as expected on a giant cached ansible inventory - I will not pastebin that | 00:08 |
| vishy | clarkb cool | 00:08 |
| clarkb | I think it may just work but I don't fully grok how that changes networking so we will see | 00:08 |
| mtreinish | jeblair: when you get a chance I pushed updates to the subunit2sql worker stuff last night, if you wanted to make another pass | 00:08 |
| mrodden | ianw: was trying to follow your and sdague's conversation from earlier... was there something merged in bashate that broke checks on devstack? | 00:08 |
| mordred | jroll: anyhow - have fun - it could almost CERTAINLY use optional debug logging and command line processing improvements and the list ... but I dont realy care righ tnow - it's time for bed | 00:09 |
| ianw | clarkb: just "has" indicates a verb that suggests a function to me. maybe just disk_image_providers may be clearer | 00:09 |
| clarkb | vishy: https://jenkins04.openstack.org/job/check-tempest-dsvm-aiopcpu/10/ is the test if you are curious | 00:09 |
| jroll | mordred: indeed, I'm wondering if there's something we could use it with, we have some large ansible inventories as well :P | 00:10 |
| *** rmcall_ has joined #openstack-infra | 00:10 | |
| ianw | mrodden: bit of both ... due to issues file matching in the tox.ini, bashate was not running on all devstack library files. when i fixed it to do that, it started failing | 00:10 |
| mrodden | oh whops... | 00:10 |
| mrodden | fixed a bug and found more | 00:11 |
| ianw | mrodden: we (I) also merged the `` v $( checking patch, which finds way too much in devstack | 00:11 |
| mrodden | yeah i saw that one | 00:11 |
| *** tsg has joined #openstack-infra | 00:11 | |
| mrodden | i haven't been paying attention to reviews much TBH | 00:11 |
| *** VijayTripathi has quit IRC | 00:11 | |
| *** VijayTripathi has joined #openstack-infra | 00:12 | |
| mrodden | i have the issue where i run off and write another python module every weekend like mordred | 00:12 |
| ianw | i think that we need a way to turn errors down to warnings if we want the `` check | 00:12 |
| mrodden | i'm all for a release soon though | 00:12 |
| mrodden | yeah | 00:12 |
| mrodden | there is an IGNORE but that isn't quite right | 00:12 |
| ianw | that way we can incrementally fix it in devstack, and turn it back to error when we're done | 00:12 |
| mrodden | need to know if its going to fail or not | 00:12 |
| ianw | yeah, if we ignore it will *never* get fixed :) | 00:12 |
| *** rmcall has quit IRC | 00:12 | |
| *** rmcall_ is now known as rmcall | 00:12 | |
| clarkb | jeblair: in http://logs.openstack.org/78/130878/10/check/gate-nodepool-python27/987ab82/console.html any idea why NodeLauncher doesn't launch the second node on provider2 before we run the assert? | 00:13 |
| ianw | mrodden: "fixed" as in cleaned up, i don't think it's a big issue ... but it would be nice to standardize | 00:13 |
| mrodden | right | 00:13 |
| mrodden | not everyone is going to be following new conventions we add | 00:13 |
| ianw | mrodden: well i think we can release it now ... it should not break devstack. i'm not 100% sure on the release process, push a tag? | 00:13 |
| mrodden | yeah ptl has to push a signed tag | 00:14 |
| mrodden | -infra magic takes care of the rest | 00:14 |
| *** zns has quit IRC | 00:14 | |
| mrodden | i'll probably make this one 0.3.0 since we added more checks | 00:14 |
| *** achanda has quit IRC | 00:15 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 00:15 | |
| mrodden | which i wanted to double check with a run against devstack to see if it passes... | 00:15 |
| *** achanda has joined #openstack-infra | 00:15 | |
| ianw | mrodden: i'm not sure if we agreed to drop the auto-find-files stuff ... i'm certainly for it. might be worth killing that before a release | 00:15 |
| *** signed8bit_ZZZzz has quit IRC | 00:15 | |
| mrodden | yeah we have 3 different approaches | 00:16 |
| *** dimtruck is now known as zz_dimtruck | 00:16 | |
| *** zz_dimtruck is now known as dimtruck | 00:16 | |
| *** VijayTripathi has quit IRC | 00:16 | |
| ianw | mrodden: the code in there i think only targets devstack, but we're using a custom find command from tox anyway | 00:16 |
| mrodden | i dont really like the auto discover, but i would like to make it take a list of directories or a list of files | 00:16 |
| mordred | jroll: I was actually thinking about it in conjunction with having ansible write out some data onto disk on a target node ... but I'm way past the point of actually thinking of useful things to do with it :) | 00:16 |
| jroll | mordred: interesting :) | 00:17 |
| clarkb | I think I must be racing image creation | 00:17 |
| ianw | mrodden: i've also got a patch out there for a man page as part of documentation | 00:17 |
| mrodden | ianw: nice, i haven't seen it yet | 00:17 |
| mrodden | so in the devstack tox.ini there is a find that is buidling the list anyway now? | 00:18 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/nodepool: Test label with snapshot and dib images https://review.openstack.org/130878 | 00:18 |
| mrodden | ah yep https://github.com/openstack-dev/devstack/blob/master/tox.ini#L13 | 00:18 |
| ianw | mrodden: yep, but it misses some stuff https://review.openstack.org/#/c/128809/ | 00:19 |
| *** achanda has quit IRC | 00:19 | |
| *** rushiagr is now known as rushiagr_away | 00:20 | |
| mrodden | man that is a big find | 00:23 |
| *** mfink has joined #openstack-infra | 00:24 | |
| *** signed8bit has joined #openstack-infra | 00:25 | |
| *** dimtruck is now known as zz_dimtruck | 00:25 | |
| *** Ryan_Lane2 has quit IRC | 00:29 | |
| *** Ryan_Lane2 has joined #openstack-infra | 00:29 | |
| mrodden | ianw: regarding the warning instead of flat ignore, whats your opinion on getting that included in the next release here or waiting until after? | 00:29 |
| *** darvon has quit IRC | 00:30 | |
| *** Ryan_Lane2 has quit IRC | 00:30 | |
| *** darvon has joined #openstack-infra | 00:30 | |
| ianw | mrodden: https://review.openstack.org/#/c/131611/ <- drop auto discover bits | 00:30 |
| ianw | mrodden: think we should release without the warning stuff, so we can get devstack checking itself properly | 00:31 |
| *** todd_dsm has joined #openstack-infra | 00:31 | |
| *** Ryan_Lane2 has joined #openstack-infra | 00:31 | |
| *** Ryan_Lane2 is now known as Ryan_Lane | 00:32 | |
| *** mmaglana has quit IRC | 00:32 | |
| *** Ryan_Lane is now known as Guest24890 | 00:32 | |
| mrodden | ianw: ok i'm cool with that | 00:32 |
| mrodden | looking over patch now | 00:32 |
| *** bhuvan has quit IRC | 00:32 | |
| *** Guest24890 is now known as Ryan_Lane | 00:32 | |
| *** Ryan_Lane has joined #openstack-infra | 00:32 | |
| *** mmaglana_ has joined #openstack-infra | 00:34 | |
| *** camunoz_mtg is now known as camunoz | 00:34 | |
| ianw | mrodden: yeah ... lesson number 1 million from devstack : use file extensions :) | 00:35 |
| *** patrickeast has quit IRC | 00:36 | |
| mrodden | yeah thats probably the simplest and cleanest solution | 00:36 |
| *** mmaglana_ has quit IRC | 00:37 | |
| *** mmaglana has joined #openstack-infra | 00:37 | |
| *** bknudson has joined #openstack-infra | 00:38 | |
| *** mmaglana has quit IRC | 00:38 | |
| *** VijayTripathi has joined #openstack-infra | 00:40 | |
| mrodden | ianw: reviewed with a comment inline | 00:40 |
| ianw | mrodden: not sure the comment made it into the review | 00:41 |
| mrodden | yeah you put a patchset 2 up | 00:41 |
| mrodden | lol | 00:41 |
| clarkb | ok multi_host = True isn't working with the gretap setup | 00:41 |
| clarkb | so that will need further debugging | 00:41 |
| *** VijayTripathi has quit IRC | 00:42 | |
| clarkb | vishy: does multi_host = True expect VM communication to be routed over the external interface on the host? | 00:42 |
| clarkb | vishy: rather than through the flat_interface/ | 00:42 |
| *** rmcall_ has joined #openstack-infra | 00:43 | |
| *** otherwiseguy has joined #openstack-infra | 00:44 | |
| mrodden | ianw: you could just 'return 1' instead of sys.exit(1) in main() | 00:44 |
| *** mattfarina has joined #openstack-infra | 00:44 | |
| *** rmcall has quit IRC | 00:44 | |
| *** rmcall_ is now known as rmcall | 00:44 | |
| *** mattfarina has quit IRC | 00:45 | |
| *** mmaglana has joined #openstack-infra | 00:46 | |
| *** Longgeek has joined #openstack-infra | 00:47 | |
| *** mmaglana has quit IRC | 00:48 | |
| *** Longgeek has quit IRC | 00:48 | |
| *** Longgeek has joined #openstack-infra | 00:48 | |
| *** david-lyle has joined #openstack-infra | 00:49 | |
| *** todd_dsm has quit IRC | 00:49 | |
| ianw | mrodden: good point; new revision avoiding the exits | 00:50 |
| mrodden | ianw: cool +2 | 00:51 |
| *** JayJ has joined #openstack-infra | 00:53 | |
| *** brad__ has joined #openstack-infra | 00:53 | |
| *** otherwiseguy has quit IRC | 00:53 | |
| mrodden | ah i have to figure out how to rename that launchpad project too | 00:53 |
| *** todd_dsm has joined #openstack-infra | 00:54 | |
| *** sarob has quit IRC | 00:54 | |
| *** sarob has joined #openstack-infra | 00:55 | |
| jogo | clarkb: nice, I assume its working too | 00:55 |
| clarkb | jogo: it isn't | 00:55 |
| jogo | clarkb: a round 2 patch: https://review.openstack.org/#/c/131557/ | 00:56 |
| clarkb | jogo: I think that multi_host may expect that VM to VM comms happen over the external network | 00:56 |
| jogo | clarkb: as in multi_node is working worse then regular? | 00:56 |
| clarkb | so br-ext instead of br100 or whatever | 00:56 |
| clarkb | jogo: ya | 00:56 |
| jogo | clarkb: hmm, it shouldn't expect it to happen over the external network | 00:56 |
| *** bknudson has left #openstack-infra | 00:56 | |
| jogo | if multi_node won't make the need for openvpn go away | 00:56 |
| jogo | we might as will wait until we get the basics working and then revisit | 00:57 |
| jogo | clarkb: I don't know how well tested devstack multi_node mode is in the first place | 00:58 |
| *** zaphod42_ has joined #openstack-infra | 00:59 | |
| clarkb | ya that works too, I just thought that maybe it would work | 00:59 |
| *** sarob has quit IRC | 00:59 | |
| *** packet has quit IRC | 00:59 | |
| *** e0ne has quit IRC | 00:59 | |
| vishy | clarkb i don’t know | 01:01 |
| vishy | :) | 01:01 |
| *** ryanpetrello has joined #openstack-infra | 01:02 | |
| *** zaphod42 has quit IRC | 01:02 | |
| *** todd_dsm has quit IRC | 01:03 | |
| *** zaphod42_ has quit IRC | 01:03 | |
| ianw | fungi: wasn't this the rocket you were going to watch take off? http://www.gizmodo.com.au/2014/10/nasas-antares-rocket-just-exploded-on-lift-off-video/ | 01:03 |
| *** yamahata has joined #openstack-infra | 01:03 | |
| *** david-lyle has quit IRC | 01:04 | |
| brad__ | so sad | 01:06 |
| mrodden | whops... | 01:06 |
| mrodden | i've done that in KSP before | 01:06 |
| fungi | ianw: yep, didn't clear the horizon from where we were standing | 01:07 |
| ianw | didn't clear much by the looks of it | 01:08 |
| *** e0ne has joined #openstack-infra | 01:13 | |
| *** ryanpetrello_ has joined #openstack-infra | 01:13 | |
| *** ryanpetrello has quit IRC | 01:13 | |
| mrodden | ianw: i'm happy to report that the latest bashate master works against devstack master | 01:14 |
| *** ryanpetrello_ is now known as ryanpetrello | 01:14 | |
| *** msabramo has joined #openstack-infra | 01:14 | |
| nibalizer | ianw: im really excited to have centos slaves, thanks for working on it | 01:14 |
| *** ayoung has joined #openstack-infra | 01:16 | |
| ianw | nibalizer: no worries, it's quite the learning experience :) | 01:19 |
| *** Longgeek has quit IRC | 01:19 | |
| *** yamamoto_ has joined #openstack-infra | 01:20 | |
| *** SumitNaiksatam has joined #openstack-infra | 01:21 | |
| *** yamahata has quit IRC | 01:22 | |
| *** yamahata has joined #openstack-infra | 01:22 | |
| ianw | jeblair / fungi / clarkb : i wasn't quite sure where we ended up with https://review.openstack.org/#/c/129496/ (Add openstack-infra/puppet-openstackci_apache) ... is it dependent on something, or people aren't sure? | 01:23 |
| *** signed8bit is now known as signed8bit_ZZZzz | 01:23 | |
| *** signed8bit_ZZZzz has quit IRC | 01:24 | |
| *** ryanpetrello has quit IRC | 01:25 | |
| *** wenlock has joined #openstack-infra | 01:25 | |
| ianw | mrodden: using the "expanded" find that gets all the files? | 01:26 |
| *** signed8bit has joined #openstack-infra | 01:26 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/nodepool: Test label with snapshot and dib images https://review.openstack.org/130878 | 01:28 |
| clarkb | finally I think that should pass testing | 01:28 |
| *** marun has quit IRC | 01:29 | |
| *** ryanpetrello has joined #openstack-infra | 01:30 | |
| *** e0ne has quit IRC | 01:33 | |
| *** Ryan_Lane has quit IRC | 01:34 | |
| *** ryanpetrello has quit IRC | 01:36 | |
| ianw | mrodden: https://review.openstack.org/#/c/131616/ <- extracted out man-page | 01:36 |
| *** wenlock has quit IRC | 01:36 | |
| *** ryanpetrello has joined #openstack-infra | 01:39 | |
| *** ivar-laz_ has joined #openstack-infra | 01:41 | |
| *** ivar-lazzaro has quit IRC | 01:44 | |
| *** david-lyle has joined #openstack-infra | 01:44 | |
| *** bitblt has quit IRC | 01:44 | |
| *** ivar-laz_ has quit IRC | 01:46 | |
| *** spzala has quit IRC | 01:47 | |
| *** weshay has quit IRC | 01:49 | |
| *** dkliban is now known as dkliban_afk | 01:51 | |
| morganfainberg | clarkb, fungi, did we lose py33/py34 testing on keystoneclient? | 01:52 |
| morganfainberg | i looked at gerrit and didn't see a test run of it. | 01:52 |
| morganfainberg | recent patch | 01:52 |
| mrodden | ianw: i did not test the expanded one | 01:52 |
| mrodden | let me grab that | 01:52 |
| * morganfainberg wants to make sure if it's expected or a gap. | 01:52 | |
| *** ryanpetrello has quit IRC | 01:53 | |
| *** signed8b_ has joined #openstack-infra | 01:54 | |
| *** signed8bit has quit IRC | 01:56 | |
| *** xchu has joined #openstack-infra | 01:56 | |
| fungi | morganfainberg: looks like the python3-jobs template is still applied to it in zuul's layout | 01:56 |
| *** ryanpetrello has joined #openstack-infra | 01:57 | |
| *** dalgaaf__ has quit IRC | 01:57 | |
| mrodden | ianw: 4 hits in projects/devstack/lib/swift of "E041: Arithmetic expansion using $[ is deprecated for $((:" | 01:58 |
| clarkb | morganfainberg: always link your example :P | 01:58 |
| morganfainberg | sorry was in the process of finding it | 01:58 |
| clarkb | but ya those py3k tests should continue to run | 01:58 |
| ianw | mrodden: https://review.openstack.org/#/c/129764/ should have fixed that? | 01:58 |
| *** xchu has quit IRC | 01:59 | |
| morganfainberg | noppe, i'm just crazy | 01:59 |
| morganfainberg | i was *sure* i saw this. | 01:59 |
| mrodden | ianw: ah, i used git checkout; probably need to rebase that tox.ini patch | 01:59 |
| *** todd_dsm has joined #openstack-infra | 01:59 | |
| morganfainberg | clarkb, fungi, don't mind me, i might just be suffering from pre-summit travel scattered brain | 01:59 |
| *** amotoki has joined #openstack-infra | 01:59 | |
| clarkb | morganfainberg: its ok we will all be joining you shortly | 01:59 |
| morganfainberg | clarkb, when you getting in? | 02:00 |
| clarkb | early sunday morning | 02:00 |
| morganfainberg | cool | 02:00 |
| mrodden | ianw: yep works fine when i cherry-pick https://review.openstack.org/#/c/128809/2 to master | 02:00 |
| clarkb | either 7:35 or 8:35 | 02:00 |
| clarkb | delta and google don't agree (I think DST is involved) | 02:00 |
| morganfainberg | haha | 02:00 |
| morganfainberg | use delta's time | 02:00 |
| clarkb | morganfainberg: you? | 02:00 |
| morganfainberg | i'm in Saturday at like 10am | 02:00 |
| morganfainberg | or 8am | 02:01 |
| morganfainberg | or something like that | 02:01 |
| clarkb | I raelized very recentlythat arriving saturday would have been best but I am dragging larissa along and she has to work friday so meh | 02:01 |
| StevenK | clarkb: I get in 0750 Paris time on Sunday | 02:01 |
| clarkb | StevenK: do you go via north america or the other way around? | 02:02 |
| StevenK | clarkb: Nope, via Dubai | 02:02 |
| clarkb | I went that way when I visited perth. its not too bad | 02:02 |
| StevenK | Sydney to Europe via the US is the wrong direction | 02:02 |
| clarkb | and you get to see that giant building out the plane window | 02:02 |
| clarkb | yes well sea to perth via dubai is wrong way too :) | 02:03 |
| *** nosnos has joined #openstack-infra | 02:06 | |
| *** msabramo has quit IRC | 02:10 | |
| *** david-lyle has quit IRC | 02:10 | |
| *** david-lyle has joined #openstack-infra | 02:13 | |
| *** bhuvan has joined #openstack-infra | 02:14 | |
| *** dims__ has quit IRC | 02:14 | |
| *** dims__ has joined #openstack-infra | 02:15 | |
| *** dims__ has quit IRC | 02:17 | |
| *** carl_baldwin has joined #openstack-infra | 02:18 | |
| *** morganfainberg is now known as only_zuul | 02:18 | |
| *** only_zuul is now known as morganfainberg | 02:18 | |
| *** david-lyle has quit IRC | 02:19 | |
| *** teran has quit IRC | 02:19 | |
| *** carl_baldwin has quit IRC | 02:21 | |
| *** signed8b_ has quit IRC | 02:30 | |
| *** carl_baldwin has joined #openstack-infra | 02:30 | |
| *** Ryan_Lane2 has joined #openstack-infra | 02:31 | |
| *** carl_baldwin has quit IRC | 02:32 | |
| *** Ryan_Lane2 has quit IRC | 02:32 | |
| *** Ryan_Lane has joined #openstack-infra | 02:33 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack/requirements: Gate python-keystoneclient-federation on global reqs https://review.openstack.org/131624 | 02:33 |
| *** ryanpetrello has quit IRC | 02:35 | |
| *** signed8bit has joined #openstack-infra | 02:36 | |
| *** mriedem has quit IRC | 02:40 | |
| *** VijayTripathi has joined #openstack-infra | 02:44 | |
| *** dims__ has joined #openstack-infra | 02:47 | |
| *** otter768 has joined #openstack-infra | 02:49 | |
| *** mmaglana has joined #openstack-infra | 02:52 | |
| *** dims__ has quit IRC | 02:52 | |
| *** carl_baldwin has joined #openstack-infra | 02:53 | |
| *** Ryan_Lane has quit IRC | 02:55 | |
| *** signed8bit has quit IRC | 02:57 | |
| *** signed8bit has joined #openstack-infra | 02:57 | |
| *** signed8bit has quit IRC | 02:57 | |
| *** chandankumar has joined #openstack-infra | 02:57 | |
| *** SumitNaiksatam has quit IRC | 02:58 | |
| *** chandankumar has quit IRC | 02:58 | |
| *** VijayTripathi has quit IRC | 02:59 | |
| *** carl_baldwin has quit IRC | 03:01 | |
| *** gyee has quit IRC | 03:03 | |
| *** carl_baldwin has joined #openstack-infra | 03:05 | |
| *** signed8bit has joined #openstack-infra | 03:06 | |
| *** bswartz has quit IRC | 03:08 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 03:11 | |
| *** JayJ has quit IRC | 03:12 | |
| *** ayoung is now known as ayoung-ZZZzz | 03:13 | |
| *** homeless has joined #openstack-infra | 03:16 | |
| *** homeless1 has quit IRC | 03:16 | |
| *** armax has quit IRC | 03:17 | |
