Tuesday, 2010-11-30

*** rsampaio has joined #openstack00:00
*** dubsquared has quit IRC00:01
*** rlucio has quit IRC00:11
*** Cybodog has quit IRC00:18
*** alekibango has quit IRC00:20
*** alekibango has joined #openstack00:21
*** kevnfx has joined #openstack00:25
*** GasbaKid has quit IRC00:26
*** miclorb has quit IRC00:26
*** trbs has quit IRC00:35
*** joearnold has quit IRC00:43
*** rnirmal has quit IRC00:49
*** Cybodog has joined #openstack00:58
*** Ryan_Lane is now known as Ryan_Lane|away01:02
*** jkakar has quit IRC01:05
*** kevnfx has left #openstack01:05
*** sophiap has joined #openstack01:07
*** miclorb_ has joined #openstack01:17
*** jkakar has joined #openstack01:17
*** DubLo7 has joined #openstack01:18
*** silassewell has joined #openstack01:25
*** silassewell has quit IRC01:33
*** daleolds has quit IRC01:36
*** maplebed has quit IRC01:40
*** jkakar has quit IRC01:45
*** BK_man has joined #openstack01:57
*** zul has quit IRC02:01
*** zul has joined #openstack02:01
*** sophiap has quit IRC02:06
*** sophiap has joined #openstack02:07
*** schisamo has quit IRC02:23
*** nati_ has quit IRC02:30
*** rchavik has joined #openstack02:37
*** rchavik has quit IRC02:42
*** jkakar has joined #openstack03:00
*** sophiap has quit IRC03:00
*** rchavik has joined #openstack03:02
*** sirp has joined #openstack03:19
*** rsampaio has quit IRC03:28
*** rsampaio has joined #openstack03:43
*** mdomsch has joined #openstack03:49
*** DubLo7 has quit IRC04:04
*** DubLo7 has joined #openstack04:04
*** sirp has quit IRC04:07
*** sirp has joined #openstack04:08
*** joearnold has joined #openstack04:20
*** ArdRigh has quit IRC04:25
*** sirp has quit IRC04:40
* BK_man wondering why nova-api is searching libvirt-templates in python lib directory instead of /usr/share04:42
*** laurensell has quit IRC05:22
*** BK_man has quit IRC05:24
*** miclorb_ has quit IRC05:25
*** dragondm has quit IRC05:28
*** kashyapc has joined #openstack05:31
*** mdomsch has quit IRC05:32
*** jimbaker has quit IRC05:34
*** gaveen has quit IRC05:37
*** ArdRigh has joined #openstack05:56
*** rsampaio has quit IRC06:03
*** AimanA is now known as HouseAway06:05
*** BK_man has joined #openstack06:25
*** f4m8_ is now known as f4m806:28
*** BK_man has quit IRC06:40
*** joearnold has quit IRC06:43
*** miclorb has joined #openstack06:48
* dillon-w sayhi06:57
*** kashyapc has quit IRC07:05
*** kashyapc has joined #openstack07:05
*** miclorb has quit IRC07:08
*** guigui1 has joined #openstack07:10
*** miclorb has joined #openstack07:13
*** omidhdl has joined #openstack07:25
*** miclorb has quit IRC07:29
*** krish has joined #openstack07:32
*** krish has left #openstack07:33
*** GasbaKid has joined #openstack07:39
*** miclorb_ has joined #openstack07:48
*** arthurc has joined #openstack07:59
*** gaveen has joined #openstack08:02
*** gaveen has quit IRC08:03
*** gaveen has joined #openstack08:04
*** almaisan-away is now known as al-maisan08:05
soreno/08:09
*** gaveen has quit IRC08:14
*** MarkAtwood has quit IRC08:25
zykes-soren: does openstack support console etc now ?08:30
*** allsystemsarego has joined #openstack08:32
*** allsystemsarego has joined #openstack08:32
*** calavera has joined #openstack08:32
*** gaveen has joined #openstack08:32
*** MarkAtwood has joined #openstack08:36
*** miclorb_ has quit IRC08:39
sorenzykes-: What do you mean, exactly?08:40
zykes-is there any console support at the momemnt ?08:41
sorenYou can get the console output using GetConsoleOutput.08:42
*** miclorb_ has joined #openstack08:42
*** miclorb_ has quit IRC08:48
*** gaveen has quit IRC08:52
*** ibarrera has joined #openstack08:52
zykes-what about vnc like console then soren ?08:54
sorenNot at the moment.08:54
zykes-k08:54
zykes-is that planned for bexar ?08:54
alekibangozykes-: there is a blueprint for console over the web08:54
alekibangousing ajax iirc08:54
zykes-iirc ?08:54
alekibangoif i remember correctly08:55
zykes-ah08:55
alekibangozykes-: i am also for vnc08:55
zykes-or you could support all ? ;p08:55
alekibangobut there is also NX and the new kid on the block, spice  from redhat...08:55
alekibangoi never actually seen spice working, but i think we should consider those08:56
alekibangovnc is deadly slow08:56
alekibangozykes-: do you know/use nx?08:56
zykes-spice was pretty ok08:56
zykes-when i used RHEV08:56
zykes-i've used nx from nomachines yes.08:57
alekibangospice is still not yet accepted standard.... and NX is ugly hack :) but works well08:57
alekibangostill there are potential security problems with it08:57
zykes-what's wrong with spice ?08:58
alekibangoits not in qemu08:58
alekibangoyou need to patch it08:58
zykes-then maybe one should get that forward08:58
alekibangowell, i would say go for it, but thats not blocked by me08:58
zykes-by who ? :p08:59
alekibangoi can only guess...08:59
alekibangobut imho it will go in soon09:00
alekibangoas redhat chooses kvm for the future09:00
alekibangohttp://www.spice-space.org/09:00
alekibangozykes-: my personal goal with openstack (not required by my client) is running desktops from cloud09:01
zykes-i know... Spice is allready in RHEV09:01
zykes-it came from Qumranet so09:01
alekibangoyes09:01
alekibangoha, they actually released some new versions i see09:02
alekibango i didnt look for almost year09:02
zykes-how you mean by that ? ;p09:02
alekibangothey had no releases for some time...09:03
alekibangonow its more alive09:03
zykes-ah09:03
alekibangowhich is good sign09:03
alekibangozykes-: can u compare nx to spice in terms of speed?09:03
zykes-don't have access to spice for the time being09:04
alekibangoi am pretty satisfied with nx  - it works like charm (freenx)09:04
alekibangostill its ugly hack on old for of X09:04
alekibango:)09:05
alekibangofor virtual machines, spice might make sense09:05
