Tuesday, 2011-06-28

*** heckj has quit IRC00:01
*** ohnoimdead has quit IRC00:10
*** dragondm has quit IRC00:10
*** negronjl has quit IRC00:10
*** s1rp has quit IRC00:10
*** ohnoimdead has joined #openstack-dev00:13
*** dragondm has joined #openstack-dev00:13
*** negronjl has joined #openstack-dev00:13
*** s1rp has joined #openstack-dev00:13
*** fyang has left #openstack-dev00:18
*** masumotok has joined #openstack-dev00:24
*** mszilagyi has quit IRC00:29
openstackjenkinsProject nova build #1,042: SUCCESS in 2 min 56 sec: http://jenkins.openstack.org/job/nova/1042/00:29
openstackjenkinsTarmac: Added floating IP support in OS API00:29
*** masumotok__ has joined #openstack-dev00:30
*** masumotok has quit IRC00:31
*** ohnoimdead has quit IRC00:31
*** masumotok has joined #openstack-dev00:34
*** masumotok__ has quit IRC00:36
*** Binbin has joined #openstack-dev00:42
*** masumotok has quit IRC00:43
*** masumotok has joined #openstack-dev00:43
*** dragondm has quit IRC00:56
*** mgius is now known as mgius-away01:02
*** cloudgroups has joined #openstack-dev01:04
*** winston-d has joined #openstack-dev01:13
*** cloudgroups has left #openstack-dev01:26
*** mattray has joined #openstack-dev01:33
*** lorin1 has quit IRC01:36
*** foxtrotgulf has joined #openstack-dev02:07
*** fyang has joined #openstack-dev02:19
*** fyang has left #openstack-dev02:20
*** alekibango has quit IRC02:36
*** cloudgroups has joined #openstack-dev02:49
*** mattray has quit IRC02:52
*** alekibango has joined #openstack-dev02:59
HugoKuo__glance 145   seems failed to build , the latest version with trunk ppa will get glance 142 ..03:01
*** Binbin has quit IRC03:23
*** Binbin has joined #openstack-dev03:24
*** ohnoimdead has joined #openstack-dev03:29
*** ohnoimdead has quit IRC03:30
*** cloudgroups has quit IRC03:36
*** cloudgroups has joined #openstack-dev03:36
*** Zangetsue has joined #openstack-dev03:40
*** foxtrotgulf has quit IRC03:40
*** nati has joined #openstack-dev04:03
*** avati_ has joined #openstack-dev04:27
*** avati has quit IRC04:27
*** _cerberus_ has quit IRC04:28
*** _cerberus_ has joined #openstack-dev04:28
*** cloudgroups has quit IRC05:01
*** avati_ is now known as avati05:43
*** Binbin has quit IRC06:15
*** Binbin has joined #openstack-dev06:20
*** nati has quit IRC06:29
*** negronjl has quit IRC06:35
*** reidrac has joined #openstack-dev07:07
*** dragondm has joined #openstack-dev07:27
*** zaitcev has quit IRC07:34
*** nati has joined #openstack-dev08:04
*** chemikadze has joined #openstack-dev08:33
*** negronjl has joined #openstack-dev08:38
*** winston-d has quit IRC08:45
*** Vasichkin has joined #openstack-dev08:49
*** chemikadze has quit IRC08:54
*** chemikadze has joined #openstack-dev08:56
*** cloudgroups has joined #openstack-dev09:08
*** nati has quit IRC09:32
*** cloudgroups has left #openstack-dev09:47
*** Binbin has quit IRC10:06
*** MarkusWPHRorg has quit IRC10:24
*** dragondm has quit IRC10:31
*** nati has joined #openstack-dev10:53
*** nati has quit IRC10:59
*** markvoelker has joined #openstack-dev11:21
ttxHugoKuo__: indeed. Fixed.11:58
HugoKuo__ttx, yes thanks for your reply ...11:59
HugoKuo__an additional point....11:59
HugoKuo__is "security group"  works now ?12:00
HugoKuo__well , I tried on r1215 , it seems no any effect ...12:00
HugoKuo__not sure if that's a bug XD12:00
ttxHugoKuo__: there are quite a few bugs around that... but nothing as definitive as "no effect"12:01
ttxHugoKuo__: filing a bug could help... worst case it will be marked as duplicate.12:02
HugoKuo__did it :>12:02
HugoKuo__https://bugs.launchpad.net/bugs/80289412:02
uvirtbotLaunchpad bug 802894 in nova "Security group no effect...." [Undecided,New]12:02
HugoKuo__I can access instance without any security rule .12:03
ttxAh! very recent, I see :)12:03
HugoKuo__got go home see ya12:06
*** negronjl has quit IRC12:09
*** spurvewt has joined #openstack-dev12:30
*** spurvewt has left #openstack-dev12:33
*** bcwaldon has joined #openstack-dev12:46
ttxbcwaldon: any reason why you did not set https://code.launchpad.net/~usc-isi/nova/instance_type_extra_specs/+merge/62728 to "Approved" ?12:51
