| *** zykes- has joined #openstack-dev | 00:19 | |
| annegentle | mtaylor: pong | 00:28 |
|---|---|---|
| *** heckj has quit IRC | 00:42 | |
| *** mattray has joined #openstack-dev | 01:05 | |
| *** mgius is now known as mgius-away | 01:05 | |
| openstackjenkins | Project nova build #955: SUCCESS in 2 min 49 sec: http://jenkins.openstack.org/job/nova/955/ | 01:14 |
| openstackjenkins | Tarmac: Logs the exception if metadata fails and returns a 500 with an error message to the client. | 01:14 |
| *** cloudgroups has joined #openstack-dev | 01:15 | |
| *** cloudgroups has left #openstack-dev | 01:15 | |
| *** sandywalsh has quit IRC | 02:30 | |
| *** dragondm has quit IRC | 02:35 | |
| *** mattray has quit IRC | 03:12 | |
| *** jkoelker has quit IRC | 03:19 | |
| *** dprince has joined #openstack-dev | 03:23 | |
| *** cloudgroups has joined #openstack-dev | 03:56 | |
| *** Binbin has joined #openstack-dev | 03:57 | |
| *** Binbin is now known as Binbin_afk | 04:01 | |
| *** openpercept_ has joined #openstack-dev | 04:02 | |
| *** cloudgroups has left #openstack-dev | 04:05 | |
| comstud | soren: might be hard to add tests for the eventlet issue, but i can try. what issue are you running into? i haven't heard anything about it yet | 04:06 |
| *** dprince has quit IRC | 04:20 | |
| *** Zangetsue has joined #openstack-dev | 04:28 | |
| *** Binbin_afk is now known as Binbin | 04:34 | |
| *** openpercept_ has quit IRC | 04:42 | |
| *** openpercept_ has joined #openstack-dev | 04:44 | |
| *** openpercept_ has quit IRC | 04:44 | |
| *** openpercept_ has joined #openstack-dev | 04:44 | |
| *** vinkeshb has joined #openstack-dev | 04:44 | |
| *** Zangetsue_ has joined #openstack-dev | 05:00 | |
| *** vinkeshb_ has joined #openstack-dev | 05:02 | |
| *** Zangetsue has quit IRC | 05:03 | |
| *** Zangetsue_ is now known as Zangetsue | 05:03 | |
| *** vinkeshb has quit IRC | 05:04 | |
| *** vinkeshb_ is now known as vinkeshb | 05:04 | |
| *** vinkeshb_ has joined #openstack-dev | 05:21 | |
| *** vinkeshb has quit IRC | 05:21 | |
| *** vinkeshb_ is now known as vinkeshb | 05:21 | |
| mtaylor | annegentle: me havey problem with docs... http://jenkins.openstack.org/job/openstack-manuals/13/console | 05:48 |
| *** RobertLaptop_ has joined #openstack-dev | 06:27 | |
| *** RobertLaptop has quit IRC | 06:30 | |
| ttx | soren: ping | 06:56 |
| ttx | vishy: still around ? | 06:56 |
| soren | soren: pong | 06:57 |
| soren | comstud: Well, the test suite as it is fails (citing buffer read-only errors). | 06:57 |
| soren | comstud: It happens in the recv_into and recvfrom_into tests. | 06:57 |
| ttx | soren: was wondering if you could give https://code.launchpad.net/~xtoddx/nova/provider-fw-rules/+merge/62517 a look before we cut the release branch | 06:57 |
| soren | ttx: Oh, I thought that landed last night. | 06:57 |
| ttx | got one approved and one needs fixing that was addressed | 06:58 |
| *** zykes- has quit IRC | 07:01 | |
| comstud | soren: oh, the evenlet test suite. yes, i have a fix for that. | 07:02 |
| comstud | soren: if you check out lp:rackspace-eventlet, i have fixes for the eventlet test | 07:02 |
| ttx | soren: otherwise Glance is ready to be cut | 07:02 |
| comstud | soren: really, it's a bug in the eventlet test suite. | 07:02 |
| ttx | soren: when you're ready to start working on that, let me know, I'll start by using common-bump-milestone, then you can create the branch and the whole shebang | 07:03 |
| comstud | sorry, I should have re-posted a diff with those fixes | 07:03 |
| *** reidrac has joined #openstack-dev | 07:03 | |
| soren | comstud: If you would, that would be great. If upstream ever gets back to looking at it, it'd be a shame if they rejected it on these grounds. | 07:05 |
| comstud | yeah | 07:05 |
| comstud | i need to do a proper pull request | 07:06 |
| comstud | - buf = buffer(array.array('B')) | 07:06 |
| comstud | + buf = array.array('B') | 07:06 |
| comstud | + buf.fromstring('\0') | 07:06 |
| comstud | that's the fix to the tests.. there's 2 tests that need it | 07:07 |
| comstud | diff coming. | 07:07 |
| comstud | they were creating a read only buffer before that passed tests for whatever reason, even though you should not pass a readonly bufer to recvfrom() | 07:07 |
| comstud | er, recvfrom_into() | 07:08 |
| comstud | and recv_into() | 07:08 |
| comstud | obviously the buffer should be writeable. | 07:08 |
| *** zykes- has joined #openstack-dev | 07:08 | |
| comstud | http://comstud.com/eventlet-socket-timeout.diff | 07:10 |
| comstud | that's the full diff, but it's against 0.9.15 | 07:10 |
| comstud | which I think you said has some issues | 07:10 |
| vishy | ttx: about to go to bed, what's up? | 07:16 |
| ttx | vishy: was wondering what to do with xs-ovs | 07:16 |
| ttx | vishy: should it get in before we cut the branch (i.e. now ?) | 07:17 |
| vishy | i looked it over, i think it is fine to go in | 07:17 |
| ttx | vishy: please approve then | 07:17 |
| vishy | k | 07:17 |
| vishy | is soren looking at the provider fw branch? | 07:17 |
| ttx | vishy: I think he will | 07:17 |
| ttx | vishy: but feel free to approve it if you still can review that | 07:19 |
| vishy | that one is fine by me also, but I'll let soren do a final once over on it | 07:19 |
| ttx | ok | 07:19 |
| comstud | soren: updated https://bitbucket.org/which_linden/eventlet/issue/87/ | 07:20 |
| comstud | 2:20am, bed for me. | 07:21 |
| * comstud & poofs | 07:22 | |
| *** Binbin has quit IRC | 07:27 | |
| openstackjenkins | Project nova build #956: SUCCESS in 2 min 49 sec: http://jenkins.openstack.org/job/nova/956/ | 07:34 |
| openstackjenkins | Tarmac: Adds hooks for applying ovs flows when vifs are created and destroyed for XenServer instances. | 07:34 |
| *** Binbin has joined #openstack-dev | 07:39 | |
| *** Binbin has quit IRC | 07:41 | |
| *** dysinger has quit IRC | 07:43 | |
| ttx | soren: and Nova will be ready to cut once you give provider-firewall the final push in the back | 07:46 |
| soren | comstud: Right, I just wondered if making it readonly was to make the tests fail if something was accidentally written to it. | 07:46 |
| ttx | soren: when that is done I'll use common-bump-milestone and you can create the milestone-proposed branches | 07:47 |
| * ttx common-bump-milestones Glance, since it's ready to branch | 07:47 | |
| ttx | soren: glance trunk set to d2, glance release set to d1. | 07:49 |
| soren | Do you want me to work on the Jenkins/branching/PPA stuff first or review provider-fw-rules first? | 07:51 |
| ttx | provider-fw-rules first | 07:52 |
| soren | ok | 07:56 |
| *** adiantum has joined #openstack-dev | 08:04 | |
| * ttx grabs coffee | 08:18 | |
| *** adiantum_ has joined #openstack-dev | 08:20 | |
| *** adiantum has quit IRC | 08:23 | |
| soren | vishy: Are you still here, by any chance? | 08:27 |
| soren | vishy: Never mind. | 08:29 |
| soren | ttx: Ok, so glance first? | 08:33 |
| ttx | soren: glance is common-milestone-bumped, so yes. | 08:33 |
| ttx | soren: for nova I'm waiting on the decision on provider-fw-rules | 08:34 |
| soren | ttx: I voted Needs Fixing on it. | 08:34 |
| soren | Just a couple of minutes ago. | 08:34 |
| ttx | ok, then I'll common-bump-milestone nova as well. | 08:34 |
| ttx | nova trunk set to d2, nova rlease set to d1. | 08:36 |
