| *** katell1 has quit IRC | 00:02 | |
| *** milner has quit IRC | 00:06 | |
| devananda | vishy: in api.py, after session.begin() and then making some change, some methods explicitly call Model.save() before returning, while others do not. Any thoughts on that? | 00:07 |
|---|---|---|
| *** spiffxp has joined #openstack-dev | 00:07 | |
| *** katell1 has joined #openstack-dev | 00:09 | |
| *** b1rkh0ff has quit IRC | 00:11 | |
| *** ewindisch has quit IRC | 00:11 | |
| *** ewindisch has joined #openstack-dev | 00:12 | |
| vishy | devananda: i think we have autocommit on | 00:12 |
| vishy | devananda: yes we create the session with autocommit = True so i think the save should only be necessary if there is an intermediate step | 00:13 |
| devananda | vishy: i think so, too, but I think it's disabled when one explicitly calls session.begin(), until the next explicit commit() or save() call. | 00:13 |
| vishy | devananda: actually I'm not sure | 00:14 |
| vishy | devananda: i think they all either call save or update | 00:14 |
| vishy | devananda: can you find one that doesn't call save that should? | 00:14 |
| devananda | vishy: ah, perhaps update() and delete() are interally calling commit? /me goes to look for that | 00:15 |
| vishy | devananda: delete does | 00:15 |
| *** blamar has quit IRC | 00:15 | |
| vishy | devananda: update does not, but all of the xxxxx_update calls do | 00:15 |
| devananda | vishy: compute_node_create() | 00:17 |
| devananda | has update but no save | 00:17 |
| *** jemartin has quit IRC | 00:17 | |
| *** troytoman is now known as troytoman-away | 00:18 | |
| *** anderstj has joined #openstack-dev | 00:19 | |
| *** milner has joined #openstack-dev | 00:20 | |
| vishy | devananda: that one is doing some funky stuff with a subtransaction | 00:21 |
| vishy | not sure what that is about | 00:21 |
| *** jakedahn is now known as jakedahn_zz | 00:21 | |
| *** rods has quit IRC | 00:22 | |
| devananda | vishy: hm, thanks. | 00:27 |
| *** torgomatic has quit IRC | 00:28 | |
| *** anderstj has quit IRC | 00:36 | |
| *** johnpostlethwait has quit IRC | 00:39 | |
| *** edygarcia has joined #openstack-dev | 00:40 | |
| *** s0mik has quit IRC | 00:43 | |
| *** matwood has quit IRC | 00:44 | |
| *** mnewby_ has joined #openstack-dev | 00:52 | |
| *** mnewby has quit IRC | 00:55 | |
| *** mnewby_ has quit IRC | 00:56 | |
| *** jdurgin has quit IRC | 00:56 | |
| *** aa has quit IRC | 00:56 | |
| *** salgado has quit IRC | 00:59 | |
| *** longhudou has joined #openstack-dev | 01:04 | |
| *** edygarcia has quit IRC | 01:12 | |
| *** rkukura has quit IRC | 01:13 | |
| *** anderstj has joined #openstack-dev | 01:20 | |
| *** e1mer has quit IRC | 01:24 | |
| *** rbasak has quit IRC | 01:27 | |
| *** Gordonz has joined #openstack-dev | 01:28 | |
| *** ayoung has quit IRC | 01:28 | |
| *** rkukura has joined #openstack-dev | 01:29 | |
| *** rbasak has joined #openstack-dev | 01:29 | |
| *** e1mer has joined #openstack-dev | 01:31 | |
| *** e1mer has joined #openstack-dev | 01:31 | |
| *** Gordonz has quit IRC | 01:34 | |
| *** _adjohn has joined #openstack-dev | 01:37 | |
| *** ywu has joined #openstack-dev | 01:37 | |
| *** maoy has joined #openstack-dev | 01:37 | |
| *** PotHix has quit IRC | 01:39 | |
| *** adjohn has quit IRC | 01:39 | |
| *** _adjohn is now known as adjohn | 01:39 | |
| *** jakedahn_zz is now known as jakedahn | 01:40 | |
| *** harlowja has quit IRC | 01:45 | |
| *** jakedahn is now known as jakedahn_zz | 01:47 | |
| *** adjohn has quit IRC | 01:48 | |
| *** adjohn has joined #openstack-dev | 01:49 | |
| *** anderstj has quit IRC | 02:03 | |
| *** jog0 has quit IRC | 02:04 | |
| *** Ryan_Lane has joined #openstack-dev | 02:14 | |
| *** Ryan_Lane1 has quit IRC | 02:16 | |
| *** bencherian has quit IRC | 02:18 | |
| *** bencherian has joined #openstack-dev | 02:19 | |
| *** bencherian has quit IRC | 02:23 | |
| *** ywu has quit IRC | 02:39 | |
| *** maoy has quit IRC | 02:39 | |
| *** pixelbeat has quit IRC | 02:42 | |
| *** novas0x2a|lapto1 has quit IRC | 02:48 | |
| *** Gordonz has joined #openstack-dev | 02:50 | |
| *** anderstj has joined #openstack-dev | 02:58 | |
| *** anderstj has quit IRC | 03:00 | |
| *** spiffxp has quit IRC | 03:11 | |
| *** garyk has quit IRC | 03:20 | |
| *** maplebed has quit IRC | 03:20 | |
| *** garyk has joined #openstack-dev | 03:22 | |
| *** e1mer has quit IRC | 03:31 | |
| *** aa has joined #openstack-dev | 03:38 | |
| *** edygarcia has joined #openstack-dev | 03:39 | |
| *** e1mer has joined #openstack-dev | 03:41 | |
| *** Gordonz has quit IRC | 03:42 | |
| *** roge has quit IRC | 03:45 | |
| *** ohnoimdead has quit IRC | 03:52 | |
| *** Mandell has joined #openstack-dev | 03:54 | |
| *** rkukura has quit IRC | 03:56 | |
| *** spiffxp has joined #openstack-dev | 03:59 | |
| *** aa has quit IRC | 04:02 | |
| *** vincentricci has joined #openstack-dev | 04:06 | |
| *** vincentricci has left #openstack-dev | 04:07 | |
| *** bencherian has joined #openstack-dev | 04:07 | |
| *** vincentricci has joined #openstack-dev | 04:08 | |
| *** vincentricci has left #openstack-dev | 04:08 | |
| *** vincentricci has joined #openstack-dev | 04:09 | |
| *** vincentricci has left #openstack-dev | 04:11 | |
| *** vincentricci has joined #openstack-dev | 04:11 | |
| *** rkukura has joined #openstack-dev | 04:15 | |
| *** spiffxp has quit IRC | 04:15 | |
| *** rkukura has quit IRC | 04:19 | |
| *** Gordonz has joined #openstack-dev | 04:20 | |
| *** garyk has quit IRC | 04:31 | |
| *** matwood has joined #openstack-dev | 04:34 | |
| *** matwood has quit IRC | 04:35 | |
| *** matwood has joined #openstack-dev | 04:35 | |
| *** vincentricci has left #openstack-dev | 04:35 | |
| *** edygarcia has quit IRC | 04:44 | |
| *** Gordonz has quit IRC | 04:50 | |
| *** dolphm has joined #openstack-dev | 04:52 | |
| *** Mandell has quit IRC | 04:54 | |
| *** ewindisch has quit IRC | 04:55 | |
| *** hattwick has quit IRC | 05:01 | |
| *** littleidea has quit IRC | 05:04 | |
| *** anderstj has joined #openstack-dev | 05:18 | |
| *** e1mer has quit IRC | 05:19 | |
| *** matwood has quit IRC | 05:23 | |
| *** e1mer has joined #openstack-dev | 05:24 | |
| *** jakedahn_zz is now known as jakedahn | 05:26 | |
| *** zaitcev has quit IRC | 05:26 | |
| *** dayou has joined #openstack-dev | 05:27 | |
| *** rkukura has joined #openstack-dev | 05:43 | |
| *** anderstj has quit IRC | 05:47 | |
| *** bencherian has quit IRC | 05:48 | |
| *** dtroyer is now known as dtroyer_zzz | 05:56 | |
| *** rkukura has quit IRC | 06:00 | |
| *** sdague_ has joined #openstack-dev | 06:04 | |
| *** sdague has quit IRC | 06:06 | |
| *** GheAway has quit IRC | 06:06 | |
| *** GheAway has joined #openstack-dev | 06:07 | |
| *** rkukura has joined #openstack-dev | 06:07 | |
| *** dolphm has quit IRC | 06:10 | |
| *** rkukura has quit IRC | 06:11 | |
| *** pmezard has joined #openstack-dev | 06:18 | |
| *** vincentricci_ has joined #openstack-dev | 06:20 | |
| *** vincentricci_ has left #openstack-dev | 06:20 | |
| *** vincentricci_ has joined #openstack-dev | 06:21 | |
| vishy | mtaylor, jeblair, LinuxJedi: this seems odd, did the version of pep8 change recently to start showing new errors? https://jenkins.openstack.org/job/gate-nova-pep8/4052/console | 06:22 |
| LinuxJedi | vishy: yes, that is the new error introduced in 1.1 | 06:23 |
| vishy | LinuxJedi: so do we need to push through a merge fixing all off those so the gating works again? | 06:24 |
| LinuxJedi | vishy: http://pypi.python.org/pypi/pep8#changelog for more details | 06:24 |
