Thursday, 2011-08-25

mtaylorif it's not, then something is broke00:00
gholtI'm guessing we'd probably never use the public PPA; control over which version is deployed when is important to each deployer.00:00
mtayloractually spoke with notmyname about some ideas for that earlier00:00
mtaylorthe "have a persistent repo for each release" isn't a bad idea there00:00
mtaylorif by "each release" we mean actual releases and not trunk pacakges... since that's a whole different kettle of fish00:01
gholtOf course, there's always pinning too, but you have to have a reasonable guarantee that old packages wouldn't disappear on you.00:01
mtayloryes00:01
mtaylormain thing though is what you said - I want to do best effort, and if you still need to diverge, then fine00:02
notmynamemtaylor: actually, the "persistent repo" thing was all gholt :-)00:02
mtaylornotmyname: was it?00:02
*** ahmed_ has joined #openstack-dev00:02
notmynamehe did it first00:02
mtaylorgholt: great- then I was having a conversatoin with YOU about that00:02
mtaylor:)00:02
gholtFor me, I didn't want to get into all that [meaning I didn't want to manage it :)] It was easier at the time to just make a separate repo. Easy peasy.00:02
gholtI only had to answer to my ops guys. You have to answer to everybody on the planet, which I'm sure is fun, fun fun. :)00:03
mtaylorI think there are some really easy and/or sensible ways to automate a good portion of this00:03
mtayloryes. yes it is00:03
mtaylorESPECIALLY when people have orthogonal and partially conflicting end goals00:03
mtayloror assumptions00:03
*** AhmedSoliman has quit IRC00:05
gholtPersonalities matter too, of course. Oil and water as it were. I also remember causing folks amazement when I said we did zero testing on anything but Lucid. Only so much time so you do what matters to you. :/00:05
gholtThe other thing that got hairy were dependencies like webob that had new packages deployed and in use by build systems, etc. without the code [Swift in this case] being tested against them first.00:08
notmynamemtaylor: no, we were talking about. but gholt is the one who came up with that idea :-)00:09
gholtWebob's been a notorious break-backwards-compatibility package.00:09
*** ahmed_ has quit IRC00:10
*** jakedahn has quit IRC00:19
gholtHmm. Another reason we may never use the OS repo is that we also have other dependencies that only really apply to us. For instance, our own kernel package for the hardware we use, swauth and slogging which not everybody would use, Rackspace's own auth middleware package, etc.00:21
gholtBut, basing our swift packages on the same packaging repo would be great.00:22
gholtEr, same packaging codebase. repo's a little overloaded when it comes to packaging.00:22
*** med_out is now known as medberry00:23
medberrysome customers/users would like the milestones to also be perpetually available.00:23
mtaylormedberry: I think that's something we can accomodate00:25
* medberry went spelunking for D2 to no avail recently00:25
gholtmtaylor: Well, I'm out for the night. Feel free to email or msg me directly if ya want. But https://github.com/crashsite/swift_debian should have exactly what we've been using so far for Swift. Seems we're not the only ones messing with packaging based on what I see at https://github.com/ohthree too. :/00:38
gholtHopefully I didn't rat my fellow coders out. :P00:39
*** gholt has left #openstack-dev00:39
creihtlol00:46
*** JStoker has joined #openstack-dev00:46
notmynamemtaylor: medberry: that may be difficult depending on what the milestone means. for nova/glance a milestone is just a checkpoint. for swift, our "milestone" releases are actually prod-ready (actually running in prod, in fact)00:47
*** rods has quit IRC00:58
*** mnour has quit IRC01:12
medberrynotmyname, yep. I was pretty sure your miles persisted01:15
medberrythey get numbered and treated more like releases.01:15
*** dragondm has quit IRC01:29
*** zul has joined #openstack-dev01:29
*** hisaharu has quit IRC01:34
*** nati has quit IRC01:35
*** vladimir3p has quit IRC01:46
*** tsuzuki_ has joined #openstack-dev01:49
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Redefining credential types.  https://review.openstack.org/33402:10
*** nati has joined #openstack-dev02:25
*** vladimir3p has joined #openstack-dev02:40
*** martine has joined #openstack-dev02:51
*** mdomsch has joined #openstack-dev02:57
openstackgerritJames E. Blair proposed a change to openstack/openstack-puppet: Remove .gitmodules.  https://review.openstack.org/33603:04