| soren | Ok, so first I'll adjust the existing -tarball jobs to the new versioning scheme. | 08:36 |
| ttx | soren: oh, I think we did that already. | 08:36 |
| ttx | oh, only for glance. | 08:37 |
| soren | 10:18 <soren> Swift change still in effet. Nova reverted, glance deferred. | 08:37 |
| soren | Hm... We must have switched glance a little bit later. | 08:38 |
| ttx | yep, and I told you Nova was ready to go as well, so I assumed it was ok | 08:39 |
| soren | ttx: No, never got around to it. | 08:39 |
| soren | ttx: Just clicked the button. | 08:39 |
| soren | Regenerate tarballs? | 08:39 |
| ttx | soren: that can't hurt and will tset the thing. | 08:39 |
| * soren does so | 08:39 | |
| soren | mv dist/nova-2011.3.tar.gz dist/nova-2011.3~d2~20110601.1130.tar.gz | 08:40 |
| ttx | soren: ack. | 08:40 |
| soren | Cool. | 08:41 |
| soren | Ok, I'll set up the extra jobs for Nova first. | 08:41 |
| soren | ttx: Did you create PPA's yet? | 08:42 |
| ttx | no. Doing so now. | 08:43 |
| soren | \o/ | 08:43 |
| ttx | https://launchpad.net/~glance-core/+archive/milestone-proposed | 08:45 |
| ttx | https://launchpad.net/~nova-core/+archive/milestone-proposed | 08:46 |
| soren | Cool. | 08:46 |
| soren | Which revno of nova should I use to create the bzr branch? | 08:46 |
| ttx | 1130 | 08:46 |
| * soren does so | 08:47 | |
| soren | And for glance? | 08:48 |
| ttx | 139 | 08:49 |
| soren | bzr branches created for both. | 08:49 |
| soren | Nova should be ready to go on Jenkins. | 08:49 |
| soren | I guess we'll know in a minute when it polls those branches. | 08:50 |
| ttx | right -- it should make tarball and PPA upload all the way ? | 08:50 |
| soren | That's the hypothesis | 08:50 |
| soren | polling bzr... | 08:51 |
| ttx | do you need extra stuff in the PPA first ? | 08:51 |
| openstackjenkins | Project nova-milestone build #1: SUCCESS in 9.5 sec: http://jenkins.openstack.org/job/nova-milestone/1/ | 08:51 |
| soren | Uh.. | 08:51 |
| soren | Heh, yeah :) | 08:51 |
| ttx | you probably still have like 30sec to do so :) | 08:51 |
| soren | I can actually make it. | 08:52 |
| * soren tries. | 08:52 | |
| soren | I have 2½ minutes. | 08:52 |
| ttx | right, it's busy pushing the trunk refresh we did before | 08:53 |
| soren | ...and the publisher only runs every 5 minutes. | 08:55 |
| soren | Damn it, didn't make it. | 08:55 |
| ttx | -ppa job didn't run yet | 08:56 |
| soren | orly? | 08:56 |
| ttx | http://jenkins.openstack.org/view/Nova/job/nova-milestone-ppa/ | 08:57 |
| soren | Maybe the first pull is special cased somehow. | 08:57 |
| soren | ...and doesn't count as a change. | 08:57 |
| ttx | ow | 08:57 |
| soren | Dunno. | 08:58 |
| ttx | nova-milestone-tarball triggers nova-ppa | 08:58 |
| soren | Although ^^^ | 08:58 |
| ttx | not nova-milestone-ppa | 08:58 |
| soren | Gah. | 08:58 |
| soren | fixed | 08:59 |
| * soren manually runs nova-milestone-ppa | 08:59 | |
| soren | Someone should whip up a tool that will take job relationships on Jenkins and turn it into a graph. | 09:00 |
| soren | Uploading nova_2011.3~d1~20110601.r1130-0ubuntu0ppa1~natty1_source.changes: done. | 09:00 |
| soren | ttx: 5 minutes, and glance should be on its way. | 09:12 |
| * soren primes that ppa, too. | 09:12 | |
| ttx | soren: nova fails to build in https://launchpad.net/~nova-core/+archive/milestone-proposed/+packages | 09:23 |
| ttx | though what should be the same tarball was partially successful in the trunk PPA. | 09:26 |
| ttx | Also glance was only partially successful in https://launchpad.net/~glance-core/+archive/milestone-proposed/+packages | 09:28 |
| * ttx retries PPA builds | 09:35 | |
| ttx | beh. | 09:45 |
| soren | Oh, right. | 09:46 |
| soren | I know why. | 09:46 |
| soren | I'll fix in a couple of minutes. | 09:47 |
| soren | brb | 09:47 |
| *** adiantum_ has quit IRC | 09:53 | |
| ttx | mtaylor: please set review team to nova-core on https://code.launchpad.net/~hudson-openstack/nova/milestone-proposed | 09:54 |
| ttx | mtaylor: and set review team to glance-core on https://code.launchpad.net/~hudson-openstack/glance/milestone-proposed | 09:54 |
| soren | ttx: Erk. Cascading failure. | 09:57 |
| soren | ttx: glance is failing to build on Maverick these days for some reason. | 09:57 |
| soren | ttx: ...so nova has no python-glance to build against. | 09:58 |
| soren | ttx: ..so nova fails on Maverick. | 09:58 |
| soren | ttx: and nova fails on Oneiric, too. I haven't quite worked out why yet, but it's been happening for a couple of days. | 09:58 |
| ttx | glance failing on maverick: "WARNING: dot command 'dot' cannot be run (needed for graphviz output), check the graphviz_dot setting" | 09:59 |
| soren | ttx: ? | 10:02 |
| soren | ttx: Where do you see that? | 10:02 |
| soren | I see tests.functional.test_logging.TestLogging failing | 10:02 |
| soren | ttx: Oh, further up. that's just a warning, I think. | 10:03 |
| ttx | https://launchpadlibrarian.net/72777801/buildlog_ubuntu-maverick-i386.glance_2011.3~d1~20110601.r139-0ubuntu0ppa1~maverick2_FAILEDTOBUILD.txt.gz | 10:03 |
| ttx | "jinja2.exceptions.UndefinedError: 'css_files' is undefined", then :) | 10:04 |
| ttx | Not seeing tests.functional.test_logging.TestLogging error there (milestone-proposed PPA) | 10:04 |
| soren | Ah, I know what that is, too :) | 10:05 |
| ttx | soren: will the successfully-built glance/milestone-proposed/maverick unblock nova/milestone-proposed/maverick build ? | 10:15 |
| ttx | (I think it would be good to have at least maverick and natty available for testing) | 10:16 |
| soren | Not directly, no. | 10:17 |
| soren | ...but I'll see what I can do. | 10:17 |
| *** sandywalsh has joined #openstack-dev | 10:39 | |
| reidrac | there's no tag in swift bzr (trunk) for version 1.4.0, that's intentionally of is it a "bug"? | 10:58 |
| jaypipes | morning folks. | 10:58 |
| jaypipes | soren, ttx: see reidrac's ? | 10:59 |
| reidrac | latest tag I can see is 1.3rc1 -- I don't know if it's my mistake (disclaimer: bzr newbie here!) | 10:59 |
| soren | It's a by-product of having these release branches. | 11:00 |
| soren | 1.4.0 doesn't actually exist on trunk. | 11:00 |
| soren | ...because 1.4.0 was released from a separate branch. | 11:00 |
| reidrac | I see | 11:01 |
| soren | ttx: Ok, only Oneiric is failing now for Nova. | 11:01 |
| reidrac | I thought the releases were a snapshot of trunk, thanks soren | 11:02 |
| soren | ttx: Trying my very best to reproduce, but failing so far. | 11:02 |
| soren | reidrac: They used to be. | 11:02 |
| soren | reidrac: It changed *very* recently. | 11:02 |
| soren | I.e. yesterday for Swift and today for glance and nova. | 11:02 |
| reidrac | can you provide me a pointer docs about that change? | 11:03 |
| soren | Nope. | 11:03 |
| soren | Well.. | 11:04 |
| soren | Actually, yes, but it'll melt your brain. You've been warned. | 11:04 |
| soren | Hang on. | 11:04 |
| soren | http://ubuntuone.com/p/vjd/ | 11:04 |
| reidrac | I see... *melting* brb | 11:05 |
| reidrac | so according to http://openstack.org/projects/storage/ -- latest release is 1.3.0 and not 1.4.0 | 11:09 |
| reidrac | I'm confused, but I don't want to waste your time | 11:09 |
| reidrac | I'll wait until there's official documentation about the new release process | 11:10 |
| reidrac | thanks again! | 11:10 |