| LinuxJedi | vishy: either that or set the version number of pep8 to 1.0.1 in tox.ini | 06:25 |
| vishy | LinuxJedi: hmm, how come other patches have been merging? | 06:25 |
| LinuxJedi | vishy: honestly, I don't know... | 06:27 |
| LinuxJedi | vishy: ah, this is stable/essex you are merging against | 06:28 |
| LinuxJedi | vishy: trunk was fixed for pep8 | 06:28 |
| vishy | ah that makes sense | 06:28 |
| * LinuxJedi guesses that wasn't backported | 06:28 | |
| LinuxJedi | vishy: for stable/essex it may be better just to fix tox.ini | 06:28 |
| LinuxJedi | vishy: to do this under "[testenv:pep8]" change "deps = pep8==1.0.1" | 06:31 |
| *** mancdaz has quit IRC | 06:31 | |
| * LinuxJedi getting kid ready for school, back later | 06:32 | |
| vishy | LinuxJedi: Ok, I'm heading to bed now, but I will give it a run in the morning if no one else tackles it first. | 06:35 |
| *** garyk has joined #openstack-dev | 06:36 | |
| *** mindpixel has joined #openstack-dev | 06:43 | |
| *** pmezard has quit IRC | 06:44 | |
| *** vincentricci_ has quit IRC | 06:47 | |
| *** cloudfly has quit IRC | 07:04 | |
| *** pmezard has joined #openstack-dev | 07:08 | |
| *** reidrac has joined #openstack-dev | 07:18 | |
| *** davidha has quit IRC | 07:26 | |
| *** dhellmann_ has joined #openstack-dev | 07:36 | |
| *** dhellmann has quit IRC | 07:36 | |
| *** dhellmann_ is now known as dhellmann | 07:36 | |
| *** e1mer has quit IRC | 07:40 | |
| *** mancdaz has joined #openstack-dev | 07:44 | |
| *** Ryan_Lane has quit IRC | 07:46 | |
| *** e1mer has joined #openstack-dev | 07:53 | |
| *** EmilienM has joined #openstack-dev | 07:58 | |
| *** EmilienM has left #openstack-dev | 07:59 | |
| *** hattwick has joined #openstack-dev | 07:59 | |
| *** davidha has joined #openstack-dev | 08:08 | |
| *** GheAway is now known as GheRivero | 08:11 | |
| *** derekh has joined #openstack-dev | 08:14 | |
| *** GheRivero has quit IRC | 08:17 | |
| *** GheRivero has joined #openstack-dev | 08:29 | |
| *** pixelbeat has joined #openstack-dev | 08:32 | |
| *** sandywalsh has quit IRC | 08:34 | |
| *** berendt has joined #openstack-dev | 08:38 | |
| *** berendt has quit IRC | 08:52 | |
| *** berendt has joined #openstack-dev | 08:52 | |
| *** Fu4ny has joined #openstack-dev | 09:03 | |
| *** Ryan_Lane has joined #openstack-dev | 09:06 | |
| *** Drakiz_ has quit IRC | 09:07 | |
| *** Drakiz has joined #openstack-dev | 09:08 | |
| *** dachary has joined #openstack-dev | 09:21 | |
| *** andrewsben is now known as andrewsben_zz | 09:26 | |
| *** troytoman-away is now known as troytoman | 09:29 | |
| *** hugokuo has quit IRC | 09:29 | |
| *** apevec has joined #openstack-dev | 09:36 | |
| *** dachary has quit IRC | 09:38 | |
| *** Fu4ny has quit IRC | 09:38 | |
| *** dachary has joined #openstack-dev | 09:39 | |
| *** b1rkh0ff has joined #openstack-dev | 09:41 | |
| *** jakedahn is now known as jakedahn_zz | 09:41 | |
| *** e1mer has quit IRC | 09:44 | |
| *** dachary has quit IRC | 09:47 | |
| *** davidha has quit IRC | 10:01 | |
| *** rods has joined #openstack-dev | 10:26 | |
| *** e1mer has joined #openstack-dev | 10:42 | |
| *** littleidea has joined #openstack-dev | 10:43 | |
| *** mjfork has joined #openstack-dev | 10:44 | |
| *** e1mer has quit IRC | 10:54 | |
| *** e1mer has joined #openstack-dev | 10:56 | |
| *** longhudou has quit IRC | 10:57 | |
| *** davidha has joined #openstack-dev | 11:00 | |
| *** davidha_who_took has joined #openstack-dev | 11:29 | |
| *** davidha has quit IRC | 11:29 | |
| *** flaviamissi has quit IRC | 11:30 | |
| *** markvoelker has joined #openstack-dev | 11:30 | |
| *** wiliam has joined #openstack-dev | 11:40 | |
| *** salgado has joined #openstack-dev | 12:00 | |
| *** alaski has joined #openstack-dev | 12:07 | |
| *** mrunge has quit IRC | 12:10 | |
| *** sandywalsh has joined #openstack-dev | 12:15 | |
| *** dprince has joined #openstack-dev | 12:18 | |
| *** lts has joined #openstack-dev | 12:24 | |
| *** alaski has quit IRC | 12:33 | |
| *** alaski has joined #openstack-dev | 12:34 | |
| *** nunosantos has joined #openstack-dev | 12:40 | |
| *** dachary has joined #openstack-dev | 12:41 | |
| *** dachary has quit IRC | 12:49 | |
| *** dolphm has joined #openstack-dev | 12:51 | |
| davidha_who_took | Hi - if function tests test the functionality, what does probtests test? | 12:53 |
| *** davidha_who_took is now known as davidha | 12:53 | |
| davidha | (In swift) | 12:54 |
| *** gael has joined #openstack-dev | 12:55 | |
| *** markmc has joined #openstack-dev | 12:59 | |
| *** GheRivero is now known as GheAway | 12:59 | |
| *** roge has joined #openstack-dev | 13:01 | |
| *** flaviamissi has joined #openstack-dev | 13:01 | |
| *** dachary has joined #openstack-dev | 13:02 | |
| *** mdomsch has quit IRC | 13:02 | |
| *** sandywalsh has quit IRC | 13:07 | |
| *** sdague_ is now known as sdague | 13:08 | |
| *** edygarcia has joined #openstack-dev | 13:11 | |
| *** rkukura has joined #openstack-dev | 13:14 | |
| *** tongli|2 has joined #openstack-dev | 13:16 | |
| *** Shrews has quit IRC | 13:20 | |
| *** maplebed has joined #openstack-dev | 13:21 | |
| garyk | ping mtaylor | 13:21 |
| garyk | jeblair: ping | 13:21 |
| *** ayoung has joined #openstack-dev | 13:24 | |
| *** Shrews has joined #openstack-dev | 13:26 | |
| *** berendt has quit IRC | 13:28 | |
| *** matwood has joined #openstack-dev | 13:29 | |
| *** edygarcia has quit IRC | 13:29 | |
| *** rkukura has quit IRC | 13:33 | |
| *** rkukura has joined #openstack-dev | 13:33 | |
| *** spiffxp has joined #openstack-dev | 13:33 | |
| *** thingee has quit IRC | 13:41 | |
| *** thingee has joined #openstack-dev | 13:41 | |
| *** thingee has quit IRC | 13:46 | |
| *** GheRivero has joined #openstack-dev | 13:46 | |
| *** e1mer has quit IRC | 13:49 | |
| davidha | chmouel | 13:51 |
| *** sstent has quit IRC | 13:53 | |
| mtaylor | morning garyk | 13:53 |
| *** sstent has joined #openstack-dev | 13:54 | |
| *** Ruetobas has quit IRC | 13:56 | |
| GheRivero | morning people | 13:57 |
| *** AlanClark has joined #openstack-dev | 13:58 | |
| *** Ruetobas has joined #openstack-dev | 14:02 | |
| *** edygarcia has joined #openstack-dev | 14:02 | |
| *** jaypipes is now known as jaypipes-afk | 14:03 | |
| garyk | mtaylor: hi, afternoon our side of the world :) | 14:03 |
| mtaylor | garyk: bah. nowhere exists outside of my immediate vicinity :) | 14:04 |
| garyk | mtaylor: :) maybe you can help - i was doing some cherry-picking for quantum into essex stable - jenkins did not like what i did - pep8 issues/ | 14:04 |
| mtaylor | garyk: uhoh! | 14:05 |
| mtaylor | garyk: there's an ongoing saga of issues related to pep8 updating upstream and our pep8 builders not honoring what's in test-requires (larger fix coming, but not there yet) | 14:05 |
| garyk | mtaylor - for example https://review.openstack.org/#/c/7967/ | 14:05 |
| garyk | mtaylor - one option would be to port all of the folsom code - but then this could be risky | 14:06 |
| mtaylor | garyk: nah - first of all, just try this: | 14:06 |
| mtaylor | garyk: under "[testenv:pep8]" change "deps = pep8==1.0.1" | 14:06 |
| mtaylor | pep8 upgraded upstream since stable/essex was cut | 14:07 |
| garyk | mtaylor: ok - silly question - in which file? | 14:07 |
| mtaylor | garyk: oh, heh | 14:09 |
| mtaylor | garyk: not silly question - tox.ini | 14:09 |
| garyk | mtaylor: thanks | 14:09 |
| *** nikhil_ has joined #openstack-dev | 14:11 | |
| *** nikhil_ is now known as Guest96786 | 14:11 | |
| *** troytoman is now known as troytoman-away | 14:14 | |