openstackgerritJames E. Blair proposed a change to openstack/openstack-puppet: Add rfc.sh.  https://review.openstack.org/33703:06
*** letterj has left #openstack-dev03:11
*** ecarlin has joined #openstack-dev03:26
*** alekibango has quit IRC03:45
*** alekibango has joined #openstack-dev03:45
*** alekibango has quit IRC03:47
*** alekibango has joined #openstack-dev03:47
*** bcwaldon has joined #openstack-dev03:51
openstackgerritJames E. Blair proposed a change to openstack/openstack-puppet: Clean up submodules.  https://review.openstack.org/33803:59
openstackgerritA change was merged to openstack/openstack-puppet: Add rfc.sh.  https://review.openstack.org/33704:01
openstackgerritA change was merged to openstack/openstack-puppet: Clean up submodules.  https://review.openstack.org/33804:04
*** martine has quit IRC04:28
*** nati has quit IRC04:39
*** nati has joined #openstack-dev04:50
*** mwhooker has quit IRC05:09
*** zaitcev has quit IRC05:09
*** thor_ has joined #openstack-dev05:35
*** sandywalsh has quit IRC05:38
*** sandywalsh has joined #openstack-dev05:50
*** vladimir3p has quit IRC05:53
*** smoser has quit IRC06:01
*** openpercept has joined #openstack-dev06:09
*** smoser has joined #openstack-dev06:16
*** tsuzuki_ has quit IRC06:23
*** ecarlin has quit IRC06:24
*** ecarlin has joined #openstack-dev06:24
* ttx sighs reading backlog06:35
*** smoser has quit IRC06:38
*** smoser has joined #openstack-dev06:41
*** thor_ has quit IRC06:45
*** reidrac has joined #openstack-dev06:56
*** ynef has joined #openstack-dev06:59
ynefSection 2.3 of the Compute manual says that "[...] OpenStack Compute keeps all the cloud system state in a distributed data store." Where can I find more information about the distributed data store used for this purpose?06:59
*** tsuzuki_ has joined #openstack-dev07:05
*** bcwaldon has quit IRC07:06
*** openpercept has quit IRC07:22
*** ecarlin_ has joined #openstack-dev07:22
*** tudamp has joined #openstack-dev07:23
*** ecarlin has quit IRC07:25
*** ecarlin_ is now known as ecarlin07:25
*** ecarlin has quit IRC07:27
*** nati has quit IRC07:36
*** tudamp has quit IRC07:38
*** raker has quit IRC07:41
*** tudamp has joined #openstack-dev07:44
*** nati has joined #openstack-dev07:46
*** AhmedSoliman has joined #openstack-dev08:02
*** nati has quit IRC08:05
openstackjenkinsProject nova-milestone build #21: SUCCESS in 16 sec: https://jenkins.openstack.org/job/nova-milestone/21/08:12
openstackjenkinsTarmac: Backport of noauth fix for admin api in D4 milestone.08:12
*** ccorrigan has joined #openstack-dev08:12
openstackjenkinsProject nova build #1,294: SUCCESS in 3 min 11 sec: https://jenkins.openstack.org/job/nova/1294/08:20
openstackjenkins* Tarmac: Once a network is associated with project, I can���t delete this network with ���nova-manage network delete���.08:20
openstackjenkinsAs you know, I can delete network by scrubbing the project with ���nova-manage project scrub���.08:20
openstackjenkinsHowever it is too much.08:20
openstackjenkinsThe cause of this problem is there is no modify command of network attribute.08:20
openstackjenkinsThis branch adds 'nova-manage network modify' command.08:20
openstackjenkinsIn this branch, we only support project and host value modifications.08:20
openstackjenkinsAnother attributes are in future work.08:20
openstackjenkins* Tarmac: Update paste config so that EC2 admin API defaults to noauth.08:20
openstackjenkinsProject nova-milestone build #22: SUCCESS in 15 sec: https://jenkins.openstack.org/job/nova-milestone/22/08:22
openstackjenkinsTarmac: Fix for iscsiadm commands.08:22
openstackjenkinsProject nova-milestone build #23: SUCCESS in 14 sec: https://jenkins.openstack.org/job/nova-milestone/23/08:32
openstackjenkinsTarmac: Fix default hostname generator so that it won't use underscores, and use minus signs instead.08:32
*** Capashen has joined #openstack-dev08:42
*** mnour has joined #openstack-dev09:29
*** darraghb has joined #openstack-dev10:10
*** rods has joined #openstack-dev10:22
*** AhmedSoliman has quit IRC10:41
*** AhmedSoliman has joined #openstack-dev10:49
*** AhmedSoliman has quit IRC11:00
sorenmtaylor, jeblair: Ok, so I'm supposed to set up a (number of?) function test job(s) in Jenkins that will test everything after it's been deployed to the test cluster.11:20