alekibangos/for of X/ fork of X/09:05
zykes-well it's made for it so09:08
alekibangozykes-: well, swine flu vaccines were made to 'make you safe' but i will not take one...09:08
alekibangobut you are right09:09
*** guigui1 has quit IRC09:13
*** soosfarm has quit IRC09:17
*** soosfarm has joined #openstack09:23
*** befreax has joined #openstack09:27
*** irahgel has joined #openstack09:35
*** miclorb has joined #openstack09:46
*** omidhdl has quit IRC09:57
*** sassyn has quit IRC09:58
*** omidhdl has joined #openstack09:59
*** irahgel has quit IRC10:22
*** rchavik has quit IRC10:45
*** vamsi has joined #openstack10:54
*** irahgel has joined #openstack10:59
*** ArdRigh has quit IRC11:02
*** guigui1 has joined #openstack11:10
zykes-http://vimeo.com/17007388 < hmmm, Ewan productize with ESXi, XenServer and Hyper-V, why not KVM i wonder11:15
*** gaveen has joined #openstack11:22
*** miclorb has quit IRC11:24
*** GasbaKid has quit IRC11:40
*** sophiap has joined #openstack11:40
*** GasbaKid has joined #openstack11:41
*** schisamo has joined #openstack11:45
*** allsystemsarego has quit IRC12:02
*** ar1_ has quit IRC12:15
*** hazmat has joined #openstack12:23
*** omidhdl has quit IRC12:28
*** DubLo7 has quit IRC12:36
*** omidhdl has joined #openstack12:42
sandywalshhttp://twistedmatrix.com/pipermail/twisted-python/2010-November/023190.html12:44
*** Ryan_Lane|away has quit IRC12:48
*** sophiap has quit IRC13:06
*** sophiap has joined #openstack13:06
*** DubLo7 has joined #openstack13:10
sorenGawd, I hate SOAP.13:13
zykes-where it that used ?13:15
*** laurensell has joined #openstack13:20
*** omidhdl has left #openstack13:20
*** johnpur has joined #openstack13:21
*** ChanServ sets mode: +v johnpur13:21
*** ppetraki has joined #openstack13:30
*** matclayton has joined #openstack13:34
*** westmaas has joined #openstack13:37
sorenzykes-: In hell.13:38
*** Podilarius has joined #openstack13:44
zulwhich level of hell though13:50
*** kashyapc has quit IRC13:50
sorenAll of them, I think.13:53
*** hazmat has quit IRC13:53
*** nelson__ has left #openstack13:56
*** BK_man has joined #openstack13:57
*** jero- has joined #openstack13:59
zykes-:p14:02
alekibango+114:03
*** guigui1 has quit IRC14:03
*** westmaas has quit IRC14:04
*** jero- is now known as jero14:10
*** BK_man has quit IRC14:12
zulsoren: i don understand commit #425 how does it simplify the packaging?14:16
sorenzul: Look in the packaging branch.14:17
*** hisaharu has joined #openstack14:18
zulsoren: ah ok14:19
*** hggdh has quit IRC14:19
sorenhttp://bazaar.launchpad.net/~openstack-ubuntu-packagers/ubuntu/natty/nova/ubuntu/revision/24 for those following at home14:19
zykes-what about that soren ?14:23
*** BK_man has joined #openstack14:25
sorenzykes-: Hm?14:26
*** hggdh has joined #openstack14:26
*** hisaharu has quit IRC14:28
*** BK_man has quit IRC14:29
*** matclayton has left #openstack14:30
*** rchavik has joined #openstack14:34
uvirtbotNew bug: #683137 in nova "CA code doesn't get installed by "setup.py install"" [Undecided,New] https://launchpad.net/bugs/68313714:38
*** Ryan_Lane has joined #openstack14:38
sandywalshis the swift mailing list going to merge with the nova mailing list, or will that be kept separate?14:41
*** westmaas has joined #openstack14:44
*** laurensell has quit IRC14:44
sandywalshDid anyone notice the reach-out from the Tahoe-LAFS team to work with Swift? https://lists.launchpad.net/swift/msg00023.html Seems like a great fit. dendrobates?14:45
dendrobatesI'll admit I didn't14:46
dendrobatessandywalsh: it is all going to merge.  Blame thanksgiving, or it would be done already.14:47
*** krish has joined #openstack14:48
*** krish has left #openstack14:48
*** dubsquared has joined #openstack14:50
sandywalshdendrobates, have you looked at Tahoe?14:56
dendrobatesno14:57
sorenI have. A while ago.14:57
sorenIt uses a clever bit of data coding magic to distribute data across N backends in a way such that you you need M of the distributed parts to reconstruct the original data. Without less than M pieces, you have nothing.15:00
sorenIt's similar to Adi Shamir's secret sharing algorithm, but uses less space.15:01
*** jdarcy has joined #openstack15:02
sorenErasure coding!15:03
*** rchavik has quit IRC15:04
*** gondoi has joined #openstack15:04
sorenThat's what it's called. Phew, good thing I finally remembered. that would have kept me awake all night pondering :)15:04
*** guigui1 has joined #openstack15:04
* soren leaves for a couple of hours15:05
*** gundlach has joined #openstack15:08
*** spectorclan has joined #openstack15:08
*** f4m8 is now known as f4m8_15:09
*** sirp has joined #openstack15:11
*** allsystemsarego has joined #openstack15:19
*** allsystemsarego has joined #openstack15:19
*** kevnfx has joined #openstack15:20
*** bryguy has joined #openstack15:26
*** ppetraki has quit IRC15:33
*** ppetraki has joined #openstack15:34
*** gaveen has quit IRC15:37
*** kevnfx has quit IRC15:38
*** Cybo has joined #openstack15:38
*** jimbaker has joined #openstack15:41
creihtsandywalsh: It seemed like less of a friendly hand, and more like spam to me15:49
*** gaveen has joined #openstack15:50
sandywalshcreiht, zooko is solid. He sees the merit of both systems working together.15:50
creihtExcept the only mails that he has sent to the lists are press releases for the latest releases of LAFS15:51
creihthttps://lists.launchpad.net/swift/15:52
sandywalshyeah, I can see how you would form that opinion. I guess the project doesn't have the weight of RS & Nasa behind it.15:52
creihtIt may be a great project, and I'm all for sharing, but I'm not keen on spaming other peoples mailing list, trying to get your project mor press15:53
zaitcevI do not want to buddy with that stupid asshole either.15:54