| *** corXi has joined #openstack-dev | 14:15 | |
| *** dayou has quit IRC | 14:16 | |
| garyk | mtaylor: 1 step forward :). jenkins still fails - ERROR: Failure: ImportError (No module named common) [https://review.openstack.org/#/c/7966/] | 14:17 |
| mtaylor | garyk: does that fail for you locally? like, if you run "tox -v -epy26" do you see the problem? and does quantum/common exist? | 14:18 |
| garyk | mtaylor: i'll check | 14:19 |
| garyk | mtaylor: quantum common does not exist :( | 14:21 |
| *** dtroyer_zzz is now known as dtroyer | 14:21 | |
| *** adalbas has joined #openstack-dev | 14:21 | |
| *** jgriffith-away is now known as jgriffith | 14:22 | |
| mtaylor | garyk: well then there's your problem :) | 14:23 |
| apevec | garyk, seems it was removed in https://github.com/openstack/quantum/commit/13222958a5c2cf861312bd6198cb0bb6e937aa6d | 14:23 |
| *** Guest96786 has quit IRC | 14:24 | |
| garyk | apevec: thanks | 14:24 |
| *** nikhil__ has joined #openstack-dev | 14:24 | |
| apevec | garyk, and then re-added on master post-essex! f6bc5d0e33616dc2ded9bacb3656694ebbab7d58 | 14:25 |
| davidkranz | jaypipes: One of the new volume tests does 'from unittest.case import SkipTest' which seems to not work in Python 2.6. Do we care? The other tests use nose.SkipTest for this. | 14:25 |
| davidkranz | jaypipes: I mean do we want to support 2.6 in tempest? | 14:26 |
| apevec | garyk, I guess that makes quantum backports more interesting | 14:26 |
| garyk | apevec: never a dull moment :) | 14:27 |
| garyk | apevec: you were right a few hours ago :) | 14:30 |
| *** utlemming has quit IRC | 14:33 | |
| *** utlemming has joined #openstack-dev | 14:33 | |
| *** dachary has quit IRC | 14:34 | |
| *** utlemming has quit IRC | 14:34 | |
| *** utlemming has joined #openstack-dev | 14:34 | |
| *** zaitcev has joined #openstack-dev | 14:37 | |
| *** stuntmachine has joined #openstack-dev | 14:38 | |
| sdague | what's the right way to specify a blueprint in a gerrit review so it gets linked? | 14:39 |
| *** dtroyer is now known as dtroyer_zzz | 14:44 | |
| *** troytoman-away is now known as troytoman | 14:45 | |
| garyk | mtaylor, apevec: for the nova i also have a cherry pick - do i also need to make the pep8 change? i feel less comfortable doing it with the nova code. what do you guys suggest? | 14:46 |
| mtaylor | garyk: the pep8 change needs to go in to nova stable/essex as well | 14:47 |
| mtaylor | garyk: vishy was going to look at doing it later on, but I'm sure it's fine if you beat him to it | 14:47 |
| garyk | mtaylor: thanks | 14:48 |
| *** rnirmal has joined #openstack-dev | 14:48 | |
| *** zaitcev_ has joined #openstack-dev | 14:49 | |
| *** pasm has joined #openstack-dev | 14:49 | |
| *** darkhorse2go has joined #openstack-dev | 14:49 | |
| *** zaitcev has quit IRC | 14:50 | |
| *** darkhorse has quit IRC | 14:52 | |
| *** maplebed has quit IRC | 14:53 | |
| belliott | morning - looking for another review on https://review.openstack.org/#/c/7763/ | 14:53 |
| garyk | mtaylor: one for the road - originally i had 4 changes. the first passes. i need to push the additional three - how do i get mr jenkind to check these (no need to make code changes) | 14:55 |
| *** blamar has joined #openstack-dev | 14:55 | |
| *** dtroyer_zzz is now known as dtroyer | 14:56 | |
| mtaylor | garyk: sorry - I don't understand the question? | 14:57 |
| mtaylor | garyk: if you push new changes to gerrit, jenkins will check them for you automatically | 14:57 |
| *** datsun180b has joined #openstack-dev | 14:59 | |
| *** JStoker has quit IRC | 14:59 | |
| *** dubsquared has joined #openstack-dev | 14:59 | |
| garyk | mtaylor: sorry for not being clear. prior to fixing the pep issue i pushed 4 mutually exclusive fixes. all were rejected by jenkins due to the pep problems. the pep is resolved. how can i push the 3 additional ones (no source code needs to be changed) | 15:00 |
| garyk | mtaylor: sorry for the odd question - i am not 100% familiar with git/gerrit and jenkins yet | 15:01 |
| *** ayoung has quit IRC | 15:02 | |
| mtaylor | garyk: oh, you need to rebase them on top of the first | 15:03 |
| mtaylor | garyk: because now the other three kind of do depend on the first | 15:04 |
| mtaylor | garyk: do you know how to do that? | 15:04 |
| garyk | mtaylor: not 100% sure, but I'll read up. Thanks! | 15:05 |
| *** ayoung has joined #openstack-dev | 15:05 | |
| mtaylor | garyk: basically - first, find the md5 of the pep8 fix | 15:05 |
| mtaylor | garyk: then, for each of the other three changes, grab them into a branch, run "git rebase -i $pep8_fix_md5" then run git review | 15:06 |
| garyk | mtaylor: great - thanks | 15:06 |
| *** galstrom has joined #openstack-dev | 15:07 | |
| *** shang has quit IRC | 15:10 | |
| *** shang has joined #openstack-dev | 15:10 | |
| *** Gordonz has joined #openstack-dev | 15:14 | |
| *** vincentricci has joined #openstack-dev | 15:14 | |
| *** JStoker has joined #openstack-dev | 15:15 | |
| *** Gordonz has quit IRC | 15:16 | |
| *** issackel_ has joined #openstack-dev | 15:16 | |
| *** jkoelker has quit IRC | 15:16 | |
| *** issackel_ has quit IRC | 15:16 | |
| *** jkoelker has joined #openstack-dev | 15:18 | |
| *** mindpixel has quit IRC | 15:19 | |
| *** mattray has joined #openstack-dev | 15:20 | |
| *** Gordonz has joined #openstack-dev | 15:21 | |
| *** vinilios has joined #openstack-dev | 15:21 | |
| *** timello has quit IRC | 15:27 | |
| *** reidrac has quit IRC | 15:28 | |
| *** pixelbeat has quit IRC | 15:28 | |
| *** darkhorse2go is now known as darkhorse | 15:31 | |
| *** derekh has quit IRC | 15:32 | |
| *** davidha has quit IRC | 15:34 | |
| *** issackelly has joined #openstack-dev | 15:37 | |
| *** maoy has joined #openstack-dev | 15:40 | |
| *** pixelbeat has joined #openstack-dev | 15:42 | |
| *** PotHix has joined #openstack-dev | 15:45 | |
| *** Gordonz has quit IRC | 15:45 | |
| *** Gordonz has joined #openstack-dev | 15:45 | |
| *** derekh has joined #openstack-dev | 15:46 | |
| *** e1mer has joined #openstack-dev | 15:46 | |
| *** jemartin has joined #openstack-dev | 15:47 | |
| *** mjfork has quit IRC | 15:49 | |
| *** thingee has joined #openstack-dev | 15:50 | |
| *** jgriffith has quit IRC | 15:51 | |
| *** thingee has quit IRC | 15:51 | |
| *** timello has joined #openstack-dev | 15:51 | |
| *** jgriffith has joined #openstack-dev | 15:51 | |
| *** thingee has joined #openstack-dev | 15:52 | |
| *** mancdaz has quit IRC | 15:53 | |
| *** dubsquared has quit IRC | 15:54 | |
| *** dachary has joined #openstack-dev | 15:54 | |
| *** garyk has quit IRC | 15:55 | |
| *** corXi has quit IRC | 15:55 | |
| *** mdomsch has joined #openstack-dev | 15:55 | |
| *** mjfork has joined #openstack-dev | 15:56 | |
| *** timello has quit IRC | 15:58 | |
| *** Ryan_Lane has quit IRC | 15:59 | |
| *** timello has joined #openstack-dev | 15:59 | |
| *** semyazz has joined #openstack-dev | 16:01 | |
| *** semyazz has left #openstack-dev | 16:01 | |
| *** wting has quit IRC | 16:02 | |
| *** maplebed has joined #openstack-dev | 16:02 | |
| *** edygarcia has quit IRC | 16:02 | |
| *** edygarcia has joined #openstack-dev | 16:03 | |
| *** dubsquared has joined #openstack-dev | 16:04 | |
| *** jdurgin has joined #openstack-dev | 16:05 | |
| belliott | heyas, looking for one more glance core https://review.openstack.org/#/c/7763/ | 16:05 |
| *** lloydde has joined #openstack-dev | 16:05 | |
| *** zaitcev_ has quit IRC | 16:07 | |
| *** dubsquared1 has joined #openstack-dev | 16:07 | |