sorenmtaylor, jeblair: To do this, I need to know some URL endpoints I can talk to.11:20
sorenmtaylor, jeblair: Are they going to be static, or will I have to somehow have them passed from the deployment job?11:21
*** Adri2000 has quit IRC11:38
*** martine has joined #openstack-dev11:45
*** lts has joined #openstack-dev11:53
*** lts has joined #openstack-dev11:56
*** bsza has joined #openstack-dev12:03
*** tsuzuki_ has quit IRC12:03
ttxjaypipes: I'll need your green light on D4, whenever you get up.12:04
ttx(there are still 8 bugs in the D4 list, but I suppose I shouldn't wait for all of them to turn green)12:05
*** mnour has quit IRC12:08
*** mnour has joined #openstack-dev12:09
openstackgerritDan Prince proposed a change to openstack/keystone: Move tools/tracer into the keystone code. Fixes ImportError's when running keystone as a .deb package.  https://review.openstack.org/33912:17
*** mfer has joined #openstack-dev12:22
*** nmistry has joined #openstack-dev12:34
*** mdomsch has quit IRC12:36
*** nmistry has quit IRC12:49
*** throughnothing has quit IRC12:51
*** mnour has quit IRC12:52
*** mnour has joined #openstack-dev12:53
*** ynef has quit IRC13:02
*** throughnothing has joined #openstack-dev13:04
openstackgerritMark McLoughlin proposed a change to openstack/glance: Switch file based logging to WatchedFileHandler for logrotate  https://review.openstack.org/34013:05
*** throughnothing has left #openstack-dev13:06
*** throughnothing has joined #openstack-dev13:06
*** markmc has joined #openstack-dev13:08
*** jsavak has joined #openstack-dev13:17
*** joesavak has joined #openstack-dev13:17
zykes-if i get an error when building the keystone docs13:18
zykes-what's wrong then ?13:18
zykes-http://paste.openstack.org/show/2272/13:21
*** jsavak has quit IRC13:44
*** joesavak has quit IRC13:44
*** joesavak has joined #openstack-dev13:45
*** jsavak has joined #openstack-dev13:45
*** coool has joined #openstack-dev13:48
*** zns has joined #openstack-dev13:50
*** coool has left #openstack-dev13:50
*** bcwaldon has joined #openstack-dev13:51
*** dprince has joined #openstack-dev13:51
*** nati has joined #openstack-dev13:54
*** ameade has joined #openstack-dev13:59
*** mtaylor has quit IRC14:02
*** troytoman-away is now known as troytoman14:03
*** mtaylor has joined #openstack-dev14:07
*** verne.freenode.net sets mode: +v mtaylor14:07
dprincesoren/mtaylor: https://code.launchpad.net/~dan-prince/keystone/ubuntu-keystone-nodoc/+merge/7290214:08
*** lorin1 has joined #openstack-dev14:13
*** Gordonz_ has joined #openstack-dev14:27
*** vladimir3p has joined #openstack-dev14:28
*** AhmedSoliman has joined #openstack-dev14:29
*** Gordonz has quit IRC14:30
*** troytoman is now known as troytoman-away14:36
*** tudamp has left #openstack-dev14:37
*** code_franco has joined #openstack-dev14:41
*** Adri2000_ has joined #openstack-dev14:44
*** rnirmal has joined #openstack-dev14:47
*** dolphm has joined #openstack-dev14:47
*** nmistry has joined #openstack-dev14:56
*** alekibango has quit IRC14:56
*** alekibango has joined #openstack-dev14:56
*** alekibango has joined #openstack-dev14:57
*** alekibango has quit IRC14:57
*** alekibango has joined #openstack-dev14:58
*** alekibango has joined #openstack-dev14:58
*** alekibango has quit IRC14:59
*** alekibango has joined #openstack-dev15:00
*** alekibango has joined #openstack-dev15:00
*** Adri2000_ has quit IRC15:03
zulmtaylor: keystone made it into the ubuntu archive15:05
*** dragondm has joined #openstack-dev15:06
openstackgerritDan Prince proposed a change to openstack/keystone: tame pylint.  https://review.openstack.org/34115:10
openstackgerritDolph Mathews proposed a change to openstack/keystone: Added backend-managed primary key to User model  https://review.openstack.org/34215:12
ttxvishy: ping15:13
vladimir3pttx: ping15:15
*** reidrac has quit IRC15:15
ttxvladimir3p: yo15:15
vladimir3pyyx: any chance to receive more reviews on volume types BP15:15
vladimir3pttx: oops ^^^15:15
ttxvladimir3p: it's almost at the top of the http://wiki.openstack.org/reviewslist/ priority -- I haven't much more influence than that.15:17
vladimir3pok, great. thanks15:17
ttxvladimir3p: it's danprince and anotherjesse review days, you can ping them15:17
dprinceSo gerrit/git question. Using the new rfc.sh alias I should be able to 'git review' to push a new patch set right?15:18
* ttx looks at the status of the review15:18