ttxlooks like 2 core approvals to me...12:52
bcwaldonttx: was waiting on another core dev reviewer12:52
bcwaldonbut I can send it in12:52
ttxbcwaldon: Vish approved it... you want a 3rd look ?12:52
bcwaldonThat was the idea. But it doesn't really matter. I can approve it12:53
ttxbcwaldon: would be good, so that it ends up in diablo-212:54
bcwaldonyep, on it12:54
bcwaldonttx: approved12:56
ttxsandywalsh: any particular reason why you marked https://bugs.launchpad.net/nova/+bug/798361 fixcommitted ? Looks like the merge proposal is still in progress ?12:56
uvirtbotLaunchpad bug 798361 in nova "Zone Aware Scheduler fails to create db entry for locally provisioned instances" [Medium,Fix committed]12:56
ttxbcwaldon: thx12:56
*** negronjl has joined #openstack-dev13:05
*** lorin1 has joined #openstack-dev13:08
openstackjenkinsProject nova build #1,043: SUCCESS in 3 min 8 sec: http://jenkins.openstack.org/job/nova/1043/13:09
openstackjenkinsTarmac: Adds support for "extra specs", additional capability requirements associated with instance types.13:09
openstackjenkinsThe instance_type dictionary now has a new extra_specs field.13:09
openstackjenkinsAdds a new table to the database: InstanceTypeExtraSpecs. This is modeled on the existing InstanceMetadata table, except that it is associated with additional capability requirements of instance types rather than generic metadata for instances.13:09
openstackjenkinsThe InstanceTypeFilter has been modified to check for extra specs.13:09
openstackjenkinsThis will ultimately be needed for supporting heterogeneous instances: we'll annotate the instance types with info about whether GPUs are present, so users will be able to request something like a "cg1.4xlarge".13:09
openstackjenkinsIncludes api support as an extension for querying and modifying this info.13:10
ttxlorin1: ^ yay13:13
*** ameade has joined #openstack-dev13:13
lorin1ttx: thanks. :)13:15
ttxlorin1: the other branch will probably not make it -- is it necessary for the feature to be considered complete ?13:15
ttxlorin1: i.e. can we mark the spec complete even if that branch hasn't landed ?13:16
lorin1ttx: Let me take a quick look at what's in the second branch.13:16
ttxlorin1: the example heterogeneous scheduler13:17
*** dprince has joined #openstack-dev13:18
lorin1ttx: I think we can declare success. The second merge prop has an example of a hetero scheduler, with support for kvm, but it's not strictly essential for supporting heterogeneity.13:18
ttxlorin1: ok, will do! thanks.13:18
lorin1ttx: The first merge prop extended the host filters, so if somebody uses those filters then they get the support for free.13:18
ttxvishy: ping me when you're around13:22
*** Zangetsue has quit IRC13:25
*** Zangetsue has joined #openstack-dev13:26
*** Zangetsue has quit IRC13:27
*** Zangetsue has joined #openstack-dev13:28
*** chemikadze has quit IRC13:39
*** mattray has joined #openstack-dev13:45
*** chemikadze has joined #openstack-dev13:54
*** jkoelker has joined #openstack-dev13:55
*** dprince has quit IRC14:07
*** fyang has joined #openstack-dev14:12
*** dprince has joined #openstack-dev14:16
*** Zangetsue has quit IRC14:23
*** chemikadze has quit IRC14:25
lorin1I'm getting some unit test failures when running inside a venv. Anybody else seeing that?14:50
*** chemikadze has joined #openstack-dev15:02
lorin1OK, looks like the tests setup() method isn't always getting called, logged it as a bug.15:04
sandywalshttx, pure stupidity :/ sorry 'bout that15:04
kpepplelorin1: my runs fine inside my virtualenv (other than pep8 errors)15:08
lorin1Try deleting clean.sqlite and just running a test in a subdir, e.g.: run_tests.sh api.test_wsgi15:09
lorin1kpepple: ^15:09
lorin1kpepple: (doesn't need to be inside a venv)15:09
*** reidrac has quit IRC15:10
kpepplelorin1: isn't that the special nova/test.py:46 magic that vishy added a while back15:12
vishylorin1: run_tests.sh -r ?15:14
vishywill recreate the db15:14
openstackjenkinsProject nova build #1,044: SUCCESS in 3 min 4 sec: http://jenkins.openstack.org/job/nova/1044/15:19