| *** dubsquared has quit IRC | 16:09 | |
| *** vinilios has quit IRC | 16:10 | |
| *** dubsquared1 has quit IRC | 16:11 | |
| *** Mandell has joined #openstack-dev | 16:12 | |
| *** jemartin has left #openstack-dev | 16:12 | |
| *** dwalleck has joined #openstack-dev | 16:15 | |
| *** maplebed is now known as maplebed|AFK | 16:15 | |
| *** Mandell has quit IRC | 16:16 | |
| *** adalbas has quit IRC | 16:16 | |
| zykes- | vishy: ping! | 16:17 |
| *** adalbas has joined #openstack-dev | 16:18 | |
| *** n0ano has joined #openstack-dev | 16:20 | |
| *** s0mik has joined #openstack-dev | 16:21 | |
| *** e1mer has quit IRC | 16:21 | |
| *** Mandell has joined #openstack-dev | 16:22 | |
| *** davidha has joined #openstack-dev | 16:22 | |
| zykes- | vishy: is https://bugs.launchpad.net/nova/+bug/997763 fixed ? | 16:23 |
| uvirtbot | Launchpad bug 997763 in nova "floating ips are not disassociated from instances on deletion" [High,Confirmed] | 16:23 |
| *** ewindisch has joined #openstack-dev | 16:30 | |
| *** gael has quit IRC | 16:30 | |
| *** mattray has quit IRC | 16:32 | |
| *** darraghb has joined #openstack-dev | 16:32 | |
| *** Mandell has quit IRC | 16:33 | |
| *** sleepsonzzz is now known as sleepsonthefloor | 16:34 | |
| *** wting has joined #openstack-dev | 16:37 | |
| *** ohnoimdead has joined #openstack-dev | 16:44 | |
| *** adjohn has joined #openstack-dev | 16:46 | |
| *** ohnoimdead has quit IRC | 16:46 | |
| vishy | zykes-: nope | 16:49 |
| *** timello has quit IRC | 16:54 | |
| *** timello has joined #openstack-dev | 16:54 | |
| *** salgado is now known as salgado-lunch | 16:57 | |
| *** hub_cap has joined #openstack-dev | 16:59 | |
| *** maoy has quit IRC | 16:59 | |
| zykes- | vishy: hard to do ? | 17:00 |
| *** GheRivero has quit IRC | 17:01 | |
| vishy | zykes-: a little tricky, but not too bad | 17:01 |
| *** edygarcia has quit IRC | 17:02 | |
| *** edygarcia has joined #openstack-dev | 17:02 | |
| *** rohitk has joined #openstack-dev | 17:03 | |
| *** andrewsben_zz is now known as andrewsben | 17:06 | |
| *** maoy has joined #openstack-dev | 17:07 | |
| *** rohit-k has joined #openstack-dev | 17:07 | |
| *** rohitk has quit IRC | 17:10 | |
| *** derekh has quit IRC | 17:14 | |
| *** darraghb has quit IRC | 17:15 | |
| *** aa has joined #openstack-dev | 17:17 | |
| *** garyk has joined #openstack-dev | 17:19 | |
| vishy | russellb: ping | 17:19 |
| *** rkukura has quit IRC | 17:20 | |
| russellb | vishy: pong | 17:20 |
| *** harlowja has joined #openstack-dev | 17:21 | |
| *** dachary has quit IRC | 17:21 | |
| vishy | russellb: so I was just doing some updates to volume api to remove a couple of extra methods | 17:22 |
| *** davidha has quit IRC | 17:22 | |
| vishy | russelb: and I noticed that all of the live migration calls are using instance_id instead of uuid | 17:22 |
| russellb | yeah, i noticed that too. | 17:22 |
| russellb | and there's one method that uses instance name :( | 17:22 |
| vishy | russellb: just wanted to get your thoughts on whether we should do that before or after versioning... | 17:22 |
| russellb | i think it's fine either way | 17:23 |
| vishy | russellb: ultimately they will all use instance instead, but should we get them consistent first? | 17:23 |
| russellb | if you do it after, it should be made to support both id and uuid | 17:23 |
| *** davidha has joined #openstack-dev | 17:23 | |
| *** pasm has quit IRC | 17:23 | |
| russellb | so it's backwards compat | 17:23 |
| vishy | russellb: I'm thinking before or else we will have in folsom 3 versions id -> uuid -> instance_object | 17:23 |
| vishy | unless we just go straight from id -> object | 17:24 |
| russellb | or you can have it all in the same API version, but make each of those parameters optional params, and have the code adapt to which one it receives | 17:24 |
| *** apevec has quit IRC | 17:24 | |
| vishy | maybe it isn't worth it to convert to now, we can just do it when we change everything to use an object | 17:25 |
| zykes- | vishy: something you could get to or ? | 17:25 |
| russellb | but yeah, if it can be done before, it helps prevent the messyness | 17:25 |
| zykes- | or point me to | 17:25 |
| zykes- | vishy: it works in latest devstack someone says but not in essex (I can verify that) | 17:26 |
| russellb | vishy: but i was hoping to get to converting everything to sending a full instance soonish | 17:26 |
| russellb | so that'll make it all moot .. | 17:27 |
| *** GheRivero has joined #openstack-dev | 17:27 | |
| russellb | well not moot, just not sure how urgent it would be to change those to uuid in between | 17:27 |
| vishy | russellb: yeah if that is happening soon, i think we can just leave it for now | 17:27 |
| russellb | ok | 17:27 |
| russellb | well "soon", certainly in folsom, hopefully folsom-2 | 17:27 |
| vishy | russellb: I'm making a conflicting change with one of your patches | 17:27 |
| *** Ryan_Lane has joined #openstack-dev | 17:28 | |
| vishy | russellb: deleting the remove_volume_connection call from volume.api | 17:28 |
| russellb | vishy: k, i was a little worried about that one since i knew that code is in heavy flux | 17:28 |
| russellb | ok, that's fine, i can just remove that patch if your patch beats mine in :) | 17:28 |
| vishy | it is silly for compute to go through volume api to make a call back to another compute | 17:28 |
| russellb | heh | 17:29 |
| vishy | russellb: well it still needs to change to use rpc | 17:29 |
| vishy | but i removed the method and put the call back in the calling side | 17:29 |
| zykes- | vishy: it's kind of bad atm, cause if in horizon you create a instance, delete it and spawn another one I think the same floating ip comes assigned. | 17:29 |
| vishy | so it is a race :) | 17:29 |
| russellb | ah got it | 17:29 |
| zykes- | s/comes/becomes | 17:29 |
| russellb | vishy: i lose races all the time, nature of the beast | 17:29 |
| russellb | i felt stupid when i realized i missed all of those compute rpc calls ... | 17:30 |
| russellb | oh well, at least i caught it and not someone else | 17:30 |
| *** mdomsch has quit IRC | 17:31 | |
| *** Ryan_Lane has quit IRC | 17:34 | |
| * russellb steps out for lunch | 17:34 | |
| *** pasm has joined #openstack-dev | 17:35 | |
| *** GheRivero has quit IRC | 17:39 | |
| zykes- | vishy: what is it that runs the deallocation of ips ? | 17:39 |
| *** mnewby has joined #openstack-dev | 17:40 | |
| vishy | zykes-: it is a call to network_api | 17:40 |
| zykes- | vishy: is it the manager ? | 17:41 |
| *** mattray has joined #openstack-dev | 17:41 | |
| vishy | jgriffith: https://review.openstack.org/7986 | 17:44 |
| vishy | russellb: ^^ that is the conflicting change | 17:44 |
| vishy | zykes-: it is a network_api call that needs to be called during terminate | 17:45 |
| vishy | zykes-: it isn't right now | 17:45 |
| zykes- | vishy: why doesn't it ? I thought this was kind of basic stuff | 17:45 |
| vishy | zykes-: it got removed with one of the network decoupling refactors | 17:45 |
| comstud | hrm | 17:46 |
| zykes- | that's kinda unsmart ? | 17:46 |
| comstud | anyone running into issues with trunk w/ the RPC changes? | 17:46 |
| comstud | exceptions doing deepcopy() of context? | 17:46 |
| comstud | russellb: ^ | 17:46 |
| vishy | zykes-: there is a call called _cleanup_volumes | 17:47 |
| vishy | zykes-: there needs to be another called _cleanup_floating_ips | 17:47 |
| vishy | or some such | 17:47 |
| vishy | zykes-: maybe I will do that now | 17:47 |
| vishy | tr3buchet: ping | 17:48 |
| *** torgomatic has joined #openstack-dev | 17:50 | |