*** mdomsch has joined #openstack15:54
sandywalshthat's fair. I guess we can slap wrists later :) I'm just trying to shed a little light on what might be a nice fit.15:54
*** Mr_T has left #openstack15:55
creihtlol15:55
sandywalsh:)15:55
sandywalshperhaps something I'm missing here15:56
johnpurbtw, we have looked at solutions involving erasure coding and Swift. it is a possible way to lessen the cost burdens of redundant copies for availability purposes.15:56
zaitcevYea, like the concept of spam.15:56
johnpurbig architectural change tho15:57
creihtjohnpur: likes erasure coding :)15:57
johnpur:)15:57
johnpur<- erasure code fanboy15:58
sandywalshwould it be beneficial to have him join us on the channel for a chat?15:59
creihtAnyone is welcome on this channel to chat :)16:00
*** ambo has joined #openstack16:00
sandywalshjohnpur, I don't really know enough about the pro's/con's (and have no agenda here ... just making introductions). But perhaps the arch change will be smaller with a partner?16:02
sandywalshor have it as a plug-in?16:02
*** guigui1 has quit IRC16:04
*** dfg_ has joined #openstack16:04
*** gaveen has quit IRC16:04
creihtsandywalsh: we have tossed around the idea of doing something like erasure coding, but it would require a complete re-architecting of the backend16:06
creihtAnd make the code/system much more complex16:06
creihtTurns out that the simple solution works quite well for us :)16:08
creihtI had looked a little a tahoe a while back, and it didn't really support our use case16:09
sandywalshcreiht, understandable. Do you think as openstack adoption increases the need for client-based encryption will increase as well?16:10
creihtThat is unknown, but I'm sure we will address it if it comes up :)16:10
*** eldarnugaev has joined #openstack16:10
sandywalshperhaps it's a similar problem to the openstack vs. ec2 api problem in nova ... could the swift api be the standard and the implementations pluggable? (I'm spit-balling here) i.e. does a consumer need to adopt the whole stack?16:12
creihtOn encryption, for swift, nothing prevents the clients from encrypting their data16:13
sandywalshtrue16:14
*** ambo has left #openstack16:14
creihtthe swift api is the "standard" storage api for openstack, of course everything else is an implementation detail :)16:15
*** kevnfx has joined #openstack16:16
creihtbut even then, as I understand it the storage stuff for nova is plugable, so that doesn't even require a swift api (if you want to write code at that layer instead)16:17
creihtor will be plugable16:17
sandywalshhmm, but wouldn't that just fragment swift (if the swift clients need drivers for each different implementation)?16:20
*** johnbergoon has joined #openstack16:22
creihtsandywalsh: depends on if you are wanting generic storage, or just storage to store your images on16:23
sandywalshshould it matter what the blob contains?16:24
creihtno16:24
creihtat least from swift's point of view16:25
*** daleolds has joined #openstack16:25
alekibangosandywalsh: swift is more likely suitable for cold storage. for images of live instances, it would be suboptimal...   sheepdog might be interesting for that..16:27
sandywalshalekibango, that's an interesting point. I clearly need to get more hands-on with swift.16:29
*** jk0 has quit IRC16:30
*** jk0 has joined #openstack16:30
*** ChanServ sets mode: +v jk016:30
*** laurensell has joined #openstack16:31
jaypipesgundlach: looks like nova/tests/api/openstack/test_auth.py in test_authorize_token is returning some stubbed out URL instead of the x-server-management-url ?16:36
gundlachjaypipes: yeah, looking at that now, thanks.16:37
gundlachx-server-mgmt-url contains 'self.host' and either the test or the code isn't setting host to 'foo' here.16:37
*** jk0 has quit IRC16:39
*** HouseAway is now known as AimanA16:39
*** jk0 has joined #openstack16:40
*** ChanServ sets mode: +v jk016:40
gundlach_cerberus_: in nova.tests.api.openstack.test_auth.Test.test_authorize_token, is the test supposed to call fake_auth_init() on something to set its self.host to 'foo'?16:43
*** eldarnugaev has quit IRC16:43
gundlach_cerberus_: i don't quite get how fake_auth_init and FakeAuthManager and stubbing out BasicApiAuthManager.__init__ work together.16:44
zykes-alekibango: what about tahoe-lafs ?16:44
*** calavera has quit IRC16:45
*** rsampaio has joined #openstack16:45
zykes-nvm16:46
*** jk0 has quit IRC16:48
*** jk0 has joined #openstack16:48
*** jk0 has joined #openstack16:48
*** ChanServ sets mode: +v jk016:48
*** AimanA is now known as HouseAway16:49
*** eldarnugaev has joined #openstack16:49
mtaylormorning eday16:51
*** GasbaKid has quit IRC16:52
*** GasbaKid has joined #openstack16:54
alekibangozykes-: tahoe-lafs might be fun, but its not a match16:58
sandywalshgundlach, the failing test is one I wrote when I had the management-flag in there. You can probably remove the test now with your changes.17:02
*** greenisus has joined #openstack17:03
gundlachsandywalsh: hm, i didn't realize your code landed in trunk.  you sure?17:05
sandywalshgundlach, it didn't, but you made your changes from my branch, yes?17:05
Ryan_LaneI'm getting an error trying to run euca-describe images in trunk: TypeError: index() takes exactly 2 arguments (3 given)17:06
gundlachsandywalsh: nope, from trunk, i thought17:06
sandywalshgundlach, hmm, don't think17:06
gundlachsandywalsh: the test refers to 'herp' and 'derp' so i'm pretty sure _cerberus_ wrote it :)17:06
*** thatsdone has joined #openstack17:06
sandywalshah17:06
gundlachsandywalsh: i'll go doublecheck17:06
Ryan_Lanehas something changed? I'm using nova.sh...17:06
edaymtaylor: morning17:10
_cerberus_gundlach: I'm here17:10
gundlach_cerberus_: see question above re fake_auth_init()17:10
_cerberus_repaste please. I don't backlog this room17:11