openstackjenkinsTarmac: Re-worked some of the WSGI and WSGIService code to make launching WSGI services easier, less error prone, and more testable. Added tests for WSGI server, new WSGI loader, and modified integration tests where needed.15:19
lorin1vishy: Oh, I didn't realize that this had changed.15:24
vishylorin1: I will make an announcement on the ml for the release so that everyone is aware of that one15:25
vishyperhaps i should have defaulted the other way?15:26
lorin1vishy: This may be a little pathological, but if I do "rm *.sqlite; ./run_tests.sh -r api.test_wsgi", it fails. Is that expected?15:26
vishylorin1: no, interesting15:26
vishylorin1: looks like we might have a bug15:26
lorin1vishy: I reported it as #803021.15:28
*** heckj has joined #openstack-dev15:36
openstackjenkinsProject nova build #1,045: SUCCESS in 3 min 3 sec: http://jenkins.openstack.org/job/nova/1045/15:39
openstackjenkinsTarmac: This branch allows LdapDriver to reconnect to LDAP server if connection is lost.15:39
ttxNova trunk switched to d315:46
*** vladimir3p has joined #openstack-dev15:55
*** ameade has quit IRC15:57
*** fyang has quit IRC16:02
openstackjenkinsProject nova-milestone build #2: SUCCESS in 13 sec: http://jenkins.openstack.org/job/nova-milestone/2/16:07
openstackjenkinsTarmac: Merge diablo-2 development work16:07
ttxGlance trunk switched to d316:15
* ttx pauses for a bit16:24
*** dragondm has joined #openstack-dev16:24
*** negronjl has quit IRC16:33
*** rnirmal has joined #openstack-dev16:40
*** chemikadze has quit IRC16:51
*** negronjl has joined #openstack-dev16:59
*** RobertLaptop has quit IRC17:00
*** RobertLaptop has joined #openstack-dev17:02
*** lorin1 has quit IRC17:07
*** negronjl has quit IRC17:19
*** MarkusWPHRorg has joined #openstack-dev17:31
*** negronjl has joined #openstack-dev17:32
*** cp16net has joined #openstack-dev18:00
*** dprince has quit IRC18:16
*** ameade has joined #openstack-dev18:28
*** johnpur has joined #openstack-dev18:32
heckjNova question about services and message queues - when a service gets initiated (like nova-api, nova-network, etc) - how does a topic get listed and generated for them rpc mechanisms (i.e. a queue defined)?18:50
heckjI'm starting to put down a little benchmark script and want to define a new "service" to run queue messages through and see what happens when we switch from in-memory queues to durable-queues.18:50
heckjThought the best way to do that would be to create a new service and use the nova rpc.call to measure things back and forth...18:51
*** foxtrotgulf has joined #openstack-dev19:09
ttxthis is a very calm day... not even enough entropy to push diablo-2 in the release process.19:22
johnpurttx: is that good?19:23
ttxjohnpur: no.19:23
ttxjohnpur: I need active core teams and PTLs in order to release a milestone :)19:25
johnpurttx: glance should be active, right?19:26
vishyheckj: services19:27
johnpurttx: what needs to be done for swift? is 1.4.2 the current milestone?19:27
vishyheckj: listen on two queues19:27
ttxjohnpur: Swift is OK, 1.4.2 is not this week.19:27
johnpuraha, there is vish!19:27
vishyheckj: there is a little magic that by default picks names based on the name of the binary, but you can overwrite it19:27
ttxvishy: did you have a chance to ping Jay ?19:28
heckjvishy: my thought was to spin up a dummy "service" using the same mechanisms that nova-network and such use, and then have one method that just returns quickly. I'm mostly interested in benchmarking the queueing mechanism and making sure there isn't a terrible performance hit if we implement durable queues19:28
heckjvishy: I found the name magic - that's fine, I can work with that19:29
vishyheckj: see init and start19:29
vishyif you define a manager it has all the logic for running as a service19:29
creiht /win 1219:29
ttxGeneral team meeting in 90 min in #openstack-meeting19:30
ttxjohnpur: glance is actually the one I have trouble with today -- can't get Jay19:30
bcwaldonttx: he's been out in SFO afaik19:30