| zykes- | vishy: if you could it would be magic | 17:50 |
| comstud | blah, trunk rpc tests fail for me too | 17:51 |
| comstud | :( | 17:51 |
| comstud | timeout on the cast test | 17:51 |
| tr3buchet | vishy: what's up? | 17:51 |
| *** mdomsch has joined #openstack-dev | 17:51 | |
| russellb | comstud: :-/ | 17:51 |
| comstud | russellb: | 17:52 |
| comstud | (sec) | 17:52 |
| comstud | russellb: http://paste.openstack.org/show/18278/ | 17:53 |
| vishy | tr3buchet: I just noticed that allocate_for_instance and deallocate_for_instance is using instance['id'] | 17:53 |
| russellb | comstud: fun | 17:53 |
| russellb | and by "fun" I mean "oh crap, what did I do" | 17:54 |
| comstud | russellb: I'm not sure, but it's that copy.deepcopy(self) in the rpc common context | 17:54 |
| * russellb nods | 17:54 | |
| tr3buchet | vishy: yeah looks like it | 17:54 |
| comstud | russellb: I have a patch to remove it.. but I suspect finding the real problem may be in a PITA... | 17:54 |
| comstud | russellb: s/remove/replace/ | 17:54 |
| tr3buchet | vishy: want me to switch to uuid only? | 17:55 |
| russellb | comstud: replace with what? | 17:55 |
| comstud | russellb: but of course, tests are failing in trunk for me too... so I need to figure that out first | 17:55 |
| vishy | tr3buchet: I think we need to switch everything to uuid right? | 17:55 |
| tr3buchet | vishy: alright, on it. | 17:55 |
| vishy | tr3buchet: I mean quantum won't be tracking the id... | 17:55 |
| tr3buchet | correct | 17:55 |
| comstud | russellb: http://paste.openstack.org/show/18277/ | 17:55 |
| zykes- | quantum what ? | 17:56 |
| vishy | tr3buchet: i guess we are passing the whole object into network.api, so it isn't too bad | 17:56 |
| tr3buchet | vishy: ok i'll get a patch together. yeah it's not a rough patch | 17:56 |
| vishy | tr3buchet: the migration of the db will be rough | 17:56 |
| tr3buchet | may nee d afew db calls | 17:56 |
| vishy | tr3buchet: so maybe hold off for the moment | 17:56 |
| comstud | russellb: kinda ugly.. but it seems that deepcopying the class instance itself is a problem | 17:56 |
| comstud | russellb: and just deepcopying the context values works fine | 17:56 |
| tr3buchet | vishy: either way i'll look into what it takes | 17:56 |
| vishy | zykes-: as far as i can tell floating ips should be deallocated on terminate | 17:56 |
| vishy | zykes-: are you sure they are still not? | 17:57 |
| comstud | russellb: I have a feeling there's a problem deepcopying 'conf' | 17:57 |
| russellb | comstud: well, works for me ... | 17:57 |
| comstud | russellb: But I can't verify it | 17:57 |
| comstud | what python version? | 17:57 |
| russellb | i meant, your patch works for me | 17:57 |
| *** mattray has quit IRC | 17:57 | |
| comstud | ahh | 17:57 |
| comstud | got it | 17:57 |
| russellb | but, tests/code pass for me, Python 2.7.3 | 17:58 |
| comstud | yeah, i have to dig into why this is failing | 17:58 |
| russellb | k | 17:59 |
| comstud | i'm getting a timeout on QUEUE.get() | 17:59 |
| comstud | in _test_cast in rpc common test on kombu | 17:59 |
| zykes- | vishy: in essex no | 17:59 |
| russellb | ok so, i was worried about that test, since another test had a comment saying fanout seemed buggy with kombu and the memory transport | 17:59 |
| comstud | yeah | 17:59 |
| russellb | but it was working for me, seemed to work on smokestack, etc ... | 17:59 |
| comstud | maybe it's my version of kombu | 17:59 |
| russellb | yeah | 18:00 |
| russellb | i have 1.1.3 | 18:00 |
| comstud | it's definitely the fanout test | 18:00 |
| comstud | that's failing for me | 18:00 |
| comstud | i'll upgrade kombu and see | 18:00 |
| russellb | if upgrading fixes it, we may be able to re-enable the other fanout test on kombu | 18:00 |
| comstud | ah crap | 18:01 |
| comstud | i'm on 2.1.8 | 18:01 |
| russellb | yeah, that's probably it | 18:01 |
| comstud | well | 18:01 |
| comstud | the fanout problem existed in 1.0.4 | 18:01 |
| comstud | i'd think it'd be a problem in 1.1.3 as well | 18:01 |
| comstud | but | 18:01 |
| comstud | kombu does go back and forth on bugs | 18:01 |
| comstud | reintroducing old ones | 18:01 |
| comstud | in my experience | 18:01 |
| comstud | i'm going to install 1.1.3 | 18:02 |
| vishy | zykes-: ok i will test on trunk, it looks like it might be fixed | 18:02 |
| comstud | figures | 18:02 |
| comstud | 1.1.3 works fine | 18:02 |
| comstud | :) | 18:02 |
| zykes- | which change vishy ? | 18:03 |
| *** dwalleck has quit IRC | 18:04 | |
| *** rohit-k has quit IRC | 18:04 | |
| russellb | comstud: pip-requires is nailed at kombu 1.0.4, too | 18:04 |
| russellb | we could probably relax that ... | 18:04 |
| comstud | i think we have to nail it to something specific | 18:05 |
| comstud | but we can pick something newer | 18:05 |
| russellb | 1.1.3 \o/ | 18:05 |
| comstud | after testin git | 18:05 |
| comstud | yea | 18:05 |
| comstud | apparently that one might be fine :) | 18:06 |
| *** ewindisch has quit IRC | 18:06 | |
| comstud | newest one is 2.1.8 | 18:06 |
| russellb | that just happens to be what is in fedora ... | 18:06 |
| comstud | which is broken | 18:06 |
| comstud | for tests anyway | 18:06 |
| *** ewindisch_ has joined #openstack-dev | 18:06 | |
| vishy | zykes-: I don't see a specific recent change | 18:06 |
| vishy | zykes-: but I don't see in the code how it isn't happening so let me verify | 18:06 |
| russellb | comstud: i don't know how to reproduce that deepcopy bug .. | 18:08 |
| *** salgado-lunch has quit IRC | 18:08 | |
| russellb | but if your patch fixes it for you, fine with me i guess | 18:08 |
| comstud | I'm not sure either. | 18:09 |
| comstud | unfortunately I'd have to dig for quite awhile i think | 18:09 |
| comstud | to discover the real problem | 18:09 |
| *** salgado has joined #openstack-dev | 18:10 | |
| *** salgado has joined #openstack-dev | 18:10 | |
| * russellb nods | 18:10 | |
| *** salgado has quit IRC | 18:10 | |
| comstud | so | 18:11 |
| *** jakedahn_zz is now known as jakedahn | 18:11 | |
| comstud | this is awesome.. that test is failing again | 18:11 |
| comstud | w/ 1.1.3 | 18:11 |
| comstud | now | 18:11 |
| comstud | nevermind. | 18:11 |
| comstud | user error. | 18:12 |
| * russellb pretends that didn't happen | 18:12 | |
| comstud | appreciate it. | 18:14 |
| comstud | hrm | 18:15 |
| comstud | tools/rfc.sh gone? | 18:15 |
| comstud | apparently i need to refresh my git alias | 18:15 |
| russellb | yes | 18:15 |
| notmyname | ttx: I haven't forgotten about you. potential issue with 1.5.0. I'm waiting on details now | 18:16 |
| comstud | https://review.openstack.org/7988 | 18:17 |
| comstud | oops | 18:18 |
| comstud | empty spaces | 18:18 |
| *** salgado has joined #openstack-dev | 18:18 | |
| *** salgado has joined #openstack-dev | 18:18 | |
| *** troytoman is now known as troytoman-away | 18:20 | |
| *** johnpostlethwait has joined #openstack-dev | 18:22 | |
| vishy | zul, adam_g, Daviey, smoser: https://bugs.launchpad.net/nova/+bug/996233 | 18:30 |
| uvirtbot | Launchpad bug 996233 in python-novaclient "nova and python-novaclient disagree on volumes API URLs" [Undecided,Confirmed] | 18:30 |
| zul | vishy: we are shipping 2012.1 nova client | 18:32 |
| *** pasm has quit IRC | 18:36 | |
| vishy | zul: hmm, maybe they have an older novaclient installed via pip | 18:37 |
| *** Mandell has joined #openstack-dev | 18:37 | |
| *** adjohn has quit IRC | 18:37 | |
| vishy | zul: I guess someone will have to verify it | 18:37 |
| zul | vishy: coould be | 18:37 |