gundlach_cerberus_: in nova.tests.api.openstack.test_auth.Test.test_authorize_token, is the test supposed to call fake_auth_init() on something to set its self.host to 'foo'?17:11
gundlach* eldarnugaev has quit (Quit: eldarnugaev)17:11
gundlach<gundlach> _cerberus_: i don't quite get how fake_auth_init and FakeAuthManager and stubbing out BasicApiAuthManager.__init__ work together.17:11
_cerberus_Lemme look17:12
gundlach(https://code.launchpad.net/~gundlach/nova/server-management-url/+merge/42194 is failing and i'm not sure why)17:12
*** befreax has quit IRC17:12
gundlach_cerberus_: it kind of looks like I need to instantiate BasicApiAuthManager in my test, since setUp() stubs out its __init__ with fake_auth_init which sets self.host to 'foo'.17:13
dubsquaredgreetings, #openstack — anyone seen this recently:  root@openstack111:/etc/nova# euca-describe-images17:14
dubsquaredTypeError: index() takes exactly 2 arguments (3 given)17:14
dubsquaredi have running instances…but i just want a listing so i can spin up others with different ami info..17:15
Ryan_Lanedubsquared: I'm having the same problem :(17:15
*** thatsdone has quit IRC17:15
dubsquaredryan_lane:  are you having any success ssh'ing to your running instances?17:16
Ryan_Lanehaven't tried lately17:16
Ryan_Lanedubsquared: did you set up the default security group?17:16
dubsquaredi can ping and ssh to them, but they are telling me password is wrong…but i snapped the image off the controller17:16
sandywalshI think there were changes to nova.sh in github that aren't in contrib yet17:17
Ryan_Laneah.17:17
dubsquaredeuca-authorize -P icmp -t -1:-1 default17:17
dubsquaredeuca-authorize -P tcp -p 22 default17:17
*** al-maisan is now known as almaisan-away17:17
dubsquaredyes17:17
dubsquarednot sure how this login info is not working :|17:17
*** guynaor1 has quit IRC17:17
Ryan_Lanedubsquared: are you connecting with a key?17:18
dubsquared nah17:18
Ryan_Laneor are you using an image that has a password?17:18
dubsquaredcorrect17:18
_cerberus_gundlach: Try creating an instance of that class, as you suggested17:21
_cerberus_Because my stub code definitely overrides the host, which is how the solution worked previously17:21
gundlach_cerberus_: don't do anything with it, just instantiate one?17:21
gundlach_cerberus_: ok17:21
_cerberus_gundlach: actually, since you replaced the url with the correct one, you probably need to find a way to stub req.url17:23
_cerberus_or at least get it to return what you want it to17:23
gundlach_cerberus_: ok, sorry for my ignorance.  looks like instantiating BasicApiAuthManager doesn't cut it, but i'll try stubbing req.url as you suggested after lunch17:23
gundlachthx!17:23
_cerberus_gundlach: see what I'm saying? The stub code would handle it IF you were still using host17:24
_cerberus_But since you're not...17:24
_cerberus_and you're welcome :-)17:24
gundlachahh, right :) ok17:24
Ryan_Laneand the github version of nova.sh doesn't work either17:28
* Ryan_Lane sighs17:28
dendrobatesvishy: I'm merging trunk back into the raw-disk-images and I have a question about the rescue mode changes to nova/virt/libvirt_conn.py17:30
dendrobatesvishy: I think changes need to be made to handle rescuing raw images with no kernel or ramdisk in the xml.17:32
zaitcev"Ryan_Lane> I'm getting an error trying to run euca-describe images in trunk: TypeError: index() takes exactly 2 arguments (3 given)" -- if there's a space between describe and images...17:32
Ryan_Lanethat space was a typo17:34
*** daleolds has quit IRC17:35
Ryan_LaneI'm typing: euca-describe-images17:35
dendrobatesvishy: actaully I looked at it again, and I think it will work unchanged.  I just didn't want to break it.17:38
zaitcevActually I am not very impressed with euca2ools, I am wondering if a fork exists.17:39
*** hornbeck has joined #openstack17:39
zaitcevFor one thing, they are far from a drop-in replacement, e.g. command line options are not compatible.17:40
*** maplebed has joined #openstack17:40
zaitcev2. retarded random bugs like https://bugzilla.redhat.com/show_bug.cgi?id=65856017:41
uvirtbotbugzilla.redhat.com bug 658560 in euca2ools "fail with a trailing slash" [Medium,New]17:41
zaitcevor plainly not working right (e.g. missing images)17:41
zaitcev3. Contribution mechanism is screwed-up with copyright assignments17:41
*** schisamo has quit IRC17:43
Ryan_Laneis no one else having this problem with trunk?17:43
JordanRinkewhat is wrong?17:44
Ryan_Laneeuca-describe-images is throwing the following error: TypeError: index() takes exactly 2 arguments (3 given)17:44
zykes- any howto guide on siwft ?17:46
zykes-swift.17:46
JordanRinkeswift.openstack.org has good doc on swift17:46
JordanRinkeincluding install instructions that I have used before and they were flawless17:46
zaitcevzykes-: Use SAIO for testing, normal config from wiki for production.17:47
JordanRinkeRyan_Lane: intersting - what OS?17:47
Ryan_Laneubuntu 10.04 lts17:47
zykes-zaitcev: SAIO ?17:48
Ryan_LaneI'm getting the same response when using aws-sdk17:48
Ryan_Lanewhich means something is likely broken in the api17:48
JordanRinkeodd, I have used those tools with no problems on that OS17:48
JordanRinkedo you ahve your novarc file setup and is it loaded?17:48
Ryan_Laneyeah. it isn't giving me a 40317:48
Ryan_Laneother commands work fine17:49
Ryan_Lanesuch as euca-describe-instances17:49
JordanRinkeinteresting17:49
zaitcevzykes-: Like http://swift.openstack.org/development_saio.html (also included into installed docs)17:49
JordanRinkeRyan_Lanen: https://bugs.launchpad.net/nova/+bug/681600 :(17:51
uvirtbotLaunchpad bug 681600 in nova "DescribeImages without image_id argument broken" [Medium,Triaged]17:51
Ryan_Lanegah17:51
Ryan_Lanethat's a PITA17:51
JordanRinkeindeed17:52
* annegentle growls at the typo on the SAIO (my bad) :) 17:58