bcwaldonttx: he's answering email off and on19:30
vishyheckj: check out test_service19:31
heckjvishy: perfect, thank you.19:31
vishyheckj: also you can make the tests run using real rabbit as well19:31
ttxbcwaldon: if he is not around by meeting time, I may ask you to ack the glance milestone-proposed merge on his behalf19:31
ttxbcwaldon: don't worry, will explain if necessary19:32
bcwaldonttx: I could probably do that. Explanation would be helpful19:32
vishyheckj: tests/fake_flags.py set fake_rabbit to default to False instead of True19:32
vishyyou could then test with different rabbit hosts using --rabbit_host if you want19:33
vishyttx: i told jay19:33
vishyhe was having internet issues19:34
vishylooks like there is a high priority bug to get in before release of glance19:34
ttxvishy: ok -- but we can add that one to the milestone-proposed branch -- would even be a good exercise.19:35
bcwaldonvishy: are you talking about the limit param bug filed today?19:35
ttxbcwaldon: ok -- so as you may know at some point in dev we branch out to a milestone-proposed branch (see http://wiki.openstack.org/BranchModel)19:35
ttxbcwaldon: we do so by merging back trunk (containing diablo-2 changes) into the milestone-proposed branch.19:36
ttxbcwaldon: so there is a merge proposal (for merging into milestone-proposed) containing all that -- and someone has to ack it19:36
bcwaldonttx: Gotcha. I saw that pop up today19:36
ttxbcwaldon: that's what i'll ask you to do if jay doesn't make it in the next 30 min.19:36
bcwaldonttx: Ok. As for this bugfix that went in today. Can we get that in milestone-proposed?19:37
ttxbcwaldon: it's actually quite simple, just needs to doublecheck I didn't fuck it up, by just comparing trunk and milestone-proposed (they should be equivalent)19:37
ttxbcwaldon: it's probably part of what I proposed19:37
bcwaldonttx: it went in no more than 20 minutes ago19:37
ttxoh, then, no.19:38
* ttx checks19:38
ttxbcwaldon: right, we'll need to push the fix to that to milestone-proposed as well.19:38
bcwaldonttx: through d2-merge, or as a seperate merge prop?19:38
ttxbcwaldon: once my branch will have refreshed it, I'll do a mergeprop for it19:39
bcwaldonttx: Ok. d2-merge looks good to me, btw. Just ran the diff19:39
ttxbcwaldon: ok, let's set the proposal to "approved" then. Once that's in, I'll prepare a mergeprop for that recnt fix19:40
bcwaldonttx: sounds good19:40
bcwaldonttx: Done. I have to step away for a few. Will be right back19:41
ttxbcwaldon: ok19:41
ttxwill take some time to go through the JenkinsMachine anyway19:41
*** jaypipes has joined #openstack-dev19:41
ttxjaypipes: ah! 2 minutes late.19:41
ttxjaypipes: just got bcwaldon to approve the d2-merge for milestone-proposed. I'm working on backporting limit-marker-lp803055 now.19:42
vishybcwaldon: yes i was19:43
jaypipesttx: k. thx for the update19:45
bcwaldonjaypipes: hey there, stranger19:46
jaypipesbcwaldon: hey sorry man, in meetings most of the day...19:47
bcwaldonjaypipes: that's fine. I've just got to give you a hard time ;)19:47
jaypipesbcwaldon: :) np19:48
vishyi want you dean and I to meet up and discuss the best way to do HA networking19:48
vishyoops wrong window19:49
vishy:)19:49
*** dolphm has joined #openstack-dev19:50
*** zaitcev has joined #openstack-dev19:51
ttxjaypipes, bcwaldon: please see https://code.launchpad.net/~ttx/glance/d2-lp803055/+merge/66202 so that diablo-2 milestone-proposed gets the fix as well.19:53
ttxarh.19:59
bcwaldonjaypipes: found one last bug that absolutely has to go in20:00
bcwaldonttx: ^20:00
ttxjaypipes: do you want API-results-ordering in as well ?20:00
bcwaldonttx: what do you mean 'in'?20:00
ttxoh, hm.20:00
* ttx was lost -- it's in.20:01
bcwaldonttx: yeah, thought so20:01
ttxbcwaldon: could you target the bug to diablo-2, and have it land in milestone-proposed first ?20:01
bcwaldonttx: yes20:01
ttxcool.20:01
ttxbcwaldon: you have until Thursday to push diablo-2 milestone-critical bugfixes in, fwiw20:02
ttxthe idea is to track them at https://launchpad.net/glance/+milestone/diablo-220:02
bcwaldonttx: that is true, disregard that20:02