| soren | 1.4.0 is definitely released. http://openstack.org/projects/storage/ is just lagging. | 11:13 |
| reidrac | but 1.4.0 doesn't exist on trunk, so there's no new development in trunk? | 11:21 |
| soren | Trunk is where new development happens. | 11:29 |
| soren | That's what "trunk" means. | 11:29 |
| soren | It's just that no new development has happened (on trunk or anywhere else). | 11:29 |
| *** nhm has quit IRC | 11:31 | |
| reidrac | but 1.4.0 is not tagged in trunk, you said that release came from a different branch | 11:33 |
| reidrac | those changes are not in trunk | 11:33 |
| *** nhm has joined #openstack-dev | 11:33 | |
| reidrac | that's the specific point I don't understand | 11:33 |
| ttx | reidrac: those changes are also in trunk. | 11:38 |
| ttx | but mixed with other changes. | 11:39 |
| ttx | reidrac: I'll write up something about the process. | 11:39 |
| ttx | When I'm done pushing the first iteration of it. | 11:39 |
| reidrac | thanks ttx | 11:39 |
| reidrac | it's clear now | 11:39 |
| ttx | reidrac: it's basically the NVIE model. | 11:40 |
| reidrac | I'll keep tracking trunk anyway, it's just I was using the tags to know which changes were in which release | 11:40 |
| ttx | http://nvie.com/posts/a-successful-git-branching-model/ | 11:40 |
| soren | ttx: I thought there weren't any changes on the release branch? | 11:40 |
| ttx | soren: there are RC bugfixes. | 11:41 |
| reidrac | ah, I know that post... nice! | 11:41 |
| ttx | soren: ...that need to be pushed to trunk as well. | 11:41 |
| ttx | soren: see for example r302 or r304 in milestone-proposed on page 1 of the PDF. | 11:43 |
| soren | ttx: Wow. That was unexpected. | 11:53 |
| soren | ttx: The Nova oneiric failure is due to a python update. | 11:53 |
| *** nhm has quit IRC | 12:07 | |
| *** yamahata_lt has joined #openstack-dev | 12:20 | |
| *** ameade has joined #openstack-dev | 12:27 | |
| *** nhm has joined #openstack-dev | 12:34 | |
| ttx | reidrac: http://wiki.openstack.org/BranchModel | 12:57 |
| reidrac | oh, sweet! thanks Thierry! | 12:58 |
| *** dprince has joined #openstack-dev | 13:00 | |
| markwash | the policy of moving merges to WIP if two nova cores agree with a needs fixing comment is in effect now, right? | 13:01 |
| soren | markwash: I've not heard of such a policy, but it sounds like a good idea. Go for it. | 13:02 |
| markwash | soren: I think vishy proposed it on the ML at some point | 13:02 |
| soren | markwash: Oh, ok. | 13:02 |
| markwash | but thanks, I will go for it :-) | 13:03 |
| ttx | reidrac: updated with pointers to the branches. | 13:04 |
| ttx | mtaylor: ping | 13:07 |
| *** mattray has joined #openstack-dev | 13:09 | |
| *** lorin1 has joined #openstack-dev | 13:11 | |
| *** Zangetsue has quit IRC | 13:14 | |
| *** Zangetsue has joined #openstack-dev | 13:15 | |
| *** foxtrotgulf has joined #openstack-dev | 13:20 | |
| reidrac | thank ttx | 13:30 |
| reidrac | *thanks | 13:30 |
| *** adiantum_ has joined #openstack-dev | 14:01 | |
| openstackjenkins | Project nova build #957: SUCCESS in 2 min 45 sec: http://jenkins.openstack.org/job/nova/957/ | 14:09 |
| openstackjenkins | Tarmac: Cleanup instances_path in the test_libvirt test_spawn_with_network_info test. Fixes issue where the nova/tests/instance-00000001/ is left in the nova source tree when running run_test.sh -N. | 14:09 |
| annegentle | so how do you control volumes in Cactus in nova - AMQP is the connector, so only the services send messages to it and there's no way for users to control volumes in nova through an API? | 14:09 |
| annegentle | trying to come up with a Cactus-oriented response to https://twitter.com/#!/ved_a/status/75867055865331712 (not Diablo since it's all changing) | 14:09 |
| annegentle | though really maybe the answer is "wait for lunr" | 14:13 |
| blamar | annegentle: The EC2 api has a lot more support for volumes than the openstack API, although there is a volumes extension for the OS API that justinsb wrote a little while ago. | 14:17 |
| blamar | annegentle: Unfortunately I don't really know how well or how either of them work haha | 14:18 |
| ttx | mtaylor: around ? | 14:18 |
| annegentle | blamar: thanks that's helpful though | 14:18 |
| *** vinkeshb has quit IRC | 14:20 | |
| *** jkoelker has joined #openstack-dev | 14:23 | |
| mtaylor | ttx, soren: shouldn't the release branch get merged back in to trunk when it has it's own changes? | 14:27 |
| mtaylor | ttx, soren: that way the tags and stuff propagate | 14:27 |
| ttx | mtaylor: yes. | 14:27 |
| ttx | mtaylor: tags propagate ? | 14:27 |
| mtaylor | ttx: ok. totally jumping in to the middle here :) | 14:27 |
| ttx | mtaylor: while you're around: | 14:28 |
| ttx | <ttx> mtaylor: please set review team to nova-core on https://code.launchpad.net/~hudson-openstack/nova/milestone-proposed | 14:28 |
| ttx | <ttx> mtaylor: and set review team to glance-core on https://code.launchpad.net/~hudson-openstack/glance/milestone-proposed | 14:28 |
| mtaylor | ttx: if you make a release branch, do stuff in there,and then merge that back to trunk, the 1.4.0 tag should wind up in trunk again | 14:28 |
| ttx | mtaylor: something must escape me. What would it tag ? Some ancestor ? | 14:32 |
| mtaylor | ttx: it would tag the revision it was defined on, which would be a sub-revision to a merge if you merged from release back in to trunk | 14:33 |
| mtaylor | ttx: so if you branched for release at rev 52 of trunk, made a release branch change which would be 53 in that branch, and then tagged that rev as 1.4.0 | 14:34 |
| mtaylor | ttx: and then merged release back in to trunk, making trunk rev 53, the tag would point to trunk rev 53.1 | 14:34 |
| ttx | mtaylor: ah, ok. Does that mean that in the case of no changes, we should still merge back to trunk with no changes ? | 14:35 |
| mtaylor | ttx: technically yes - although mechanically that might get odd with tarmac and the way bzr tags work - but certainly could be solved | 14:36 |
| annegentle | does our Jenkins installation run on Ubuntu 10.10? Trying to replicate the error mtaylor's seeing | 14:37 |
| mtaylor | annegentle: yes | 14:37 |
| mtaylor | ttx: those branches are fixed | 14:37 |
| mtaylor | annegentle: yeah - it's a weird one isn't it? | 14:38 |
| annegentle | mtaylor: k, thanks. I don't like that you see an X11 error message so I'm gonna build non-locally. | 14:38 |
| ttx | mtaylor: thx | 14:44 |
| *** vladimir3p has joined #openstack-dev | 14:49 | |
| *** dragondm has joined #openstack-dev | 14:55 | |
| mtaylor | annegentle: seems to be related to batik - trying to figure out why we're triggering it | 15:00 |
| *** Zangetsue has quit IRC | 15:04 | |
| *** antonyy has quit IRC | 15:07 | |
| annegentle | mtaylor: ya I can build successfully on a fresh 10.10 install. Batik does the SVG covers or some such? | 15:09 |
| mtaylor | annegentle: found it - and it's fun | 15:09 |
| annegentle | heh, do tell | 15:09 |
| mtaylor | annegentle: because we install jenkins via apt, we ssh in to the box to do upgrades/restarts | 15:09 |
| mtaylor | annegentle: this is causing our personal DISPLAY env var from our ssh connection to get sucked in to the env of the running jenkins process | 15:10 |
| mtaylor | annegentle: even though there is, in fact, no valid display for jenkins | 15:10 |
| annegentle | mtaylor: ohhhh. now what? | 15:10 |
