Monday, 2011-11-28

« Sunday, 2011-11-27 Index Tuesday, 2011-11-29 »
*** chomping has joined #openstack-dev00:30
*** pixelbeat has quit IRC00:33
*** chomping has quit IRC00:34
*** lzyeval has joined #openstack-dev00:35
*** lzyeval has left #openstack-dev00:35
*** lorin1 has joined #openstack-dev00:38
*** lorin1 has quit IRC00:38
*** lorin1 has joined #openstack-dev00:39
*** rods has quit IRC01:00
*** zns has joined #openstack-dev01:03
*** dolphm has joined #openstack-dev01:06
*** mnour has quit IRC01:46
openstackgerritZiad Sawalha proposed a change to openstack/keystone: Changes to support /extensions call. - Introduced a new extension reader to read static extension content. - Added additional rst files explaining extensions. - Removed functionality from  additional middleware that used to support /extensions call.ie RAX  https://review.openstack.org/189101:47
*** lloydde has quit IRC01:52
*** dolphm has quit IRC02:00
*** dolphm has joined #openstack-dev02:14
*** zns has quit IRC02:51
*** dolphm has quit IRC03:08
*** lorin1 has quit IRC03:10
*** mjfork has quit IRC03:14
*** dolphm has joined #openstack-dev03:42
*** nati2 has joined #openstack-dev03:43
openstackgerritdan wendlandt proposed a change to openstack/openstack-manuals: Q-Admin Guide: Improve info about using Melange  https://review.openstack.org/191303:54
*** map_nw_ has joined #openstack-dev03:57
*** map_nw has quit IRC03:59
*** yshh has joined #openstack-dev04:06
openstackgerritEwan Mellor proposed a change to openstack/nova: Bug #897054: stack crashes with AttributeError on e.reason if the server returns an error  https://review.openstack.org/191404:09
uvirtbotLaunchpad bug 897054 in nova "stack crashes with AttributeError on e.reason if the server returns an error" [Undecided,In progress] https://launchpad.net/bugs/89705404:09
*** deshantm_laptop has joined #openstack-dev04:14
*** deshantm_laptop has quit IRC04:14
*** bengrue has joined #openstack-dev04:28
*** koolhead17 has quit IRC04:32
*** yshh has quit IRC04:41
*** nati2 has quit IRC04:42
*** yshh has joined #openstack-dev04:44
*** dolphm has quit IRC05:55
*** koolhead17 has joined #openstack-dev06:01
*** sleepsonthefloo has joined #openstack-dev06:03
*** negronjl has quit IRC06:07
*** negronjl has joined #openstack-dev06:23
*** negronjl has quit IRC06:38
*** negronjl has joined #openstack-dev06:40
*** negronjl has joined #openstack-dev06:41
*** bengrue has quit IRC06:45
*** negronjl has quit IRC06:48
*** GheRivero has joined #openstack-dev07:03
*** sleepsonthefloo has quit IRC07:24
*** ben_duyujie has joined #openstack-dev07:26
openstackgerritEwan Mellor proposed a change to openstack/nova: Bug #897091: "nova actions" fails with HTTP 400 / TypeError if a server action has been performed  https://review.openstack.org/191507:31
uvirtbotLaunchpad bug 897091 in nova ""nova actions" fails with HTTP 400 / TypeError if a server action has been performed" [Undecided,In progress] https://launchpad.net/bugs/89709107:31
*** koolhead17 has quit IRC07:46
*** jeffjapan has quit IRC08:03
*** map_nw_ has quit IRC08:18
*** map_nw has joined #openstack-dev08:19
*** reidrac has joined #openstack-dev08:29
*** pixelbeat has joined #openstack-dev08:35
*** map_nw has quit IRC08:43
*** map_nw has joined #openstack-dev08:44
*** zns has joined #openstack-dev08:47
*** map_nw has quit IRC08:52
*** Arminder-Office has left #openstack-dev08:52
*** zns1 has joined #openstack-dev08:59
*** popux has joined #openstack-dev09:00
*** zns has quit IRC09:02
*** Rajaram has joined #openstack-dev09:13
openstackgerritLikitha Shetty proposed a change to openstack/nova: Fix Bug #891718  https://review.openstack.org/181709:21
uvirtbotLaunchpad bug 891718 in nova "nova components always log to standard error, even when syslog is turned on" [Wishlist,Confirmed] https://launchpad.net/bugs/89171809:21
*** ben_duyujie has left #openstack-dev09:35
*** LinuxJedi has joined #openstack-dev09:45
*** scottjg has quit IRC09:48
*** scottjg has joined #openstack-dev10:00
*** yshh has quit IRC10:06
*** scottjg has quit IRC10:07
*** ollie1 has left #openstack-dev10:12
*** darraghb has joined #openstack-dev10:18
*** apevec has joined #openstack-dev10:21
*** ollie1 has joined #openstack-dev10:24
*** map_nw has joined #openstack-dev10:24
*** map_nw has quit IRC10:27
*** map_nw has joined #openstack-dev10:28
comstudvishy: https://bugs.launchpad.net/nova/+bug/89715510:30
uvirtbotLaunchpad bug 897155 in nova "rpc calls are broken when remote procedure returns None" [Undecided,In progress]10:30
comstudvishy: just stumbled upon this while working on something else10:30
comstudvishy: This could very well be the 'exchange not found' issue.  I dunno about the 'second channel.open' thing tho.10:35