| *** leopoldj_ has joined #openstack-infra | 03:19 | |
| *** teran has joined #openstack-infra | 03:19 | |
| tchaypo | StevenK: it depends on your criteria; i believe it can sometimes be cheaper | 03:20 |
| tchaypo | does the infra team mirror the ubuntu cloud images? | 03:20 |
| *** otter768 has quit IRC | 03:20 | |
| clarkb | nope | 03:20 |
| *** yamahata has quit IRC | 03:22 | |
| *** leopoldj has quit IRC | 03:22 | |
| tchaypo | and that would be because | 03:22 |
| tchaypo | the dib elements for the OSes do their own caching | 03:23 |
| tchaypo | I'm guessing? | 03:23 |
| tchaypo | possibly there's an upstream squid involved too? | 03:23 |
| tchaypo | it seems like an upstream squid combined with the element caches should do 90% of the work with 10% the effort of trying to mirror the actual cloud image servers | 03:24 |
| *** zz_naotok is now known as naotok | 03:24 | |
| *** teran has quit IRC | 03:24 | |
| clarkb | neither | 03:27 |
| clarkb | we dont dib properly yet and dont squid | 03:27 |
| clarkb | we do put the images that devstack useson the slaves though | 03:28 |
| *** VijayTripathi has joined #openstack-infra | 03:31 | |
| ianw | clarkb: you got a minute to log into one of these building hp centos7 images to help me debug | 03:33 |
| ianw | clarkb: it's bailing out running "sudo grub2-mkconfig" ... it's a path issue, i can easily send a patch to make it /usr/sbin/grub2-mkconfig | 03:33 |
| clarkb | ianw not really on phone for the night | 03:33 |
| ianw | but i would like to figure out why this is failing on hpcloud and passing on rax | 03:34 |
| ianw | clarkb: ok | 03:35 |
| *** VijayTripathi has quit IRC | 03:37 | |
| *** todd_dsm has quit IRC | 03:37 | |
| *** armax has joined #openstack-infra | 03:40 | |
| *** armax has quit IRC | 03:41 | |
| *** todd_dsm has joined #openstack-infra | 03:41 | |
| *** msabramo has joined #openstack-infra | 03:43 | |
| *** ryanpetrello has joined #openstack-infra | 03:43 | |
| *** bswartz has joined #openstack-infra | 03:45 | |
| *** nosnos has quit IRC | 03:46 | |
| *** nosnos has joined #openstack-infra | 03:46 | |
| *** bswartz has quit IRC | 03:47 | |
| *** todd_dsm has quit IRC | 03:48 | |
| *** bhuvan has quit IRC | 03:49 | |
| *** nosnos has quit IRC | 03:51 | |
| *** todd_dsm has joined #openstack-infra | 03:51 | |
| *** changbl has joined #openstack-infra | 03:53 | |
| *** ujuc has joined #openstack-infra | 03:54 | |
| *** koolhead17 has joined #openstack-infra | 03:56 | |
| *** koolhead17 has joined #openstack-infra | 03:56 | |
| *** carl_baldwin has quit IRC | 03:56 | |
| *** mbacchi has quit IRC | 03:57 | |
| *** todd_dsm has quit IRC | 03:57 | |
| *** homeless has quit IRC | 03:59 | |
| *** bswartz has joined #openstack-infra | 04:01 | |
| *** bhuvan has joined #openstack-infra | 04:01 | |
| jogo | clarkb: how is the aiopncpu | 04:03 |
| clarkb | its fine we have a plan to use "private" addresses | 04:04 |
| jogo | using openvpn? | 04:05 |
| *** chandankumar has joined #openstack-infra | 04:05 | |
| *** yamahata has joined #openstack-infra | 04:05 | |
| clarkb | no it is already setup to gre | 04:06 |
| clarkb | but gre wasnt working on nodes that didnt know about their public IPs | 04:07 |
| jogo | so you got it working for both providers? | 04:08 |
| clarkb | ya it always worked | 04:08 |
| clarkb | we just had to sort out if other circumstances worked too | 04:09 |
| *** garyh has quit IRC | 04:09 | |
| jogo | very exciting | 04:10 |
| *** signed8bit has quit IRC | 04:11 | |
| *** armax has joined #openstack-infra | 04:20 | |
| *** harlowja is now known as harlowja_away | 04:21 | |
| *** nosnos has joined #openstack-infra | 04:27 | |
| *** chandankumar has quit IRC | 04:32 | |
| *** ryanpetrello has quit IRC | 04:34 | |
| *** teran has joined #openstack-infra | 04:36 | |
| *** pcrews has quit IRC | 04:37 | |
| *** teran has quit IRC | 04:40 | |
| *** yfried has joined #openstack-infra | 04:40 | |
| *** todd_dsm has joined #openstack-infra | 04:41 | |
| tchaypo | clarkb: neither meaning neither the image-create caching nor squid? | 04:41 |
| *** pcrews has joined #openstack-infra | 04:41 | |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/project-config: Allow jenkins to log into the host as root https://review.openstack.org/131634 | 04:44 |
| *** achanda has joined #openstack-infra | 04:49 | |
| *** rmcall has quit IRC | 04:51 | |
| *** rmcall has joined #openstack-infra | 04:51 | |
| *** morganfainberg is now known as mightbestevemar | 04:55 | |
| *** koolhead17 has quit IRC | 04:57 | |
| *** pcrews has quit IRC | 04:57 | |
| *** links has joined #openstack-infra | 05:03 | |
| *** mightbestevemar is now known as morganfainberg | 05:04 | |
| *** rmcall has quit IRC | 05:06 | |
| *** todd_dsm has quit IRC | 05:08 | |
| *** todd_dsm has joined #openstack-infra | 05:08 | |
| *** rmcall has joined #openstack-infra | 05:08 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 05:14 | |
| openstackgerrit | Ian Wienand proposed a change to openstack-infra/project-config: Use full path to grub2-mkconfig https://review.openstack.org/131636 | 05:15 |
| ianw | clarkb / jeblair : ^ fix to hopefully get hpcloud centos7 images building. if i can get access to a host, i might be able to come up with a more satisfactory answer on how PATH is being reset... | 05:16 |
| afazekas | jeblair: Is a Db schema change ok without a migrate script? | 05:19 |
| *** sarob has joined #openstack-infra | 05:19 | |
| clarkb | tchaypo: right. currently none of our hosts are built with dib so no dib caching | 05:20 |
| clarkb | tchaypo: and we run no squid | 05:20 |
| clarkb | afazekas: yes, we basically manually apply those changes and indicate the need for change in the commit somewher | 05:20 |
| clarkb | afazekas: we really should just add alembic to nodepool though | 05:20 |
| clarkb | afazekas: also, do you know why https://review.openstack.org/#/c/131604/ fails? | 05:21 |
| clarkb | I think it should work because the subnode has the pub_if interface for the flat_interface which should bridge it together with the other node | 05:22 |
| *** todd_dsm has quit IRC | 05:22 | |
| afazekas | clarkb: multihost might require other options when the network created (or a bug) | 05:24 |
| *** yfried has quit IRC | 05:25 | |
| *** sarob has quit IRC | 05:25 | |
| afazekas | also n-net needs to start on every node | 05:25 |
| *** pcrews has joined #openstack-infra | 05:26 | |
| *** sarob has joined #openstack-infra | 05:26 | |
| clarkb | ya my change adds n-net to the sub node | 05:26 |
| clarkb | I actually think I may need to override public_interface as it should be pub_if and not br100 | 05:27 |
| clarkb | let me try that | 05:27 |
| clarkb | in the case of multi_node = false br100 is correct ecause we are routing everything through to that br100 on the controller iirc | 05:27 |
| clarkb | but in mutli_node = true we don't do that and need to hop into the bridge on each node | 05:27 |
| clarkb | * hop onto the gretap bridge | 05:28 |
| *** rushiagr_away is now known as rushiagr | 05:29 | |
| clarkb | oh nevermind its set properly | 05:29 |
| afazekas | http://docs.openstack.org/havana/install-guide/install/apt/content/nova-network.html AFAIK we do not have the '--multi-host=T' option | 05:29 |
| *** sarob has quit IRC | 05:30 | |
| clarkb | afazekas: that would be in devstack? | 05:30 |
| *** pcrews has quit IRC | 05:32 | |
| afazekas | clarkb: devstack can use a NETWORK_CREATE_ARGS as an extra arg (it is still using nova-manage) | 05:32 |
| *** reed has quit IRC | 05:33 | |
| clarkb | ya my greps for network-create showed nothing but looks like near the end of stack.sh is nova-manage stuff for the fixed range | 05:34 |
| clarkb | I will add that to my change really quickly to see what happens | 05:35 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 05:36 |
| *** rmcall has quit IRC | 05:39 | |
| *** rmcall has joined #openstack-infra | 05:39 | |
| *** mrmartin has joined #openstack-infra | 05:40 | |
| *** chandankumar has joined #openstack-infra | 05:44 | |
| *** loki184 has joined #openstack-infra | 05:44 | |
| afazekas | jeblair, clarkb: after the np change a ready script also need to be changed, by adding the fixed addresses to the iptables INPUT -j ACCEPT | 05:47 |
| clarkb | afazekas: your d-g change does that today right? | 05:47 |
| clarkb | (I am not sure where it should go yet, just double checking that you are currently doing it somewhere) | 05:48 |
| afazekas | clarkb: yes, I have and abandoned change for the ready script as well | 05:49 |
| *** garyh has joined #openstack-infra | 05:49 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 05:51 | |
| clarkb | afazekas: part of me thinks that individual jobs should worry about their own network requirements, but allowing all communication on the private addresses is probably something we can just do for the multinode setups | 05:53 |
| *** mrmartin has quit IRC | 05:57 | |
| *** Ryan_Lane has joined #openstack-infra | 05:57 | |
| *** unicell has quit IRC | 06:02 | |
| *** ildikov has quit IRC | 06:02 | |
| clarkb | afazekas: wouldn't multi_host = True in nova.conf get picked up by the command at http://logs.openstack.org/04/131604/2/experimental/check-tempest-dsvm-aiopcpu/90708c6/logs/devstacklog.txt.gz#_2014-10-29_05_54_29_395 | 06:04 |
| clarkb | I am not sure this is the issue | 06:04 |
| openstackgerrit | afazekas proposed a change to openstack-infra/project-config: multinode_setup allow fixed ip https://review.openstack.org/131646 | 06:05 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 06:06 |
| * afazekas moved https://review.openstack.org/#/c/114841/2 -> https://review.openstack.org/131646 | 06:06 | |
| *** teran has joined #openstack-infra | 06:06 | |
| *** afazekas is now known as __afazekas | 06:11 | |
| *** rmcall has quit IRC | 06:11 | |
| *** rmcall_ has joined #openstack-infra | 06:11 | |
| *** achanda has quit IRC | 06:12 | |
| *** teran has quit IRC | 06:13 | |
| *** achanda has joined #openstack-infra | 06:14 | |
| *** rmcall_ has quit IRC | 06:15 | |
| *** rmcall has joined #openstack-infra | 06:16 | |
| *** Ryan_Lane has quit IRC | 06:18 | |
| *** Ryan_Lane has joined #openstack-infra | 06:18 | |
| *** Ryan_Lane has joined #openstack-infra | 06:18 | |
| *** brad__ has quit IRC | 06:20 | |
| *** rmcall has quit IRC | 06:20 | |
| *** achanda has quit IRC | 06:20 | |
| *** rmcall has joined #openstack-infra | 06:23 | |
| *** achanda has joined #openstack-infra | 06:25 | |
| *** rmcall has quit IRC | 06:28 | |
| *** unicell has joined #openstack-infra | 06:30 | |
| *** rmcall has joined #openstack-infra | 06:31 | |
| *** mrmartin has joined #openstack-infra | 06:31 | |
| tchaypo | clarkb: ah. i think your idea of "our hosts" is a bit different from what i had in mind | 06:32 |
| tchaypo | doesn't toci_devtest invoke dib at some point? | 06:33 |
| *** armax has quit IRC | 06:33 | |
| clarkb | something invokes dib from within a test iirc | 06:34 |
| clarkb | because i wrote a dib change once that failed on a ramdisk image build | 06:34 |
| clarkb | but we don't cache any of the images used for that | 06:34 |
| clarkb | aiui | 06:34 |
| *** yfried has joined #openstack-infra | 06:37 | |
| *** Ryan_Lane has quit IRC | 06:38 | |
| *** achanda has quit IRC | 06:39 | |
| *** mmaglana has quit IRC | 06:42 | |
| *** amcrn has quit IRC | 06:45 | |
| *** ashaeron has joined #openstack-infra | 06:50 | |
| *** rlandy has joined #openstack-infra | 06:50 | |
| *** teran has joined #openstack-infra | 06:54 | |
| *** ildikov has joined #openstack-infra | 06:55 | |
| *** e0ne has joined #openstack-infra | 06:56 | |
| *** bhuvan has quit IRC | 06:57 | |
| *** bhuvan has joined #openstack-infra | 07:01 | |
| *** camunoz has quit IRC | 07:03 | |
| *** afazekas has joined #openstack-infra | 07:03 | |
| *** jerryz has joined #openstack-infra | 07:04 | |
| *** ildikov has quit IRC | 07:15 | |
| *** yfried is now known as yfried|afk | 07:19 | |
| *** yfried|afk is now known as yfried | 07:19 | |
| *** rmcall has quit IRC | 07:22 | |
| *** nosnos has quit IRC | 07:25 | |
| *** nosnos has joined #openstack-infra | 07:27 | |
| *** leopoldj_ has quit IRC | 07:29 | |
| *** HeOS has quit IRC | 07:33 | |
| *** changbl has quit IRC | 07:36 | |
| *** rmcall has joined #openstack-infra | 07:40 | |
| *** yamamoto_ has quit IRC | 07:42 | |
| *** bhuvan has quit IRC | 07:42 | |
| *** rmcall has quit IRC | 07:45 | |
| *** katyafervent is now known as katyafervent_awa | 07:47 | |
| *** arosen has joined #openstack-infra | 07:47 | |
| *** yaguang has joined #openstack-infra | 07:49 | |
| arosen | Hi, i'm having a problem with pbr it seems. I'm getting these two different traces when i try to install the congressclient via pypi any ideas? http://paste.openstack.org/show/126347/ | 07:49 |
| *** unicell has quit IRC | 07:51 | |
| *** koolhead17 has joined #openstack-infra | 07:52 | |
| *** koolhead17 has quit IRC | 07:53 | |
| *** katyafervent_awa is now known as katyafervent | 07:55 | |
| *** arxcruz has joined #openstack-infra | 08:06 | |
| *** jcoufal has joined #openstack-infra | 08:07 | |
| *** gokrokve has joined #openstack-infra | 08:09 | |
| *** andreykurilin_ has joined #openstack-infra | 08:12 | |
| *** jlibosva has joined #openstack-infra | 08:12 | |
| *** notstevemar has quit IRC | 08:14 | |
| *** katyafervent is now known as katyafervent_awa | 08:14 | |
| *** slashme has joined #openstack-infra | 08:17 | |
| *** e0ne has quit IRC | 08:19 | |
| *** koolhead17 has joined #openstack-infra | 08:19 | |
| *** yamahata has quit IRC | 08:22 | |
| *** ildikov has joined #openstack-infra | 08:23 | |
| *** katyafervent_awa is now known as katyafervent | 08:23 | |
| *** bcourt has joined #openstack-infra | 08:26 | |
| *** rushiagr is now known as rushiagr_away | 08:27 | |
| *** jgallard has joined #openstack-infra | 08:28 | |
| *** k4n0 has joined #openstack-infra | 08:28 | |
| *** bcourt has quit IRC | 08:33 | |
| *** bcourt has joined #openstack-infra | 08:33 | |
| openstackgerrit | Marton Kiss proposed a change to openstack-infra/system-config: Add groups portal openid url https://review.openstack.org/131667 | 08:34 |
| *** IvanBerezovskiy has joined #openstack-infra | 08:35 | |
| *** rushiagr_away is now known as rushiagr | 08:39 | |
| *** unicell has joined #openstack-infra | 08:40 | |
| openstackgerrit | Sergey Reshetnyak proposed a change to openstack/requirements: Update python-saharaclient to 0.7.5 version https://review.openstack.org/131668 | 08:41 |
| SergeyLukjanov | sdague, when you'll be online | 08:42 |
| SergeyLukjanov | sdague, there are a few comments from jhesketh on your's fix for hidejs https://review.openstack.org/#/c/131249/2 | 08:42 |
| *** andreykurilin_ has quit IRC | 08:47 | |
| *** e0ne has joined #openstack-infra | 08:48 | |
| *** jpich has joined #openstack-infra | 08:50 | |
| *** MaxV_ has joined #openstack-infra | 08:51 | |
| *** IaKing has quit IRC | 08:51 | |
| *** nikesh_vedams_ has joined #openstack-infra | 08:51 | |
| nikesh_vedams_ | i am trying to deploy devstack on RHEL-6.5 | 08:52 |
| nikesh_vedams_ | but i am getting below error | 08:52 |
| nikesh_vedams_ | http://paste.openstack.org/show/126342/ | 08:52 |
| *** talluri has joined #openstack-infra | 08:52 | |
| nikesh_vedams_ | did any-one face this issue | 08:52 |
| nikesh_vedams_ | ? | 08:52 |
| *** hdd has joined #openstack-infra | 08:53 | |
| *** lttrl has joined #openstack-infra | 08:54 | |
| *** vdo has quit IRC | 08:55 | |
| *** e0ne has quit IRC | 08:56 | |
| *** BharatK has joined #openstack-infra | 08:57 | |
| *** ihrachyshka has joined #openstack-infra | 08:57 | |
| *** jule has joined #openstack-infra | 09:01 | |
| *** teran has quit IRC | 09:04 | |
| *** jistr has joined #openstack-infra | 09:06 | |
| *** derekh has joined #openstack-infra | 09:07 | |
| *** HeOS has joined #openstack-infra | 09:09 | |
| *** ujuc has quit IRC | 09:12 | |
| *** leopoldj has joined #openstack-infra | 09:13 | |
| *** slashme has quit IRC | 09:13 | |
| *** dims__ has joined #openstack-infra | 09:15 | |
| *** dims__ has quit IRC | 09:19 | |
| *** dizquierdo has joined #openstack-infra | 09:22 | |
| *** jedimike has joined #openstack-infra | 09:24 | |
| *** Guest74700 has joined #openstack-infra | 09:25 | |
| jule | I'm playing with Zuul but now I'm confused. How is zuul merger (gating, dependent pipeline manager) supposed to work when there is multiple independent changes ready to be integrated? | 09:27 |
| *** sarob has joined #openstack-infra | 09:27 | |
| jule | According to my observation it will merge only the first change and not even try to merge the other changes. Is my configuration somehow broken or is this the expected behaviour? | 09:28 |
| *** ZZelle has quit IRC | 09:30 | |
| *** ZZelle has joined #openstack-infra | 09:30 | |
| *** sarob has quit IRC | 09:32 | |
| *** achanda has joined #openstack-infra | 09:40 | |
| *** afazekas_beer is now known as psedlak | 09:40 | |
| *** teran has joined #openstack-infra | 09:41 | |
| *** _nadya_ has joined #openstack-infra | 09:44 | |
| *** pelix has joined #openstack-infra | 09:44 | |
| *** achanda has quit IRC | 09:44 | |
| *** teran has quit IRC | 09:46 | |
| *** gokrokve has quit IRC | 09:47 | |
| *** Guest74700 has quit IRC | 09:48 | |
| *** Guest74700 has joined #openstack-infra | 09:48 | |
| *** mkerrin has quit IRC | 09:49 | |
| *** mkerrin has joined #openstack-infra | 09:54 | |
| mordred | jule: it merges them in the order they were approved assuming each will pass | 09:59 |
| *** gokrokve has joined #openstack-infra | 10:00 | |
| mordred | jule: so if you're only seeing it do something with the first one, then yeah, I'd say there's something off in your config | 10:00 |
| mordred | jule: pastebin your config? | 10:00 |
| *** gokrokve has quit IRC | 10:06 | |
| *** gokrokve has joined #openstack-infra | 10:06 | |
| *** yfried is now known as yfried|afk | 10:07 | |
| jule | mordred: yes, the config: http://pastebin.com/8kUXuDBX and the layout: http://pastebin.com/FZifDrZn | 10:09 |
| *** marun has joined #openstack-infra | 10:09 | |
| *** yamahata has joined #openstack-infra | 10:10 | |
| *** gokrokve has quit IRC | 10:11 | |
| *** yfried|afk is now known as yfried | 10:12 | |
| *** BharatK has quit IRC | 10:15 | |
| *** teran has joined #openstack-infra | 10:16 | |
| *** e0ne has joined #openstack-infra | 10:18 | |
| *** gokrokve has joined #openstack-infra | 10:25 | |
| *** gokrokve has quit IRC | 10:25 | |
| *** gokrokve has joined #openstack-infra | 10:26 | |
| *** amotoki has quit IRC | 10:27 | |
| *** mpaolino has joined #openstack-infra | 10:28 | |
| mordred | jule: that looks fine ... do you have the extra approved status added to your gerrit? | 10:29 |
| *** dims__ has joined #openstack-infra | 10:30 | |
| *** gokrokve has quit IRC | 10:30 | |
| *** gokrokve has joined #openstack-infra | 10:30 | |
| *** jp_at_hp has joined #openstack-infra | 10:32 | |