| mtaylor | annegentle: so - passing -Djava.awt.headless=true fixes it | 15:10 |
| annegentle | mtaylor: who knew? Oh wait, you did. Nice! | 15:11 |
| mtaylor | annegentle: so now they're building just fine and I'll get them ftping over and stuff | 15:12 |
| annegentle | whoohoo | 15:12 |
| *** reidrac has quit IRC | 15:13 | |
| *** rnirmal has joined #openstack-dev | 15:15 | |
| *** openstackjenkins has quit IRC | 15:23 | |
| *** openstackjenkins has joined #openstack-dev | 15:23 | |
| *** openpercept_ has left #openstack-dev | 15:25 | |
| *** heckj has joined #openstack-dev | 15:35 | |
| *** RobertLaptop_ has quit IRC | 15:36 | |
| ttx | jaypipes: diablo-1 looking good ? | 15:39 |
| ttx | jaypipes: unless told otherwise, I'll consider lp:~hudson-openstack/glance/milestone-proposed ready to ship tomorrow morning. | 15:40 |
| ttx | vishy: same for you, with lp:~hudson-openstack/nova/milestone-proposed :) | 15:40 |
| ttx | vishy: though you might still be around when I do so :) | 15:41 |
| *** openstackjenkins has quit IRC | 15:44 | |
| *** openstackjenkins has joined #openstack-dev | 15:45 | |
| jaypipes | ttx: still looking through an issue with the S3 bug... | 15:53 |
| jaypipes | ttx: I *really* want that to go in... bear with me. | 15:53 |
| jaypipes | dprince is helping me on it.. | 15:53 |
| jaypipes | and I mean "helping" in terms of showing me the code is broken :P | 15:56 |
| *** openstackjenkins has quit IRC | 15:57 | |
| *** openstackjenkins has joined #openstack-dev | 15:57 | |
| *** bcwaldon has joined #openstack-dev | 16:15 | |
| *** antonyy has joined #openstack-dev | 16:18 | |
| markwash | ttx: when do you tag diablo-1 off of lp:nova? | 16:20 |
| markwash | ttx: specifically, merges that get in today, do they go into d1? | 16:21 |
| *** deshantm is now known as deshantm_away | 16:23 | |
| *** vinkeshb has joined #openstack-dev | 16:30 | |
| *** vinkeshb has left #openstack-dev | 16:30 | |
| * jaypipes stabs boto | 16:32 | |
| jaypipes | markwash: yes. in about 3 hours I think. | 16:32 |
| jaypipes | markwash: maybe a little later than that. | 16:32 |
| markwash | jaypipes: cool thanks | 16:33 |
| *** johnpur has joined #openstack-dev | 16:33 | |
| *** ChanServ sets mode: +v johnpur | 16:33 | |
| jaypipes | johnpur: afternoon, boss. | 16:34 |
| markwash | anybody particularly familiar with setting up a carrot consumer for messages coming out of nova? | 16:34 |
| johnpur | jaypipes: good morning/afternoon/evening! | 16:45 |
| *** zaitcev has joined #openstack-dev | 16:49 | |
| *** jdurgin has joined #openstack-dev | 16:57 | |
| *** RobertLaptop has joined #openstack-dev | 16:57 | |
| jaypipes | dprince: ok, narrowed it down... it's a bug in bin/glance. It's not picking up stdin redirection into a file object properly. Working on a fix... | 16:58 |
| dprince | jaypipes: Sure. | 16:59 |
| jaypipes | dprince: I'm a little puzzled as to how this got past the functional test case test_bin_glance... | 17:00 |
| jaypipes | dprince: hmm, I probably shouldn't have used a 150M image to test ;) | 17:06 |
| jaypipes | dprince: waiting... | 17:06 |
| *** mgius-away is now known as mgius | 17:08 | |
| mtaylor | jaypipes, dprince: was about to send out the email about testing group, and then realized if I re-ordered a couple of priorities it would be WAY easier for all of us to play ball together... | 17:23 |
| dprince | jaypipes: I'm back. | 17:23 |
| mtaylor | jaypipes, dprince: just didn't want you to think I'd forgotten about you :) | 17:23 |
| *** adiantum_ has quit IRC | 17:23 | |
| jaypipes | dprince: wasn't waiting for you :) was waiting for 150M to upload to S3 ;) | 17:23 |
| jaypipes | dprince: last time I do that! | 17:23 |
| dprince | jaypipes: Ha. | 17:24 |
| dprince | mtaylor: Sure. What do you have in mind? | 17:24 |
| *** bcwaldon has quit IRC | 17:24 | |
| mtaylor | dprince: mainly just that, one of my tasks for a few weeks from now is to get the jenkins openid auth provider configured to use single-sign-on mode with our existing launchpad teams. if I do that _now_ then folks who need access to do things and have sensible conversations will all have it | 17:25 |
| mtaylor | dprince: so anyway - I'm going to get that sorted real quick so that I can include instructions in the email I send out. rather than "um, hey guys - I'll have something for you to look at _soon_ ... hang out for a bit until I do" :) | 17:27 |
| dprince | mtaylor: Sure. Can't wait to see what you've got in the works. I will admit that I'm not entirely sold on the Jenkins approach for this type of testing (giving a larger group of people the keys to a Jenkins server, etc.) That is what you are setting up right? | 17:32 |
| mtaylor | dprince: understood - I'll do what I can to convince you. :) we've been doing this for drizzle dev for a couple of years now to quite a large amount of success. you don't actually have to give everyone the keys to the whole server as it were - you can basically just give folks ability to launch parameterized jobs | 17:34 |
| mtaylor | but the most important thing to me, implementation aside, is that devs are given the ability to run the same tests in the same way that whatever the trunk gatekeeper is going to do | 17:35 |
| mtaylor | because otherwise they have no way to test that their work in a predictable and reproducable manner | 17:36 |
| * mtaylor expects we're going to have all sorts of fun conversations | 17:36 | |
| mtaylor | soren: ping? | 17:36 |
| *** adiantum_ has joined #openstack-dev | 17:38 | |
| dprince | mtaylor: Agreed. I have similar goals. Devs have to have access. Having a gatekeeper 'torpedo' branches without giving someone access to reproduce things would just get frustrating. | 17:38 |
| mtaylor | dprince: sweet. as long as we have similar underlying assumptions, the rest is just details | 17:39 |
| mtaylor | it's when the underlying goals are out of whack that conversations just can never progress | 17:40 |
| mtaylor | vishy: were you the one chatting with soren the other day about the libvirt/nova issues? | 17:42 |
| vladimir3p | Guys, could you pls tell if there are plans to get rid of centralized DB model for scheduling/managing resources. I'm talking about scheduler within AZ (and not distributed scheduler cross zones). Can we assume in our custom scheduler that CPU/disk resources will continue to be available from the DB (which is updated periodically)? Thanks. | 17:43 |
| *** adiantum_ has quit IRC | 17:44 | |
| *** Shyam_ has joined #openstack-dev | 17:48 | |
| mtaylor | jaypipes: ola? | 17:50 |
| jaypipes | mtaylor: ya | 17:51 |
| mtaylor | jaypipes: do you know anything about the nova/python-glance depend situation at the moment? | 17:52 |
| jaypipes | mtaylor: other than it should exist? :) | 17:53 |
| mtaylor | jaypipes: trying to install debs built from the latest trunk borks me on libvirt versions and on missing python-glance (the reason the smoketests job is failing) | 17:53 |
| mtaylor | jaypipes: I may just need soren | 17:53 |
| jaypipes | mtaylor: yeah, sorry, I think soren might be best for that one. | 17:53 |
| mtaylor | jaypipes: k | 17:54 |
| mtaylor | soren: ? | 17:54 |
| mtaylor | soren: ? | 17:54 |
| mtaylor | soren: ? | 17:54 |
| mtaylor | I miss you soren | 17:54 |