*** popux has quit IRC10:57
*** popux has joined #openstack-dev11:27
*** markvoelker has joined #openstack-dev12:07
openstackgerritJimmy Bergman proposed a change to openstack/nova: Fix for EC2 API part of bug 897164  https://review.openstack.org/191612:09
uvirtbotLaunchpad bug 897164 in nova "neither EC2 api nor openstack API expose the hostname" [Undecided,New] https://launchpad.net/bugs/89716412:09
*** popux_ has joined #openstack-dev12:16
*** popux has quit IRC12:17
zykes-viraptor: or anyone from hp here that knows swift?12:18
*** akscram has joined #openstack-dev12:20
viraptorzykes-: I don't know swift, but can forward to the right people... what's up?12:21
zykes-i wonder what nodes there are12:21
zykes-that you use for storage etc if possible + management capabilities12:21
zykes-most people i work here with is click "Next"12:21
*** deshantm_laptop has joined #openstack-dev12:37
*** mnour has joined #openstack-dev12:38
*** GheRivero has quit IRC12:39
*** map_nw has quit IRC12:39
*** GheRivero has joined #openstack-dev12:42
*** zigo has joined #openstack-dev12:46
*** Rajaram_ has joined #openstack-dev13:02
*** Rajaram has quit IRC13:05
*** Rajaram_ is now known as Rajaram13:05
*** mjfork has joined #openstack-dev13:05
*** ben_duyujie has joined #openstack-dev13:09
openstackgerritKei Masumoto proposed a change to openstack/nova: First commit "deleting-compute-node-table"  https://review.openstack.org/191713:11
*** lts has joined #openstack-dev13:21
Davieyhola, Quantum isn't *required* by nova in essex, is it?13:28
Davieyhttp://pb.daviey.com/pvZO/ made me wonder..13:29
KiallDaviey: I'm pretty sure its not. But I would imagine the test cases should test the integration...13:31
*** dolphm has joined #openstack-dev13:32
DavieyKiall: right, i'd expect a skip..13:34
DavieyIt's not clear to me if it provides a stub Quantum server, need to deep deeper.13:35
KiallI have no clue how nova's test suite works, but I would imagine it.. heh beat me to mentioning a Stub ;)13:36
*** kbringard has joined #openstack-dev13:41
*** markmc has joined #openstack-dev13:46
*** ccorrigan has joined #openstack-dev13:50
zykes-viraptor: ?13:55
*** hugokuo has joined #openstack-dev13:57
*** HugoKuo__ has quit IRC13:58
openstackgerritLorin Hochstein proposed a change to openstack/openstack-manuals: Added info on how to build docs to the README.  https://review.openstack.org/181314:03
*** bsza has joined #openstack-dev14:08
*** rods has joined #openstack-dev14:12
*** dprince has joined #openstack-dev14:13
*** dolphm has quit IRC14:14
openstackgerritZiad Sawalha proposed a change to openstack/keystone: Updated testing  https://review.openstack.org/191814:15
*** zns1 has quit IRC14:15
*** crobinso has joined #openstack-dev14:17
viraptorzykes-: sorry, was away - what did you mean exactly? software configuration, hardware, or ... - what did you mean by "nodes"14:18
openstackgerritoliver-leahy-l proposed a change to openstack/nova: Fixes bug 888649  https://review.openstack.org/190014:18
uvirtbotLaunchpad bug 888649 in nova "Snapshots left in undeletable state" [Undecided,New] https://launchpad.net/bugs/88864914:18
openstackgerritA change to openstack/nova has been rejected: Fix for EC2 API part of bug 897164  https://review.openstack.org/191614:21
uvirtbotLaunchpad bug 897164 in nova "neither EC2 api nor openstack API expose the hostname" [Undecided,New] https://launchpad.net/bugs/89716414:21
*** sandywalsh has joined #openstack-dev14:22
zykes-viraptor: hardware14:23
zykes-for storage nodes + proxy14:23
*** lloydde has joined #openstack-dev14:24
*** bcwaldon has joined #openstack-dev14:26
viraptornot my area, sorry... I'm not sure anyone would publish this either :(14:26
viraptorbut you can try emailing people from hp who submit to openstack/swift on gerrit14:26
*** popux_ has quit IRC14:38
zykes-;p14:39
*** AlanClark has joined #openstack-dev14:41
*** zns has joined #openstack-dev14:44
dprincezul/Davey/soren: https://code.launchpad.net/~dan-prince/keystone/ubuntu-postinst-bash-fixes/+merge/8361814:44
sorendprince: Gosh.14:44
dprincezul/Davey/soren: that fixes the keystone installer for me which has been broken for awhile now...14:45
sorendprince: Yeah, I can imagine.14:45
Davieydprince: crikey, thanks for that.  Will make sure that is uploaded today14:46
sorenDaviey: Are you merging it already?14:46
Davieysoren: no, are you?14:46
sorenYes.14:46
dprinceThanks guys. I'd like to switch over and use the keystone trunk installer again.14:47
sorendprince: Oh, thank *you*.14:47
Daviey+114:49
sorenMerged.14:50