*** laurensell has quit IRC18:06
*** abecc has joined #openstack18:10
*** abecc has joined #openstack18:11
*** ibarrera has quit IRC18:11
*** theron has joined #openstack18:11
*** kevnfx has left #openstack18:12
*** aliguori has quit IRC18:14
*** dantti_work has joined #openstack18:16
*** joearnold has joined #openstack18:16
*** laurensell has joined #openstack18:17
dantti_workhi, I've just installed openstack but adding iso does not work (at least not with http and file:/), with http the item shows but after refresh it's gone, which log should I look to get what is going on?18:18
*** laurensell has quit IRC18:23
xtoddxhow do i make sense of a "bzr praise FILENAME" revision number?  It looks like 504.1.7, but it doesn't show up in revision 504 on launchpad web view.18:23
JordanRinkedantti_work: can you explain a little more?18:23
edayxtoddx: are you looking at the same branch? each branch could be numbered differently18:24
xtoddxeday: yea same branch ~anso/nova/deploy in both places18:24
openstackhudsonProject nova build #269: SUCCESS in 1 min 11 sec: http://hudson.openstack.org/job/nova/269/18:24
xtoddxi just plugged the revision number into the URL with the dots and at seems to work, but I hate that it isn't discoverable18:25
dantti_workJordanRinke: sure, well I go to "Templates" -> "ISO" -> add an Iso, fill the fields, pass an url to the iso (on a http server) click create. It then shows up on the list but when I click refresh the iso item is gone18:25
dantti_workand I can't create a new vm18:25
zykes-dantti_work: what GUI ?18:26
dantti_workdunno it's called cloud.com console, which looking around the web drove me to this irc channel :P18:27
dantti_workmaybe it has nothing to do :P18:27
JordanRinkedantti_work: I don't think that is an actual part of OpenStack18:27
JordanRinkeas far as I know openstack currently has no UI18:27
dantti_workhmm sorry then... ah it's so hard to find a goot ui to kvm18:28
* dantti_work feels stupid :P18:28
*** nati has joined #openstack18:33
*** arthurc has quit IRC18:35
*** aliguori has joined #openstack18:36
annegentledantti_work: nah, don't feel stupid, we're working on a Live ISO to play around with OpenStack but I don't know if you can use it on Cloud.com, might be a good place to test though.18:41
dantti_workannegentle: right, so is openStack another cloud management ui?18:42
dantti_workI have tried so many guis to get rid of vm server terreble ui :P but no lucky still18:42
*** ctennis has joined #openstack18:43
dantti_workopen.cloud.com was the first ui that installed and seems powerfull enough but I can't create a vm in it...18:43
annegentledantti_work: no GUI, yet.18:43
vishydendrobates: I would think the rescue mode code would work, although I think we should add a flag to images specifying whether they are whole disk.18:43
annegentledantti_work: and yeah, seems like the UI should cover the basics :)18:44
*** ctennis has quit IRC18:44
dantti_workannegentle: sure, hope you can provide a good alternative though, another gui that seems promising is Archipel but it never gets released :P18:45
annegentledantti_work: we're working on it I assure you :)18:46
dantti_worknice thanks18:46
alekibangodantti_work: web ui and desktop ui will be there for openstack, few months later18:46
*** joearnold has quit IRC18:48
xtoddxalekibango dantti_work: greenisus is also working on an iOS/mobile UI18:49
alekibangoyes i know18:50
alekibangoi categorized that under 'desktop' hehe18:50
alekibangoyou are right, its different18:50
dantti_workwell if I had time I could try to help with maybe a Qt plugin for desktops but I alread have too many os projects to care about :P18:51
alekibangodantti_work: this project is not like any other. you will see that soon :)18:52
greenisusxtoddx alekibango dantti_work yeah, that's right.  i'm working on the iOS client right now, and we just started the openstack-client-developers in launchpad.  you should join to keep up with it!18:52
alekibangogreenisus: i dont use such tools... i am more interested in desktop18:52
alekibangobut i will watch18:52
greenisusalekibango we're also working on web clients.  no desktop app plans at the moment18:53
alekibangoah, thats my area of interest18:53
alekibangobut i have heard some other web management console will be released soon, right vishy?18:53
*** Ryan_Lane is now known as Ryan_Lane|food18:54
dantti_workyep, I also don't have any iHardware ...18:54
*** joearnold has joined #openstack18:54
notmynamegreenisus: where did you hear the top 2% stat?18:56
greenisusfrom dendrobates on the phone :)18:56
alekibango? what 2% stat?18:57
dendrobatesnotmyname: http://www.ohloh.net/p/novacc/factoids/391101618:57
JordanRinkeyeah, I was about to ask the same thing, what 2%?18:57
alekibangowow18:57
alekibangoi think when we will talk about importance, it will be at least top 0.3% :)18:58
dendrobatestrue, I don't know their methodology18:58
notmynameseems that their search is broken19:00
*** sophiap has quit IRC19:00
*** sophiap has joined #openstack19:00
greenisusyeah, sorry for not clarifying there :)19:02
notmynameand as a small nit to pick, openstack != nova ;-)19:03
*** hornbeck has quit IRC19:09
openstackhudsonProject nova build #270: SUCCESS in 1 min 13 sec: http://hudson.openstack.org/job/nova/270/19:10
*** hornbeck has joined #openstack19:10
*** kevnfx has joined #openstack19:11
*** GasbaKid has quit IRC19:15
*** abecc has quit IRC19:15
*** GasbaKid has joined #openstack19:16
*** laurensell has joined #openstack19:17
*** dendrobates is now known as dendro-afk19:18
*** laurensell has quit IRC19:18
*** hazmat has joined #openstack19:19
*** kevnfx has quit IRC19:19
*** dendro-afk is now known as dendrobates19:20
dendrobatesnotmyname: they list swift separately19:23
notmynameya, I saw that19:23