| *** gokrokve has quit IRC | 10:32 | |
| *** gokrokve has joined #openstack-infra | 10:33 | |
| *** gokrokve has quit IRC | 10:35 | |
| *** gokrokve has joined #openstack-infra | 10:35 | |
| sdague | SergeyLukjanov: looking... | 10:37 |
| sdague | SergeyLukjanov: jhesketh missed that the outer parens are escaped | 10:37 |
| *** yfried is now known as yfried|afk | 10:38 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Use full path to grub2-mkconfig https://review.openstack.org/131636 | 10:38 |
| *** rushiagr is now known as rushiagr_away | 10:39 | |
| *** gokrokve has quit IRC | 10:40 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Refactor puppet-apply jobs into template https://review.openstack.org/131548 | 10:41 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Use common integrated jobs for sahara https://review.openstack.org/131543 | 10:41 |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add Juno stable-compat-jobs to trunk sahara client https://review.openstack.org/131542 | 10:42 |
| *** pfalleno1 has quit IRC | 10:42 | |
| *** darvon has quit IRC | 10:43 | |
| *** yaguang has quit IRC | 10:44 | |
| *** bcourt has quit IRC | 10:46 | |
| sdague | jhesketh: you still awake, wanted to chat with you about zuul status things | 10:47 |
| openstackgerrit | Flavio Percoco proposed a change to openstack-infra/devstack-gate: Set ZAQAR_BACKEND in localrc if present https://review.openstack.org/128224 | 10:50 |
| *** bcourt has joined #openstack-infra | 10:51 | |
| openstackgerrit | Ihar Hrachyshka proposed a change to openstack-infra/project-config: Don't run python33 job for oslo.messaging/icehouse https://review.openstack.org/131701 | 10:53 |
| *** cdent has joined #openstack-infra | 10:56 | |
| openstackgerrit | Flavio Percoco proposed a change to openstack-infra/project-config: Add a gate job for zaqar-redis https://review.openstack.org/125563 | 10:57 |
| flaper87 | clarkb: hey, can I get your eyes on https://review.openstack.org/128224 and https://review.openstack.org/125563 :) | 10:58 |
| *** lttrl has quit IRC | 10:58 | |
| *** katyafervent is now known as katyafervent_awa | 10:58 | |
| *** Guest74700 has quit IRC | 10:59 | |
| *** gokrokve has joined #openstack-infra | 11:01 | |
| *** loki184 has quit IRC | 11:01 | |
| *** gokrokve has quit IRC | 11:05 | |
| *** gokrokve has joined #openstack-infra | 11:06 | |
| *** jgallard has quit IRC | 11:09 | |
| *** marun has quit IRC | 11:10 | |
| *** gokrokve has quit IRC | 11:10 | |
| *** boris-42 has quit IRC | 11:11 | |
| *** marun has joined #openstack-infra | 11:13 | |
| *** darvon has joined #openstack-infra | 11:15 | |
| *** unicell has quit IRC | 11:16 | |
| *** yamahata has quit IRC | 11:17 | |
| jule | mordred: Thanks for checking the config. No, we do not have any extra approved status. | 11:19 |
| *** eharney has quit IRC | 11:21 | |
| jule | I analysed the problem more and it seems really weird. The Jenkins pipeline is being triggered for all 3 changes. But then the status for the last jobs in the pipeline is "None" for the two last changes | 11:23 |
| *** aysyd has joined #openstack-infra | 11:25 | |
| *** rushiagr_away is now known as rushiagr | 11:25 | |
| jule | Anyway, is it so that a separate pipeline is always triggered for independent changes? I.e. the merger does not try to group independent changes and test them together (to optimize the integration time)? | 11:27 |
| *** sarob has joined #openstack-infra | 11:28 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Add translation jobs for manila https://review.openstack.org/130989 | 11:30 |
| *** vhoward has joined #openstack-infra | 11:31 | |
| *** sarob has quit IRC | 11:32 | |
| *** talluri_ has joined #openstack-infra | 11:33 | |
| *** talluri has quit IRC | 11:33 | |
| *** andreaf_ has joined #openstack-infra | 11:33 | |
| *** doug-fish has joined #openstack-infra | 11:34 | |
| *** andreaf_ is now known as andreaf | 11:36 | |
| *** IaKing has joined #openstack-infra | 11:36 | |
| *** dizquierdo has quit IRC | 11:36 | |
| openstackgerrit | A change was merged to openstack-infra/project-config: Promote ironic sideways grenade to voting https://review.openstack.org/128725 | 11:37 |
| mordred | jule: nope. we optimize integration time in a dependent pipeline by running changes in parallel but using optimistic branch prediction to do merges as they would occur if all things merged successful | 11:38 |
| mordred | jule: that way if the tests fail, we know how to eject a single change and retest the other changes automatically | 11:39 |
| mordred | jule: so it's expected that the jenkins jobs will run for each test | 11:39 |
| mordred | gah | 11:39 |
| mordred | for each change | 11:39 |
| *** IaKing_ has joined #openstack-infra | 11:42 | |
| *** doug-fish1 has joined #openstack-infra | 11:44 | |
| jule | Ok. The reason I am asking this is, that in our project we are testing on real HW and cannot affort running "infinitely" parellel pipelines. Maybe Zuul does not fit in our case then. | 11:44 |
| *** IaKing has quit IRC | 11:44 | |
| *** doug-fish has quit IRC | 11:46 | |
| jule | I expected that it would first merge all open changes and test them at once. And then back off to testing smaller batches if this fails. | 11:46 |
| *** gokrokve has joined #openstack-infra | 11:47 | |
| *** bcourt has quit IRC | 11:47 | |
| *** mwagner_lap has quit IRC | 11:49 | |
| *** matel has joined #openstack-infra | 11:52 | |
| marun | anteaya: I'm assuming this is your paste? http://paste.openstack.org/show/118809/ | 11:52 |
| marun | anteaya: I hadn't used gertty in a while and now I'm seeing the same error. Did you manage to diagnose the problem? | 11:52 |
| sdague | mordred: can I ask you for an uber permissions change in gerrit? | 11:53 |
| sdague | oh, never mind, I think I can do this with review | 11:53 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/devstack-gate: Remove nosetests.xml reference https://review.openstack.org/131714 | 11:53 |
| *** ryanpetrello has joined #openstack-infra | 11:54 | |
| mordred | jule: yah - zuul's dependent pipeline manager may not help you for that use case | 11:55 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/project-config: give the entire bashate-core team release authority https://review.openstack.org/131715 | 11:55 |
| jule | mordred: Ok. Thank you very much for the clarification. | 11:55 |
| mordred | jule: sure thing! | 11:55 |
| sdague | mordred: / SergeyLukjanov ^^^ | 11:55 |
| mordred | jule: now, that said, I imagine it should be possible to write a zuul pipeline manager that implements the logic you just described | 11:56 |
| mordred | jule: and although we don't really have a need for such a thing in the openstack project directly, I can imagine a set of people who would be well served by such a thing | 11:56 |
| mordred | sdague: +@ | 11:57 |
| mordred | gah | 11:57 |
| mordred | sdague: +2 | 11:57 |
| *** _nadya_ has quit IRC | 11:57 | |
| jule | mordred: Sounds like an idea :) I think we are going to have a look on it | 11:57 |
| mordred | woot! | 11:57 |
| *** yfried|afk is now known as yfried | 11:59 | |
| mordred | sdague: so, the devstack-gate patch I just pushed up is actually something I saw while beginning to noodle on another thought in my brain - which is essential how can we remove more logic from devstack_gate and make it explicitly about env setup and not sets of things like: | 12:00 |
| mordred | if [[ "$DEVSTACK_GATE_TEMPEST_DISABLE_TENANT_ISOLATION" -eq "1" ]]; then | 12:00 |
| mordred | echo "TEMPEST_ALLOW_TENANT_ISOLATION=False" >>localrc | 12:00 |
| mordred | fi | 12:00 |
| sdague | yeh, while I don't love the var setting gymnastics, it does have the value of being explicit | 12:01 |
| sdague | but I agree, we could probably do a better thing | 12:01 |
| mordred | sure - but there is a bunch of stuff in there that really should be in devstack and not devstack-gate | 12:01 |
| sdague | yep, probably, periodically we sync some of it back | 12:02 |
| mordred | yah | 12:02 |
| sdague | though.... we then run into issues of the fact that devstack should work fine in a 4G vm, which d-g does not | 12:02 |
| mordred | I think there are definitely settings that are specific to d-g | 12:02 |
| mordred | but things like if [[ "$DEVSTACK_GATE_VIRT_DRIVER" == "ironic" ]]; then | 12:03 |
| *** _nadya_ has joined #openstack-infra | 12:03 | |
| mordred | describe a set of follow on settings like "DEFAULT_INSTANCE_TYPE" :) | 12:03 |
| mordred | anyway - I don't have anything specific of note - just starting to noodle in the space | 12:04 |
| sdague | yep, that's fair | 12:05 |
| sdague | honestly, a huge part of the current d-g refactor into functions was done by me on an airplane to hong-kong :) | 12:05 |
| mordred | :) | 12:05 |
| sdague | pre summit times are great times to think about stuff like this | 12:05 |
| mordred | sdague: did you see my crazypants new lib from yesterday? | 12:05 |
| sdague | .... I guess not | 12:06 |
| mordred | it's called "explore" | 12:06 |
| mordred | gah | 12:06 |
| mordred | it's called "explode" | 12:06 |
| mordred | you can pip install it | 12:06 |
| mordred | it also comes with a command line utility | 12:06 |
| mordred | it's for taking a python data structure, or a yaml file or a json file and exploding it into a bunch of directories and flat text files | 12:06 |
| sdague | oh, nice | 12:06 |
| mordred | for ease of consumption by shell things | 12:06 |
| mordred | this is the sort of thing we do pre-summit :) | 12:07 |
| *** naotok is now known as zz_naotok | 12:07 | |
| sdague | though, jq makes some of that fine with shell. I wonder if there is a yaml equiv | 12:07 |
| *** MaxV_ has quit IRC | 12:07 | |
| *** MaxV_ has joined #openstack-infra | 12:08 | |
| sdague | 318 downloads already? wow | 12:08 |
| mordred | I'm popular I guess | 12:08 |
| *** MaxV_ has quit IRC | 12:09 | |
| *** MaxV_ has joined #openstack-infra | 12:09 | |
| mordred | sdague: you could always do "flamel foo.yaml | jq" if you wanted to go that route | 12:09 |
| *** bcourt has joined #openstack-infra | 12:10 | |
| *** dprince has joined #openstack-infra | 12:10 | |
| sdague | hmmm flamel is not an apt-get installable for me | 12:10 |
| sdague | oh, it's a SpamapS git tree :) | 12:11 |
| jhesketh | sdague: just logged on, but really need to get some sleep before my flight | 12:11 |
| *** weshay has joined #openstack-infra | 12:11 | |
| jhesketh | sdague: if I missed something re the zuul javascript I'm all good for you/others to go ahead and merge it :-) | 12:11 |
| *** mfink has quit IRC | 12:12 | |
| mordred | sdague: it's also pip installable | 12:12 |
| sdague | jhesketh: so 2 quick things | 12:13 |
| sdague | oh, you already did - https://review.openstack.org/#/c/131249/ | 12:13 |
| sdague | jhesketh: the other thing was jeblair wanted your overall pov on - https://review.openstack.org/#/c/125944/ - because you were working on a different version of similar things | 12:14 |
| jhesketh | sdague: right, so my comments there should explain what I was doing | 12:15 |
| sdague | so if you are ok with that going in if I fix it, would be good to specify in the comments | 12:15 |
| jhesketh | oh yeah, absolutely | 12:15 |
| *** nosnos has quit IRC | 12:15 | |
| sdague | ah, ok, great | 12:15 |
| sdague | yeh, I think your comment is pretty clear. I'll respin for fixes today then | 12:15 |
| jhesketh | sdague: jeblair seemed to like the overlay I had though, so maybe we should chat about it at the summit? | 12:15 |
| sdague | sure | 12:15 |
| *** dizquierdo has joined #openstack-infra | 12:16 | |
| sdague | I was personally liking the compact view | 12:16 |
| jhesketh | sdague: my preference would be your approach and then improving on it to pull in actual jobs and progress bars | 12:16 |
| sdague | ok, yeh, that works as well | 12:16 |
| jhesketh | if we pull in the jobs running it should be no longer than after the results are posted | 12:16 |
| jhesketh | but this is a great start towards that | 12:16 |
| sdague | yep | 12:16 |
| sdague | ok, great | 12:16 |
| sdague | thanks for the reviews | 12:16 |
| jhesketh | no worries | 12:17 |
| jhesketh | I'm off then, see you in a few days :-) | 12:17 |
| *** JayJ has joined #openstack-infra | 12:20 | |
| *** bcourt has quit IRC | 12:20 | |
| *** gokrokve has quit IRC | 12:21 | |
| *** bcourt has joined #openstack-infra | 12:23 | |
| *** dims__ has quit IRC | 12:24 | |
| *** dims__ has joined #openstack-infra | 12:24 | |
| *** mbacchi has joined #openstack-infra | 12:24 | |
| *** JayJ has quit IRC | 12:25 | |
| *** mfink_ has joined #openstack-infra | 12:26 | |
| *** JayJ has joined #openstack-infra | 12:26 | |
| *** JayJ has quit IRC | 12:30 | |
| *** pc_m_ has joined #openstack-infra | 12:31 | |
| *** pc_m__ has joined #openstack-infra | 12:32 | |
| *** isviridov is now known as isviridov_away | 12:32 | |
| *** bswartz has quit IRC | 12:33 | |
| openstackgerrit | Mikhail S Medvedev proposed a change to openstack-infra/puppet-kibana: Add .gitreview https://review.openstack.org/131732 | 12:33 |
| *** ayoung-ZZZzz is now known as ayoung | 12:33 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: yum non interactive https://review.openstack.org/130719 | 12:35 |
| *** dizquierdo has quit IRC | 12:35 | |
| *** pc_m_ has quit IRC | 12:36 | |
| *** baoli has joined #openstack-infra | 12:36 | |
| sdague | jhesketh: see you in paris | 12:41 |
| *** mase_x200 has joined #openstack-infra | 12:43 | |
| *** gokrokve has joined #openstack-infra | 12:44 | |
| *** gokrokve has quit IRC | 12:45 | |
| *** Krast has joined #openstack-infra | 12:48 | |
| * fungi heads to the airport... mostly afk for the next few days while flying and conferencing. i look forward to seeing many of you in paris! | 12:49 | |
| *** IaKing_ has quit IRC | 12:50 | |
| *** unicell has joined #openstack-infra | 12:51 | |
| *** radez_g0n3 is now known as radez | 12:51 | |
| *** IaKing has joined #openstack-infra | 12:52 | |
| *** tomoe has quit IRC | 12:52 | |
| *** dkranz has joined #openstack-infra | 12:53 | |
| *** jerryz has quit IRC | 12:54 | |
| *** IaKing has quit IRC | 12:58 | |
| *** IaKing has joined #openstack-infra | 12:58 | |
| mrmartin | have a nice trip fungi | 12:59 |
| *** kgiusti has joined #openstack-infra | 12:59 | |
| dims__ | fungi: see you in paris | 12:59 |
| *** kgiusti has quit IRC | 13:01 | |
| *** kgiusti has joined #openstack-infra | 13:01 | |
| *** mattfarina has joined #openstack-infra | 13:03 | |
| *** yfried is now known as yfried|afk | 13:05 | |
| *** eharney has joined #openstack-infra | 13:06 | |
| *** talluri_ has quit IRC | 13:06 | |
| *** dkliban_afk is now known as dkliban | 13:07 | |
| *** dizquierdo has joined #openstack-infra | 13:07 | |
| *** koolhead17 has quit IRC | 13:08 | |
| *** mestery has quit IRC | 13:08 | |
| *** yfried|afk is now known as yfried | 13:08 | |
| *** koolhead17 has joined #openstack-infra | 13:09 | |
| *** yamahata has joined #openstack-infra | 13:09 | |
| *** amotoki has joined #openstack-infra | 13:09 | |
| *** markmcclain has joined #openstack-infra | 13:09 | |
| *** teran has quit IRC | 13:10 | |
| *** teran has joined #openstack-infra | 13:10 | |
| *** marun_ has joined #openstack-infra | 13:10 | |
| *** mestery has joined #openstack-infra | 13:11 | |
| *** marun has quit IRC | 13:12 | |
| *** mestery has quit IRC | 13:13 | |
| *** koolhead17 has quit IRC | 13:13 | |
| *** mestery has joined #openstack-infra | 13:13 | |
| *** bswartz has joined #openstack-infra | 13:15 | |
| *** marun_ is now known as marun | 13:15 | |
| *** Hal has joined #openstack-infra | 13:16 | |
| *** spzala has joined #openstack-infra | 13:16 | |
| matel | Hello, I would like to ask for some help with a service account, is anyone around? | 13:16 |
| *** Hal is now known as Guest99028 | 13:16 | |
| *** unicell has quit IRC | 13:17 | |
| *** jerryz has joined #openstack-infra | 13:18 | |
| *** pc_m__ has quit IRC | 13:18 | |
| *** mjturek has joined #openstack-infra | 13:18 | |
| openstackgerrit | Aaron Rosen proposed a change to openstack-infra/nodepool: libffi-dev/libssl-dev is required to install nodepool https://review.openstack.org/125515 | 13:19 |
| *** mwagner_lap has joined #openstack-infra | 13:21 | |
| *** dustins has joined #openstack-infra | 13:22 | |
| *** dizquierdo has quit IRC | 13:23 | |
| *** mriedem has joined #openstack-infra | 13:23 | |
| *** bknudson has joined #openstack-infra | 13:24 | |
| *** gokrokve has joined #openstack-infra | 13:24 | |
| *** IaKing has quit IRC | 13:31 | |
| *** amotoki has quit IRC | 13:31 | |
| *** talluri has joined #openstack-infra | 13:32 | |
| *** miqui has joined #openstack-infra | 13:32 | |
| *** IaKing has joined #openstack-infra | 13:32 | |
| *** hdd has quit IRC | 13:33 | |
| *** nikesh_vedams_ has quit IRC | 13:33 | |
| *** david-lyle has joined #openstack-infra | 13:37 | |
| *** hdd has joined #openstack-infra | 13:38 | |
| *** homeless has joined #openstack-infra | 13:38 | |
| *** yamahata has quit IRC | 13:41 | |
| *** yamahata has joined #openstack-infra | 13:41 | |
| *** zns has joined #openstack-infra | 13:42 | |
| *** david-lyle has quit IRC | 13:43 | |
| *** dmsimard_away is now known as dmsimard | 13:43 | |
| *** cdent has quit IRC | 13:43 | |
| *** Krast has quit IRC | 13:43 | |
| *** pfallenop has joined #openstack-infra | 13:46 | |
| *** cdent has joined #openstack-infra | 13:47 | |
| *** rushiagr is now known as rushiagr_away | 13:47 | |
| *** dangers_away is now known as dangers | 13:48 | |
| openstackgerrit | afazekas proposed a change to openstack-infra/nodepool: Add private ip to /etc/nodepool https://review.openstack.org/114840 | 13:49 |
| *** groknix has quit IRC | 13:49 | |
| *** lttrl has joined #openstack-infra | 13:49 | |
| *** groknix has joined #openstack-infra | 13:49 | |
| *** koolhead17 has joined #openstack-infra | 13:50 | |
| *** nelsnelson has joined #openstack-infra | 13:52 | |
| afazekas | jeblair: ^ | 13:53 |
| matel | Hello, I would like to ask for some help with a service account, is anyone around? | 13:53 |
| openstackgerrit | A change was merged to openstack-infra/project-config: give the entire bashate-core team release authority https://review.openstack.org/131715 | 13:54 |
| *** xyang0 has joined #openstack-infra | 13:54 | |
| openstackgerrit | Mikhail S Medvedev proposed a change to openstack-infra/puppet-kibana: Add required puppet module files and .gitreview https://review.openstack.org/131771 | 13:54 |
| mordred | afazekas: thanks - lgtm | 13:54 |
| *** talluri has quit IRC | 13:55 | |
| *** amitgandhinz has joined #openstack-infra | 13:59 | |
| *** zns has quit IRC | 13:59 | |
| *** sarob has joined #openstack-infra | 13:59 | |
| *** hdd has quit IRC | 14:00 | |
| *** sarob_ has joined #openstack-infra | 14:04 | |
| *** talluri has joined #openstack-infra | 14:04 | |
| *** zns has joined #openstack-infra | 14:04 | |