*** rkukura has joined #openstack-dev14:50
dprinceGreat. I'm happy to test out any major installer refactorings 'pre-merge' if you guys want an extra set of eyes on things. SmokeStack is parameterized to take custom installer branches (directory from LaunchPad) so holler if you want me to give anything a go.14:51
*** Rajaram has quit IRC14:53
openstackgerritMark Washenberger proposed a change to openstack/nova: Use instance uuids for consoles and diagnostics.  https://review.openstack.org/191914:56
Davieydprince: I will do that!14:58
*** popux_ has joined #openstack-dev15:01
*** mattray has joined #openstack-dev15:02
*** zns has quit IRC15:10
*** deshantm_laptop has quit IRC15:10
*** dprince has quit IRC15:13
*** bsza has quit IRC15:21
*** cp16net has joined #openstack-dev15:23
*** GheRivero_ has joined #openstack-dev15:23
openstackgerritRainer Toebbicke proposed a change to openstack/swift: add content-range headers to partial HTTP responses triggered by "Range:" requests as required by the HTTP protocol  https://review.openstack.org/192015:24
*** cp16net has quit IRC15:25
*** cp16net has joined #openstack-dev15:25
*** bsza has joined #openstack-dev15:33
*** rkukura has left #openstack-dev15:34
*** lorin1 has joined #openstack-dev15:38
*** rkukura has joined #openstack-dev15:39
*** vladimir3p has joined #openstack-dev15:43
*** zigo has quit IRC15:47
*** reed has joined #openstack-dev15:47
*** rods has quit IRC15:48
*** mnour has quit IRC15:51
dfgmtaylor: jeblair: I am trying to follow the instruction on http://wiki.openstack.org/GerritWorkflow but when I run 'git review -s' I get: http://paste.openstack.org/show/3540/15:55
*** code_franco has joined #openstack-dev15:56
*** adjohn has joined #openstack-dev15:58
*** aclark has joined #openstack-dev16:01
*** AlanClark has quit IRC16:01
*** dragondm has joined #openstack-dev16:02
*** ben_duyujie has quit IRC16:03
*** aclark is now known as AlanClark16:03
*** ben_duyujie has joined #openstack-dev16:03
*** troytoman-away is now known as troytoman16:04
openstackgerritAnne Gentle proposed a change to openstack/openstack-manuals: Fixes bug 897042  https://review.openstack.org/192116:14
uvirtbotLaunchpad bug 897042 in openstack-manuals "openstack-compute-admin fails to build" [Medium,Confirmed] https://launchpad.net/bugs/89704216:14
*** rnirmal has joined #openstack-dev16:15
*** _adjohn has joined #openstack-dev16:15
*** adjohn has quit IRC16:15
*** _adjohn is now known as adjohn16:15
*** reidrac has quit IRC16:16
*** zns has joined #openstack-dev16:19
*** negronjl has joined #openstack-dev16:21
*** AlanClark has quit IRC16:22
*** AlanClark has joined #openstack-dev16:23
*** GheRivero_ has quit IRC16:29
*** GheRivero_ has joined #openstack-dev16:30
dfgmtaylor: jeblair: apparently it was because I was using an old version (1.7.0) of git and the 'git branch --color=never' command in the review script was failing. I upgraded to 1.7.7 and it seems to be working.16:33
*** GheRivero_ has quit IRC16:33
*** GheRivero_ has joined #openstack-dev16:35
notmynamemtaylor: jeblair: seems that the root issue with dfg's problem (git version issue) is that the dev target is lucid, but git 1.7.0 is all that's available there in the packages for lucid. perhaps the git review tool can not require a later version of git that what is provided in our target environment?16:35
*** zns has quit IRC16:37
*** GheRivero_ has quit IRC16:46
*** n0ano has joined #openstack-dev16:46
openstackgerritVerification of a change to openstack/glance failed: Fixes LP Bug#850377  https://review.openstack.org/165616:46
*** GheRivero_ has joined #openstack-dev16:47
*** jaypipes has quit IRC16:54
*** popux_ has quit IRC16:57
*** rkukura has quit IRC16:58
*** adjohn has quit IRC16:59
*** rkukura has joined #openstack-dev17:02
*** zns has joined #openstack-dev17:04
*** jaypipes has joined #openstack-dev17:04
*** ben_duyujie has left #openstack-dev17:04
*** vladimir3p has quit IRC17:09
*** code_franco has quit IRC17:10
*** code_franco has joined #openstack-dev17:14
openstackgerritDavid Goetz proposed a change to openstack/swift: Fix object replicator to handle Timeouts fixes: lp 814263  https://review.openstack.org/192217:14
uvirtbotLaunchpad bug 814263 in swift "Bug in exception handling with obj replicator" [Undecided,New] https://launchpad.net/bugs/81426317:14
*** sleepsonthefloo has joined #openstack-dev17:16
*** sleepsonthefloo_ has joined #openstack-dev17:17
*** sleepsonthefloo has quit IRC17:21
*** sleepsonthefloo_ is now known as sleepsonthefloo17:21
vishycomstud: interesting17:28
vishycomstud: if you have a fix I will try it in the lab, not sure what could be returning None some small percentage of the time though17:29
*** koolhead17 has joined #openstack-dev17:39
mtaylornotmyname: ah. that's a great point. I'll fix that.17:43