*** sophiap has quit IRC19:24
dendrobatesbut if nova > 2% than openstack > 2%19:24
notmynameagreed19:27
ttxTeam meeting in 90 minutes in #openstack-meeting !19:31
uvirtbotNew bug: #683308 in nova "AttributeError: 'FlatManager' object has no attribute 'allocate_network'" [Undecided,New] https://launchpad.net/bugs/68330819:32
*** jfluhmann has quit IRC19:32
*** jfluhmann has joined #openstack19:32
* creiht still wonders why nova and swift are not on the launchpad.net featured projects list on the front page ;)19:33
* ttx doesn't trust ohloh a lot.19:33
* ttx looks up the lp.net front page for the first time in centuries.19:34
*** spectorclan has quit IRC19:34
ttxcreiht: ah! even Drizzle is in there.19:34
*** HouseAway is now known as AimanA19:52
sorenIt's going to be a while before I get tired of being able to just "apt-get update && apt-get dist-upgrade" and have bleeding edge packages.19:53
sorenI wonder why more projects aren't doing it. It wasn't /that/ hard.19:53
*** eldarnugaev has quit IRC19:57
ttxmtaylor: around ?19:58
*** Ryan_Lane|food is now known as Ryan_Lane19:59
edaysoren: could I bug you for an approve or needs fixing with specifics? https://code.edge.launchpad.net/~eday/nova/compute-abstraction/+merge/4180519:59
*** westmaas has quit IRC20:02
*** westmaas has joined #openstack20:02
edayvishy: any feedback on this one before merging? https://code.edge.launchpad.net/~rlane/nova/ldapimprovements/+merge/4099320:03
*** MarkAtwood has quit IRC20:06
gundlacheday: mind re-approving https://code.launchpad.net/~gundlach/nova/novash/+merge/42195 ?  i merged from trunk and hudson apparently now doesn't let ninja-patches through.20:10
*** gundlach is now known as gundlach_bbiab20:10
mtaylorttx: ola!20:15
*** jfluhmann has quit IRC20:15
ttxmtaylor: I had a few questions on tarmac for you20:15
*** jfluhmann has joined #openstack20:15
mtaylorttx: yes. I have all of the answers20:15
mtaylorttx: you don't want to do some java hacking, do you?20:16
ttxmtaylor: I'll probably have to plug a CLA verifier into the workflow20:16
ttxmtaylor: i.e. check that the person submitting the branch actually is listed in a CLA signers list20:16
ttxmtaylor: sounds like a job for a tarmac plugin20:16
mtaylorttx: ok. oh - so I should send you the write up on how I was planning on having things hang together20:16
ttxmtaylor: that would be great.20:17
ttxmtaylor: is the config we use stored in bzr somewhere ?20:17
soreneday: Let me have a look.20:17
mtaylorttx: no. but it should be20:17
*** jfluhmann has quit IRC20:17
mtaylorttx: ACTUALLY - it sounds like a job for a hudson plugin - we need tarmac's functionality more integrated with hudson to be able to sensibly do more complex things20:17
mtaylorttx: lemme dig up my email about it20:18
ttxmtaylor: ok20:18
*** joearnold has quit IRC20:18
mtaylorttx: you will find my current setup of tarmac to be done wrong in just about every way possible - it's installed from source from bzr trunk, and the config file is just edited directly on disk :(20:19
ttxmtaylor: about Java, I wouldn't mind if the last Java I ever did was on a good project like Hudson.20:19
edaymtaylor: you following the potential hudson fork?20:19
ttxeday: I've seen that. I like "Alfred".20:19
mtayloreday: no. is someone going to fork it from kohsuke?20:20
ttxmtaylor: from Oracle, in fact.20:20
mtayloroh - well, that's already done20:20
edaymtaylor: http://www.hudson-labs.org/content/whos-driving-thing20:20
mtaylorkohsuke == hudson - if oracle decides they also want to develop hudson, they can suck it20:20
ttxOracle apparently owns the name.20:21
*** jfluhmann has joined #openstack20:21
edayI like the suggestion of Mr. Belvedere20:21
mtaylorah.20:23
mtaylorfucking oracle20:23
mtaylorI think oracle can suck a fat dick20:23
* mtaylor should perhaps keep those things on #drizzle20:23
*** brd_from_italy has joined #openstack20:23
* mtaylor apologizes to folks here perhaps not used to his tendency towards obscenity20:23
ttxmtaylor: obscenity towards Oracle is always accepted.20:24
* ttx used to be forced to use it.20:24
mtaylorttx: do you have a new rackspace email you'd prefer I use?20:26
ttxmtaylor: thierry@openstack.org20:26
ttxMeeting in 30 minutes in #openstack-meeting20:31
sandywalshI'm just going through the create code in openstack api. Sometimes it calls nova.compute.manager directly and sometimes it uses the queue. Is direct vs. indirect completely based on time-for-function-to-complete?20:34
sorengundlach_bbiab: With your recently landed change to the openstack API, is python-cloudservers supposed to work? I've tried changing AUTH_URL to http://127.0.0.1:8774/v1.0/, but I just get 404's back.20:37
soreneday: Ok, now I'm /actually/ taking a look. I got distracted by a somewhat delicate e-mail exchange that needed my attention.20:38
*** burris has quit IRC20:38
*** burris has joined #openstack20:39
*** westmaas has quit IRC20:39
edaysoren: no worries, and thanks :)20:41
*** burris has quit IRC20:41
edaysandywalsh: can you explain with more context?20:42
*** westmaas has joined #openstack20:42
edaysandywalsh: you might want to take a look at my oustanding branch if your touching code in there20:42
mtaylorttx: k. sent20:43
edaysandywalsh: https://code.edge.launchpad.net/~eday/nova/compute-abstraction/+merge/41805  this might overlap with what you're doing20:43
mtaylorttx: I couldn't find what I wrote before, so I just wrote it again :)20:43
ttxheh20:44
sandywalsheday, I was wondering why sometimes api can call compute directly and other times it has to go through the queue.20:45
edaysandywalsh: depends on if it should use the worker or a function in-process20:45
sandywalsheday, right, and is that decision purely based on time-to-complete?20:46