| mtaylor | jaypipes: (it may be that the smoketests job actually caught a real problem, but I didn't have the output set up right so that it was apparent. :) ) | 17:55 |
| jaypipes | mtaylor: understood. | 17:55 |
| *** openstackjenkins has quit IRC | 18:04 | |
| *** openstackjenkins has joined #openstack-dev | 18:04 | |
| *** openstackjenkins has quit IRC | 18:06 | |
| *** openstackjenkins has joined #openstack-dev | 18:07 | |
| *** openstackjenkins has quit IRC | 18:12 | |
| *** openstackjenkins has joined #openstack-dev | 18:13 | |
| *** bcwaldon has joined #openstack-dev | 18:14 | |
| mtaylor | jaypipes, dprince: if you're bored, can you try going to jenkins.openstack.org and verify that after you log in you have administrative privileges? | 18:14 |
| *** openstackjenkins has quit IRC | 18:15 | |
| *** openstackjenkins has joined #openstack-dev | 18:16 | |
| *** openstackjenkins has joined #openstack-dev | 18:17 | |
| jaypipes | mtaylor: sure, one minute. | 18:18 |
| mtaylor | jaypipes: (you should also notice that it autmatically logs you in using lauchpad openid service) | 18:18 |
| *** openstackjenkins has quit IRC | 18:20 | |
| *** openstackjenkins has joined #openstack-dev | 18:21 | |
| *** antonyy has quit IRC | 18:25 | |
| *** antonyy has joined #openstack-dev | 18:25 | |
| *** vladimir3p has quit IRC | 18:30 | |
| *** vladimir3p has joined #openstack-dev | 18:31 | |
| mtaylor | vishy, devcamcar: if either of you get a chance, could you do me a favor and make sure that you can see a link on the side of http://jenkins.openstack.org/job/nova-param/ titled "Build Now" - and that you do _not_ see a link on http://jenkins.openstack.org/ titled "Manage Jenkins" | 18:34 |
| mtaylor | anotherj1sse: ^^ or you | 18:36 |
| openstackjenkins | Project nova build #958: SUCCESS in 2 min 51 sec: http://jenkins.openstack.org/job/nova/958/ | 18:43 |
| openstackjenkins | Tarmac: - move osapi-specific wsgi code from nova/wsgi.py to nova/api/openstack/wsgi.py | 18:43 |
| openstackjenkins | - refactor wsgi modules to use more object-oriented approach to wsgi request handling: | 18:43 |
| openstackjenkins | - Resource object steps up to original Controller position | 18:43 |
| openstackjenkins | - Resource coordinates deserialization, dispatch to controller, serialization | 18:43 |
| openstackjenkins | - serialization and deserialization broken down to be more testable/flexible | 18:43 |
| openstackjenkins | - this will definitely help fixing current serialization-related bugs | 18:43 |
| openstackjenkins | - this paves the way for schema validation to be implemented elegantly | 18:43 |
| openstackjenkins | Project nova-tarball-bzr-delta build #206: FAILURE in 9.9 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/206/ | 18:48 |
| openstackjenkins | Tarmac: - move osapi-specific wsgi code from nova/wsgi.py to nova/api/openstack/wsgi.py | 18:48 |
| openstackjenkins | - refactor wsgi modules to use more object-oriented approach to wsgi request handling: | 18:48 |
| openstackjenkins | - Resource object steps up to original Controller position | 18:48 |
| openstackjenkins | - Resource coordinates deserialization, dispatch to controller, serialization | 18:48 |
| openstackjenkins | - serialization and deserialization broken down to be more testable/flexible | 18:48 |
| openstackjenkins | - this will definitely help fixing current serialization-related bugs | 18:49 |
| openstackjenkins | - this paves the way for schema validation to be implemented elegantly | 18:49 |
| mtaylor | creiht, gholt, notmyname: if you get a chance, could you do me a favor and make sure that you can see a link on the side of http://jenkins.openstack.org/job/swift-param/ titled "Build Now" - and that you do _not_ see a link on http://jenkins.openstack.org/ titled "Manage Jenkins" | 18:52 |
| bcwaldon | devcamcar: you around? | 18:53 |
| notmyname | mtaylor: I do not see a "manage jenkins" link for the second. I also do not see a "build now" link on the first | 18:54 |
| mtaylor | notmyname: hrm. that's exciting and then less exciting | 18:55 |
| mtaylor | notmyname: thanks! | 18:55 |
| *** Shyam_ has quit IRC | 18:56 | |
| creiht | mtaylor: I see the same thing as notmyname | 18:57 |
| *** openstackjenkins has quit IRC | 18:59 | |
| *** openstackjenkins has joined #openstack-dev | 18:59 | |
| *** openstackjenkins has quit IRC | 19:00 | |
| *** openstackjenkins has joined #openstack-dev | 19:00 | |
| openstackjenkins | Project nova build #959: SUCCESS in 2 min 44 sec: http://jenkins.openstack.org/job/nova/959/ | 19:03 |
| openstackjenkins | Tarmac: This just fixes a bunch of pep8 issues that have been lingering around for a while and bothering me :) | 19:03 |
| *** markvoelker has quit IRC | 19:04 | |
| soren | mtaylor: what? | 19:05 |
| soren | mtaylor: what? | 19:05 |
| soren | mtaylor: what? | 19:05 |
| soren | mtaylor: what? | 19:05 |
| mtaylor | soren: YAY! | 19:05 |
| mtaylor | soren: so what's the current deal with trunk packages, python-glance and libvirt? | 19:06 |
| mtaylor | soren: I don't seem to be able to install recent debs due to libvirt and python-glance depend problems - even if I add the nova-core ppa | 19:06 |
| soren | mtaylor: Is there a deal with all those tings? | 19:06 |
| soren | Err. | 19:07 |
| soren | Whuh? | 19:07 |
| mtaylor | soren: if I copy debs built by nova-binaries to a pristine box, add them to a local apt-ftparchive archive, and then try to install them via apt, they are uninstallable | 19:07 |
| dprince | mtaylor: There have recently been additions to glance/nova. | 19:07 |
| dprince | mtaylor: python-glance now supports filtering. | 19:07 |
| mtaylor | soren: http://jenkins.openstack.org/job/nova-smoketests/96/console | 19:08 |
| soren | mtaylor: How does that relate to anything I'm doing? | 19:08 |
| mtaylor | soren: well, if there is a packaging depend, there should be some way to satisfy it, yeah? I guess I'm just wondering what that is | 19:08 |
| dprince | http://jenkins.openstack.org/job/nova-smoketests/96/console | 19:08 |
| dprince | mtaylor: Is this on a clean machine? | 19:10 |
| vishy | mtaylor: verified both | 19:10 |
| mtaylor | dprince: yup | 19:10 |
| mtaylor | vishy: and it _did_ work for you? sweet | 19:10 |
| mtaylor | (hrm. why does it work for nova-core people and not swift-core people... me goes digging) | 19:11 |
| vishy | mtaylor: radioedit on cloudservers we have to force an older version of libvirt | 19:11 |
| vishy | perhaps you need to do the same? | 19:12 |
| mtaylor | vishy: well, the package has a hard depend on a newer one | 19:12 |
| vishy | mtaylor: ah, yes we aren't installing from packages | 19:12 |
| vishy | i don't know why the new libvirt won't install on cloudserver | 19:13 |
| mtaylor | vishy: indeed... but it's kind of a problem if we produce unusable packages :) | 19:13 |
| vishy | mtaylor: correct, although it works on other installs. Could it be the lack of a kernel module in the cloudservers basic image | 19:14 |
| mtaylor | vishy: possibly - but it's not a code-level failure - it's a lack of existence of a package | 19:14 |
| mtaylor | I'm not even as far as figuring out why the code may be unhappy | 19:15 |
| *** openstackjenkins has quit IRC | 19:16 | |
| *** openstackjenkins has joined #openstack-dev | 19:17 | |
| soren | mtaylor: How can I reproduce this? | 19:18 |