| *** sarob has quit IRC | 14:05 | |
| *** zz_dimtruck is now known as dimtruck | 14:08 | |
| *** todd_dsm has joined #openstack-infra | 14:08 | |
| clarkb | matel it is usually best to just ask. whats u | 14:09 |
| clarkb | *whats up? | 14:09 |
| *** andreaf has quit IRC | 14:11 | |
| *** andreaf has joined #openstack-infra | 14:12 | |
| *** markmcclain has quit IRC | 14:12 | |
| *** teran_ has joined #openstack-infra | 14:14 | |
| *** IaKing has quit IRC | 14:14 | |
| *** talluri has quit IRC | 14:14 | |
| *** IaKing has joined #openstack-infra | 14:15 | |
| *** dizquierdo has joined #openstack-infra | 14:16 | |
| *** msabramo has quit IRC | 14:16 | |
| *** teran has quit IRC | 14:17 | |
| sdague | matel: oh hey, you are around, can you fix your devstack patch here - https://review.openstack.org/#/c/129208/ | 14:17 |
| *** nikhil_k has joined #openstack-infra | 14:18 | |
| *** freemanbrandon has joined #openstack-infra | 14:18 | |
| nikhil_k | fungi: hi! | 14:18 |
| nikhil_k | wanted to understand this patch https://review.openstack.org/#/c/128928/1 | 14:18 |
| nikhil_k | whenever you've a couple of mins | 14:19 |
| *** dizquierdo has quit IRC | 14:19 | |
| clarkb | nikhil_k fungi is traveling to paris and afk for a couple days | 14:19 |
| *** mase_x200 has quit IRC | 14:19 | |
| *** dizquierdo has joined #openstack-infra | 14:19 | |
| nikhil_k | clarkb: ah! can you please help? | 14:20 |
| clarkb | that change "merges" juno back into kilo so that kilo is aware of juno in its history | 14:20 |
| clarkb | it should just construct the git tree necessary to make that happen and change no code in kilo (master) | 14:20 |
| clarkb | you can confirmthis with git diff | 14:21 |
| *** yfried is now known as yfried|afk | 14:21 | |
| *** zz_jgrimm is now known as jgrimm | 14:22 | |
| *** nimrodsun___ has quit IRC | 14:22 | |
| *** dizquierdo has quit IRC | 14:22 | |
| *** talluri has joined #openstack-infra | 14:23 | |
| *** talluri has quit IRC | 14:23 | |
| nikhil_k | clarkb: Thank you! Got the idea, approving | 14:25 |
| *** links has quit IRC | 14:26 | |
| *** nikil89_ has joined #openstack-infra | 14:27 | |
| *** nikil89_ is now known as naggappan | 14:27 | |
| *** naggappan has left #openstack-infra | 14:28 | |
| *** naggappan has joined #openstack-infra | 14:28 | |
| *** stevemar has joined #openstack-infra | 14:28 | |
| *** habib has joined #openstack-infra | 14:30 | |
| naggappan | Hi i am trying to set up the CI system. Now when ever changes happens in openstack-dev/sandbox, my jenkins job is getting triggered. Just for test purpose how can i publish the result back for same code/branch/commitID for which the job is triggered | 14:30 |
| *** doude has quit IRC | 14:30 | |
| *** doude has joined #openstack-infra | 14:31 | |
| clarkb | naggappan you should be able to comment but not vote | 14:31 |
| clarkb | so set it to comment with "success" or "failure" comments that do not +1 or -1 | 14:32 |
| *** yfried|afk is now known as yfried | 14:33 | |
| naggappan | clarkb: should we need to use any plugin to comment back? currently i am using gerrint plugin in jenkins to get events and trigger the job. | 14:33 |
| *** jerryz has quit IRC | 14:34 | |
| openstackgerrit | Eric Harney proposed a change to openstack-infra/project-config: Add centos7 experimental job for Cinder https://review.openstack.org/131783 | 14:34 |
| openstackgerrit | afazekas proposed a change to openstack-infra/project-config: Repo caching in parallel https://review.openstack.org/131784 | 14:34 |
| clarkb | naggappan that should work | 14:35 |
| *** ZZelle has quit IRC | 14:35 | |
| *** dizquierdo has joined #openstack-infra | 14:37 | |
| *** signed8bit has joined #openstack-infra | 14:37 | |
| openstackgerrit | Sergey Skripnick proposed a change to openstack-infra/project-config: Add performance check job for murano https://review.openstack.org/131790 | 14:42 |
| *** emagana has joined #openstack-infra | 14:42 | |
| *** yfried is now known as yfried|afk | 14:45 | |
| *** mpaolino has quit IRC | 14:45 | |
| pleia2 | good morning | 14:46 |
| *** habib has quit IRC | 14:46 | |
| *** zns has quit IRC | 14:48 | |
| mordred | pleia2: morning | 14:51 |
| *** david-lyle has joined #openstack-infra | 14:52 | |
| jeblair | good morning | 14:53 |
| *** wenlock has joined #openstack-infra | 14:53 | |
| *** todd_dsm has quit IRC | 14:55 | |
| *** thedodd has joined #openstack-infra | 14:56 | |
| *** hdd has joined #openstack-infra | 14:56 | |
| *** carl_baldwin has joined #openstack-infra | 14:58 | |
| *** yamahata has quit IRC | 14:58 | |
| *** yamahata has joined #openstack-infra | 14:58 | |
| *** jcoufal has quit IRC | 14:58 | |
| *** changbl has joined #openstack-infra | 14:59 | |
| *** mpaolino has joined #openstack-infra | 14:59 | |
| *** sarob_ has quit IRC | 14:59 | |
| *** jaypipes is now known as jaypipes-afk | 15:00 | |
| *** sarob has joined #openstack-infra | 15:00 | |
| sdague | jeblair: if you want to spin through the hideci changes, you'll get pipeline grouping, and fixes for new screen - https://review.openstack.org/#/c/131249 https://review.openstack.org/#/c/131500/ both already have some +2s | 15:00 |
| *** jcoufal has joined #openstack-infra | 15:01 | |
| matel | sdague: sure, will look at the devstack change - I was a bit distracted. | 15:01 |
| matel | Our gerrit account seems to be disabled, could someone take a look at it? The username is: citrix_xenserver_ci | 15:03 |
| *** IaKing has quit IRC | 15:03 | |
| *** spzala has quit IRC | 15:03 | |
| *** todd_dsm has joined #openstack-infra | 15:03 | |
| clarkb | matel are you subscribed to third-party-announce? fungi mentioned it there | 15:04 |
| clarkb | http://lists.openstack.org/pipermail/third-party-announce/2014-October/000084.html | 15:05 |
| openstackgerrit | Mark Vanderwiel proposed a change to openstack-infra/project-config: Missed repo gate job for zuul layout filter. https://review.openstack.org/131467 | 15:06 |
| *** sarob has quit IRC | 15:06 | |
| *** sarob has joined #openstack-infra | 15:07 | |
| matel | clarkb: thanks, I thought that it only means that voting disabled, but a second read revealed that it's causing the issues. | 15:07 |
| matel | clarkb: What's the protocol to re -enable the account, should I just get in touch with fungi? | 15:07 |
| *** rushiagr_away is now known as rushiagr | 15:08 | |
| *** imcsk8 has quit IRC | 15:08 | |
| *** mrmartin has quit IRC | 15:08 | |
| *** imcsk8 has joined #openstack-infra | 15:08 | |
| clarkb | protocol is typicaly address the specific issues and we can reenable | 15:08 |
| *** imcsk8 has quit IRC | 15:09 | |
| *** imcsk8 has joined #openstack-infra | 15:09 | |
| *** yfried has joined #openstack-infra | 15:09 | |
| jeblair | sdague: please find someone to test, review, and merge those changes soon. i do not have time to review them before the summit. since fungi, jhesketh, and SergeyLukjanov approved your original change, i think they are most familiar with the code and should be able to. | 15:09 |
| matel | clarkb: Okay, I already fixed the issues, and upstreamed them: https://review.openstack.org/#/c/131438/ | 15:10 |
| jeblair | sdague: if they don't get merged soon, i think we need to merge the revert because i do not think it's good to have the new change screen broken. | 15:10 |
| matel | clarkb: should I get in touch with fungi, or could you enable the account as well? | 15:10 |
| sdague | jeblair: sergey tested them | 15:10 |
| clarkb | any infra rooter can enable it | 15:11 |
| openstackgerrit | afazekas proposed a change to openstack-infra/project-config: Repo caching in parallel https://review.openstack.org/131784 | 15:11 |
| jeblair | sdague: but didn't review or approve? | 15:11 |
| sdague | I'll find another tester | 15:11 |
| sdague | he +2ed the 2nd patch | 15:11 |
| sdague | he did not hit +A, I don't know why | 15:11 |
| *** sarob has quit IRC | 15:12 | |
| *** yfried_ has joined #openstack-infra | 15:12 | |
| *** yfried|afk has quit IRC | 15:12 | |
| SergeyLukjanov | sdague, I've not approved it because the dependent patch isn't approved yet | 15:12 |
| *** pcrews has joined #openstack-infra | 15:12 | |
| sdague | SergeyLukjanov: can you look at the dependent patch? | 15:12 |
| sdague | it has a +2 from jhesketh | 15:12 |
| SergeyLukjanov | sdague, yup, I'll check it in 15 mins | 15:12 |
| sdague | cool, great | 15:12 |
| *** yamamoto has joined #openstack-infra | 15:13 | |
| matel | fungi: ping | 15:13 |
| *** yfried has quit IRC | 15:15 | |
| clarkb | fungi is traveling | 15:15 |
| *** dizquierdo has quit IRC | 15:15 | |
| clarkb | matel I am poking around (reviewed that change) and can enable in a bit | 15:16 |
| matel | clarkb: thanks | 15:16 |
| *** dane_leblanc has joined #openstack-infra | 15:19 | |
| *** isviridov_away is now known as isviridov | 15:20 | |
| *** yfried_ has quit IRC | 15:22 | |
| openstackgerrit | Cyril Roelandt proposed a change to openstack-infra/project-config: sqlalchemy-migrate: enable the py33 gate. https://review.openstack.org/131797 | 15:22 |
| *** rmcall has joined #openstack-infra | 15:24 | |
| clarkb | matel: ok before I enable anything, you guys have fixed the pypi thing? | 15:26 |
| clarkb | matel: I see the commit on github. is that deployed and in use? | 15:26 |
| matel | Yes, we are using the python one. | 15:26 |
| clarkb | matel: (you will likely need to have nodepool build new images for that to happen) | 15:26 |
| matel | clarkb: https://github.com/citrix-openstack/project-config/commits/xenserver-ci <- this one is deployed | 15:26 |
| clarkb | matel: and you rebuilt your image? | 15:27 |
| matel | clarkb: Images are already re-built. | 15:27 |
| clarkb | perfect | 15:27 |
| matel | clarkb: yep | 15:27 |
| clarkb | gah fungi didn't specifically mention the username of the disabled account /me looks at database | 15:29 |
| matel | clarkb: citrix_xenserver_ci | 15:30 |
| *** hemna__ is now known as hemna | 15:30 | |
| clarkb | matel: all done | 15:31 |
| *** reed has joined #openstack-infra | 15:33 | |
| matel | clarkb: thank you! | 15:33 |
| *** atiwari has joined #openstack-infra | 15:36 | |
| clarkb | I really need a whiteboard in home office. This linux networking stuff for multinode deserves a giant picture drawn in dry erase marker | 15:37 |
| BobBall | Use your fridge. And a CD marker. | 15:37 |
| BobBall | That's what my daughter did. | 15:37 |
| pleia2 | haha | 15:38 |
| BobBall | (although admittedly not to draw a digram of multinode linux networking...) | 15:38 |
| clarkb | I think the reason multi_host is not working is that with multi_host you do the floating_range NAT on each individual compute node rather than on the controller. This means the controller needs to know to send packets across the gretap bridge to the other node to get to 172.24.4.0/24 | 15:38 |
| clarkb | on the non multi_host this works fine because all of the 172.24.4.0/24 addresses are NATed on the controller so the controller says go locally to hit this stuff | 15:38 |
| clarkb | but with multi_host=True there may not be any 172.24.4.0/24 address on the controller | 15:39 |
| clarkb | I think this can be fixed by adding addresses on the pub_if interfaces for the hosts themselves on that network (also possibly need to mess with iptables so that packets can get around) | 15:39 |
| *** Sukhdev has joined #openstack-infra | 15:39 | |
| clarkb | BobBall: at least you have a one of a kind fridge :) | 15:41 |
| *** msabramo has joined #openstack-infra | 15:41 | |
| *** oomichi has joined #openstack-infra | 15:41 | |
| BobBall | I should sell it on ebay. I'm sure that being one-of-a-kind makes it super valuable. | 15:42 |
| *** oomichi has quit IRC | 15:42 | |
| *** dizquierdo has joined #openstack-infra | 15:42 | |
| sileht | fungi, the bandersnatch bug strike again: http://pypi.iad.openstack.org/packages/source/S/SQLAlchemy-Utils/SQLAlchemy-Utils-0.27.6.tar.gz#md5=e03689838a6087d551fa08ee7a228e98 is missing | 15:43 |
| openstackgerrit | Elizabeth K. Joseph proposed a change to openstack-infra/infra-manual: Add Working with Bugs section to Dev guide https://review.openstack.org/107741 | 15:43 |
| *** zaphod42 has joined #openstack-infra | 15:46 | |
| *** pblaho has joined #openstack-infra | 15:46 | |
| *** patrickeast has joined #openstack-infra | 15:47 | |
| *** dizquierdo has quit IRC | 15:47 | |
| *** todd_dsm has quit IRC | 15:48 | |
| *** todd_dsm has joined #openstack-infra | 15:49 | |
| *** gokrokve has quit IRC | 15:50 | |
| *** todd_dsm has quit IRC | 15:52 | |
| *** armax has joined #openstack-infra | 15:52 | |
| *** spzala has joined #openstack-infra | 15:52 | |
| carl_baldwin | Hi infra: I have a gerrit question. I’m wondering why I’m unable to add “Oleg Bondarev <obondarev@mirantis.com>” as a reviewer to a patch. Gerrit offers three identical choices in the auto-complete box. When I select any of the three it says: “Oleg Bondarev <obondarev@mirantis.com> does not identify a registered user or group” | 15:54 |
| sdague | clarkb: does that just mean setting flat_interface correctly? | 15:54 |
| *** todd_dsm has joined #openstack-infra | 15:55 | |
| *** ashaeron has quit IRC | 15:56 | |
| *** kgiusti has quit IRC | 15:56 | |
| *** bswartz has quit IRC | 15:59 | |
| clarkb | sdague: flat interface is correct | 16:00 |
| afazekas | jeblair: can you have a look on this: https://review.openstack.org/#/c/114840/ | 16:00 |
| *** bswartz has joined #openstack-infra | 16:00 | |
| clarkb | sdague: but if I have no address on a network linux should use my default route to get there | 16:00 |
| *** msabramo has quit IRC | 16:00 | |
| clarkb | sdague: so I think that is happening is flat_interface is pub_if (correctly so) but traffic from controller destined for 172.24.4.0/24 is headed out eth0 | 16:00 |
| clarkb | sdague: if however pub_if always has an addr on 172.24.4.0/24 (adjust floating range appropriately) then we should always send that traffic over pub_if | 16:01 |
| clarkb | I am drawing ascii pictures | 16:01 |
| sdague | ok | 16:02 |
| *** rmcall has quit IRC | 16:03 | |
| *** k4n0 has quit IRC | 16:03 | |
| *** chandankumar has quit IRC | 16:03 | |
| *** rmcall has joined #openstack-infra | 16:04 | |
| *** sarob has joined #openstack-infra | 16:07 | |
| *** mmaglana has joined #openstack-infra | 16:07 | |
| *** mwagner_lap has quit IRC | 16:08 | |
| *** todd_dsm has quit IRC | 16:09 | |
| *** sarob_ has joined #openstack-infra | 16:09 | |
| *** msabramo has joined #openstack-infra | 16:09 | |
| jeblair | afazekas: lgtm, thanks! | 16:10 |
| jeblair | clarkb: ^ | 16:10 |
| jeblair | mordred: also you | 16:10 |
| clarkb | jeblair: awesome I shall review as soon as my picture drawing is complete | 16:11 |
| clarkb | (this is actually helping me a lot and I shoudl have done this last night) | 16:11 |
| *** denis_makogon has quit IRC | 16:12 | |
| *** sarob has quit IRC | 16:12 | |
| mordred | jeblair: oh - I reviewed that earlier and didn't click submit :) | 16:12 |
| jeblair | clarkb: i learned emacs has "artist mode", which due to the extremely sad state of linux ascii drawing programs, is one of the better ways of doing ascii art | 16:12 |
| *** tsg has quit IRC | 16:12 | |
| clarkb | huh I might have to look at that. I am mostly done though | 16:13 |
| mordred | I'm not goign to +A just in case clarkb wants to look | 16:13 |
| mordred | (and since I wrote a chunk of the code) | 16:13 |
| jeblair | we also probably want to make the db change and restart around the same time as +A | 16:13 |
| jeblair | http://www.cinsk.org/emacs/emacs-artist.html | 16:14 |
| jeblair | screencast | 16:14 |
| *** vhoward has left #openstack-infra | 16:14 | |
| mordred | jeblair: it would be cooler if that was a jeblair-telnet-screencast style screencast | 16:14 |
| clarkb | sdague: jeblair afazekas http://paste.openstack.org/show/126564/ shows the two setups | 16:14 |
| mordred | jeblair: do we have a name for that yet? | 16:15 |
| clarkb | primary node is the controller node and is where tempest runs yet | 16:15 |
| jeblair | mordred: oh yeah, we should brand it so it becomes a thing | 16:15 |
| mordred | jeblair: oh - I guess it's using a mouse | 16:15 |
| *** reed has quit IRC | 16:15 | |
| mordred | jeblair: have you tried out my explode program yet? I'm curious to see what you think of the result | 16:15 |
| jeblair | mordred: it sounds like a mess! :P | 16:16 |
| clarkb | and after drawing that I am 99% sure the reason mutli_host = true does not work is traffic bound for 172.24.4.0/24 on the controller originating with tempest is going to go out eth0 not pub_if | 16:16 |
| *** hdd has quit IRC | 16:16 | |
| clarkb | if however I put some 172.24.5.x/23 addresses on pub_if I think that will fix it (assuming iptables lets that traffic through) | 16:16 |
| jeblair | mordred: apparently joey hess took the name termcast: https://joeyh.name/termcast/ | 16:17 |
| clarkb | I don't want floating_range and the hypervisor host IPs to overlap so we can making floating range 172.24.4.0/24 within a /23 network with the same prefix | 16:17 |
| clarkb | sdague: ^ does that make sense? | 16:17 |
| *** hdd has joined #openstack-infra | 16:18 | |
| *** markmcclain has joined #openstack-infra | 16:18 | |
| clarkb | mordred: can you review the parent of 114840? | 16:18 |
| jeblair | clarkb: that is a diagram of what devstack does currently based on the value of that parameter? | 16:18 |
| sdague | clarkb: I think so, but I'll warn you my brain doesn't have full context on this one, and probably won't until after summit | 16:19 |
| *** tsg has joined #openstack-infra | 16:19 | |
| clarkb | jeblair: within the devstack-gate aiopcpu setup yes | 16:19 |
| *** rlandy has quit IRC | 16:19 | |
| *** ildikov has quit IRC | 16:19 | |
| *** IvanBerezovskiy has left #openstack-infra | 16:19 | |
| clarkb | jeblair: so the multi_host = true case fails tempest tests when trying to ssh to hosts on 172.24.4.0/24 | 16:19 |
| *** armax has quit IRC | 16:19 | |
| jeblair | clarkb: ok, so both of those are two-node setup, right? | 16:20 |
| clarkb | jeblair: yes | 16:20 |
| afazekas | clarkb: cool, maybe the other nic should get 10.0.z.q , just to show it is different than x.y | 16:20 |
| clarkb | multi_host true is basically run nova network on all your compute nodes and they will not all tunnel through a spof | 16:20 |
| jeblair | clarkb: so is there a route on the primary node for 172.24? | 16:21 |
| clarkb | jeblair: no I do not think so (I am waiting for a run to finish on a held env so I can double check) | 16:21 |
| afazekas | jeblair, clarkb, mordred: https://review.openstack.org/#/c/131646/ | 16:22 |
| mordred | clarkb: on it | 16:22 |
| clarkb | jeblair: afazekas: I also think I need something like 131646 for the 172.24 addresses if NAT can be hosted on either node | 16:23 |
| clarkb | because traffic will cross nodes with a dest of 172.24.stuff | 16:23 |
| *** tsg has quit IRC | 16:23 | |