edaysandywalsh: but a lot of this is in the process of shuffling around, we're trying to make the API do almost nothing but ship events, I think :)20:46
sandywalsheday, that's the answer I was looking for20:46
edaysandywalsh: nope, really just depends on what you want to do20:46
soreneday: Followed up.20:46
sorenbbiab20:46
vishyeday: I haven't tested the ldap stuff, but it all looks fine to me20:47
sandywalsheday, I'm in the process of trying to get openstack calls to work with xenserver on my dev machine and just studying the code at this point.20:47
Ryan_LaneI tested it with and without the config option set. seems to work reliably20:48
ttxmtaylor: so you would get rid of tarmac completely.20:48
sandywalsheday, but that just struck me as a little arbitrary. Thanks for the branch link I'll look at it.20:48
Ryan_LaneI didn't test it with openldap, but it shouldn't matter which ldap server is used20:49
mtaylorttx: yes. once we can have it actually integrated into hudson20:49
ttxmtaylor: I'm not familiar with the merge queue concept and how you go from "approved merge request" to "in the merge queue".20:49
mtaylorttx: they overlap in functionality too much20:49
edayvishy: cool, I'll go ahead and approve then20:50
*** rsampaio has quit IRC20:50
mtaylorttx: oh - so they're adding a new feature to launchpad called a merge queue - this keeps the approved/denied status of a merge request separate from its life of actually being merged or not20:50
ttxmtaylor: If the "build" checks succeed, you push the approved merge request to the merge queue ?20:50
mtayloryes20:50
ttxok, that was the missing bit.20:50
mtayloroops :)20:51
mtaylorttx: the idea being the the 'build' checks are quick and cheap to run - so you could have 20 hudson workers doing 'build' checks on all outstanding merge requests simultaneously if you wanted to20:52
*** gundlach_bbiab is now known as gundlach20:52
ttxmtaylor: ok, I need to figure out if that plan is compatible with the timeframe we want for CLA verification.20:52
mtaylorttx: but then once you're actually going to be doing integration testing, those need to be serialized more20:53
*** gondoi has quit IRC20:53
*** irahgel has quit IRC20:53
mtaylorttx: we certainly don't have to do that whole plan to get CLA verification into the mix ... you could certainly write a tarmac plugin for now and then we could just add that as a thing adrian needs to do for the hudson plugin20:54
ttxmtaylor: yes20:54
mtaylorttx: of course, if we're going to start doing CLA checking - I would recommend that we go full-bore and require signed commits too20:54
* mtaylor signs all of his commits already, so it wouldn't change anything for him :)20:55
*** jbryce has joined #openstack20:55
*** spectorclan has joined #openstack20:55
ttxTeam meeting in 5 minutes in #openstack-meeting20:56
gundlachsoren: dunno yet -- as i get GroundControl working on OpenStack I may uncover other bugs.20:56
ttxmtaylor: let me know if I can access the current tarmac config in some way, to see how it's currently set up20:58
mtaylorttx: do you have ssh access to hudson.openstack.org?20:59
ttxmtaylor: no.20:59
*** rlucio has joined #openstack20:59
mtaylorttx: ah - well, that'll be first step - OR ... http://paste.openstack.org/show/211/21:00
ttxTeam meeting starting NOW in #openstack-meeting !21:00
mtaylorttx: I've also got a script half-done to add ssh account for all members of a launchpad team to a box21:00
openstackhudsonProject nova build #271: SUCCESS in 1 min 15 sec: http://hudson.openstack.org/job/nova/271/21:00
*** eldarnugaev has joined #openstack21:02
ttxjaypipes: meeting ?21:05
gundlachlast seen 4 hrs ago21:05
openstackhudsonProject nova build #272: SUCCESS in 1 min 18 sec: http://hudson.openstack.org/job/nova/272/21:06
*** irahgel has joined #openstack21:07
*** gundlach_ has joined #openstack21:07
*** gundlach has quit IRC21:11
*** gundlach_ is now known as gundlach21:11
*** joearnold has joined #openstack21:12
*** algorist has joined #openstack21:13
*** adjohn has joined #openstack21:14
*** dai__ has joined #openstack21:15
sorenttx: I've configured the bug trackers for Nova, Swift, and Glance to say "Please include the exact version of X with which you're experiencing this issue.".21:16
jaypipesttx: /me thought it was in an hour... sorry.21:17
sorenttx: It's not as prominent as I'd hoped, but it's there.21:17
ttxsoren: ah ok. Wasn't obvious last bug I filed ;)21:17
ttxjaypipes: no problem. I just assumed teller-api was on track.21:17
sorenttx: No no, I just did now.21:18
jaypipesttx: yes, it is.21:18
sorenttx: I saw your BugLifeCycle page and saw that people should do that, so I added it on Launchpad.21:18
ttxsoren: ah, that explains why I missed it, two days ago.21:18
zykes-is it meeting time ?21:18
gundlachyes21:18
gundlach#openstack-meeting21:18
ttxsoren: BugsLifecycle is still a draft, but I still think people should do that anyway.21:18
*** hornbeck has quit IRC21:18
* ttx realiazes his obnoxious meeting reminders are not working well enough21:19
*** hornbeck has joined #openstack21:19
gundlachttx: perhaps set the channel topic a couple hours ahead of time?21:19
sorenttx: I think I have an idea for that..21:19
ttxsoren: I planned to piggyback on some channel bot21:20
sorenttx: Let me just try it out.21:20
sorenttx: that could work.21:20
*** kevnfx_ has joined #openstack21:22
sorenttx: How about a Google Calendar. Anyone using a calendar on their computer (be it in Outlook, Evolution, on Google, whatever) can import that and use whatever notification method they're used to?21:34
*** infernixx is now known as infernix21:34
sorenApologies for my poor punctuation.21:34