KiallIf anyone is interested in nightly .deb builds of the stable/diablo branches: https://launchpad.net/~managedit/+archive/openstack-nightly17:49
*** coli_ has joined #openstack-dev17:49
*** dragondm has quit IRC17:51
*** GheRivero_ has quit IRC17:51
*** GheRivero_ has joined #openstack-dev17:51
*** hugokuo has quit IRC17:58
*** jdurgin has joined #openstack-dev18:00
comstudvishy: Yeah.. kind of a long shot, I guess. I have a fix to prop when I get a second to do some final clean up18:01
*** reed has quit IRC18:08
comstudvishy: you could add logging in _process_data().  Where the ctxt.reply(x, None) and ctxt.reply(rval, None) lines are around 623 thru 63018:10
comstudvishy: LOg something is x is None or rval is None18:10
comstudis/if18:10
vishycomstud: not sure why returning none is a problem?18:11
vishycomstud: wouldn't it just stop the multicall right away?18:11
vishycomstud: since we aren't actually using multicall for anything I mean18:11
*** GheRivero_ has quit IRC18:12
*** GheRivero_ has joined #openstack-dev18:12
comstudvishy: The calling side will stop at the first None and close the exchange18:13
comstudvishy: The responding side will try to add another None reply for the end of marker18:13
comstudthe problem is we use 'None' for end of marker18:13
vishycomstud: ah interesting18:13
comstudwhen 'None' is also a valid response.18:13
vishycomstud: got it18:13
*** code_franco has quit IRC18:14
comstudWe need to signify end of response a different way18:14
comstudI'm adding a 'ending': True to the response dictionary in my proposed fix18:14
comstudlast msg will have that18:14
comstudall for suggestions on better fix :)18:14
comstudi have it done and working... i just need to prop it18:15
comstudI spotted this while writing a new 'fake' rpc implementation18:15
*** rkukura has quit IRC18:16
*** bengrue has joined #openstack-dev18:16
comstudcouple things to clean up there before I prop18:16
*** adjohn has joined #openstack-dev18:17
*** code_franco has joined #openstack-dev18:17
comstudvishy: And this thing.. wouldn't be a problem in carrot, because in carrot we don't destroy the queue (diff bug)18:17
vishycomstud: yes this sounds very promising18:18
* vishy hopes that it fixes the annoying rpc bug18:18
comstudIt certainly would cause the 'exchange not found' issue.18:19
vishycomstud: although it sounds like we might need to track down what might make calls return none18:19
comstudI just don't know if this is _it_ :)18:19
comstudvishy: Yeah.. that's why I suggest adding some logging around 623-630 in impl_kombu18:19
comstudat least to verify this happens18:19
comstudi don't think we use multicall today, so you could probably just check the 'rval = node_func' response for None18:20
comstudand throw a log message that's easy to grep18:20
comstudprint out node_func and we'll know where it came from18:20
vishyok18:21
vishyi will try and get that change up somewhere18:21
comstudi'm at castle, so i'll be in and out18:21
*** GheRivero_ has quit IRC18:22
*** lloydde has quit IRC18:23
*** zaitcev has joined #openstack-dev18:24
*** rkukura has joined #openstack-dev18:33
openstackgerritBrad Hall proposed a change to openstack/quantum: Second round of packaging changes  https://review.openstack.org/188618:33
*** darraghb has quit IRC18:34
*** scottjg has joined #openstack-dev18:35
openstackgerritNaveed Massjouni proposed a change to openstack/nova: Updating {add,remove}_fixed_ip_from_instance in compute.api and compute.manager to use instance uuid instead of instance id. blueprint internal-uuids  https://review.openstack.org/185518:56
*** zul has quit IRC19:13
*** novas0x2a|laptop has joined #openstack-dev19:13
*** koolhead17 has quit IRC19:14
*** zul has joined #openstack-dev19:19
openstackgerritDavid Goetz proposed a change to openstack/swift: Fix object replicator to handle Timeouts fixes: lp 814263  https://review.openstack.org/192219:25
uvirtbotLaunchpad bug 814263 in swift "Bug in exception handling with obj replicator" [Undecided,New] https://launchpad.net/bugs/81426319:25
openstackgerritA change to openstack/nova has been rejected: Fix for EC2 API part of bug 897164  https://review.openstack.org/191619:25
uvirtbotLaunchpad bug 897164 in nova "neither EC2 api nor openstack API expose the hostname" [Undecided,New] https://launchpad.net/bugs/89716419:25
*** bcwaldon has quit IRC19:38
*** zns has quit IRC19:44
*** mwhooker has joined #openstack-dev19:47
*** novas0x2a|laptop has quit IRC19:51
*** GheRivero_ has joined #openstack-dev19:51
*** timr has joined #openstack-dev19:51
*** novas0x2a|laptop has joined #openstack-dev19:51
*** bcwaldon has joined #openstack-dev19:53
*** troytoman has quit IRC19:53
*** rkukura has quit IRC19:54
*** lloydde has joined #openstack-dev19:55
*** troytoman-away has joined #openstack-dev19:55
*** negronjl has quit IRC19:58
notmynamettx: I noticed that there are 2 PPB meetings on the openstack-meetings calendar. I think the 2nd one is in error. can you remove it? (I don't have permissions to do so)20:01