| vishy | mtaylor: checking why we had to force an older libvirt. I'm thinking it is because the newer one depends on a kernel module that isn't available on cloudservers | 19:18 |
| soren | No. | 19:18 |
| vishy | so it fails to install, which i think is the issue you are seeing? | 19:18 |
| soren | It depends being able to find a PCI bus. | 19:18 |
| * soren searches for bug reference | 19:19 | |
| openstackjenkins | Project nova-tarball-bzr-delta build #207: STILL FAILING in 9.7 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/207/ | 19:19 |
| openstackjenkins | Tarmac: This just fixes a bunch of pep8 issues that have been lingering around for a while and bothering me :) | 19:19 |
| soren | https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/790837 | 19:19 |
| uvirtbot | Launchpad bug 790837 in libvirt "Fails to start inside PV domU" [Medium,Triaged] | 19:19 |
| mtaylor | so - the issue I'm seeing is purely at the apt-get install step - it fails because nova-core depends on libvirt-bin >= 0.8.8 | 19:20 |
| soren | What the hell is nova-core? | 19:20 |
| mtaylor | haha. I meant nova-compute | 19:20 |
| mtaylor | nova-core is a team :) | 19:20 |
| soren | Ah. | 19:20 |
| vishy | soren: ah that explains it | 19:21 |
| vishy | soren: thx | 19:21 |
| soren | Right, so it depends on libvirt 0.8.8. | 19:21 |
| mtaylor | but there is no source for libvirt-bin >= 0.8.8 on maverick as far as I can tell. on the other hand, if it installs, I'm guessing I'll have operational problems because libvirt-bin >= 0.8.8 doesn't work on cloud servers | 19:21 |
| soren | mtaylor: ?!? | 19:21 |
| soren | It's *right* there. | 19:21 |
| mtaylor | where? | 19:21 |
| soren | libvirt - 0.8.8-1ubuntu3~ppamaverick1 | 19:21 |
| soren | nova-trunk ppa. | 19:21 |
| soren | nova-core/trunk ppa, that is. | 19:22 |
| mtaylor | ah | 19:22 |
| mtaylor | nova-trunk | 19:22 |
| mtaylor | I was installing nova-core/ppa, not nova-core/trunk. thanks. that was actually the answer I was looking for all along :) | 19:22 |
| soren | vishy: So you guys aren't installing from packages? | 19:22 |
| soren | nova-core/ppa needs to *die*. | 19:22 |
| mtaylor | great! | 19:23 |
| mtaylor | good to know | 19:23 |
| soren | As soon as we find a good home for whatever extra packages we need for Jenkins, it's gone. | 19:23 |
| vishy | soren: for radioedit we are running using nova.sh | 19:23 |
| soren | I have *no* idea what radioedit is. | 19:24 |
| *** openstackjenkins has quit IRC | 19:24 | |
| soren | openstackjenkins.. I miss you already. | 19:24 |
| *** openstackjenkins has joined #openstack-dev | 19:24 | |
| soren | whoo! | 19:24 |
| *** openstackjenkins has quit IRC | 19:25 | |
| soren | aw | 19:25 |
| *** openstackjenkins has joined #openstack-dev | 19:25 | |
| mtaylor | soren: excellent | 19:26 |
| mtaylor | creiht: could I trouble you again to check http://jenkins.openstack.org/job/swift-param/ and see if you have the build now link now? | 19:27 |
| creiht | mtaylor: yes, I see it now | 19:28 |
| mtaylor | creiht: woohoo! | 19:28 |
| creiht | :) | 19:28 |
| mtaylor | ladies and gentlemen... we finally have sensible group-controls on being able to do stuff on jenkins! | 19:29 |
| *** openstackjenkins has quit IRC | 19:29 | |
| vishy | mtaylor: \o/ | 19:30 |
| *** openstackjenkins has joined #openstack-dev | 19:30 | |
| mtaylor | soren: so, also, I want to take the scripts we have on the filesystem on the jenkins server (serveral shell scripts and a python program) and check them in to bzr ... are there any of them you can think of off the top of your head that have passwords or keys in them? | 19:37 |
| mtaylor | soren: I don't think there are - well, other than my python program which I will fix - but I thought I'd check | 19:37 |
| soren | mtaylor: No secret sauce. | 19:39 |
| mtaylor | wootr | 19:39 |
| soren | mtaylor: Only ugly hacks that never should see the light of day, but that's a luxury I've given up many years ago. | 19:39 |
| mtaylor | soren: yup. the glory of working in the sunlight all the time :) | 19:40 |
| heckj | I'd love to get a copy of the jenkins setup and associated scripts - I'd like to replicate the testing/building locally | 19:42 |
| *** openstackjenkins has quit IRC | 19:45 | |
| *** openstackjenkins has joined #openstack-dev | 19:46 | |
| *** dprince has quit IRC | 19:48 | |
| ttx | markwash: no they don't. The release branch was cut this morning. | 19:53 |
| ttx | markwash: only critical bugfixes go into the release branch. | 19:53 |
| soren | Man... It's June now, and when I read "2011" it still looks wrong. | 19:54 |
| ttx | markwash: so it depends on what you mean by "in". | 19:55 |
| ttx | markwash: critical bugfixes that land in the milestone-proposed branch today will be in d1. Features and other bugfixes that land in trunk won't. | 19:56 |
| ttx | markwash: see http://wiki.openstack.org/BranchModel | 19:57 |
| *** vladimir3p has quit IRC | 19:58 | |
| *** vladimir3p has joined #openstack-dev | 19:59 | |
| *** johnpur has quit IRC | 19:59 | |
| mtaylor | heckj: so, that's possible - but is there any reason that would be a benefit as opposed to just launching builds on the openstack jenkins? (not trolling, curious) | 20:00 |
| openstackjenkins | Project swift build #272: SUCCESS in 27 sec: http://jenkins.openstack.org/job/swift/272/ | 20:01 |
| openstackjenkins | Tarmac: replicator: Log timings for replicator-rpc sync method if debug mode is on. | 20:01 |
| ttx | jaypipes: sure, I want it in too. | 20:03 |
| markwash | ttx thanks! | 20:20 |
| ttx | markwash: no pb :) | 20:20 |
| vishy | ttx: we confirmed the switch to 2 reviewers a day yes? | 20:21 |
| ttx | vishy: yes, soren is supposed to have implemented it. | 20:21 |
| markwash | vishy: its there | 20:21 |
| ttx | indeed. | 20:21 |
| heckj | mtalyor: figured having the builds running in parallel would keep from (or reduce) any accidental "environment" setups. Plus I'd like to learn more about how it's set up and contribute additional testing mechanisms in the future (functional/smoke - potentially against my own HW infrasructure) | 20:22 |
| ttx | vishy: d1 looking good so far ? | 20:23 |
| ttx | vishy: my limited testing was pretty successful -- haven't touched the new features though. | 20:23 |
| vishy | ttx: haven't done testing yet...catching up on other things. so we'll see this afternoon :) | 20:26 |
| vishy | ttx: if we find a critical bug, what is the process for merging into release? Is it automated? Or just a manual push? | 20:27 |
| ttx | you propose the fix to the milestone-proposed branch and vreiew/approve it there | 20:28 |
| openstackjenkins | Project nova build #960: SUCCESS in 2 min 44 sec: http://jenkins.openstack.org/job/nova/960/ | 20:29 |
| openstackjenkins | Tarmac: Support SSL AMQP connections. | 20:29 |
| ttx | vishy: if it's "approved" it will get picked up by the Jenkins magic | 20:29 |
| ttx | and a new candidate tarball built. | 20:29 |
| ttx | vishy: see my email to the list for the pointer to the milestone-proposed branch. | 20:30 |
| vishy | ttx: cool | 20:30 |
| ttx | vishy: next milestone we should have at least two days of baking | 20:30 |
| ttx | one day is a bit short. | 20:30 |
| vishy | agreed :) | 20:32 |
| vishy | i'm hoping that by next milestone monty's testing stuff is so stellar that we don't even need to bake :) | 20:32 |
| ttx | vishy: oh right ! I forgot about that :) | 20:32 |