ttxbcwaldon: currently there is a bit of noise in there, but we'll fix it during the meeting20:02
*** MarkusWPHRorg has quit IRC20:08
*** fyang has joined #openstack-dev20:14
*** ymht has joined #openstack-dev20:40
ttxMeeting in 15 minutes in #openstack-meeting20:45
*** johnpur has quit IRC20:56
*** masumotok has quit IRC20:56
openstackjenkinsProject nova build #1,046: SUCCESS in 3 min 5 sec: http://jenkins.openstack.org/job/nova/1046/20:59
openstackjenkinsTarmac: This adds system usage notifications using the notifications framework.20:59
openstackjenkinsThese are designed to feed an external billing or similar system that subscribes to the nova feed and does the analysis.20:59
*** gholt has left #openstack-dev21:00
ttxMeeting starts now! Join #openstack-meeting21:02
*** alekibango has quit IRC21:06
*** alekibango has joined #openstack-dev21:07
*** dolphm has quit IRC21:08
*** alekibango has quit IRC21:10
*** alekibango has joined #openstack-dev21:10
*** troytoman-away is now known as troytoman21:12
*** ymht has quit IRC21:33
*** heckj_ has joined #openstack-dev21:42
*** alekibango has quit IRC21:42
*** alekibango has joined #openstack-dev21:42
*** mgius-away is now known as mgius21:43
*** alekibango has quit IRC21:44
*** alekibango has joined #openstack-dev21:44
*** heckj has quit IRC21:45
*** heckj_ is now known as heckj21:45
*** rods has quit IRC21:47
*** johnpur has joined #openstack-dev21:51
*** bcwaldon has quit IRC21:56
*** ameade has quit IRC21:58
*** heckj has quit IRC22:00
*** cp16net has quit IRC22:01
jk0https://code.launchpad.net/~blamar/nova/wsgi-service-cleanup/+merge/65258 <-- this seriously breaks running nova services without install nova to the python path22:03
jk0now instead of making a branch and running the nova services directly from there, they must be installed22:03
jk0was this intended?22:03
jk0http://paste.openstack.org/show/1789/ <-- that's the part that was removed, but is still needed22:04
jk0cc vishy ^22:04
*** alekibango has quit IRC22:06
*** fyang has quit IRC22:06
vishydon't think so22:06
vishyugh it took out possible topdir?22:07
vishyblamar: ping?22:07
jk0yes22:07
*** alekibango has joined #openstack-dev22:08
vishyjk0: lets report a bug and fix it before milestone22:09
vishythat is annoying22:09
jk0*very*22:09
jk0seems that just putting it back doesn't fix it though22:09
jk0wonder what else is going on22:09
vishymaybe he replaced it with something22:10
jk0out of api, compute, network and scheduler, only API seems to be missing it22:10
jk0haven't checked anything other than those22:10
jk0trying to figure out API right now22:10
jk0ok, fixed it22:11
comstudhm22:11
jk0vishy: bug/fix coming22:12
* comstud waits to merge trunk22:15
comstudheh.22:15
vishycool22:16
jk0vishy: https://code.launchpad.net/~jk0/nova/lp803186/+merge/6621422:18
vishyoh did you make it off of trunk?22:20
jk0yeah22:20
*** rnirmal has quit IRC22:21
vishycan you actually make it off of rev 1221?22:21
jk0ok, one sec22:21
vishythat way we can propose it into trunk and milestone without any issue22:21
*** mattray has quit IRC22:29
jk0vishy: https://code.launchpad.net/~jk0/nova/lp803186/+merge/6621522:30
*** foxtrotgulf has quit IRC22:40
*** markvoelker has quit IRC22:43
*** pasik has joined #openstack-dev22:49
openstackjenkinsProject nova build #1,047: SUCCESS in 3 min 1 sec: http://jenkins.openstack.org/job/nova/1047/22:49
openstackjenkinsTarmac: Put possible_topdir back in nova-api.22:49
jk0vishy: would it be appropriate for me to review your merge? :)22:50
*** troytoman is now known as troytoman-away22:52
*** jaypipes has quit IRC22:52
*** vladimir3p has quit IRC23:08
s1rpjk0: https://code.launchpad.net/~rconradharris/glance/lp803226/+merge/6622123:11
s1rp  ^^ anyone else mind taking a ganders at that23:12
*** dragondm has quit IRC23:16
*** fyang has joined #openstack-dev23:23
*** dragondm has joined #openstack-dev23:25
*** jkoelker has quit IRC23:37
*** jaypipes has joined #openstack-dev23:52
*** Vasichkin has quit IRC23:53

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!