*** scottjg has quit IRC20:04
*** adjohn has quit IRC20:04
*** sleepsonthefloo_ has joined #openstack-dev20:05
*** mattray has quit IRC20:06
*** sleepsonthefloo has quit IRC20:07
*** sleepsonthefloo_ is now known as sleepsonthefloo20:07
*** adjohn has joined #openstack-dev20:08
*** sleepsonthefloo has quit IRC20:24
*** zaitcev has quit IRC20:26
*** zns has joined #openstack-dev20:34
*** GheRivero_ has quit IRC20:36
*** GheRivero_ has joined #openstack-dev20:37
*** GheRivero_ has quit IRC20:38
*** GheRivero_ has joined #openstack-dev20:38
*** dragondm has joined #openstack-dev20:39
*** sleepsonthefloo has joined #openstack-dev20:42
openstackgerritMark Washenberger proposed a change to openstack/nova: start/stop in compute manager to use uuids  https://review.openstack.org/185620:43
*** GheRivero_ has quit IRC20:46
*** andrewbogott has joined #openstack-dev20:46
*** GheRivero_ has joined #openstack-dev20:47
openstackgerritMonty Taylor proposed a change to openstack-ci/git-review: Only use color on systems that support it.  https://review.openstack.org/192320:49
mtaylornotmyname: I think that ^^^ should fix the earlier git/lucid/color issue. seem sane to you?20:50
notmynamemtaylor: thanks :-)20:50
*** GheRivero_ has quit IRC20:50
*** Daviey has quit IRC20:50
notmynamethe commit message seems nice20:50
notmyname(I didn't look at the code)20:51
*** GheRivero_ has joined #openstack-dev20:51
annegentlewhat's the git command for fetching a remote branch specifically? I want to work on a branch that's at git@github.com:annegentle/openstack-manuals.git (as far as I can tell).20:54
annegentleBut I'm not sure how to pull it down?20:54
notmynamegit pull <remoteurl>20:54
notmynamegit pull <remoteurl> <remote branch name>20:54
openstackgerritNaveed Massjouni proposed a change to openstack/nova: Updating inject_file in compute.api and compute.manager to use instance uuids instead of instance ids. blueprint: internal-uuids  https://review.openstack.org/192420:56
annegentlenotmyname: ok thanks120:56
notmynameannegentle: of course, if you want to pull it in more permanently, there is a slightly better way20:57
notmynameannegentle: git remote add <remote url>20:57
notmynameannegentle: oops20:57
notmynameannegentle: git remote add <remote name> <remote url>20:57
notmynameannegentle: git checkout -b new_local_branch_name <remote name>/<remote branch name>20:58
*** Guest913 is now known as medberry20:58
annegentlenotmyname: ok yeah I guess checkout is what I was thinking of20:58
*** GheRivero_ has quit IRC20:59
annegentlenotmyname: oh did you get the PPB calendar duplicate sorted? I have access and I'm looking at it and I don't see 2.20:59
*** GheRivero_ has joined #openstack-dev20:59
notmynameannegentle: no, I still see a PPB meeting at 3 (central)21:00
andrewbogottjkoelker:  thank you for the review!  I am brand new to the project, so this will be like pulling teeth :)21:00
*** statik has joined #openstack-dev21:00
jkoelkerandrewbogott: no probs21:02
jkoelkerandrewbogott: welcome to the phun! ;)21:02
andrewbogottCan you point me in the right direction re: the test suite?  Does each module have its own tests, or is there a separate git repo with all the tests?21:02
jkoelkerso in nova/tests/ there is a test_network.py file, you could just create a new class there21:03
*** GheRivero_ has quit IRC21:03
jkoelkerits kinda all haphazard right now as far as how the tests are layed out21:03
andrewbogottIs there a wiki page about test formatting &c, or should I just imitate what's there now?21:03
mwhookerhaving a tough time getting devstack to work. stack.sh runs without error, but euca, floating_ips, and swift fail when running exercise.sh21:03
*** GheRivero_ has joined #openstack-dev21:04
mwhookeranyone have any ideas?21:04
jkoelkereach file is layed out different ;(21:04
jkoelkerso i usually try to emulate whatever is already there21:04
andrewbogottok, that keeps the bar low.21:04
*** GheRivero_ has quit IRC21:05
*** GheRivero_ has joined #openstack-dev21:06
annegentlemtaylor: can you poke at the OpenStack planet job on Jenkins to get it unstuck? There's a blog entry that should have published yesterday.21:06
mtaylorannegentle: yup21:07
jkoelkerandrewbogott: let me know if you get stuck, the network tests are particularly fun with some of the stubbing that is going on21:09
*** GheRivero_ has quit IRC21:09
andrewbogottthanks.  atm I don't have a working master branch install working, so I have lots of admin work before I can return to coding.21:10