*** GasbaKid has quit IRC21:35
*** GasbaKid has joined #openstack21:35
*** gundlach has quit IRC21:35
*** gundlach has joined #openstack21:36
*** abecc has joined #openstack21:39
*** abecc has joined #openstack21:39
*** dai__ has quit IRC21:41
*** ddumitriu has joined #openstack21:42
*** nati has quit IRC21:43
*** hornbeck has quit IRC21:43
*** hornbeck has joined #openstack21:43
*** pothos_ has joined #openstack21:44
*** pothos has quit IRC21:46
*** pothos_ has quit IRC21:47
*** pothos has joined #openstack21:47
*** spectorclan has quit IRC21:50
*** kevnfx_ has left #openstack21:51
sorenttx: ^21:52
*** laurensell has joined #openstack21:54
sorenvishy: My e-mail reveals that you set the commit message on  https://code.launchpad.net/~soren/nova/lp681600/+merge/4218621:56
sorenvishy: Any particular reason?21:56
vishysoren: I approved for merge, but it needed a commit message to merge properly21:56
sorenvishy: Did something fail, or did you just believe that to be true? :)21:56
vishyum I believed that to be true from past attempts21:57
sorenvishy: Tarmac uses the description if there is no commit message set. It only fails if neither is set.21:57
vishysoren: nice, it didn't used to do that21:57
sorenvishy: This has been so for... quite a while.21:57
sorenNot sure when. Months for sure.21:57
vishyok21:58
*** anotherjesse has joined #openstack21:58
sorenvishy: There. I just saved you 12 seconds per day. You can pay me back in beer.21:58
*** anotherjesse has quit IRC21:58
*** anotherjesse has joined #openstack21:59
*** miclorb_ has joined #openstack21:59
*** laurensell has quit IRC22:06
*** laurensell has joined #openstack22:08
*** Cybo has quit IRC22:09
*** jdarcy has quit IRC22:13
*** hornbeck has quit IRC22:13
*** hornbeck has joined #openstack22:14
*** adjohn has quit IRC22:15
*** masumotok_ has joined #openstack22:16
*** holoway has quit IRC22:18
*** holoway has joined #openstack22:20
*** kevnfx_ has joined #openstack22:20
*** kevnfx_ has left #openstack22:21
*** allsystemsarego has quit IRC22:21
*** DubLo7 has quit IRC22:22
*** anotherjesse has quit IRC22:26
*** anotherjesse has joined #openstack22:27
*** gundlach has quit IRC22:29
*** westmaas has quit IRC22:29
*** anotherjesse has quit IRC22:36
dubsquaredvishy:  any recent changes that cause dnsmasq to not come back up on a nova-network restart?22:41
*** GasbaKid has quit IRC22:41
uvirtbotNew bug: #683413 in swift "Improve error handling in swift-auth-add-user" [Low,Confirmed] https://launchpad.net/bugs/68341322:41
*** eldarnugaev has quit IRC22:42
vishydubsquared: it should come back22:43
*** hornbeck has quit IRC22:43
dubsquaredvishy: hrm, when i kill it, and restart nova-network/libvirt-bin only one spawns22:44
vishyaer22:44
vishyif libvirt-bin is restarted22:44
vishyit will start dnsmasq22:44
vishyyou have to kill libvirts dnsmasq22:44
*** maplebed has quit IRC22:45
dubsquaredyeah, that is what im killing22:45
*** maplebed has joined #openstack22:45
dubsquared dnsmasq --strict-order --bind-interfaces --pid-file=/var/run/libvirt/network/default.pid --conf-file=  --listen-address 192.168.122.1 --except-interface lo --dhcp-range 192.168.122.2,192.168.122.254 --dhcp-lease-max=25322:45
dubsquaredWARNING:root:Starting network node22:46
dubsquared   ...done.22:46
dubsquaredroot@openstack110:/etc/init.d# ps aux | grep dns22:46
dubsquaredroot      7722  0.0  0.0   7620   912 pts/0    S+   16:46   0:00 grep --color=auto dns22:46
vishyhmm22:46
dubsquared2010-11-30 16:46:31-0600 [-] Log opened.22:47
dubsquared2010-11-30 16:46:31-0600 [-] twistd 10.0.0 (/usr/bin/python 2.6.5) starting up.22:47
dubsquared2010-11-30 16:46:31-0600 [-] reactor class: twisted.internet.selectreactor.SelectReactor.22:47
dubsquared2010-11-30 16:46:31-0600 [-] set uid/gid 108/6553422:47
dubsquared2010-11-30 16:46:31-0600 [-] (root): INFO backend <module 'nova.db.sqlalchemy.api' from '/usr/lib/pymodules/python2.6/nova/db/sqlalchemy/api.pyc'>22:47
vishyare you in flatdhcp22:47
vishy?22:47
vishyI'm just looking and it looks like it doesn't actually recreate it on reboot22:47
vishyhmm22:47
dubsquared--network_manager=nova.network.manager.FlatManager22:47
*** ddumitriu has quit IRC22:48
vishyuh FlatManager doesn't do dhcp22:48
vishy..22:48
*** aliguori has quit IRC22:49
dubsquaredha….i should be referring to my documentation that defines the network managers22:49
dubsquareddoes that exist anywhere :(22:49
vishythere are pretty descriptive docstrings now22:50
vishy:)22:50
dubsquaredsweeet22:50
*** theron has quit IRC22:50
*** theron has joined #openstack22:50
vishyhttp://nova.openstack.org/devref/network.html#module-nova.network.manager22:51
vishyalthough I just noticed a typo22:51
*** jimbaker has quit IRC22:51
vishydubsquared: It may not actually reload dnsmasq on reboot, I have some code for that that I will be proposing soon22:53
dubsquaredvishy:  cool, would be nice :D22:54
dubsquaredlooking to see what manager i should switch to22:54
vishyFlatDHCPManager probably22:55
*** theron has left #openstack22:55
*** laurensell has quit IRC22:57
*** anotherjesse has joined #openstack22:58
*** brd_from_italy has quit IRC23:00
*** jbryce has quit IRC23:03
*** kevnfx_ has joined #openstack23:05
*** ppetraki has quit IRC23:06
*** kevnfx_ has quit IRC23:06
notmynameanotherjesse: I think I can help with your email + iPad issue23:11
*** jimbaker has joined #openstack23:12
*** jbaker has joined #openstack23:12
*** jimbaker has quit IRC23:12
*** jbaker is now known as jimbaker23:12
*** anotherjesse has quit IRC23:19
*** aliguori has joined #openstack23:19
*** eldarnugaev has joined #openstack23:20
*** jakedahn has joined #openstack23:33
*** KyleM1 has joined #openstack23:43
*** rlucio has quit IRC23:43
*** johnbergoon has quit IRC23:48
*** abecc has quit IRC23:49
*** rchavik has joined #openstack23:52
*** rchavik has joined #openstack23:52
*** johnpur has quit IRC23:57
*** irahgel has left #openstack23:57

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