| vishy | zykes-: verified that floating works on trunk, checking essex | 18:37 |
| zykes- | kk | 18:39 |
| *** sandywalsh has joined #openstack-dev | 18:40 | |
| *** asisin has joined #openstack-dev | 18:40 | |
| asisin | howdy folks | 18:40 |
| asisin | Just got a brand new ubuntu laptop and am installing OS using devstack but it is hanging at a particular point in the install everytime | 18:41 |
| asisin | http://paste.openstack.org/show/18281/ | 18:43 |
| smoser | zul, i can verify that. | 18:46 |
| *** jakedahn is now known as jakedahn_zz | 18:46 | |
| zul | smoser: nope | 18:48 |
| vishy | zykes-: also works in stable/essex | 18:51 |
| zykes- | vishy: hmmm | 18:52 |
| zykes- | what piece of code handles it ? | 18:52 |
| smoser | vishy, it would appear to me to be broken in upstream | 18:55 |
| smoser | at least i just made this change http://paste.ubuntu.com/1016793/ against my installed version of that file, and it went from exhibiting that bug report to functioning | 18:56 |
| *** zaitcev has joined #openstack-dev | 18:57 | |
| asisin | does anyone know if this step in the devstack install - Downloading/unpacking django>=1.4 - takes a looong time or am I hitting a bug | 18:59 |
| *** epim has joined #openstack-dev | 18:59 | |
| *** GheRivero has joined #openstack-dev | 19:01 | |
| smoser | but that seems to be reverting the change that you made at https://github.com/openstack/python-novaclient/commit/03f54c57e13f27324f0ac5aa556bf4ca94e3ed0e | 19:01 |
| smoser | so maybe the server is what is messed up | 19:02 |
| zykes- | vishy: found anything ? | 19:06 |
| zykes- | cause it's not working in my setup (quantum) | 19:06 |
| *** sandywalsh has quit IRC | 19:07 | |
| notmyname | mtaylor: ping | 19:08 |
| notmyname | or jeblair | 19:08 |
| russellb | vishy: https://review.openstack.org/7993 ... regression in that volume patch | 19:09 |
| *** jakedahn_zz is now known as jakedahn | 19:10 | |
| mtaylor | notmyname: whazzup? | 19:10 |
| notmyname | mtaylor: how do submit something for review to a different branch (specifically, the milestone-proposed branch) | 19:11 |
| mtaylor | notmyname: "git review milestone-proposed" | 19:11 |
| mtaylor | notmyname: or | 19:11 |
| mtaylor | notmyname: git push gerrit HEAD:refs/for/milestone-proposed | 19:11 |
| mtaylor | if you wanted to do it by hand | 19:11 |
| notmyname | mtaylor: ah cool. the first is fine. thanks :-) | 19:11 |
| mtaylor | notmyname: sure nuff! | 19:11 |
| *** sandywalsh has joined #openstack-dev | 19:12 | |
| *** openstackjenkins has quit IRC | 19:14 | |
| *** openstackjenkins has joined #openstack-dev | 19:15 | |
| *** jakedahn is now known as jakedahn_zz | 19:15 | |
| notmyname | ttx: running some tests on some patches that are needed for milestone-proposed. will propose and merge them when I have confirmation | 19:17 |
| notmyname | ttx: I don't have a timeframe yet, but worst-case we may need to delay until tomorrow. I still hope to get it taken care of this afternoon (US). sorry it's getting late for you | 19:19 |
| zykes- | vishy: ? | 19:21 |
| *** dolphm has quit IRC | 19:23 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:25 | |
| *** mdomsch has quit IRC | 19:31 | |
| *** davidha has quit IRC | 19:35 | |
| *** davidha has joined #openstack-dev | 19:35 | |
| davidkranz | vishy: A tempest test is failing which does 'allocate floating-ip; delete floating-ip; list floating ips and make sure it is not there. There seems to be a time lag between when the delete returns 202 and when list floating ips no longer shows it. Is this a nova bug? | 19:36 |
| *** rkukura has joined #openstack-dev | 19:41 | |
| *** apevec has joined #openstack-dev | 19:46 | |
| *** rkukura has quit IRC | 19:46 | |
| davidha | notmyname hi | 19:47 |
| davidha | Is there some tool to test swift performance? | 19:48 |
| notmyname | davidha: swift ships with swift-bench. that's a very good start since it does test various aspects of the system and allows for better comparisons between clusters | 19:51 |
| *** markmc has quit IRC | 19:51 | |
| davidha | And what are probtests? | 19:52 |
| notmyname | davidha: unit tests test the code. functional tests make sure it works as a black box. probetests do some amount of integration testing for the different parts of swift | 19:54 |
| *** datsun180b has quit IRC | 19:55 | |
| *** jeremydei has quit IRC | 19:58 | |
| *** jeremydei has joined #openstack-dev | 19:59 | |
| *** openstackjenkins has quit IRC | 20:01 | |
| *** openstackjenkins has joined #openstack-dev | 20:01 | |
| *** mdomsch has joined #openstack-dev | 20:02 | |
| zykes- | vishy: ? | 20:02 |
| *** rnirmal has quit IRC | 20:02 | |
| *** andrewsben is now known as andrewsben_zz | 20:04 | |
| *** wiliam has quit IRC | 20:05 | |
| *** sannes has joined #openstack-dev | 20:10 | |
| davidha | notmyname: is there info somewhere about swift-bench and the results pepole should expect when running it with different parameters? | 20:11 |
| davidha | (e.g. when running in SAIO env) | 20:12 |
| *** apevec has quit IRC | 20:17 | |
| *** dubsquared has joined #openstack-dev | 20:21 | |
| *** apevec has joined #openstack-dev | 20:22 | |
| *** andrewsben_zz is now known as andrewsben | 20:24 | |
| *** mnaser has quit IRC | 20:27 | |
| *** lts has quit IRC | 20:38 | |
| *** jakedahn_zz is now known as jakedahn | 20:42 | |
| *** GheRivero has quit IRC | 20:50 | |
| *** torgomatic_ has joined #openstack-dev | 20:51 | |
| bcwaldon | markwash: what do you think about this: https://review.openstack.org/#/c/7991/ | 20:52 |
| *** torgomatic has quit IRC | 20:54 | |
| *** torgomatic_ is now known as torgomatic | 20:54 | |
| notmyname | davidha: no where official. I actually bought swiftbench.com a while back, but it's so far down on my list of stuff to do that I haven't even thought about it in a while | 20:54 |
| *** dprince has quit IRC | 20:55 | |
| notmyname | davidha: on prod hardware, I'd expect swift bench to give you somewhere in the neighborhood of 1k/sec pretty easily. probably closer to 75-100/sec on a SAIO VM | 20:55 |
| vishy | tr3buchet: any idea why this might be happening on stop/start of an instance? http://paste.openstack.org/show/18285/ | 20:56 |
| tr3buchet | vishy: looking | 20:56 |
| tr3buchet | vishy: appears the instance doesn't have any ips | 20:57 |
| *** alaski has quit IRC | 20:58 | |
| tr3buchet | vishy: otherwise it could be a bug from my last patch. i didn't have a great way to test libvirt other than what the unittests do | 20:58 |
| vishy | tr3buchet: well somehow allocate_network seems to be returning bad info | 20:58 |
| *** relateable has joined #openstack-dev | 20:58 | |
| tr3buchet | vishy: have a paste of that? | 20:58 |
| sdague | vishy: I was running into exactly that issue on my box just now, ips was an empty array | 21:00 |
| *** markvoelker has quit IRC | 21:00 | |
| sdague | it only was on the test machine that had 2 nics, one of which had no address of it's own, and was only used for guests | 21:01 |
| tr3buchet | vishy: sdague: i betcha this is my fault | 21:01 |
| sdague | on the 1 nic box, where the host network and guests shared the same card and bridge, it was fine | 21:01 |
| sdague | if that helps narrow down | 21:01 |
| vishy | tr3buchet: http://paste.openstack.org/show/18286/ | 21:02 |
| vishy | tr3buchet: it looks like it is assigning a new ip when the start_instance commnad runs again | 21:04 |
| tr3buchet | vishy: both have ip | 21:04 |
| tr3buchet | yeah but ip is changing | 21:04 |
| tr3buchet | which is weird | 21:04 |