*** rkukura has joined #openstack-dev21:11
openstackgerritKevin L. Mitchell proposed a change to openstack/nova: Templatize extension handling.  https://review.openstack.org/192521:11
openstackgerritAdrian Smith proposed a change to openstack/nova: Fix for bug 887712  https://review.openstack.org/192621:12
uvirtbotLaunchpad bug 887712 in openstack-qa "instance_update with uuid as instance_Id and metadata fails" [Medium,In progress] https://launchpad.net/bugs/88771221:12
*** anotherjesse has joined #openstack-dev21:16
*** jakedahn has joined #openstack-dev21:22
*** donaldngo_hp has joined #openstack-dev21:25
openstackgerritVerification of a change to openstack/nova failed: Fix deprecation warnings  https://review.openstack.org/189821:27
*** GheRivero_ has joined #openstack-dev21:27
openstackgerritMark Washenberger proposed a change to openstack/nova: Use uuids for file injection.  https://review.openstack.org/192721:28
*** lorin1 has quit IRC21:31
*** Daviey has joined #openstack-dev21:32
*** timr has quit IRC21:39
*** reed has joined #openstack-dev21:40
openstackgerritDaniel Salinas proposed a change to openstack/nova: Adds openvz support to nova.  https://review.openstack.org/162921:43
*** GheRivero_ has quit IRC21:44
*** zns has quit IRC21:46
*** zns has joined #openstack-dev21:47
*** negronjl has joined #openstack-dev21:48
*** benner has quit IRC21:54
*** benner has joined #openstack-dev21:54
openstackgerritBrian Waldon proposed a change to openstack/nova: Convert compute manager delete methods to objects  https://review.openstack.org/192821:55
*** zns has quit IRC22:01
vishycomstud: 2011-11-28 15:56:14,950 WARNING nova.rpc [-] NONE CALL from: _rpc_allocate_fixed_ip - {u'instance_id': 1, u'network_id': 1, u'vpn': False, u'address': None} -22:01
vishycomstud: also it appears to happen with every lease / release22:02
*** zns has joined #openstack-dev22:03
vishycomstud: oh nm, those are casts, so it shouldn't matter that they return None22:04
*** andrewbogott has quit IRC22:06
*** gyee_ has joined #openstack-dev22:08
*** markvoelker has quit IRC22:08
annegentleso I get a "updating paths is incompatible with switching branches" when I try to checkout and make a new branch of my https://annegentle@github.com/annegentle/openstack-manuals.git/112311 branch… what should I do to retrieve and work on my remote branch?22:11
notmynameannegentle: is that repo using the gerrit review process or not?22:13
annegentleopenstack-manuals uses gerrit, yup22:13
vishycomstud: i think i have the bug.  Just testing now22:13
annegentlebut seems like I should be able to throw a one-off up on my own remote? Or no?22:13
comstudvishy: ok22:13
comstudvishy: Ah, yeah..forgot that'd catch the casts too22:14
vishycomstud: there is a greenthread spawning an rpc call to _rpc_allocate_fixed_ip22:14
vishywhich wasn't returning a value22:14
annegentlenotmyname: is it trying to save me from myself? :)22:14
notmynameannegentle: ya, you can22:15
notmynameannegentle: I have a similar situation with swift. I use the openstack repo but also have a fork on my personal github account22:15
comstudvishy: I think i've seen that greenthread.. IIRC, it was something i didn't quite understand (the need)22:15
vishycomstud: so my theory is that was blowing up when it couldn't tear down properly, causing an exception22:15
notmynameannegentle: this is what I would do22:15
notmynameannegentle: git remote add github git://annegentle@github.com/annegentle/openstack-manuals.git22:16
notmynamegit fetch github22:16
notmynameannegentle: git checkout -b <local branch name> github/<starting point branch>22:17
notmynameannegentle: then, when you are ready to push the changes to your local github repo, do this:22:17
notmynameannegentle: (assuming you are on the branch you want to push) git push -u github <branch name>22:18
notmynamethen you can use push and pull to update that branch as normal22:18
vishycomstud: and this is why you weren't seeing failures22:18
vishycomstud: because you don't use flatdhcp network22:19
annegentlenotmyname: ok I think I have done git remote add annegentle at some point so I was able to do git checkout -b <local branch name> annegentl/<starting point branch>22:20
comstudvishy: Yep22:20
vishytesting…22:20
vishysuccess!!22:20
notmynameannegentle: ok, that sounds good22:20
comstudand carrot leaves the queue around22:20
annegentlenotmyname: huzzah, that was it!22:20
comstudso that's why you don't see it there22:20
notmynameannegentle: yay! (umm..which part?)22:20
notmynamethe checkout -b?22:20
vishycomstud: 30 instances run in groups of 1022:21
annegentleyup, finally got checkout -b for the branch where I had moved a bunch of stuff around22:21