| *** todd_dsm has joined #openstack-infra | 16:24 | |
| *** zz_gondoi is now known as gondoi | 16:24 | |
| *** marun has quit IRC | 16:24 | |
| afazekas | clarkb: the bridge has an allow everything rule | 16:25 |
| mordred | jeblair: for the provider information blocks - do we think that any of those values might have spaces? | 16:25 |
| jeblair | 131646 lgtm and i believe it will not fail even without the new files present | 16:25 |
| jeblair | mordred: i think we are assuming not. i don't know if nova supports that (to answer whether they possibly could in future) | 16:26 |
| mordred | kk. just thought I'd ask | 16:26 |
| jeblair | mordred: we certainly assumed not when we stuck them in dns :) | 16:26 |
| clarkb | afazekas: ok so only unbridged traffic (the hpcloud or rackspace "private" networks) need the additional rules? | 16:26 |
| clarkb | afazekas: basically things on eth0 as in my diagram | 16:27 |
| jeblair | mordred: explode needs '--help' | 16:28 |
| *** _nadya_ has quit IRC | 16:28 | |
| mordred | jeblair: yeah - that's on my todo list. "explode <infile> <outputdir>" | 16:28 |
| jeblair | mordred: yeah, i guessed that | 16:28 |
| clarkb | ya there is no 172.24.4.0/24 related route on either node in either situation | 16:29 |
| jeblair | mordred: explode works well with a semi-complicate yaml file i'm working on | 16:29 |
| mordred | jeblair: woot | 16:29 |
| clarkb | so I think next step is whip up a change that puts some addresses on taht network to get a route with source on the same l2 network | 16:29 |
| clarkb | er | 16:29 |
| mordred | jeblair: I tried it on the most complex thing I had | 16:29 |
| clarkb | same l3 network that is hosted on same l2 lan | 16:30 |
| jeblair | clarkb: sounds like it | 16:30 |
| *** pblaho has quit IRC | 16:30 | |
| afazekas | clarkb: Maybe some other rule required, but the FIP traffic should be enabled by nova. IMHO if an additional rule is nova multohost specific it should be in d-g | 16:30 |
| clarkb | afazekas: in this case I think it needs to go in d-g because we are constructing this virtual network from scratch | 16:31 |
| clarkb | afazekas: normally you would use your eth0 which has addresses and routes properly configured, but we are missing the route bit | 16:32 |
| clarkb | jeblair: 131646 will fail because it is set -e | 16:32 |
| clarkb | (er fail until we have the new stuff in nodepool) | 16:32 |
| jeblair | clarkb: it did not fail in my local testing | 16:32 |
| clarkb | I just did `cat file_that_exists file_that_does_not_exist` and it returns 1 | 16:33 |
| jeblair | clarkb: i did this: http://paste.openstack.org/show/126585/ | 16:33 |
| jeblair | and it echoed 'hi' | 16:33 |
| jeblair | with output: http://paste.openstack.org/show/126586/ | 16:33 |
| clarkb | oh does for/if/while protect against the set -e? | 16:34 |
| jeblair | that's what i'm guessing | 16:34 |
| clarkb | I guess that makes sense otherwise you wouldn't have conditions that worked | 16:34 |
| *** gokrokve has joined #openstack-infra | 16:35 | |
| afazekas | false | true ; echo $? # 0 | 16:35 |
| jeblair | i extended that to one file that exists and one that does not, and it also works | 16:35 |
| *** gokrokve has quit IRC | 16:36 | |
| *** zhiyan has quit IRC | 16:36 | |
| afazekas | clarkb: set -e does not fails if only the left side of the pipe was failing | 16:37 |
| *** jraim has quit IRC | 16:37 | |
| *** simonmcc has quit IRC | 16:37 | |
| *** ameade has quit IRC | 16:37 | |
| *** serverascode__ has quit IRC | 16:37 | |
| *** ctracey has quit IRC | 16:38 | |
| jeblair | mtreinish: around? | 16:38 |
| *** tomoe has joined #openstack-infra | 16:38 | |
| mtreinish | jeblair: yep | 16:38 |
| jeblair | clarkb, mordred, SergeyLukjanov: subunit2sql system-config changes lgtm if you want to review. terminating at 127060 | 16:38 |
| *** gokrokve has joined #openstack-infra | 16:39 | |
| *** gyee has joined #openstack-infra | 16:39 | |
| *** tsg has joined #openstack-infra | 16:40 | |
| jeblair | mtreinish: ^ if someone else has time to review those soon, we could merge them today; but if it's more than a few hours, we're getting into the zone where i can't commit to fixing any problems that creep up due to going afk for ny/paris | 16:40 |
| *** dkliban is now known as dkliban_brb | 16:40 | |
| *** gokrokve has quit IRC | 16:40 | |
| clarkb | afazekas: mordred https://review.openstack.org/#/c/114840/ noted two things inline on that change | 16:40 |
| clarkb | jeblair: its such a big change :/ | 16:41 |
| mtreinish | clarkb: it's mostly copy and pasted from your worker script :) | 16:41 |
| * clarkb looks to see how crazy it is :) | 16:41 | |
| clarkb | mtreinish: thats the scariest thing :) | 16:41 |
| mtreinish | jeblair: sure, no worries | 16:41 |
| jeblair | mordred: you may want to look at clarkb's second question in 114840 | 16:42 |
| *** tonytan4ever has joined #openstack-infra | 16:42 | |
| *** Swanson has joined #openstack-infra | 16:43 | |
| *** reed has joined #openstack-infra | 16:43 | |
| *** ivar-lazzaro has joined #openstack-infra | 16:45 | |
| *** tomoe has quit IRC | 16:46 | |
| *** ivar-lazzaro has quit IRC | 16:46 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/storyboard: Add documentation for getting openid URL to setup super users. https://review.openstack.org/131276 | 16:46 |
| *** ivar-lazzaro has joined #openstack-infra | 16:47 | |
| *** bcourt has quit IRC | 16:47 | |
| zaro | morning | 16:49 |
| zaro | pelix: jjb release next week. anything you want to get in? | 16:50 |
| clarkb | mtreinish: jeblair comments on https://review.openstack.org/#/c/108003/ didn't vote though as I wanted to hear if you guys felt that cleanup would be helpful | 16:53 |
| openstackgerrit | afazekas proposed a change to openstack-infra/nodepool: Add private ip to /etc/nodepool https://review.openstack.org/114840 | 16:54 |
| jeblair | clarkb: all things equal, i think addressing those 3 points would be good | 16:54 |
| clarkb | mtreinish: jeblair: it is also not clear to me why openstack_project::subunit2sql_server is a thing are we going to run that on a special host? | 16:55 |
| *** jcoufal has quit IRC | 16:55 | |
| jeblair | clarkb: agreed, that seems excessive | 16:57 |
| mtreinish | clarkb: so I responded I can respin the patch quickly to drop the unused vars | 17:00 |
| *** arxcruz has quit IRC | 17:00 | |
| *** bhuvan has joined #openstack-infra | 17:00 | |
| mtreinish | but for the ini change I'd like to do that as a follow-on patch so I can do some manual testing | 17:00 |
| pelix | zaro: https://review.openstack.org/#/c/127370/ seems to be the most important one, maybe https://review.openstack.org/#/c/118170/ as well as it fixes an ordering problem. | 17:00 |
| mtreinish | clarkb: as for openstack_project::subunit2sql_server I can drop that, I only added because I thought we might need to do extra config at some point on the sql server | 17:01 |
| mtreinish | but that's not really a good justification | 17:01 |
| clarkb | mtreinish: follow up patch is fine | 17:03 |
| pelix | zaro: think it's too late in the cycle for looking at performance improvements, but would be good to get some opinions on whether https://review.openstack.org/45371 or https://review.openstack.org/#/c/75514/ should be worked on to allow parallel uploads. | 17:03 |
| clarkb | mtreinish: oh you use oslo.config | 17:03 |
| clarkb | that'll learn you | 17:03 |
| clarkb | mtreinish: I am fine to +2 the current state of that but can wait if you want to cleanup the output* vars | 17:04 |
| mtreinish | clarkb: heh, only because I used oslo.db | 17:04 |
| *** ivar-laz_ has joined #openstack-infra | 17:04 | |
| clarkb | ok progress on mutli_host. if I add 172.24.5.[12]/23 addresses on either node the traffic for that /23 network crosses the bridge because a route is added | 17:04 |
| clarkb | current issue is neither side is responding to arp requests for the other... | 17:05 |
| clarkb | not sure why that is happening yet | 17:05 |
| *** thedodd has quit IRC | 17:05 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql gearman workers https://review.openstack.org/108003 | 17:05 |
| mtreinish | clarkb: ^^^ dropped the unused variables | 17:06 |
| clarkb | I probably didn't set the appropriate type with ip address add | 17:06 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuration https://review.openstack.org/127060 | 17:06 |
| *** ivar-lazzaro has quit IRC | 17:07 | |
| mtreinish | clarkb: did you want me to respin the top patch to drop openstack_project::subunit2sql_server ? | 17:09 |
| *** hdd has quit IRC | 17:09 | |
| clarkb | mtreinish: ya I would put that on logstash.o.o or something | 17:10 |
| clarkb | mtreinish: and just consolidate the command and control gearman client portion of that stuff on one node | 17:10 |
| zaro | mgagne: have time for jjb review? | 17:10 |
| mgagne | zaro: unfortunately not, but depends if it's complex or not | 17:11 |
| *** HeOS has quit IRC | 17:11 | |
| *** ivar-laz_ has quit IRC | 17:11 | |
| *** ivar-lazzaro has joined #openstack-infra | 17:11 | |
| mtreinish | clarkb: ok, will do. I'll move it into openstack_project::logstash | 17:11 |
| *** dkliban_brb is now known as dkliban | 17:12 | |
| *** mwagner_lap has joined #openstack-infra | 17:12 | |
| zaro | mgagne: i would say medium. but no big deal, next time. see you soon :) | 17:13 |
| *** dane_leblanc has quit IRC | 17:13 | |
| *** _shaps_ has quit IRC | 17:13 | |
| mgagne | zaro: can't wait to get at the summit =) | 17:13 |
| *** tsg has quit IRC | 17:14 | |
| zaro | pelix: we will focus on those performance ones for next release. i'm trying to be more regular with releases, doing them ~ monthly. | 17:14 |
| *** tsg has joined #openstack-infra | 17:15 | |
| zaro | pelix: those were not on my review radar because of merge conflict. how about you get them in shape and then i will review? | 17:15 |
| *** VijayTripathi has joined #openstack-infra | 17:15 | |
| *** yamamoto has quit IRC | 17:15 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuration https://review.openstack.org/127060 | 17:16 |
| *** harlowja_away is now known as harlowja | 17:16 | |
| mtreinish | clarkb: ^^^ | 17:16 |
| *** tgohad has joined #openstack-infra | 17:16 | |
| *** jpich has quit IRC | 17:16 | |
| zaro | pelix: how about apprv on these doc changes? https://review.openstack.org/#/c/130614 and https://review.openstack.org/#/c/130391 already got +1 from the English maven. | 17:17 |
| *** Sincler has joined #openstack-infra | 17:17 | |
| *** emagana has quit IRC | 17:17 | |
| *** ihrachyshka has quit IRC | 17:18 | |
| matel | sdague: I hav updated the devstack changes. | 17:18 |
| *** tsg has quit IRC | 17:18 | |
| zaro | fungi: re-review? https://review.openstack.org/#/c/127370/ | 17:20 |
| *** jlibosva has quit IRC | 17:20 | |
| *** harlowja_ has joined #openstack-infra | 17:21 | |
| *** harlowja has quit IRC | 17:21 | |
| *** talluri has joined #openstack-infra | 17:22 | |
| nelsnelson | howdy channel. So I went back and re-did my devstack-gate deployment following the instructions from https://github.com/openstack-infra/devstack-gate again. This time with the proper Ubuntu 14.04 performance image. 8gig. Still hanging now, just seems to hang earlier for some reason. https://gist.github.com/nelsnelson/901c055acd13a7689d74 | 17:22 |
| nelsnelson | That gist is just a sample of the log for my most recent execution of safe-devstack-vm-gate-wrap.sh | 17:23 |
| nelsnelson | Sampling latest entries. | 17:23 |
| *** achanda has joined #openstack-infra | 17:23 | |
| *** baoli has quit IRC | 17:23 | |
| nelsnelson | Now 5 minutes after those log entries, with no other output. | 17:24 |
| *** emagana has joined #openstack-infra | 17:24 | |
| pelix | zaro: they look good to me, do we have another core around to approve them? | 17:25 |
| *** david-lyle has quit IRC | 17:25 | |
| *** derekh has quit IRC | 17:25 | |
| *** amitgandhinz has quit IRC | 17:25 | |
| zaro | pelix: i would consider pursehouse a core on anything jjb doc related :) | 17:26 |
| *** hdd has joined #openstack-infra | 17:27 | |
| *** ivar-lazzaro has quit IRC | 17:27 | |
| * zaro wishes gerrit had jj doc core option | 17:27 | |
| *** Ryan_Lane has joined #openstack-infra | 17:27 | |
| *** jcoufal has joined #openstack-infra | 17:27 | |
| *** e0ne has quit IRC | 17:28 | |
| *** Sukhdev has quit IRC | 17:28 | |
| *** mrmartin has joined #openstack-infra | 17:28 | |
| *** ivar-lazzaro has joined #openstack-infra | 17:28 | |
| *** leopoldj has quit IRC | 17:29 | |
| *** thedodd has joined #openstack-infra | 17:30 | |
| *** markmcclain has quit IRC | 17:31 | |
| *** jraim has joined #openstack-infra | 17:33 | |
| openstackgerrit | Ramy Asselin proposed a change to openstack-infra/project-config: Add puppet-jenkins as split out module https://review.openstack.org/131302 | 17:34 |
| *** jaypipes-afk is now known as jaypipes | 17:34 | |
| *** ivar-lazzaro has quit IRC | 17:34 | |
| *** achanda has quit IRC | 17:35 | |
| zaro | pelix: i think these are also ready to go.. https://review.openstack.org/#/c/108281/ and https://review.openstack.org/#/c/108281/ | 17:35 |
| *** andreaf has quit IRC | 17:35 | |
| *** ivar-lazzaro has joined #openstack-infra | 17:35 | |
| *** skata has joined #openstack-infra | 17:35 | |
| *** andreaf has joined #openstack-infra | 17:35 | |
| *** hdd has quit IRC | 17:36 | |
| asselin | nelsnelson, not 100% famailiar, but those errors seem ok... | 17:36 |
| *** rmcall_ has joined #openstack-infra | 17:36 | |
| *** jistr has quit IRC | 17:37 | |
| pelix | zaro: same change repeated ;) | 17:37 |
| skata | Hello .. I am running into a issue and would appreciate any help in debugging it .. | 17:37 |
| *** rmcall has quit IRC | 17:37 | |
| *** rmcall_ is now known as rmcall | 17:37 | |
| nelsnelson | Yeah, that's what I was thinking, asselin. | 17:37 |
| nelsnelson | I'm still trying to figure out what is causing the safe-devstack-vm-gate-wrap.sh script to stall, but I don't have much else to go on. | 17:37 |
| skata | The gerrit trigger gives the following error when I try connecting to the review.openstack.org server "Connection error : com.jcraft.jsch.JSchException: ProxyHTTP: java.io.IOException: proxy error: Forbidden" | 17:38 |
| nelsnelson | It just sits there, and eventually seems to timeout, maybe, an hour later or so. | 17:38 |
| asselin | nelsnelson, which log file is that? | 17:39 |
| nelsnelson | Good question. There are like, three. | 17:39 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Fix ant-name usage for tests https://review.openstack.org/106892 | 17:39 |
| zaro | pelix: oops, this one https://review.openstack.org/#/c/122463/ | 17:39 |
| pleia2 | zaro: looking to submit this css change for gerrit upstream, I've reported the issue on their tracker, do they link issue #s in the code review somehow like we do? (browsing through docs and examples, not getting any answers) | 17:39 |
| *** jamielennox is now known as jamielennox|away | 17:40 | |
| asselin | nelsnelson, check the others. perhaps that's the end of the log file :) | 17:40 |
| nelsnelson | It is really hard to determine which log file is actually recording output for a given run, because devstack refuses to log in anything other than UTC, apparently. | 17:40 |
| nelsnelson | Everytime I kick off a new execution of the | 17:41 |
| nelsnelson | safe-devstack-vm-gate-wrap.sh script | 17:41 |
| nelsnelson | I can't seem to see any current output of in /opt/stack/new | 17:41 |
| nelsnelson | or /opt/stack/logs | 17:41 |
| zaro | pleia2: yes, example: https://gerrit-review.googlesource.com/#/c/57350/ | 17:41 |
| pleia2 | zaro: thanks :) | 17:41 |
| pleia2 | ooh, I also think I just found the proper docs page | 17:42 |
| *** simonmcc has joined #openstack-infra | 17:42 | |
| *** ctracey has joined #openstack-infra | 17:42 | |
| nelsnelson | plus, there are devstack-gate-setup-workspace-new.txt, and devstack-gate-setup-host.txt, and devstacklog.txt in various locations. | 17:42 |
| *** arosen has quit IRC | 17:42 | |
| asselin | nelsnelson, take a look at devstacklog.tx | 17:42 |
| *** mrmartin has quit IRC | 17:42 | |
| nelsnelson | None of which have current time stamps | 17:42 |
| *** ryanpetrello has quit IRC | 17:43 | |
| nelsnelson | /opt/stack/new/devstacklog.txt is a simlink to a file from 4 hours ago, and it isn't changing | 17:43 |
| *** fifieldt has quit IRC | 17:43 | |
| nelsnelson | despite having just kicked off another run of the safe-devstack-vm-gate-wrap.sh script | 17:43 |
| *** beekneemech has joined #openstack-infra | 17:44 | |
| *** bnemec has quit IRC | 17:44 | |
| nelsnelson | I'm honestly quite baffled as to what safe-devstack-vm-gate-wrap.sh is doing with respect to logs. | 17:44 |
| pleia2 | zaro: do you use git-review to interface with this? (sorry for all the questions, feel free to send me to #gerrit if you're busy :)) | 17:44 |
| asselin | nelsnelson, I usually run it on clean vms... | 17:44 |
| nelsnelson | Okay, I'll start it over again from scratch... | 17:44 |
| nelsnelson | brb | 17:44 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Cleaning up index.rst file https://review.openstack.org/108281 | 17:45 |
| asselin | nelsnelson, but if you've already run it before, I don't think you need to run it again. | 17:45 |
| nelsnelson | It never worked once correctly. | 17:45 |
| nelsnelson | I'm just trying to get the thing to complete once successfully. | 17:45 |
| asselin | nelsnelson, it just sets up the vm, and then kicks off the tests.... | 17:45 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Add additional documentation for passing objects to templates https://review.openstack.org/130391 | 17:45 |
| zaro | pleia2: no, they don't have the .gitreview files on their projects so you can't use it. | 17:46 |
| asselin | nelsnelson, how do you know it was not successful? | 17:46 |
| pleia2 | zaro: yeah, I noticed they didn't exist, wasn't sure if you created one | 17:46 |
| pleia2 | zaro: thanks, I'll do gerrit the old-fashioned way | 17:46 |
| zaro | pleia2: to push you do 'push origin HEAD:refs/for/master' | 17:46 |
| nelsnelson | asselin, it timed out with a bunch of weird error messages splatted all over the terminal screen after an hour or so | 17:46 |
| pleia2 | zaro: and do the change-id stuff | 17:46 |
| nelsnelson | And none of the logs had any errors. | 17:46 |
| nelsnelson | And the logs had stopped recording any new information at all about 10 minutes into the script execution. | 17:47 |
| zaro | pleia2: change it should auto generate for you. that's just gerrit | 17:47 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Add documentation and example on the usage of yaml anchors and aliases. https://review.openstack.org/130614 | 17:47 |
| asselin | nelsnelson, I'd look into those weird error messages.... | 17:47 |
| asselin | nelsnelson, you have a copy? | 17:47 |
| *** achanda has joined #openstack-infra | 17:48 | |