| openstackjenkins | Project nova-tarball-bzr-delta build #208: STILL FAILING in 10 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/208/ | 20:33 |
| openstackjenkins | Tarmac: Support SSL AMQP connections. | 20:33 |
| vishy | ruh ro...tarball build fail | 20:34 |
| *** ameade has quit IRC | 20:34 | |
| vishy | Resolving jenkins.openstack.org... 173.203.104.204 | 20:35 |
| vishy | Reusing existing connection to hudson.openstack.org:80. | 20:35 |
| vishy | HTTP request sent, awaiting response... 403 Forbidden | 20:35 |
| ttx | mtaylor: looks like your auth stuff broke some jobs | 20:36 |
| ttx | mtaylor: sounds like a very bad idea on a milestone brewing day. | 20:37 |
| bcwaldon | Can someone make a decision on this MP? | 20:44 |
| bcwaldon | https://code.launchpad.net/~dan-prince/nova/chown_vdb_device/+merge/57214 | 20:44 |
| bcwaldon | It's been in the queue for quit a while | 20:44 |
| *** lorin1 has quit IRC | 20:54 | |
| openstackjenkins | Project nova build #961: SUCCESS in 2 min 44 sec: http://jenkins.openstack.org/job/nova/961/ | 20:59 |
| openstackjenkins | Tarmac: Updates to the 018_rename_server_management_url migration to avoid adding and dropping a column. Just simply rename the column. | 20:59 |
| vishy | bcwaldon: does that code run in nova-compute or in xen dom0? | 20:59 |
| bcwaldon | vishy: no idea, that's why I didn't review it at all | 20:59 |
| bcwaldon | devcamcar: you here? | 21:00 |
| vishy | if it is running in domU I'm totally ok with it going in as is then trying to fix the chown stuff later | 21:00 |
| bcwaldon | vishy: I suggest putting that on the official MP :) | 21:00 |
| mtaylor | ttx: really? | 21:01 |
| vishy | :) | 21:01 |
| mtaylor | ttx: can you point me to the jobs? | 21:01 |
| mtaylor | ttx: (there is no reason that it should affect anything in that way...) | 21:01 |
| bcwaldon | vishy: I haven't heard from ewan lately, either. Not sure if he is still in a state of disapproval | 21:01 |
| soren | mtaylor: You'll be surprised. | 21:01 |
| mtaylor | soren: I do not like being surprised | 21:01 |
| soren | mtaylor: I haven't looked, but I imagine it's my pulling of tarballs from /lastSuccessfulBuild/.../dist.zip | 21:02 |
| soren | mtaylor: ...that no longer works. | 21:02 |
| soren | ttx: Am I right? | 21:02 |
| mtaylor | soren: AH. yes. gotcha. lemme fix that real quick | 21:02 |
| soren | \o/ | 21:02 |
| mtaylor | soren: any idea which jobs do that? | 21:02 |
| * soren .oO{ how? } | 21:02 | |
| mtaylor | soren: by using the copy artifacts feature | 21:03 |
| soren | Yeah, all the PPA ones. | 21:03 |
| mtaylor | ok. looking right now | 21:03 |
| mtaylor | sorry ttx | 21:03 |
| soren | Oh, that. | 21:03 |
| openstackjenkins | Project nova-tarball-bzr-delta build #209: STILL FAILING in 9.3 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/209/ | 21:03 |
| openstackjenkins | Tarmac: Updates to the 018_rename_server_management_url migration to avoid adding and dropping a column. Just simply rename the column. | 21:03 |
| soren | mtaylor: Oh, heh. | 21:06 |
| bcwaldon | soren: Can you address Todd's comment in https://code.launchpad.net/~xtoddx/nova/provider-fw-rules/+merge/62517 | 21:06 |
| soren | mtaylor: The nova-tarball-bzr-delta one, too. | 21:06 |
| soren | bcwaldon: sure thing. | 21:07 |
| mtaylor | yup. fixing that one first (it's smaller) | 21:07 |
| bcwaldon | soren: Thank you, sir | 21:07 |
| openstackjenkins | Project nova-tarball-bzr-delta build #210: STILL FAILING in 8.8 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/210/ | 21:08 |
| openstackjenkins | Project nova-tarball-bzr-delta build #211: STILL FAILING in 8.8 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/211/ | 21:11 |
| vishy | happy day smoketests pass for checkout of milestone-proposed. Now to try from packages | 21:14 |
| mtaylor | vishy: yay! I'm still getting various failures in the smoketests job, but as we've posited, I'm guessing that's due to config assumptions... | 21:15 |
| vishy | well this should be exciting | 21:21 |
| vishy | trying using vagrant/chef | 21:21 |
| *** ohnoimdead has joined #openstack-dev | 21:23 | |
| vishy | mtaylor, soren: any chance of packaging nova-adminclient so i can install it from packages? | 21:29 |
| mtaylor | vishy: it's already done | 21:29 |
| mtaylor | vishy: so, yes! | 21:29 |
| vishy | o rly? | 21:29 |
| mtaylor | yup | 21:30 |
| vishy | is it hiding in a ppa somewhere? | 21:30 |
| mtaylor | package is python-nova.adminclient and it's in the trunk ppa | 21:31 |
| vishy | cool thx | 21:31 |
| mtaylor | soren: I updated the nova-ppa job and it seems to be happy... can you verify all looks good? | 21:39 |
| soren | mtaylor: Just click "build now". It should be safe. | 21:41 |
| mtaylor | soren: I did. it worked... but I thought I'd have you eyeball to make sure | 21:41 |
| soren | mtaylor: I'm all out of eyeballs for the day. | 21:52 |
| mtaylor | k | 21:53 |
| vishy | mtaylor: I don't see it here: https://launchpad.net/~nova-core/+archive/trunk | 21:58 |
| vishy | are you sure you are not thinking of python-novaclient? | 21:59 |
| *** foxtrotgulf has quit IRC | 22:00 | |
| mtaylor | vishy: hrm.... well, I'm installing it from packages in the smoketests job | 22:02 |
| vishy | weird it does seem to work | 22:04 |
| vishy | even though i don't see it listed | 22:04 |
| vishy | :) | 22:04 |
| *** troytoman-away is now known as troytoman | 22:05 | |
| vishy | i was hoping to copy it into the milestone-proposed ppa | 22:05 |
| mtaylor | vishy: yes. I'm very confused... I don't even see it browsing through the package pool by hand | 22:05 |
| vishy | is it possible it is in another package? | 22:06 |
| soren | I can explain. | 22:07 |
| soren | It's a leftover from when it was still included in Nova. | 22:07 |
| vishy | soren: 2011.2~bzr934-0ubuntu0ppa1~maverick1 | 22:07 |
| vishy | yeah i just noticed that :) | 22:08 |
| soren | Right. Ancient. | 22:08 |
| *** ecarlin has joined #openstack-dev | 22:08 | |
| vishy | that explaings some of the smoketest failures that mony was seeing :) | 22:08 |
| soren | mtaylor: You'll find it in the nova directory. | 22:08 |
| soren | mtaylor: Probably. | 22:08 |
| mtaylor | vishy: AH. that would explain that | 22:09 |
| vishy | considering we are pretty much deprecating it, I don't know if it is a huge issue to get it packaged | 22:09 |
| vishy | but it does cause failures in smoketests/test_admin if it isn't installed | 22:09 |
| soren | I do seem to remember ttx working on the packaging. | 22:10 |
| mtaylor | vishy: would it be easier for current smoketest testing to just install it from elsewhere? | 22:11 |
| bcwaldon | Quick question for core-devs in the room | 22:11 |
| bcwaldon | What do we do about making this MP go away? | 22:11 |
| bcwaldon | https://code.launchpad.net/~reldan/nova/lp776679/+merge/59832 | 22:11 |
| mtaylor | vishy: (as in, if I easy_install it or something, will it make my tests go greener?) | 22:11 |
| bcwaldon | Assuming the Disapproves are not going to change | 22:11 |
| vishy | yes install via pip works | 22:11 |
| mtaylor | bcwaldon: you can always set the status of it to rejected or to work in progress, that will take it out of the to-be-reviewed list | 22:12 |
| vishy | bcwaldon: mark it rejected | 22:12 |