vishyzero failures22:21
notmynameannegentle: cool22:21
comstudgreat!22:21
annegentlenotmyname: whew.22:21
comstudvery strange how I stumbled upon this22:21
vishycomstud: submitting patch, nice find on that one22:21
comstudthnx22:21
vishyshould I actually make it throw an exception if a method returns none and there is a msg_id?22:22
vishyor will your fix take care of it?22:22
comstudi'm fixing the real problem22:23
comstudthe rpc problem22:23
comstudi've added a different way to detect end-of-response22:23
comstudNone should be a valid response22:23
vishyok22:23
comstudeven an iterator that can return Nones should be valid22:23
vishyi will take out the log method then22:23
comstud(multicall)22:23
comstudyea22:23
comstudi'll have this fix propped within an hour22:24
comstudit has some unrelated changes... which is how I found this22:24
comstuda fake-rabbit/kombu memory transport alternative that's faster.22:24
comstudspeeds tests by 10%22:24
comstudor so22:24
Vekis jenkins wedged or something?  Saw a change approved over an hour ago that hasn't been verified yet...22:27
mtaylorVek: this is running currently: https://jenkins.openstack.org/job/gate-nova-unittests/98/22:31
mtaylorVek: there were several things queued up a little while ago - it might have just been a busy period22:31
Vekmtaylor: OK, just checking.22:33
Vekheh.  That job is related to my change :)22:34
*** mattray has joined #openstack-dev22:34
*** bcwaldon has quit IRC22:35
*** rnirmal has quit IRC22:39
*** crobinso has quit IRC22:40
*** apevec has quit IRC22:40
Vekmtaylor: I dunno, but I think that job is actually wedged.22:42
Vekaccording to the progress bar, it's been running for 48 minutes...22:43
Veknow I know our test suite can take a while to run, but 48 minutes is a bit absurd even for it!  :)22:43
mtaylorhrm22:43
openstackgerritVish Ishaya proposed a change to openstack/nova: Makes rpc_allocate_fixed_ip return properly  https://review.openstack.org/192922:43
* Vek wonders if it's grinding swap or something...22:45
mtaylorVek: "17691 jenkins   20   0  859m 525m 7404 R  100 52.7  51:32.70 python          "22:47
openstackgerritVerification of a change to openstack/nova failed: Templatize extension handling.  https://review.openstack.org/192522:47
mtaylorVek: I killed the test job. when the queue is cleared out, we can re-trigger it and see if it sticks again22:48
*** gyee_ has quit IRC22:48
Vek'k.22:48
vishytr3buchet: ping22:48
*** gyee has joined #openstack-dev22:48
*** vladimir3p has joined #openstack-dev22:48
vishymtaylor: another one stuck earlier22:49
vladimir3pguys, is there a repo for stable/diablo keystone packages (Ubuntu 11.04)?22:50
Vekvishy: where?  Looks like mine wedged running XenAPIVolumeTestCase.test_attach_volume22:50
vishyVek: not sure22:50
vishyVek: i just saw one that ended aborted earlier22:50
Vekmight be interesting to compare notes :)22:50
mtaylorjeblair: we might want to enable the "this job is taking too long" plugin on gate-nova-unittests22:52
jeblairmtaylor: perhaps so22:53
*** apevec has joined #openstack-dev22:53
jeblairwe turned that on for keystone recently.22:53
Vekjust pick the right timeout.  If the test suite digs into swap at all, there's no telling how long it could run...22:53
jeblairi believe the normal run time is about 9 minutes22:54
VekRan 2023 tests in 395.455s22:54
tr3buchetvishy: what's up?22:54
Vekon my dev box, with the suite digging into swap...22:54
vishytr3buchet: just wanted to give a heads up on this: https://review.openstack.org/192922:55
*** apevec has quit IRC22:55
vishysince it is network related22:55
mtaylorVek: well, considering that the jobs all run on boxes that do nothing else but run one copy of the test suite at a time...22:55
vishytr3buchet: I was trying to figure out how to write a test for it.22:55
Vek(that's up from ~341s before I had to enable swap to get the suite to run)22:55
mtaylorVek: if it ever dug in to swap, that would be a pretty huge bug22:55
Vekmtaylor: how much memory does the box have?22:55
jeblairif that does happen, we should deploy bigger test boxes.  but right now, with the suite running, the server is at about 60% used22:56
mtaylorVek: 1G22:56
vishytr3buchet: and I am at a loss for how to do it in a unit test aside from the naive stub the call, and make sure the wrapping call returns None22:56
vishytr3buchet: which seems mostly useless22:56
tr3buchetvishy: interesting22:56
Vekmtaylor: that's what I've got in my dev box, and I *had* to enable swap just to get the thing to run.22:56
mtaylorVek: but you're likely doing other things on your dev box?22:56
Vekwe've got a *massive* memory leak somewhere in the test suite.22:56