| nelsnelson | Had one earlier. I'll run the whole thing over fresh again. It'll be a very long time before those time out errors occur though. | 17:48 |
| pleia2 | zaro: well, docs say it's a local git hook, so I'll grab that | 17:48 |
| openstackgerrit | therese-mchale proposed a change to openstack/requirements: Add os-net-config dependency https://review.openstack.org/131841 | 17:48 |
| zaro | pleia2: it will give you the link when you attempt to push wihout the change id | 17:48 |
| *** serverascode__ has joined #openstack-infra | 17:49 | |
| pleia2 | zaro: ah, ok | 17:49 |
| nelsnelson | asselin, here's an older example from a run I did a couple days ago: https://gist.github.com/nelsnelson/f43b987a5d4e94a69277 | 17:49 |
| krotscheck | zaro: So, I’m going to approve the sysinfo patch of yours, but would you mind putting a functional test together for it as well? Api coverage => consistent client contract => happy client libraries. | 17:50 |
| krotscheck | You can do it as a separate patch. | 17:51 |
| asselin | nelsnelson, this means look at the devstack.log file: https://gist.github.com/nelsnelson/f43b987a5d4e94a69277#file-gistfile1-txt-L121 | 17:51 |
| *** ameade has joined #openstack-infra | 17:51 | |
| zaro | krotscheck: will do. | 17:51 |
| *** isviridov is now known as isviridov_away | 17:51 | |
| *** achanda has quit IRC | 17:51 | |
| krotscheck | zaro: Thanks! There’s tons of examples. | 17:52 |
| asselin | nelsnelson, you should setup ntp service for accurate timestamps: https://gist.github.com/nelsnelson/f43b987a5d4e94a69277#file-gistfile1-txt-L5 | 17:52 |
| zaro | pelix: around? | 17:52 |
| *** yamahata has quit IRC | 17:53 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Add a rest endpoint to retreive system information https://review.openstack.org/131599 | 17:53 |
| openstackgerrit | A change was merged to openstack-infra/storyboard-webclient: Allow assigning tasks without expanding the edition box https://review.openstack.org/125643 | 17:54 |
| asselin | nelsnelson, you can override the server sync used by the script by export NTP_SERVER=<your ntp server> | 17:54 |
| *** tgohad is now known as tsg | 17:54 | |
| mtreinish | jeblair: I'm going to go find some food, I'll bbiab | 17:54 |
| *** rmcall has quit IRC | 17:54 | |
| *** rmcall has joined #openstack-infra | 17:55 | |
| *** harlowja_ has quit IRC | 17:55 | |
| pleia2 | zaro: I did it! https://gerrit-review.googlesource.com/61142 | 17:56 |
| *** miqui has quit IRC | 17:56 | |
| *** harlowja has joined #openstack-infra | 17:56 | |
| *** afazekas has quit IRC | 17:57 | |
| nelsnelson | asselin, I did setup ntp and synced the server time. However, it did not adjust the log file times for devstacklog.txt. | 17:57 |
| nelsnelson | That log continued to log in UTC. | 17:57 |
| asselin | nelsnelson, yes, but at least you can sync it manually to your timezone | 17:57 |
| *** spzala has quit IRC | 17:57 | |
| krotscheck | So, for some reason a patch of mine didn’t clear the CR bugs - likely because it was a rebase that removed a bunch of files, but didn’t change any file content. Does anyone know of a ‘clean’ way of removing those? | 17:57 |
| nelsnelson | Furthermore, it did not get rid of that weird ntp.conf error at the top. | 17:57 |
| nelsnelson | I did sync the server manually to my timezone. | 17:58 |
| krotscheck | Other than a whitespace change? | 17:58 |
| *** Sincler has quit IRC | 17:58 | |
| nelsnelson | But it did not change the timestamps in the devstacklog.txt file | 17:58 |
| nelsnelson | Nor did it correct the error at the top of that output for the safe-devstack-vm-gate-wrap.sh execution. | 17:58 |
| asselin | nelsnelson, then I think you can ignore the error. If ntp is steup, then the UTC timestamps will be accurate. | 17:59 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Added application cache https://review.openstack.org/130649 | 17:59 |
| krotscheck | Ehn, n/m. I’ll just change the comment. | 17:59 |
| nelsnelson | They're accurate, yes. But very confusing when trying to pin down exactly when an error is occurring. | 17:59 |
| nelsnelson | Especially across multiple runs. | 17:59 |
| *** zhiyan has joined #openstack-infra | 18:00 | |
| nelsnelson | But apparently, once a run is borked, I need to start over from scratch from a fresh VM? | 18:00 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Plugins may now register cron workers. https://review.openstack.org/129609 | 18:00 |
| pleia2 | zaro: one last question, does the issue tracker auto-update with the patch link, or is that a manual thing? (I noticed your bug had a link) | 18:00 |
| *** emagana has quit IRC | 18:00 | |
| nelsnelson | Which of course, tacks another 5 minutes or so onto my iteration on this stuff. | 18:00 |
| *** emagana has joined #openstack-infra | 18:01 | |
| zaro | pleia2: manual, it's not very well integrated at all. | 18:01 |
| pleia2 | zaro: thanks :) | 18:01 |
| *** winston-d is now known as winston-d_zZZ | 18:02 | |
| *** todd_dsm has quit IRC | 18:02 | |
| *** tsg has quit IRC | 18:03 | |
| *** tsg has joined #openstack-infra | 18:03 | |
| *** emagana_ has joined #openstack-infra | 18:04 | |
| asselin | nelsnelson, UTC is convenient so anyone can look at the logs from their own timezone. | 18:04 |
| *** emagana has quit IRC | 18:04 | |
| nelsnelson | is there a flag I can pass to less that automatically converts those time stamps? | 18:04 |
| asselin | nelsnelson, you can manually run devstack on your vm. what is at the end of the devstack.log file? | 18:04 |
| asselin | nelsnelson, I don't know | 18:05 |
| *** armax has joined #openstack-infra | 18:05 | |
| nelsnelson | asselin, well, I rekicked the vm, so I can't tell you what used to be at the end of the devstack.log file, at this point. | 18:05 |
| mordred | nelsnelson: most of the tools we have around here, like devstack-gate, assume single-use slaves - so there's very little work to do things like handle re-running | 18:06 |
| nelsnelson | asselin, are you talking about a devstack.log from a manual run of devstack? Or are you talking about some file called "devstack.log" which contains output from the script safe-devstack-vm-gate-wrap.sh? | 18:06 |
| asselin | nelsnelson, safe-devstack-vm-gate-wrap.sh eventually calls stack.sh. | 18:07 |
| *** pcrews has quit IRC | 18:07 | |
| jeblair | mtreinish: syntax error in 127060 | 18:07 |
| nelsnelson | understood, mordred, that makes sense | 18:08 |
| *** mrmartin has joined #openstack-infra | 18:08 | |
| asselin | nelsnelson, it does so here: https://gist.github.com/nelsnelson/f43b987a5d4e94a69277#file-gistfile1-txt-L118 | 18:08 |
| SergeyLukjanov | sdague, re https://review.openstack.org/#/c/131249/2 you've not responded to comment about === vs == | 18:12 |
| SergeyLukjanov | sdague, the patch itself is lgtm | 18:12 |
| *** MaxV_ has quit IRC | 18:13 | |
| sdague | oh, from jhesketh's review. We can switch it to ===, semantically they aren't hugely different, and it's sort of nit level comment | 18:13 |
| sdague | his only -1 was over the regex | 18:13 |
| SergeyLukjanov | sdague, yeah, I understand, just to ensure that you saw the comment | 18:14 |
| sdague | yep | 18:14 |
| SergeyLukjanov | sdague, I'm performing last checks and will aprove both changes | 18:14 |
| sdague | great | 18:15 |
| *** jp_at_hp has quit IRC | 18:16 | |
| *** amitgandhinz has joined #openstack-infra | 18:16 | |
| *** andreaf has quit IRC | 18:17 | |
| *** rmcall has quit IRC | 18:19 | |
| mrmartin | sdague, hi may I get a review on this: https://review.openstack.org/#/c/131667/ ? | 18:19 |
| *** rmcall has joined #openstack-infra | 18:20 | |
| SergeyLukjanov | sdague, done | 18:20 |
| *** dprince has quit IRC | 18:20 | |
| mrmartin | and SergeyLukjanov if you have two free minutes ^^^^^ plz. | 18:20 |
| openstackgerrit | Brian Saville proposed a change to openstack-infra/jenkins-job-builder: Add support for multiple port allocations https://review.openstack.org/131845 | 18:21 |
| *** Guest99028 has quit IRC | 18:21 | |
| *** spzala has joined #openstack-infra | 18:23 | |
| *** markmcclain has joined #openstack-infra | 18:25 | |
| *** AJaeger has joined #openstack-infra | 18:26 | |
| *** todd_dsm has joined #openstack-infra | 18:28 | |
| openstackgerrit | A change was merged to openstack-infra/system-config: only show most recent results per pipeline https://review.openstack.org/131249 | 18:28 |
| *** jedimike has quit IRC | 18:29 | |
| *** tonytan4ever has quit IRC | 18:29 | |
| *** thedodd has quit IRC | 18:29 | |
| mtreinish | jeblair: sigh... One sec I'll respin it | 18:29 |
| openstackgerrit | A change was merged to openstack-infra/system-config: fix hideci for new change screen https://review.openstack.org/131500 | 18:31 |
| *** SumitNaiksatam has joined #openstack-infra | 18:31 | |
| *** bhunter71 has joined #openstack-infra | 18:31 | |
| fungi | checking in from an airport. any emergencies which need my help? | 18:32 |
| *** baoli has joined #openstack-infra | 18:32 | |
| *** baoli has quit IRC | 18:33 | |
| *** baoli has joined #openstack-infra | 18:34 | |
| *** gondoi is now known as zz_gondoi | 18:34 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuration https://review.openstack.org/127060 | 18:34 |
| fungi | looks like the hideci fixes got tested and approved. awesome | 18:34 |
| mtreinish | jeblair: ^^^ damn commas... | 18:34 |
| *** bitblt has joined #openstack-infra | 18:36 | |
| *** thedodd has joined #openstack-infra | 18:37 | |
| *** dprince has joined #openstack-infra | 18:37 | |
| mrmartin | fungi, flying directly to Paris? | 18:40 |
| AJaeger | fungi: have a good flight! | 18:40 |
| mrmartin | this will be my first Summit without a jetlag. priceless | 18:41 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: fix delete job command and add tests https://review.openstack.org/127370 | 18:41 |
| fungi | mrmartin: yeah, for a fairly absurd definition of directly which involves going hundreds of miles in the wrong direction and then sitting in another airport for a few hours | 18:41 |
| fungi | AJaeger: thanks! | 18:41 |
| fungi | mrmartin: i was lucky with the atlanta summit since it was just a hour flight south of me. won't be that lucky again for a while i expect | 18:42 |
| mrmartin | fungi, I always used to do that, we usually don't have direct flights since to national airways bankrupted (stolen.) | 18:42 |
| fungi | mrmartin: in some ways, so is ours | 18:42 |
| pelix | zaro: back, was afk for a while | 18:43 |
| *** mpaolino has quit IRC | 18:43 | |
| pelix | looking at the other change now | 18:43 |
| *** freemanbrandon has quit IRC | 18:44 | |
| openstackgerrit | Brian Saville proposed a change to openstack-infra/jenkins-job-builder: Add doxygen builder support https://review.openstack.org/131852 | 18:45 |
| lifeless | so | 18:45 |
| AJaeger | fungi: I'm lucky with Paris ;) | 18:45 |
| pleia2 | AJaeger: hey, have a few minutes to chat translations? | 18:45 |
| *** freemanbrandon has joined #openstack-infra | 18:46 | |
| *** tomoe has joined #openstack-infra | 18:47 | |
| mordred | lifeless: buttons | 18:47 |
| lifeless | mordred: see #openstack-qa | 18:47 |
| *** ildikov has joined #openstack-infra | 18:47 | |
| mordred | mrmartin: I'm in paris early, so I should be over the jetlag by summit time. fingers crossed | 18:47 |
| mrmartin | :) next summer you could do a Budapest trip :) | 18:48 |
| *** skata has quit IRC | 18:48 | |
| mrmartin | just +one and half our of flight from there | 18:48 |
| mtreinish | hmm: http://logs.openstack.org/60/127060/11/check/gate-infra-puppet-apply-precise/1f5fdf9/console.html#_2014-10-29_18_40_37_741 I wonder what I'm doing wrong | 18:50 |
| *** freemanbrandon has quit IRC | 18:51 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: New search parameters. https://review.openstack.org/127660 | 18:51 |
| *** tomoe has quit IRC | 18:51 | |
| *** achanda has joined #openstack-infra | 18:52 | |
| mordred | sdague, jeblair, clarkb, fungi, SergeyLukjanov: brainstorm time - if lifeless were about to cut a release of unittest2 upstream, which testtools depends on, which ALL of our unittests use | 18:52 |
| mordred | how would we test that before he does? | 18:52 |
| mordred | cause, you know, it seems like the type of thing that could completely break the world | 18:53 |
| lifeless | right now testools doesn't depend on it, but we co-install them always to get discover functionality from subunit.run | 18:53 |
| AJaeger | pleia2: yeah | 18:53 |
| mordred | lifeless: we do? | 18:53 |
| mordred | lifeless: what depends on it? | 18:53 |
| mtreinish | mordred: maybe you could push up a devstack-gate or devstack patch that pulls it from source | 18:53 |
| lifeless | mordred: 2.6 doesn't have discover built-in | 18:53 |
| openstackgerrit | Sahdev Zala proposed a change to openstack-infra/project-config: Rename the heat-translator project repo https://review.openstack.org/131558 | 18:53 |
| lifeless | mordred: we may be stopping as we stop testing 2.6 | 18:53 |
| mordred | mtreinish: we don't run unittests in devstack though | 18:53 |
| lifeless | tempest is unittest built | 18:54 |
| mordred | lifeless: oh - so does testtools install unittest2 only on 2.6? | 18:54 |
| mtreinish | mordred: it get's used if py26 is there | 18:54 |
| pleia2 | AJaeger: so we have a translations tools session on the schedule for Infra, I'd like to sketch out some of what we need from 1) the rest of infra 2) the rest of the translators, figure we can put it in our existing pad: https://etherpad.openstack.org/p/kilo-translations | 18:54 |
| mtreinish | so I guess on centos | 18:54 |
| lifeless | mordred: testtools doesn't yet pull it in. | 18:54 |
| lifeless | mordred: next release of testtools does. The current pull-in is done by openstack glue | 18:54 |
| mordred | then I'm going to disagree with you | 18:54 |
| lifeless | mordred: not sure where the glue is | 18:54 |
| mordred | nova does not install it | 18:54 |
| mordred | at least, it does not declare that it does | 18:54 |
| mordred | maybe oslo test? | 18:55 |
| mordred | nope | 18:55 |
| *** zaphod42 has left #openstack-infra | 18:55 | |
| *** dizquierdo has joined #openstack-infra | 18:55 | |
| mtreinish | mordred: it's used in tempest if run with py26 (which we don't support :) ) | 18:55 |
| lifeless | grep unittest2 ../openstack -r | grep requirements | 18:56 |
| lifeless | sahara, sahara-dashboard and tuskar today explicitly specify it | 18:56 |
| mordred | k. but most of the rest of openstack does not | 18:56 |
| fungi | mordred: maybe a tox.ini patch which downloads clones and preinstalls it into the venv? | 18:56 |
| lifeless | mordred: ../openstack-dev/devstack/tools/fixup_stuff.sh: install_package python-unittest2 patch | 18:56 |
| AJaeger | pleia2: I continue chatting using the etherpad | 18:56 |
| mordred | I mean, I guess what I'm saying is that VERY little of openstack will be broken if this breaks something | 18:57 |
| mordred | as opposed to everything | 18:57 |
| mordred | so that's good | 18:57 |
| *** __TheDodd__ has joined #openstack-infra | 18:57 | |
| *** achanda has quit IRC | 18:57 | |
| lifeless | I love that we're installing a file and then patching it | 18:58 |
| mtreinish | lifeless: oh they apply the patch to fix: https://code.google.com/p/unittest-ext/issues/detail?id=79 in devstack | 18:58 |
| mtreinish | heh, that's weird | 18:58 |
| pleia2 | AJaeger: thanks | 18:58 |
| *** thedodd has quit IRC | 18:58 | |
| *** teran_ has quit IRC | 18:59 | |
| *** freemanbrandon has joined #openstack-infra | 18:59 | |
| sdague | lifeless: well... rhel refused to fix it upstream, and that's what the redhat guys proposed | 18:59 |
| clarkb | is it only noon? | 19:00 |
| clarkb | I should eat something | 19:00 |
| mtreinish | sdague: the patch is from an open unittest2 bug, see the link I pasted above | 19:00 |
| *** freemanbrandon has quit IRC | 19:00 | |
| *** tgohad has joined #openstack-infra | 19:01 | |
| lifeless | sdague: yeah | 19:01 |
| *** markmcclain1 has joined #openstack-infra | 19:01 | |
| mtreinish | it's still crazy to do it in devstack like that | 19:01 |
| lifeless | hmm | 19:01 |
| lifeless | we should remove the discover package | 19:01 |
| lifeless | we install it everywhere | 19:01 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 19:02 |
| lifeless | but its obsolete/about to be obsoleted | 19:02 |
| clarkb | __afazekas: ^ I kinda want ot use that even if it doesn't fix the multi_host setup | 19:02 |
| clarkb | __afazekas: it makes each node have a mostly identical bridge setup | 19:02 |
| mordred | lifeless: then we should absolutely not install it | 19:02 |
| *** markmcclain2 has joined #openstack-infra | 19:02 | |
| clarkb | I need to regeneralize it htough so we shoulnd't merge it like that | 19:02 |
| *** sabeen1 has joined #openstack-infra | 19:02 | |
| *** sabeen2 has quit IRC | 19:03 | |
| *** markmcclain has quit IRC | 19:03 | |
| *** tsg has quit IRC | 19:04 | |
| *** rmcall has quit IRC | 19:05 | |
| *** sabeen1 has quit IRC | 19:06 | |
| *** markmcclain1 has quit IRC | 19:06 | |
| *** sabeen1 has joined #openstack-infra | 19:06 | |
| *** pcrews has joined #openstack-infra | 19:07 | |
| *** andreykurilin_ has joined #openstack-infra | 19:08 | |
| *** emagana_ has quit IRC | 19:08 | |
| *** rushiagr is now known as rushiagr_away | 19:10 | |
| *** tgohad is now known as tsg | 19:12 | |
| sdague | mtreinish: yeh, well rhel wouldn't work right without it | 19:13 |
| *** Sukhdev has joined #openstack-infra | 19:13 | |
| *** dizquierdo has quit IRC | 19:13 | |
| sdague | it's things like that which is why I am skeptical of using rhel as a first class test platform in our system | 19:13 |
| *** MaxV has joined #openstack-infra | 19:14 | |
| mtreinish | fungi: if you've got a sec, I need some puppet help. On the subunit2sql stuff i definied the package in init.pp, and when I'm referencing it in the server class the apply test is blowing up | 19:16 |
| mtreinish | I'm sure I'm just doing something stupid, but if you've got any suggestions | 19:16 |
| fungi | mtreinish: saw your link... lookingnow | 19:16 |
| *** emagana has joined #openstack-infra | 19:17 | |
| *** MaxV has quit IRC | 19:18 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 19:19 |
| fungi | mtreinish: got a link for the review that log came from? we apparently don't echo it in that job | 19:19 |
| mtreinish | fungi: https://review.openstack.org/#/c/127060/ | 19:20 |
| fungi | mtreinish: heh, yep, just found openstackgerrit mentioning it in scrollback | 19:20 |
| reed | mordred, do you have two accounts in gerrit? which one do you use most? | 19:21 |
| mordred | reed: mordred@inaugust.com - I created the hp.com one a while ago to test signup process ... and never got aroudn to killing it | 19:21 |
| reed | mordred, I guessed right :) | 19:22 |