| *** Aaton_off is now known as Aaton | 21:05 | |
| vishy | tr3buchet: so stop/start shouldn't be doing that | 21:05 |
| tr3buchet | no way | 21:05 |
| tr3buchet | why is it even logging the nw_info on stop start.. | 21:06 |
| tr3buchet | vishy: ^ | 21:06 |
| *** tongli|2 has quit IRC | 21:06 | |
| vishy | tr3buchet: stop does a terminate of the instance but leaves stuff in place | 21:06 |
| vishy | tr3buchet: so that is the problem it should not be deallocating / reallocating on stop start | 21:07 |
| tr3buchet | damn | 21:07 |
| tr3buchet | why does reboot need nw_info | 21:07 |
| vishy | tr3buchet: what is the purpose of is_first_time? | 21:08 |
| tr3buchet | looks like it's to update some instance metadata | 21:09 |
| tr3buchet | vishy: ^ | 21:09 |
| tr3buchet | vishy: someone added a new column to instance looks like access_ip_v4 and access_ip_v6 | 21:10 |
| vishy | comstud: ^^ is_first_time look familiar? | 21:10 |
| *** stuntmachine has quit IRC | 21:10 | |
| comstud | i'm familiar with that | 21:10 |
| tr3buchet | oh good | 21:11 |
| comstud | :) | 21:11 |
| tr3buchet | please to be educating | 21:11 |
| tr3buchet | cause this looks nasty | 21:11 |
| comstud | access* ips are normally set by the user | 21:11 |
| comstud | but in event they don't pass it in... | 21:11 |
| comstud | we're being nice and setting it on initial build only | 21:11 |
| *** dachary has joined #openstack-dev | 21:12 | |
| tr3buchet | that code is inefficiently looping through all IPs even though it looks like only the first one is needed | 21:12 |
| vishy | tr3buchet, comstud: also i really dislike that we have stop/start and poweroff / poweronn | 21:12 |
| tr3buchet | vishy: yep clear as day | 21:13 |
| tr3buchet | vishy: nova/compute/manager.py:716 | 21:14 |
| tr3buchet | vishy: 716 self._deallocate_network(context, instance) | 21:14 |
| tr3buchet | that's in shutdown instance which is called by stop | 21:14 |
| tr3buchet | that makes no sense | 21:14 |
| tr3buchet | hmm | 21:15 |
| tr3buchet | i added that line but when i did so it wasn't in shutdown.. it was delete | 21:15 |
| tr3buchet | looks like a refactor there went wrong | 21:16 |
| tr3buchet | and they put the deallocate into shutdown instead of delete | 21:16 |
| tr3buchet | vishy: ^ | 21:16 |
| tr3buchet | vishy: comstud: even the comments are wonky... "# NOTE(vish) get bdms before destroying the instance" that's in _shutdown_instance() | 21:17 |
| vishy | tr3buchet: its ok I'm adding a boolean to check it | 21:17 |
| tr3buchet | we should probably just move that stuff out of shutdown and into delete | 21:17 |
| tr3buchet | vishy: ^ | 21:17 |
| vishy | tr3buchet: that doesn't fix the reallocation on start | 21:17 |
| vishy | tr3buchet: unless it is ok to allocate multiple times? | 21:18 |
| tr3buchet | vishy: it's not | 21:18 |
| tr3buchet | i'm looking at that part now | 21:18 |
| *** apevec has quit IRC | 21:18 | |
| *** sandywalsh has quit IRC | 21:18 | |
| tr3buchet | vishy: i can't figure how this happened... | 21:18 |
| *** ncode has joined #openstack-dev | 21:19 | |
| tr3buchet | vishy: also i don't understand the differece between start_instance and run_instance | 21:19 |
| vishy | start is to restart stopped instance | 21:19 |
| vishy | tr3buchet: hold up there is more that has been messed up | 21:19 |
| tr3buchet | vishy: yeah | 21:20 |
| vishy | tr3buchet: this happened because we don't have an exercise for start/stop | 21:20 |
| tr3buchet | also power_on_instance | 21:20 |
| tr3buchet | vishy: talk about convoluted | 21:20 |
| vishy | tr3buchet: there was a cleanup=false being passed in to the driver as well | 21:21 |
| vishy | tr3buchet: sigh | 21:21 |
| tr3buchet | yeah | 21:21 |
| tr3buchet | wtf | 21:21 |
| tr3buchet | i can't even find where build starts.. | 21:21 |
| tr3buchet | thought i found it | 21:22 |
| tr3buchet | but now | 21:22 |
| tr3buchet | no* | 21:22 |
| vishy | tr3buchet: you know what, I think stop and start should just use poweroff and power onn | 21:22 |
| tr3buchet | vishy: yeah i agree | 21:22 |
| *** crobinso has quit IRC | 21:23 | |
| vishy | tr3buchet: or they should be consolidated into one call | 21:23 |
| tr3buchet | run_instance and start_instance especially | 21:23 |
| tr3buchet | vishy: and neither of those should be calling _run_instance | 21:24 |
| tr3buchet | vishy: we really need to keep a closer eye on what is happening in the compute manager | 21:24 |
| vishy | tr3buchet: unfortunately power_off and power_on do not have block_device_mapping or network_info | 21:24 |
| vishy | i will have to add them | 21:25 |
| tr3buchet | why does power on/off even need nw_info ? | 21:25 |
| vishy | tr3buchet: libvirt regenerates the xml for the instance | 21:25 |
| vishy | at least that is how it is implemented | 21:26 |
| vishy | tr3buchet: let me see if there is another way i could do it | 21:26 |
| tr3buchet | vishy: that seems silly. is that unavoidable? | 21:26 |
| vishy | tr3buchet: does power_off in xen always do a hard power_off? | 21:27 |
| vishy | tr3buchet: is there a way to do a soft shutdown? or just a reboot? | 21:27 |
| tr3buchet | vishy: i don't think so | 21:27 |
| *** openstackjenkins has quit IRC | 21:28 | |
| tr3buchet | vishy: i don't think so to your first question. there should be a way yes' | 21:28 |
| *** openstackjenkins has joined #openstack-dev | 21:28 | |
| tr3buchet | vishy: power_off and power_on don't use nw_infpo | 21:29 |
| vishy | tr3buchet: so even hard reboot is done by destroying and recreating the instance | 21:29 |
| tr3buchet | driver.power_off() | 21:29 |
| tr3buchet | that should be doing self.driver.reboot | 21:29 |
| tr3buchet | vishy: looking | 21:29 |
| vishy | tr3buchet: I can do some refactoring to make it write out the xml before reboot and then use the existing xml instead of regenerating it | 21:30 |
| maoy | when an instance is in ERROR state, what's the CLI way to get the error message? | 21:31 |
| tr3buchet | vishy: probably a better way to do it | 21:31 |
| tr3buchet | reboot in xen takes a reboot_type parameter which can be HARD | 21:32 |
| vishy | tr3buchet: i understand that | 21:32 |
| vishy | tr3buchet: i was asking about power_off | 21:32 |
| tr3buchet | oh | 21:32 |
| davidha | notmyname: 75-100 is the rate of GETs or PUTs? | 21:33 |
| tr3buchet | vishy: there are options for hard shutdown as well | 21:33 |
| tr3buchet | it defaults to hard | 21:33 |
| davidha | (On SAIO) | 21:33 |
| *** mnaser has joined #openstack-dev | 21:33 | |
| vishy | tr3buchet: but that isn't exposed to nova right? | 21:33 |
| tr3buchet | i'm looking | 21:33 |
| notmyname | davidha: PUTs. what are you seeing? | 21:33 |
| tr3buchet | looks like power_off is always hard | 21:34 |
| tr3buchet | vishy: yeah configuring that part isn't exposed | 21:34 |
| vishy | tr3buchet: so the libvirt driver needs some refactoring here. I think it should keep the domain definition instead of recreating it on hard reboot / poweroff / poweron | 21:34 |
| tr3buchet | vishy: it's only in the virt layer | 21:35 |
| *** hub_cap has quit IRC | 21:35 | |
| vishy | tr3buchet: I'm not totally sure if it should destroy and recreate the vifs on shutdown / hard reboot | 21:35 |
| tr3buchet | vishy: agree. and i think it absolutely should not | 21:35 |
| vishy | tr3buchet: i suppose it is safe to keep the firewall rules / vifs in place | 21:35 |
| tr3buchet | you want new vifs do a rebuild | 21:35 |
| *** datsun180b has joined #openstack-dev | 21:36 | |
| *** aa has quit IRC | 21:37 | |
| *** dachary has quit IRC | 21:37 | |