vishytr3buchet: comstud has a more complicated patch that will fix the underlying problem of returning None in a call22:56
tr3buchetvishy: looking into that bug22:56
mtaylorVek: hehe. we should find it22:57
VekI could not get the suite to complete without swap with virtually everything else shut off.22:57
mtaylorVek: those sorts of things are, well, bad22:57
mtaylorwow. that's interesting22:57
vishytr3buchet: but as it is causing failures as is, I feel like the rpc call should return something22:57
Vekindeed.  And I've tried, but so far been unsuccessful.22:57
jeblair             total       used       free     shared    buffers     cached22:57
jeblairMem:       1022536    1003496      19040          0      51240     26070022:57
jeblair-/+ buffers/cache:     691556     33098022:57
jeblairSwap:      2097148      41772    205537622:57
jeblairthat's right at the end of the run22:57
tr3buchetright22:57
jeblair-/+ buffers/cache:     219644     80289222:57
tr3bucheta cast would be fine, except we need to wait22:57
jeblairand that's after the run finished22:57
Vekmy memory usage builds up to just shy of 50% of my 1G dev box by the end of the run.22:57
tr3buchetso return it and disregard22:57
tr3buchetseems like a hacky fix, but it works22:58
jeblairso yeah, it looks like it uses about .5G22:58
vishytr3buchet: right. If the other fix is in, it should work just returning normally22:58
Vek(it was causing the openssl tests to fail, because it couldn't allocate memory to start external processes)22:58
vishytr3buchet: but i felt that that rpc should return a value22:58
wwkeyboardtr3buchet: vishy: Are you talking about the issue with network manager falling over if quantum spits up a bad value(i.e. something that does not deserialize properly, or a 500) ?22:58
tr3buchetvishy: so this isn't necessary once comstud's patch hits?22:58
vishytr3buchet: since the underlying call does22:58
tr3buchetah i see22:58
vishyand we are using call instead of cast22:59
vishywwkeyboard: no22:59
wwkeyboardok, thanks though, I'll dig into this22:59
jeblairmtaylor: how about a 20 minute timout?22:59
tr3buchetvishy: test seems easy, make a test that ensures it doesn't return None23:00
Veksounds reasonable to me, if you want my opinion :)23:00
tr3bucheti mean sortof useless but, yeah23:00
jeblairVek: absolutely!23:00
vishytr3buchet: yeah, ok seems completely silly but can't hurt i suppose23:00
*** kbringard has quit IRC23:01
jeblairokay 20 minute timeout set23:01
tr3buchetvishy: if allocate_fixed_ip(...) is already tested, then all this is is a db call and a return statement. simple test for a simple function. works right?23:01
tr3buchetvishy: ping once you get the test setup and i'll approve23:02
*** lts has quit IRC23:02
tr3buchetvishy: i can't see how it could affect anything. nothing depends on the return value of that functin23:03
vishyyup adding23:04
*** zns has quit IRC23:07
openstackgerritVerification of a change to openstack/nova failed: Convert compute manager delete methods to objects  https://review.openstack.org/192823:09
*** reed_ has joined #openstack-dev23:14
Vekby the way, in terms of memory leaking during the test suite--I tried modifying the global tearDown() method to additionally call gc.collect(), and that had no effect.23:16
*** reed has quit IRC23:16
Vekmy best guess is that something is leaking at the C level, rather than the python level; just haven't been able to figure out what.  A lot of it tends to happen during ServersTest, though...23:18
openstackgerritVish Ishaya proposed a change to openstack/nova: Makes rpc_allocate_fixed_ip return properly  https://review.openstack.org/192923:19
vishytr3buchet: test is in23:20
*** bsza has quit IRC23:25
tr3buchetvishy: sweet23:27
*** andrewbogott has joined #openstack-dev23:29
openstackgerritmcgrue proposed a change to openstack/nova: Removing obsolete bzr-related clauses in setup.py  https://review.openstack.org/193123:32
openstackgerritmcgrue proposed a change to openstack/nova: Adding an install_requires to the setup call. Now you can pip install nova on a naked machine.  https://review.openstack.org/193223:32
*** jog0 has joined #openstack-dev23:32
*** ben_duyujie has joined #openstack-dev23:33
bengrueI should've been bengrue on launchpad too.23:34
tr3buchetprobably23:34
bengruefaaaugh, Authors file.23:39
vishyanyone know if Lorin Hochstein is on irc?23:43
vishyhmm launchpad says he is lorin1, maybe he just isn't around23:44
vishysoren: can you make sure https://blueprints.launchpad.net/nova/+spec/consolidate-testing-infrastructure lines up with your plans and approve and target it if so?23:57
*** zul has quit IRC23:58
« Sunday, 2011-11-27 Index Tuesday, 2011-11-29 »

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