| mordred | fungi: if you're ever deleting things from gerrit db and feel like deleting monty.taylor@hp.com I would not be opposed | 19:22 |
| krotscheck | So, I have a bug in storyboard prod which I fixed with unit tests, so the review’s a bit meaty. Anyone able to take a look? https://review.openstack.org/#/c/131564/ | 19:22 |
| krotscheck | It was one of those “Well, why don’t I write tests on how this thing SHOULD work, and oh crap I just reimplemented this didn’t I" | 19:23 |
| krotscheck | things | 19:23 |
| fungi | mtreinish: commented inline | 19:24 |
| fungi | mordred: i can mark it inactive an it should cease being available in text completion | 19:25 |
| mordred | ooh! | 19:26 |
| mtreinish | fungi: do I still need that, since I put it in the parent patch's init.pp: https://review.openstack.org/#/c/108003/21/modules/subunit2sql/manifests/init.pp | 19:26 |
| reed | I added the wrong person to a review: how can I remove him? | 19:26 |
| clarkb | reed: there should be a little x next to the name | 19:27 |
| clarkb | just click that | 19:27 |
| reed | clarkb, that's what I remembered... can't see it | 19:27 |
| reed | https://review.openstack.org/#/c/131854/ wrong sergey | 19:27 |
| fungi | mordred: i just marked 12706 (monty.taylor@hp.com) inactive in gerrit. it did not seem to have been used | 19:28 |
| *** _nadya_ has joined #openstack-infra | 19:28 | |
| mordred | fungi: nope. never was | 19:29 |
| *** jcoufal has quit IRC | 19:30 | |
| SergeyLukjanov | reed, in the new screen view there is a "X" button near the names | 19:30 |
| __afazekas | clarkb: we do not have STP | 19:30 |
| *** talluri has quit IRC | 19:30 | |
| *** __afazekas is now known as afazekas | 19:30 | |
| SergeyLukjanov | reed, I've replaced with me | 19:30 |
| reed | SergeyLukjanov, thanks, you were the Sergey I wanted to add :) | 19:30 |
| *** emagana has quit IRC | 19:30 | |
| reed | I'm emailing you all with a bit more background on that document | 19:30 |
| reed | i've added to the review all PTLs and the TC | 19:31 |
| clarkb | afazekas: do we need it? /me is looking for loops on the paper he drew on | 19:31 |
| afazekas | clarkb: I am not sure is it wise to connect each bridge with each other , it does not makes any difference with 2 node, but it does with 2+ | 19:31 |
| fungi | mtreinish: possible that it's installing those classes separately or in the wrong order | 19:31 |
| *** emagana has joined #openstack-infra | 19:31 | |
| fungi | mtreinish: i don't think package resource names are scoped... | 19:31 |
| clarkb | afazekas: I don't think we have any loops | 19:32 |
| clarkb | afazekas: we are spoked | 19:32 |
| afazekas | ok | 19:32 |
| clarkb | afazekas: I could be completely wrong though :) | 19:32 |
| mordred | reed: I LOVE THAT YOU JUST SUBMITTED LaTeX code | 19:32 |
| mordred | reed: don't kill me - there are a bunch of trailing spaces in that patch ... | 19:32 |
| * mordred hides | 19:32 | |
| reed | LOL | 19:33 |
| *** _nadya_ has quit IRC | 19:33 | |
| reed | I'm writing you all :) | 19:33 |
| reed | adding a new item to the 'known issues' :) | 19:33 |
| mordred | jeblair: can we please start producing and publishing more LaTeX? | 19:33 |
| clarkb | afazekas: I think as long as the primary node is the only place we bridge >1 gretap device then we are ok | 19:33 |
| mordred | annegentle: ^^ | 19:33 |
| *** carl_baldwin has quit IRC | 19:33 | |
| clarkb | afazekas: this means we have a spof and isn't how you would normally set up your bridges ut for testing I think it is ok | 19:33 |
| mtreinish | fungi: hmm, ok. So any suggestions on how to resolve. Just add a requires to the exec? | 19:34 |
| fungi | mtreinish: oh, i wonder if the subunit2sql::server class is getting instantiated directly without the init for the module being included | 19:34 |
| clarkb | afazekas: if however every node's gre_br had all the gretap_X devices on it then ya we would need stp | 19:34 |
| clarkb | (and it isn't required ot have all we just need some greater set that turns the spoke into some loops) | 19:34 |
| fungi | clarkb: ^ does puppet not automatically include the module's init class if you instantiate another class from it? | 19:34 |
| clarkb | fungi: if you do subunit2sql::server and not subunit2sql you will not get init.pp | 19:35 |
| fungi | mtreinish: ^ aha | 19:35 |
| *** cdent has quit IRC | 19:35 | |
| *** emagana has quit IRC | 19:35 | |
| mtreinish | fungi: ah ok, so I'll add a include subunit2sql to logstash.pp | 19:36 |
| mtreinish | thanks | 19:36 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 19:36 |
| openstackgerrit | Ricardo Carrillo Cruz proposed a change to openstack-infra/system-config: Expose jenkins_ssh_public_key parameter in openstack_project::jenkins https://review.openstack.org/131861 | 19:36 |
| clarkb | fungi: you may want to double check my assertions about loops in 131604 at some point | 19:37 |
| clarkb | fungi: but no rush (I don't think it even works yet | 19:37 |
| *** carl_baldwin has joined #openstack-infra | 19:37 | |
| fungi | mtreinish: yep, marked the same inline | 19:37 |
| *** VijayTripathi has quit IRC | 19:37 | |
| fungi | clarkb: i'll need a whiteboard (or a bigger emacs) | 19:38 |
| *** sabeen1 has quit IRC | 19:38 | |
| clarkb | I have a blank sheet of paper that I have scribbled all over. Basically primary node has a bridge that is "hub" for all of the sub node spokes | 19:38 |
| fungi | clarkb: but if you're asking about ethernet topology loops, then with only two nodes and one link between them, unlikely | 19:38 |
| clarkb | for subnodes to talk to each other they must go through the primary node "hub" | 19:39 |
| clarkb | fungi: ya right now in a 2 node config its fine | 19:39 |
| fungi | clarkb: also, linuxbridge can have stp enabled | 19:39 |
| *** esker has joined #openstack-infra | 19:39 | |
| clarkb | but in >2 we introduce at least one new link but I think as long as we only do one "upstream" link per subnode to the primary node we are good | 19:39 |
| *** lttrl has quit IRC | 19:39 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/system-config: Add subunit2sql server configuration https://review.openstack.org/127060 | 19:39 |
| fungi | no idea if nova/devstack has support to do that directly (yet) but the driver does | 19:39 |
| *** lttrl has joined #openstack-infra | 19:39 | |
| clarkb | fungi: oh well we are creating these bridges | 19:39 |
| clarkb | so we can enable stp if we need it then | 19:39 |
| fungi | ahh, yep, anyway if it's needed it's all of a couple of parameters in the cli or a line in the interface config | 19:40 |
| *** sabeen1 has joined #openstack-infra | 19:40 | |
| *** baoli has quit IRC | 19:40 | |
| afazekas | clarkb: why do you configure ip addresses for the NICs here? | 19:41 |
| fungi | and if it truly is a star/hub+spoke topology then shouldn't matter anyway (but stp is hardly ever a poor choice on switch-to-switch links) | 19:41 |
| clarkb | afazekas: I think that will fix tempest in multi_host= true | 19:41 |
| clarkb | afazekas: without IPs on a network with the floating range tempest does not know how to get to floating IPs hosted on the subnode | 19:41 |
| *** baoli has joined #openstack-infra | 19:42 | |
| *** teran has joined #openstack-infra | 19:42 | |
| clarkb | afazekas: because there is no route for 172.24.4.0/24 the tempest packets use the default route which is out eth0 | 19:42 |
| *** marcusvrn_ has quit IRC | 19:42 | |
| clarkb | if we slap those addresses on pub_if I believe it will set up the 172.24.4.0/23 route to go via pub_if | 19:42 |
| clarkb | with a source of the address I set (172.24.5.x) | 19:42 |
| afazekas | clarkb: what about adding a route rule ? | 19:42 |
| *** baoli has quit IRC | 19:43 | |
| clarkb | afazekas: we could do that too but then the source address will be something like 10.1.0x right? which would be asymetric routing | 19:43 |
| clarkb | and I hate asymetric routing | 19:43 |
| clarkb | (but I am open to refining this, it is still all very hackish) | 19:43 |
| *** baoli has joined #openstack-infra | 19:43 | |
| sdague | did something go wrong with puppet deploy of hideci? | 19:44 |
| clarkb | sdague: it requires a gerrit restart iirc | 19:44 |
| sdague | it looks like after SergeyLukjanov approved it, it's now 404 | 19:45 |
| clarkb | fungi: ^ seen that before? | 19:45 |
| SergeyLukjanov | I'm looking on it too | 19:45 |
| *** dimtruck is now known as zz_dimtruck | 19:46 | |
| *** zz_dimtruck is now known as dimtruck | 19:46 | |
| SergeyLukjanov | sdague, clarkb, due to the puppetboard hidejs was successfully updated | 19:47 |
| sdague | yeh, so if you hit the file directly, it seems fine | 19:47 |
| sdague | https://review.openstack.org/static/hideci.js?e=243560feb251e64e9b9b272eec54a151 - 404 | 19:48 |
| sdague | https://review.openstack.org/static/hideci.js there | 19:48 |
| fungi | clarkb: SergeyLukjanov: sdague: i thought jhesketh had determined that doing a touch on that file to update its timestamp got gerrit to reload it, and then he added puppet to do that on subscribe/notify or something | 19:48 |
| fungi | testing | 19:48 |
| *** beekneemech is now known as bnemec | 19:48 | |
| *** baoli has quit IRC | 19:49 | |
| SergeyLukjanov | fungi, I've already "touched" it | 19:49 |
| fungi | ahh, no, that was GerritSiteHeader.html anyway | 19:49 |
| * fungi curses his failing memory | 19:49 | |
| *** baoli has joined #openstack-infra | 19:50 | |
| fungi | perhaps it's caching it? i wonder if flush-caches affects that at all | 19:50 |
| *** markmcclain2 has quit IRC | 19:50 | |
| *** markmcclain has joined #openstack-infra | 19:51 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Enable nova multi host on aiopcpu tests https://review.openstack.org/131604 | 19:51 |
| fungi | flushed. any better now? | 19:51 |
| SergeyLukjanov | fungi, nope atm | 19:51 |
| *** esker has quit IRC | 19:52 | |
| *** arborism has joined #openstack-infra | 19:52 | |
| *** arborism is now known as amcrn | 19:52 | |
| fungi | i wonder how we managed to get the previous changes to take effect | 19:53 |
| lifeless | -> doctor and stuff | 19:53 |
| *** pcrews has quit IRC | 19:53 | |
| fungi | welp, i have to stop troubleshooting for a bit... boarding a plane now | 19:54 |
| afazekas | clarkb: can we give 172.24.4.1/24 to the tempest node ? | 19:54 |
| *** esker has joined #openstack-infra | 19:54 | |
| *** esker has quit IRC | 19:54 | |
| fungi | back soon if i have in-flight wifi, otherwise not until i get to the next airport in a few hours | 19:54 |
| clarkb | afazekas: maybe? do we have to change the floating range if we do that? | 19:54 |
| *** esker has joined #openstack-infra | 19:54 | |
| openstackgerrit | Brian Saville proposed a change to openstack-infra/jenkins-job-builder: Add doxygen builder support https://review.openstack.org/131330 | 19:55 |
| afazekas | clarkb: if it is not reserved by default, yes | 19:55 |
| *** esker has quit IRC | 19:56 | |
| openstackgerrit | Brian Saville proposed a change to openstack-infra/jenkins-job-builder: Add support for multiple port allocations https://review.openstack.org/131335 | 19:56 |
| clarkb | afazekas: also I tried just adding 172.24.4.1 and 172.24.5.1 to pub_if as is in your change and it would not respond to who has arp requests | 19:57 |
| *** esker has joined #openstack-infra | 19:57 | |
| *** esker has quit IRC | 19:57 | |
| *** pelix has quit IRC | 19:57 | |
| clarkb | afazekas: I think because pub_if sits between the two bridge (pub_if_br and br100) | 19:57 |
| clarkb | which is what first started my changes on the bridge stuff I needed a pub_if that wasn't in a bridge directly | 19:58 |
| clarkb | afazekas: in any case I think we should move forward with your change and multi_host = false | 19:58 |
| clarkb | afazekas: then refine and figure out if multi_host = true is doable in a not insane way | 19:58 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: DB Migration config read https://review.openstack.org/131863 | 19:59 |
| afazekas | There several systl options about how aggressive is the arp response (by default it usually response very frequently) | 19:59 |
| *** tsg has quit IRC | 19:59 | |
| *** tsg has joined #openstack-infra | 19:59 | |
| *** achanda has joined #openstack-infra | 19:59 | |
| afazekas | http://kb.linuxvirtualserver.org/wiki/Using_arp_announce/arp_ignore_to_disable_ARP | 20:00 |
| mtreinish | fungi: yep, adding the include made it pass the apply job | 20:00 |
| clarkb | afazekas: the issue was it didn't respond at all | 20:01 |
| clarkb | I had tcpdumps on every single interface and if you take the pub_if on the primary node after it does your change and add 172.24.4.1/24 to it and have the subnode ping that it will never arp | 20:01 |
| afazekas | iptables ? | 20:01 |
| *** rlandy has joined #openstack-infra | 20:02 | |
| clarkb | I set everything to be wide open (at least I though, basically INPUT, OUTPUT and FORWARD rules that allow 172.24.4.1/23 to anywhere and anywhere to 172.24.4.1/23) | 20:02 |
| clarkb | but its possible I missed something | 20:02 |
| afazekas | we also have ebtables, but I would be surprised if it would cause the issue.. | 20:04 |
| *** Ryan_Lane1 has joined #openstack-infra | 20:04 | |
| *** _nadya_ has joined #openstack-infra | 20:05 | |
| *** kgiusti has joined #openstack-infra | 20:06 | |
| *** yfried_ has joined #openstack-infra | 20:06 | |
| *** yfried_ is now known as yfried|afk | 20:06 | |
| clarkb | afazekas: part of the motivation for doing it this way for multihost is all the docs I can see say put fixed ips on an "internal" network and floating ips on your normal interface. So this gives us two interfaces that match docs (though both networks are bridged teogher which may or may not be a problem) | 20:07 |
| *** Ryan_Lane has quit IRC | 20:07 | |
| clarkb | I just want to see if I can get it to work at all :) | 20:07 |
| clarkb | SergeyLukjanov: anything I can do to help with the js stuff? | 20:08 |
| *** andreykurilin_ has quit IRC | 20:08 | |
| *** aysyd has quit IRC | 20:11 | |
| openstackgerrit | Ramy Asselin proposed a change to openstack-infra/project-config: Add puppet-jenkins as split out module https://review.openstack.org/131302 | 20:12 |
| openstackgerrit | afazekas proposed a change to openstack-infra/nodepool: Add private ip to /etc/nodepool https://review.openstack.org/114840 | 20:12 |
| *** baoli has quit IRC | 20:12 | |
| *** markmcclain has quit IRC | 20:13 | |
| *** MaxV has joined #openstack-infra | 20:15 | |
| *** yfried has joined #openstack-infra | 20:18 | |
| *** yfried|afk has quit IRC | 20:18 | |
| *** MaxV has quit IRC | 20:19 | |
| *** dprince has quit IRC | 20:19 | |
| *** carl_baldwin has quit IRC | 20:21 | |
| *** esker has joined #openstack-infra | 20:21 | |
| *** lttrl has quit IRC | 20:22 | |
| *** yfried_ has joined #openstack-infra | 20:23 | |
| *** yfried_ is now known as yfried|afk | 20:23 | |
| *** spzala has quit IRC | 20:23 | |
| *** pcrews has joined #openstack-infra | 20:24 | |
| *** yfried has quit IRC | 20:24 | |
| *** ihrachyshka has joined #openstack-infra | 20:25 | |
| *** yfried|afk has quit IRC | 20:25 | |
| *** mrmartin has quit IRC | 20:26 | |
| clarkb | woot my multi_host change is down to only two failing tests | 20:27 |
| clarkb | both related to volumes (snapshot test and boot from volume test) | 20:27 |
| clarkb | I am going to guess that cinder needs to be taught to get on the 10.1.0.0/24 network | 20:28 |
| clarkb | but this is progress | 20:28 |
| clarkb | afazekas: ^ | 20:28 |
| *** esker has quit IRC | 20:28 | |
| mtreinish | clarkb: awesome | 20:28 |
| *** garyh_ has joined #openstack-infra | 20:29 | |
| clarkb | jogo: ^ | 20:29 |
| *** adam_g_gone is now known as adam_g | 20:30 | |
| openstackgerrit | Sahdev Zala proposed a change to openstack-infra/project-config: Rename the heat-translator project repo https://review.openstack.org/131558 | 20:31 |
| *** garyh has quit IRC | 20:31 | |
| *** garyh_ is now known as garyh | 20:31 | |
| openstackgerrit | Sahdev Zala proposed a change to openstack-infra/project-config: Rename the heat-translator project repo https://review.openstack.org/131558 | 20:33 |
| *** david-lyle has joined #openstack-infra | 20:34 | |
| *** atiwari has quit IRC | 20:35 | |
| *** afazekas_ has joined #openstack-infra | 20:36 | |
| *** carl_baldwin has joined #openstack-infra | 20:36 | |
| SergeyLukjanov | clarkb, after updating hideci.js in gerrit (changes approved) it's 404 https://review.openstack.org/static/hideci.js?e=243560feb251e64e9b9b272eec54a151 | 20:36 |
| *** _nadya_ has quit IRC | 20:37 | |
| jgriffith | clarkb: that's great! | 20:38 |
| jgriffith | clarkb: check that provider_location debug statement I put in and make sure you can ping it from compute node | 20:39 |
| jgriffith | should be EZ/PZ after that :) | 20:39 |
| jogo | clarkb: nice, very exciting | 20:39 |
| *** david-lyle has quit IRC | 20:40 | |
| jgriffith | clarkb: suspect it's the iscsi_ip_address setting | 20:40 |
| *** koolhead17 has quit IRC | 20:42 | |
| *** david-lyle has joined #openstack-infra | 20:42 | |
| *** david-lyle_ has joined #openstack-infra | 20:43 | |
| clarkb | jgriffith: thank you | 20:43 |
| *** dkliban is now known as dkliban_biab | 20:44 | |
| jaypipes | clarkb: did we revert the change to the Gerrit UI that lists the CI systems separately? | 20:44 |
| clarkb | we broke it somehow | 20:44 |
| *** Shrews has quit IRC | 20:44 | |
| clarkb | sergey is/was debugging | 20:44 |
| jaypipes | clarkb: ok, no worries. was just curious. | 20:44 |
| *** Ryan_Lane1 has quit IRC | 20:45 | |
| SergeyLukjanov | clarkb, the hideci.js isn't asseccible with hash (https://review.openstack.org/static/hideci.js?e=243560feb251e64e9b9b272eec54a151 | 20:45 |
| SergeyLukjanov | clarkb, but by direct url it's ok | 20:45 |
| SergeyLukjanov | clarkb, any ideas? | 20:45 |
| clarkb | ya I think we need to restart gerrit | 20:45 |
| SergeyLukjanov | clarkb, yeah, sounds like this | 20:45 |
| *** Shrews has joined #openstack-infra | 20:46 | |
| *** Shrews has quit IRC | 20:46 | |
| *** Ryan_Lane has joined #openstack-infra | 20:47 | |
| *** david-lyle_ has quit IRC | 20:47 | |
| *** tomoe has joined #openstack-infra | 20:48 | |
| clarkb | I am being dragged to lunch. I can restart it later | 20:49 |
| clarkb | or you can do it | 20:49 |
| clarkb | just stop then start the gerrit service | 20:50 |
| *** achanda has quit IRC | 20:50 | |
| SergeyLukjanov | clarkb, do we need to send an announce to the openstack-dev for it? | 20:51 |
| *** MaxV has joined #openstack-infra | 20:51 | |
| jeblair | SergeyLukjanov: nope; you could "#status notice restarting gerrit" or something if you want | 20:52 |
| SergeyLukjanov | jeblair, okay | 20:52 |
| *** tomoe has quit IRC | 20:53 | |
| SergeyLukjanov | like #status notice Restarting gerrit to get fixed CI javascript | 20:53 |
| jeblair | SergeyLukjanov: yep | 20:54 |
| *** Shrews has joined #openstack-infra | 20:54 | |