| bcwaldon | rejected it is, work in progress implies it still has hope | 22:12 |
| mtaylor | vishy: one last one on that - using the adminclient from packages, I was having to do this: s/nova_adminclient/nova.adminclient/ on test_admin.py ... can I skip that now? | 22:14 |
| vishy | yup | 22:14 |
| soren | bcwaldon: i don't understand their objection, though. | 22:15 |
| markwash | vishy: the merge prop to switch from MyISAM to InnoDB. . should that setting be a deployers decision? | 22:15 |
| bcwaldon | soren: would you prefer I set it back to needs review and allow new discussion to take place? | 22:16 |
| soren | bcwaldon: I think so, yeah. I think the change makes sense, I don't understand the objection. | 22:17 |
| bcwaldon | soren: Ok, will do. | 22:17 |
| soren | bcwaldon: ta | 22:17 |
| soren | Isn't it funny that half the chatter in the *net*stack meeting is about people losing their connections? | 22:18 |
| vishy | so i found out why we had pushd / popd in the novarc file | 22:24 |
| vishy | :) | 22:24 |
| vishy | so it works on a mac :) | 22:25 |
| soren | How does the new stuff fail? | 22:26 |
| soren | (which shell do you have on a mac?) | 22:26 |
| *** bcwaldon has quit IRC | 22:30 | |
| devcamcar | mtaylior: just got your message about jenkins, but everything looks good | 22:30 |
| mtaylor | devcamcar: sweet. thanks man | 22:31 |
| devcamcar | bcwaldon: I'm around now | 22:31 |
| mtaylor | markwash: jumping in where I have no context, but there is honestly zero reason in a sane system to use MyISAM... so it _can_ be a deployer's decision, but the default should be the sane one and if they want to set themselves up to shoot themselves in the foot, we should make them consciously make that decision | 22:32 |
| mtaylor | markwash: not to mention, innodb is the mysql default as of 5.5 anyway | 22:32 |
| * mtaylor goes back to his hole | 22:32 | |
| markwash | mtaylor: agree (at least for our use case) | 22:33 |
| *** bcwaldon has joined #openstack-dev | 22:33 | |
| bcwaldon | devcamcar: excellent. There were a couple of branches I would appreciate you looking at | 22:33 |
| markwash | mtaylor: so if sqlalchemy is somehow doing this (setting up myisam) then I guess we need to clean up after it | 22:34 |
| devcamcar | bcwaldon: sure, which ones? | 22:34 |
| bcwaldon | devcamcar: i found at least one where you had a needs information | 22:34 |
| bcwaldon | devcamcar: give me one sec, i'll find it | 22:34 |
| devcamcar | bcwaldon: thanks, been traveling a lot so reviews have been a bit sporadic :) | 22:34 |
| mtaylor | markwash: it may just be _not_ explicitly setting anything, which on a default install will default to myisam ... but I believe there is a sqlalchemy setting somewhere to override that | 22:34 |
| *** ecarlin has quit IRC | 22:35 | |
| bcwaldon | devcamcar: Here's one: https://code.launchpad.net/~rackspace-titan/nova/osapi_v11_resize/+merge/62786 | 22:35 |
| devcamcar | bcwaldon: oh yea, i remember that one. i am still seeing some weirdness in the diff that lauchpad is putting up | 22:36 |
| devcamcar | bcwaldon: i still see two versions of the _action_resize method | 22:37 |
| mtaylor | vishy: wow. changing to easy_install nova-adminclient had ZERO effect :) | 22:37 |
| devcamcar | i'll pull down source and visually inspect | 22:37 |
| bcwaldon | devcamcar: there should be three, actually. One in Controller, one in ControllerV10 and one in ControllerV11 | 22:37 |
| vishy | hehe :) | 22:37 |
| vishy | oh i noticed some duplicated methods in tests as well recently | 22:37 |
| devcamcar | bcwaldon: yep, confusing from the diff, going to visually inspect and then approve | 22:37 |
| vishy | mtaylor: output? | 22:38 |
| mtaylor | http://jenkins.openstack.org/job/nova-smoketests/105/#showFailuresLink | 22:38 |
| mtaylor | vishy: but don't spend too much time looking at it - I can sort it out (as I'm sure it's a monty/config issue) | 22:39 |
| devcamcar | bcwaldon: marked it as approved :) | 22:40 |
| *** mattray has quit IRC | 22:40 | |
| devcamcar | were there any others in particular you can think of? | 22:40 |
| bcwaldon | It may have just been that one | 22:40 |
| bcwaldon | Thanks for addressing that, though | 22:40 |
| vishy | mtaylor: test_admin test 1 looks like a timing issue since the rest of the commands work. Perhaps nova-manage user admin hasn't run yet? | 22:41 |
| bcwaldon | waldon out | 22:42 |
| *** bcwaldon has quit IRC | 22:42 | |
| mtaylor | vishy: it should have - unless there is lag between running it and being able to use it | 22:42 |
| vishy | the other ones look like a config error, really need traceback from nova-api to see what is up | 22:42 |
| vishy | is your code starting nova-objectstore? | 22:42 |
| *** bcwaldon has joined #openstack-dev | 22:44 | |
| *** bcwaldon has quit IRC | 22:46 | |
| vishy | good news | 22:47 |
| vishy | smoketests pass from packages as well | 22:47 |
| openstackjenkins | Project nova build #962: SUCCESS in 2 min 46 sec: http://jenkins.openstack.org/job/nova/962/ | 22:54 |
| openstackjenkins | Tarmac: Implement the v1.1 style resize action with support for flavorRef. | 22:54 |
| openstackjenkins | Project nova-tarball-bzr-delta build #213: FAILURE in 9.4 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/213/ | 22:58 |
| openstackjenkins | Tarmac: Implement the v1.1 style resize action with support for flavorRef. | 22:58 |
| *** rnirmal has quit IRC | 23:01 | |
| *** troytoman is now known as troytoman-away | 23:03 | |
| mtaylor | vishy: sweet. | 23:12 |
| mtaylor | vishy: I had a problem running the smoketests against packages also because of a flags conflict... | 23:13 |
| mtaylor | vishy: is there anything special you're doing to run them? | 23:13 |
| vishy | i install with chef | 23:13 |
| vishy | or nova.sh | 23:13 |
| vishy | chef for packages | 23:13 |
| vishy | then run them normally | 23:13 |
| *** jkoelker has quit IRC | 23:13 | |
| vishy | although I run them from outside the vm | 23:13 |
| mtaylor | sure- but from where do you get the smoketests code themselves - a tarball/checkout of the tree, yeah? | 23:13 |
| mtaylor | AH | 23:14 |
| vishy | it should work either way | 23:14 |
| vishy | i do it from trunk checkout | 23:14 |
| mtaylor | k. I'll keep poking. | 23:15 |
| *** RobertLaptop has quit IRC | 23:17 | |
| *** jtran has joined #openstack-dev | 23:30 | |
| openstackjenkins | Project nova build #963: SUCCESS in 2 min 43 sec: http://jenkins.openstack.org/job/nova/963/ | 23:34 |
| openstackjenkins | Tarmac: Fixes nova-manage bug. When a nova-network host has allocated floating ips *AND* some associated, the nova-manage floating list <host> would throw exception because was expecting hash with 'ec2_id' key , however, the obj returned is a sqlalchemy obj and the attr we need is 'hostname'. | 23:34 |
| openstackjenkins | Project nova-tarball-bzr-delta build #214: STILL FAILING in 9.9 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/214/ | 23:38 |
| openstackjenkins | Tarmac: Fixes nova-manage bug. When a nova-network host has allocated floating ips *AND* some associated, the nova-manage floating list <host> would throw exception because was expecting hash with 'ec2_id' key , however, the obj returned is a sqlalchemy obj and the attr we need is 'hostname'. | 23:38 |
| *** antonyy has quit IRC | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!