| vishy | tr3buchet: this hard reboot is a little annoying, it is used for rescue and unrescue as well | 21:38 |
| tr3buchet | vishy: hmm.. | 21:38 |
| vishy | tr3buchet: ok i'm going to have to refactor a few things in libvirt | 21:39 |
| * vishy rolls up his sleeves | 21:39 | |
| vishy | tr3buchet: I will remove stop start and convert them to use power_off power_on | 21:39 |
| vishy | tr3buchet: this will be a nasty patch, I'm sure i will break a bunch of tests | 21:40 |
| maoy | vishy: while you are at it.. i'd like to complain about some code duplication | 21:40 |
| vishy | maoy: shoot | 21:40 |
| tr3buchet | vishy: yeah it surely will :( | 21:40 |
| maoy | vishy: spawn, hard_reboot, finish_migration, finish_revert_migration | 21:40 |
| maoy | vishy: all "kind of" create a VM, but slightly different | 21:40 |
| vishy | maoy: ok | 21:41 |
| *** flaviamissi has quit IRC | 21:41 | |
| vishy | maoy: are you refering to the duplicated calls to firewall_driver create image etc.? | 21:41 |
| maoy | vishy: i'm not sure it's intentional or not.. e.g. spawn does a little more. anyway, could use a refactor. | 21:41 |
| maoy | vishy: yes | 21:42 |
| vishy | maoy: ok I will tackle that in this patch too | 21:42 |
| maoy | vishy: cool | 21:42 |
| vishy | maoy: I'm going to allow you to create an instance with xml or from an existing domain | 21:44 |
| vishy | maoy: so things like hard_reboot / power_off / power_on can leave the domain in place | 21:44 |
| *** ewindisch_ has quit IRC | 21:44 | |
| *** aa has joined #openstack-dev | 21:44 | |
| vishy | I will only do the firewall rule setup if the xml is passed in. | 21:45 |
| vishy | maoy: and I will have a way to destroy an instance without deleting the domain | 21:45 |
| maoy | vishy: sounds good. | 21:46 |
| vishy | maoy: cancel that, I may make it a flag | 21:46 |
| vishy | * boolean | 21:46 |
| *** b1rkh0ff has quit IRC | 21:46 | |
| *** rkukura has joined #openstack-dev | 21:47 | |
| maoy | vishy: change of topic a bit: stop() start() doesn't exist in python-novaclient? | 21:49 |
| maoy | vishy: is it EC2 specific? | 21:50 |
| *** openstackjenkins has quit IRC | 21:51 | |
| *** openstackjenkins has joined #openstack-dev | 21:52 | |
| *** salgado is now known as salgado-afk | 21:58 | |
| vishy | maoy: no it needs to be added | 21:58 |
| vishy | maoy: i'm going to make it do power_off and power_on instead as well | 21:58 |
| vishy | brb | 21:58 |
| *** b1rkh0ff has joined #openstack-dev | 21:59 | |
| comstud | vishy: agree w/ start/stop/poweron/poweroff | 22:00 |
| comstud | sorry, had to disappear for a bit | 22:00 |
| comstud | tr3buchet: need anything else? | 22:00 |
| comstud | :) | 22:00 |
| tr3buchet | yeah ts if you have a minute | 22:01 |
| tr3buchet | comstud: ^ | 22:01 |
| maoy | vishy: i c. | 22:01 |
| comstud | give me 5 minutes? | 22:01 |
| tr3buchet | sure | 22:01 |
| comstud | alrighty | 22:01 |
| maoy | vishy: there is a shutdown_terminate flag. is that EC2 specific? | 22:01 |
| tr3buchet | take all the time you need | 22:01 |
| maoy | vishy: when it's true, stop() == terminate(). i'm not sure what's the default value, and if it's effective if OS API is called | 22:02 |
| *** ayoung has quit IRC | 22:03 | |
| maoy | vishy: I just did a test against trunk. when I call euca-stop-instances. the vm state is set to STOPPED, but the VM is destroyed. and futher euca-start-instances will make vm state to ERROR | 22:04 |
| comstud | ya, i guess there's that difference | 22:04 |
| maoy | comstud: seems like a bug to me. shouldn't it be set to DELETED directly instead of STOPPED? | 22:05 |
| *** negronjl has quit IRC | 22:06 | |
| comstud | not sure about EC2 | 22:06 |
| *** relateable_ has joined #openstack-dev | 22:06 | |
| maoy | comstud: the OS API code doesn't seem to check this flag. so stop is always stop | 22:06 |
| harlowja | qq, for someone who knows the iptables code, i am looking at essex libvirt.xml, and i see that it uses the following as a filterref, "nova-instance-${name}-${nic.id}" | 22:07 |
| harlowja | but then in the firewall code it uses 'inst-%s' % (instance['id'],) as the chain-name, how are those 2 actually connectd? | 22:08 |
| *** datsun180b has quit IRC | 22:08 | |
| harlowja | ah, nm | 22:10 |
| *** Gordonz has quit IRC | 22:12 | |
| *** maoy has quit IRC | 22:17 | |
| *** nikhil__ has quit IRC | 22:21 | |
| *** ywu has joined #openstack-dev | 22:25 | |
| *** hub_cap has joined #openstack-dev | 22:26 | |
| *** pmezard has quit IRC | 22:28 | |
| *** galstrom has quit IRC | 22:30 | |
| *** dubsquared has quit IRC | 22:39 | |
| *** ywu has quit IRC | 22:40 | |
| *** adjohn has joined #openstack-dev | 22:40 | |
| *** negronjl has joined #openstack-dev | 22:43 | |
| *** spiffxp has quit IRC | 22:48 | |
| *** AlanClark has quit IRC | 22:49 | |
| johnpostlethwait | I'm having some issues with the Jenkins tests for a review of mine and I'm not entirely sure what to do about it: https://jenkins.openstack.org/job/gate-integration-tests-devstack-vm/4455/console | 22:56 |
| johnpostlethwait | The error is "ImportError: No module named versiontools.setuptools_hooks" | 22:56 |
| mtaylor | johnpostlethwait: looking | 22:58 |
| mtaylor | johnpostlethwait: perhaps some module didn't make it in to the pip-requires for something? | 22:59 |
| mtaylor | hrm. looks like it installed versiontools... | 22:59 |
| *** lloydde has quit IRC | 23:00 | |
| johnpostlethwait | In my change I added django_compressor as a pip-require | 23:00 |
| *** andrewsben is now known as andrewsben_zz | 23:05 | |
| johnpostlethwait | Yeah, this is strange, the version of versiontools it installs definitely has that function | 23:05 |
| johnpostlethwait | http://versiontools.readthedocs.org/en/latest/reference.html#versiontools-setuptools-hooks | 23:06 |
| *** dtroyer is now known as dtroyer_zzz | 23:07 | |
| *** jakedahn is now known as jakedahn_zz | 23:09 | |
| *** blamar has quit IRC | 23:10 | |
| *** openstackjenkins has quit IRC | 23:18 | |
| *** openstackjenkins has joined #openstack-dev | 23:19 | |
| *** anderstj has joined #openstack-dev | 23:21 | |
| *** edygarcia has quit IRC | 23:22 | |
| *** e1mer has joined #openstack-dev | 23:22 | |
| *** flaviamissi has joined #openstack-dev | 23:22 | |
| *** davidha has quit IRC | 23:26 | |
| *** blamar has joined #openstack-dev | 23:29 | |
| *** jgriffith is now known as jgriffith_away | 23:29 | |
| *** davidha has joined #openstack-dev | 23:32 | |
| *** rods has quit IRC | 23:36 | |
| *** spiffxp has joined #openstack-dev | 23:38 | |
| *** jdurgin has quit IRC | 23:46 | |
| *** jdurgin has joined #openstack-dev | 23:47 | |
| mikal | Has anyone else experienced the projects table in nova going missing? | 23:50 |
| vishy | mikal: no but it shouldn't be needed/used anywhere | 23:53 |
| mikal | I just sent an email to the list about it | 23:54 |
| mikal | The table appears in "show tables", but a describe and a repair both fail on it | 23:54 |
| mikal | Essex nova-api is complaining about the table being missing | 23:54 |
| mikal | Ahhh, there is a suggestion this is https://bugs.launchpad.net/ubuntu/+source/nova/+bug/975085 from a guy at work | 23:56 |
| uvirtbot | Launchpad bug 975085 in nova "nova mysql DB can't be restored from backups (dup-of: 993663)" [Medium,Confirmed] | 23:56 |
| uvirtbot | Launchpad bug 993663 in nova "dns_domains table mysql charset is 'latin1'. Should be 'utf8'" [Medium,Fix released] | 23:56 |
| *** johnpostlethwait has quit IRC | 23:56 | |
| *** jdurgin has quit IRC | 23:57 | |
| *** roge has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!