| jeblair | i assume that was a question :) (start the line with #status to actually execute) | 20:54 |
| *** hdd has joined #openstack-infra | 20:55 | |
| *** emagana has joined #openstack-infra | 20:55 | |
| SergeyLukjanov | jeblair, yup, it was a question, I'm not so ill yet :) | 20:55 |
| SergeyLukjanov | #status notice Restarting gerrit to get fixed CI javascript | 20:56 |
| openstackstatus | SergeyLukjanov: sending notice | 20:56 |
| *** MaxV has quit IRC | 20:56 | |
| -openstackstatus- NOTICE: Restarting gerrit to get fixed CI javascript | 20:57 | |
| *** ZZelle has joined #openstack-infra | 20:57 | |
| openstackstatus | SergeyLukjanov: finished sending notice | 20:58 |
| *** hdd has quit IRC | 20:59 | |
| *** patrickeast has quit IRC | 20:59 | |
| *** Shrews has quit IRC | 20:59 | |
| *** patrickeast has joined #openstack-infra | 20:59 | |
| *** Shrews has joined #openstack-infra | 20:59 | |
| SergeyLukjanov | done | 21:00 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/storyboard: setup for executing and debugging in eclipse https://review.openstack.org/131870 | 21:00 |
| *** tgohad has joined #openstack-infra | 21:00 | |
| SergeyLukjanov | jeblair, fungi, clarkb, sdague, I've restarted gerrit and it looks like everything was fixed and working | 21:00 |
| *** emagana has quit IRC | 21:00 | |
| sdague | SergeyLukjanov: yay | 21:00 |
| *** emagana has joined #openstack-infra | 21:01 | |
| fungi | back now. glad to see it's resolved. i guess we should hold further hideci patches for maintenance windows | 21:01 |
| SergeyLukjanov | sdague, I like the new "(N rechecks)" info | 21:01 |
| sdague | yeh, good to know that can't be a live update | 21:01 |
| SergeyLukjanov | fungi, yeah, I've been thinking that it could be applied w/o restarting gerrit | 21:02 |
| *** tsg has quit IRC | 21:02 | |
| jeblair | i thought someone worked out how the hash codes work and how to do a live update... | 21:02 |
| SergeyLukjanov | fungi, or at least apply changes in batches on weekdays for example | 21:02 |
| SergeyLukjanov | jeblair, the file "touch" has been added for some other file | 21:02 |
| jeblair | like if you touch the header file, it recalculates the hash | 21:03 |
| SergeyLukjanov | jeblair, I've tried it for hideci.js and it doesn't help | 21:03 |
| jeblair | SergeyLukjanov: yeah, that's what i was thinking of. did that get implemented but just not work this time? | 21:03 |
| SergeyLukjanov | jeblair, it was implemented only for header | 21:03 |
| fungi | jeblair: it was GerritSiteHeader.html we saw it work on | 21:03 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 21:03 | |
| *** patrickeast_ has joined #openstack-infra | 21:03 | |
| *** Ryan_Lane has quit IRC | 21:03 | |
| SergeyLukjanov | jeblair, I've tried to do it manually for hideci.js but no result | 21:03 |
| *** patrickeast has quit IRC | 21:03 | |
| *** patrickeast_ is now known as patrickeast | 21:03 | |
| jeblair | SergeyLukjanov: did you touch GerritSiteHeader.html today? | 21:04 |
| fungi | oh, wait, touching GerritSiteHeader.html is what causes hideci.js to get updated? | 21:04 |
| SergeyLukjanov | jeblair, nope, I've touched hideci.js | 21:05 |
| jeblair | fungi: that's what i thought i read that someone wrote | 21:05 |
| SergeyLukjanov | fungi, the same q. :) | 21:05 |
| *** Sukhdev has quit IRC | 21:05 | |
| fungi | i guess i had it in my head that was only for changes to the GerritSiteHeader.html file | 21:05 |
| SergeyLukjanov | fungi, +1 | 21:05 |
| *** mfink_ has quit IRC | 21:05 | |
| *** emagana has quit IRC | 21:05 | |
| *** emagana_ has joined #openstack-infra | 21:05 | |
| jeblair | fungi: i have it in my head that the person who explored that said that it caused gerrit to recalculate the hash for all the deps | 21:05 |
| fungi | but now that you mention it, we did that for a js update which didn't take | 21:05 |
| *** mfink has joined #openstack-infra | 21:05 | |
| fungi | so yeah, that rings a bell | 21:06 |
| *** annegent_ has joined #openstack-infra | 21:06 | |
| *** mattfarina has quit IRC | 21:06 | |
| *** amcrn has quit IRC | 21:06 | |
| *** hdd has joined #openstack-infra | 21:06 | |
| asselin | I have this error on my nodepool image-builds. anyone seen it before? | 21:07 |
| asselin | No distributions at all found for py>=1.4.17 (from tox | 21:07 |
| *** _nadya_ has joined #openstack-infra | 21:07 | |
| *** todd_dsm has quit IRC | 21:07 | |
| *** Shrews has quit IRC | 21:07 | |
| ianw | clarkb / jeblair: what do we have to do to deploy https://review.openstack.org/#/c/131636/ ... hpcloud centos7 builds are still failing | 21:08 |
| fungi | asselin: is it trying to install py>=1.4.17 from pypi.python.org or some mirror? | 21:08 |
| asselin | fungi, I'm using a mirror | 21:08 |
| *** Shrews has joined #openstack-infra | 21:08 | |
| fungi | asselin: is it present on your mirror? | 21:08 |
| *** sarob_ has quit IRC | 21:08 | |
| *** _nadya_ has quit IRC | 21:09 | |
| *** _nadya_ has joined #openstack-infra | 21:09 | |
| asselin | fungi, no, it's not! :/ | 21:10 |
| *** mfink has quit IRC | 21:10 | |
| fungi | sounds like you found the problem | 21:10 |
| asselin | not sure why. image builds were working this morning... | 21:11 |
| *** carl_baldwin has quit IRC | 21:12 | |
| fungi | asselin: it's possible some project started depending on that (possibly transitively) since then | 21:12 |
| *** alexpilotti has quit IRC | 21:13 | |
| fungi | ianw: it merged before today's image updates would have fired, but perhaps something else is wrong... seeing if i can tell over this laggy connection | 21:13 |
| *** mbacchi has quit IRC | 21:13 | |
| asselin | fungi, any reason why when you set a mirror, pip doesn't try the non-mirror if the mirror doesn't have it? | 21:14 |
| *** alexpilotti has joined #openstack-infra | 21:14 | |
| fungi | asselin: depends on whether you configure it as your only index or an extra index | 21:14 |
| ianw | fungi: thanks, it seems like the script hasn't deployed? the centos7 builds are just looping around over and over | 21:14 |
| *** todd_dsm has joined #openstack-infra | 21:14 | |
| *** Ryan_Lane has joined #openstack-infra | 21:14 | |
| *** dimtruck is now known as zz_dimtruck | 21:15 | |
| ianw | fungi: i'm trying to see where we grab the latest scripts, it doesn't seem to be in nodepool puppet... | 21:15 |
| asselin | fungi, which one should I use? I have pypi as an extra index now | 21:15 |
| *** weshay has quit IRC | 21:15 | |
| *** alexpilotti has quit IRC | 21:16 | |
| *** JayJ has joined #openstack-infra | 21:17 | |
| fungi | asselin: oh, also possible you ran into a network connection hitting pypi.python.org after not finding the package on your mirror. that can also cause the error | 21:17 |
| *** bswartz has quit IRC | 21:17 | |
| *** dkranz has quit IRC | 21:17 | |
| ianw | fungi: ahh, i guess cloned by modules/project_config/manifests/init.pp | 21:17 |
| afazekas | https://github.com/openstack-infra/project-config/blob/master/nodepool/scripts/configure_mirror.sh I wonder would't be better to specify the the full mirror url in env-var, instead of expected all pypi mirror is under openstack.org | 21:18 |
| *** rlandy has quit IRC | 21:18 | |
| *** zz_dimtruck is now known as dimtruck | 21:18 | |
| fungi | ianw: we puppet-manage /etc/nodepool/scripts on nodepool.o.o but it hasn't updated with that change. checking it out | 21:18 |
| fungi | ianw: http://puppetboard.openstack.org/ | 21:19 |
| fungi | ianw: nodepool.o.o hasn't run a puppet update for more than a day | 21:19 |
| mtreinish | jeblair, fungi, clarkb: I think the subunit2sql patches are ready now, fixed them up w/ clarkb's suggestions and it's passing everything now | 21:19 |
| *** marun has joined #openstack-infra | 21:19 | |
| *** ihrachyshka has quit IRC | 21:19 | |
| fungi | ianw: clarkb: "Skipping run of Puppet configuration client; administratively disabled" | 21:19 |
| fungi | clarkb: are you okay with me reenabling puppet on nodepool.o.o now? | 21:20 |
| *** stevemar has quit IRC | 21:20 | |
| *** VijayTripathi has joined #openstack-infra | 21:20 | |
| ianw | fungi: thanks ... that explains a lot :) | 21:21 |
| *** annegent_ has quit IRC | 21:21 | |
| *** ihrachyshka has joined #openstack-infra | 21:22 | |
| *** _nadya_ has quit IRC | 21:23 | |
| *** achanda has joined #openstack-infra | 21:24 | |
| *** mase_x200 has joined #openstack-infra | 21:25 | |
| *** patrickeast has quit IRC | 21:25 | |
| *** todd_dsm has quit IRC | 21:26 | |
| *** todd_dsm has joined #openstack-infra | 21:29 | |
| *** dustins has quit IRC | 21:30 | |
| *** tgohad has quit IRC | 21:31 | |
| *** hdd has quit IRC | 21:31 | |
| *** patrickeast has joined #openstack-infra | 21:32 | |
| openstackgerrit | afazekas proposed a change to openstack-infra/project-config: Allow to use custom mirror https://review.openstack.org/131878 | 21:33 |
| *** patrickeast has quit IRC | 21:34 | |
| *** patrickeast has joined #openstack-infra | 21:35 | |
| clarkb | fungi yup go for it | 21:36 |
| fungi | clarkb: thanks! reenabled now. ianw: keep an eye on puppetboard but i expect it to update there in about 10 minutes | 21:37 |
| *** mjturek has quit IRC | 21:38 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/storyboard: add test for system_info.py https://review.openstack.org/131880 | 21:38 |
| *** amitgandhinz has quit IRC | 21:41 | |
| ianw | fungi: great, thanks. i'm afk for about an hour helping out in kindergarten reading groups, i'll check it out. probably one of the only things that makes nodepool image creation & deployment seem sane in comparison :) | 21:41 |
| fungi | ha! | 21:43 |
| *** kgiusti has quit IRC | 21:46 | |
| *** gyee has quit IRC | 21:49 | |
| *** openstackgerrit has quit IRC | 21:50 | |
| *** MaxV has joined #openstack-infra | 21:52 | |
| ZZelle | clarkb, fungi, hi | 21:53 |
| *** lttrl has joined #openstack-infra | 21:55 | |
| *** MaxV has quit IRC | 21:56 | |
| fungi | landing now... bbl | 21:58 |
| *** AJaeger has quit IRC | 22:00 | |
| *** baoli has joined #openstack-infra | 22:00 | |
| *** carl_baldwin has joined #openstack-infra | 22:00 | |
| *** emagana_ has quit IRC | 22:02 | |
| *** emagana has joined #openstack-infra | 22:03 | |
| *** achanda_ has joined #openstack-infra | 22:04 | |
| *** emagana_ has joined #openstack-infra | 22:06 | |
| *** baoli has quit IRC | 22:06 | |
| *** emagana has quit IRC | 22:07 | |
| *** eharney has quit IRC | 22:08 | |
| *** achanda has quit IRC | 22:08 | |
| *** carl_baldwin has quit IRC | 22:10 | |
| *** mfink has joined #openstack-infra | 22:11 | |
| *** doug-fish1 has left #openstack-infra | 22:15 | |
| *** todd_dsm has quit IRC | 22:17 | |
| *** andreykurilin_ has joined #openstack-infra | 22:22 | |
| *** xyang0 has quit IRC | 22:22 | |
| *** IaKing has joined #openstack-infra | 22:22 | |
| *** dmsimard is now known as dmsimard_away | 22:23 | |
| *** _buzz_ has quit IRC | 22:25 | |
| *** ociuhandu has quit IRC | 22:25 | |
| *** bknudson has quit IRC | 22:26 | |
| *** todd_dsm has joined #openstack-infra | 22:27 | |
| *** dims_ has joined #openstack-infra | 22:28 | |
| *** ociuhandu has joined #openstack-infra | 22:28 | |
| *** dims_ has quit IRC | 22:29 | |
| jogo | clarkb: so talked to PhilD about https://review.openstack.org/#/c/113754/ | 22:29 |
| jogo | clarkb: and its not on HP public cloud yet | 22:29 |
| *** dims_ has joined #openstack-infra | 22:30 | |
| jogo | clarkb: would getting that landed have a big impact? if so there may be a way | 22:30 |
| *** dims__ has quit IRC | 22:31 | |
| *** kgiusti has joined #openstack-infra | 22:31 | |
| *** david-lyle has quit IRC | 22:32 | |
| *** bswartz has joined #openstack-infra | 22:33 | |
| *** ChuckC has quit IRC | 22:33 | |
| *** achanda_ has quit IRC | 22:34 | |
| *** __TheDodd__ has quit IRC | 22:37 | |
| *** ChuckC has joined #openstack-infra | 22:45 | |
| *** baoli has joined #openstack-infra | 22:45 | |
| *** todd_dsm has quit IRC | 22:46 | |
| *** kgiusti has quit IRC | 22:48 | |
| *** tomoe has joined #openstack-infra | 22:48 | |
| *** david-lyle has joined #openstack-infra | 22:49 | |
| *** mase_x200 has quit IRC | 22:51 | |
| *** hdd has joined #openstack-infra | 22:52 | |
| *** MaxV has joined #openstack-infra | 22:53 | |
| *** tomoe has quit IRC | 22:53 | |
| *** gyee has joined #openstack-infra | 22:54 | |
| clarkb | jogo: the impact would be we would stop leaking floating ips | 22:55 |
| clarkb | arosen's nodepool patch seemed to slow it down a bit but we definitely do leak still | 22:55 |
| *** mriedem has quit IRC | 22:55 | |
| *** hemna is now known as hemna__ | 22:57 | |
| *** msabramo has quit IRC | 22:57 | |
| *** achanda has joined #openstack-infra | 22:58 | |
| *** MaxV has quit IRC | 22:58 | |
| jogo | clarkb: hmm ok that should be enough info | 22:58 |
| *** VijayTripathi has quit IRC | 22:58 | |
| clarkb | jgriffith: "Set provider_location to: 10.0.3.8:3260,1" that is on eth0 but flat_interface is flat_if and public_interface is pub_if for nova | 22:58 |
| clarkb | jgriffith: should provider_location be on the same network as nova's flat_interface? | 22:59 |
| *** ociuhandu_ has joined #openstack-infra | 22:59 | |
| clarkb | hrm it actually looks like things failed due to ssh timeouts again | 23:00 |
| *** ociuhandu has quit IRC | 23:00 | |
| *** ociuhandu_ is now known as ociuhandu | 23:00 | |
| clarkb | so maybe I haven't quite fixed tempest's ability to ssh to the nodes /me tries spinnig up a node outside of tempest | 23:00 |
| *** armax has quit IRC | 23:04 | |
| jogo | top gate failure is http://status.openstack.org/elastic-recheck/index.html#1338742 | 23:04 |
| jogo | git.o.o fail | 23:05 |
| *** aysyd has joined #openstack-infra | 23:05 | |
| *** armax has joined #openstack-infra | 23:05 | |
| *** dims_ has quit IRC | 23:05 | |
| *** dims__ has joined #openstack-infra | 23:06 | |
| *** msabramo has joined #openstack-infra | 23:08 | |
| *** lttrl has quit IRC | 23:09 | |
| *** hdd has quit IRC | 23:12 | |
| *** baoli has quit IRC | 23:12 | |
| *** baoli has joined #openstack-infra | 23:13 | |
| *** andreykurilin_ has quit IRC | 23:18 | |
| *** baoli has quit IRC | 23:21 | |
| *** bhuvan has quit IRC | 23:22 | |
| *** ZZelle has quit IRC | 23:24 | |
| *** VijayTripathi has joined #openstack-infra | 23:25 | |
| *** JayJ has quit IRC | 23:31 | |
| *** wenlock has quit IRC | 23:31 | |
| *** david-lyle has quit IRC | 23:32 | |
| ianw | jogo: yeah, my CI has been pretty unstable getting through to git.o.o as well | 23:36 |
| *** signed8bit is now known as signed8bit_ZZZzz | 23:36 | |
| *** msabramo has quit IRC | 23:37 | |
| *** msabramo has joined #openstack-infra | 23:37 | |
| clarkb | its nova's fault >_> | 23:37 |
| clarkb | but I think we need to maybe add more nodes. it is probably worth looking at updating to newer operating system fornewer git too | 23:38 |
| clarkb | but we are running into memory pressure doing git ops on the mirror nodes | 23:38 |
| mordred | clarkb: oh - we can go to centos7 now | 23:38 |
| clarkb | and nova appeas to be the bad repo | 23:38 |
| mordred | bad. bad repo | 23:40 |
| * mordred wags finger at nova repo | 23:40 | |
| *** dimtruck is now known as zz_dimtruck | 23:41 | |
| ianw | mordred: i think moving to centos7 will be quite the exercise in puppet-wrangling | 23:42 |
| *** annegentle has quit IRC | 23:42 | |
| *** annegentle has joined #openstack-infra | 23:42 | |
| * ianw got into postgresql and gave up, let alone mysql v maria etc | 23:43 | |
| *** yjiang5 is now known as yjiang5_away | 23:43 | |
| clarkb | ianw: we only need selinux and apache to work | 23:43 |
| clarkb | everything else is basic ifle manipulation | 23:43 |
| mordred | so - honestly - I know this is a hackneyed meme - but I just realized - if git works via afs AT ALL | 23:44 |
| mordred | then it means we don't need the git server process on the git farm to scale | 23:44 |
| mordred | because each node would be running basic file manipulations | 23:44 |
| clarkb | mordred: except that file listings don't work | 23:44 |
| mordred | file listings work fine | 23:44 |
| mordred | the pypi mirror file listing happened to be a directory with close to 64k listings | 23:45 |
| *** ihrachyshka has quit IRC | 23:45 | |
| mordred | entries | 23:45 |
| clarkb | git is good at that too :) | 23:45 |
| mordred | sure - but git is not good at scaling that because it needs a server process | 23:45 |
| clarkb | no I mean git repos with thousnands of dir ents is not uncommon | 23:46 |
| clarkb | we currently pack things constantly to avoid it as best we can | 23:46 |
| mordred | so the question is - which is more efficient - git over AFS with each git process doing the file thigns but over an AFS network location, or a git farm with a bunch of server proceses | 23:46 |
| *** Shrews has quit IRC | 23:46 | |
| *** annegentle has quit IRC | 23:46 | |
| *** annegentle has joined #openstack-infra | 23:47 | |
| *** Shrews has joined #openstack-infra | 23:47 | |
| clarkb | mordred: considering that you can't really get away with not having http(s) served git repos it may be more of a question of do we need both things | 23:47 |
| mordred | yes | 23:47 |
| clarkb | also this nova networking crap is really annoying | 23:47 |
| mordred | I totally agre | 23:47 |
| mordred | with both things :) | 23:47 |
| clarkb | I still cannot hit floating ip of VM on subnode from the primary node | 23:47 |
| clarkb | I do now have working arp though | 23:48 |
| clarkb | so that is progress | 23:48 |
| *** annegentle has quit IRC | 23:48 | |
| *** annegentle has joined #openstack-infra | 23:48 | |
| *** annegentle has quit IRC | 23:49 | |
| *** annegentle has joined #openstack-infra | 23:49 | |
| *** msabramo has quit IRC | 23:50 | |
| clarkb | but my icmp echo requests are dying between gretap_1 and gre_br | 23:50 |
| clarkb | tcpdump shows them just disappear | 23:51 |
| *** msabramo has joined #openstack-infra | 23:51 | |
| clarkb | guessing iptables is ruining something somewhere | 23:51 |
| *** kgiusti has joined #openstack-infra | 23:51 | |
| *** dangers is now known as dangers_away | 23:53 | |
| clarkb | I blame nat | 23:53 |
| *** MaxV has joined #openstack-infra | 23:54 | |
| *** MaxV has quit IRC | 23:55 | |
| *** MaxV has joined #openstack-infra | 23:55 | |
| *** bknudson has joined #openstack-infra | 23:56 | |
| *** asselin has quit IRC | 23:58 | |
| ianw | clarkb: i spent about a week once chasing down intermittent connection issues, through gre tunnels, down into ovs flows, tcpdump running in god knows how many different name-spaces etc. turned out to be two neutron-agents running at the same time. both consumed incoming messages, and *occasionally* they would interleave and totally screw up the flow routing | 23:58 |
| ianw | i'm actually more surprised when packets manage to get out, rather than expecting anything to work :) | 23:59 |
| *** msabramo has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!