ttxdprince: never tried, but that's how it's supposed to work, I think15:18
jeblairdprince: yes, ammend your latest commit, leave the change-id line intact, and git review15:18
dprinceI just did it and it pushed an totally new review (not a new patchset)15:18
ttxvladimir3p: dprince is right here ! Don't let him run !15:19
dprincevladimir3p: I can run pretty fast too! (I'll look at it in a minute). Not sure I'm the expert on volume matters however.15:19
ttxdprince: the expert on volume matters already reviewed it :)15:20
jeblairdprince: probably didn't ammend the commit then?  if you made a new commit with a new change-id, you'll get a new change in gerrit.15:20
dprincejeblair: all I did is 'git commit' my new change. and then 'git review' again.15:20
jeblairdprince: you can abandon the new change in gerrit, squash your new commit onto the old one (make sure to keep the change-id from the old on in the revised commit message) and git review to add your new patchset15:20
vladimir3pdprice: thanks! I hope you will not run away ;-)15:20
vladimir3p*dprince15:21
jeblairdprince: yep, that's what happened.  Next time, here are the instructions for uploading a new patchset to an existing change:15:21
jeblairdprince: http://wiki.openstack.org/GerritWorkflow#Updating_a_Change15:21
jeblairfor this time, if you want to update the change, you'll need to squash and upload.15:22
jeblairdprince: the process for fixing this is similar to "long-lived topic branches" section of the instructions:15:23
jeblairdprince: http://wiki.openstack.org/GerritWorkflow#Long-lived_Topic_Branches15:23
jeblair(that talks about how to squash commits)15:23
*** nati has quit IRC15:24
openstackgerritDolph Mathews proposed a change to openstack/keystone: Added backend-managed primary key to User model  https://review.openstack.org/34215:25
dprincejeblair: thank you sir.15:26
*** alekibango has quit IRC15:27
dolphmdprince: ping me if you need help!15:27
*** alekibango has joined #openstack-dev15:27
openstackgerritDan Prince proposed a change to openstack/keystone: Move tools/tracer into the keystone code. Fixes ImportError's when running keystone as a .deb package.  https://review.openstack.org/33915:27
dprincedolphm: should be fixed now. new patchset is ready.15:29
dolphmdprince: cool :)15:30
dolphmdprince: do you use tracer?15:31
dprincedolphm: most of the keystone binaries do.15:31
dprincedolphm: I'm fine w/ it I guess. My immediate interest was the ability to deploy keystone from a .deb package.15:32
dolphmdprince: only if you use the --enable-tracing argument ... i was just wondering if you find tracer *useful*15:32
dprincedolphm: importing 'tools/tracer' seems like a bad way to get it.15:32
dolphmdprince: i don't personally consider it to be a part of the keystone codebase anyway... it's not like it's tested, and i don't know anyone who uses it15:33
dprincedolphm: So you'de rather remove the imports then?15:34
dolphmdprince: personally, yes... but that's just me15:34
dprincedolphm: Sure. I'll go either way. We shouldn't leave the imports like they were however. Its unusable that way.15:35
jeblairsoren: the endpoints should be static.  configurable via jenkins would probably be nice, but they aren't going to change soon.15:35
dolphmdprince: but then again, i don't use it... so if there's people that *do* use it, their opinion needs to be weighed as well :)15:35
dprincedolphm: Okay. Who else would you like me to ask?15:36
dolphmdprince: i have no clue! lol15:38
dolphmdprince: i don't know if ziad wrote it, but i know he used to use it (at least)... i'll make sure he reviews the change15:39
dprincedolphm: yeah. Cool. he approved the original so we'll see what happens.15:41
dprincevladimir3p: your branch looks good to me. smokes fine. If the unit tests run I'll send it.15:44
*** nmistry has quit IRC15:46
vladimir3pdprince: thanks, great !15:48
*** mnour has quit IRC15:50
*** alekibango has quit IRC15:51
*** alekibango has joined #openstack-dev15:51
bcwaldonttx, vishy: this should probably go in d4 -> https://code.launchpad.net/~rackspace-titan/nova/boot-failure-lp833856/+merge/7291815:51
ttxbcwaldon: I'm ready to give a few extra hours to vishy, if he shows up :)15:52
bcwaldonsounds good15:52
*** kbringard has joined #openstack-dev15:58
* ttx will be back later, hopefully mtaylor and vishy will be around.16:02
openstackgerritA change was merged to openstack/openstack-ci-puppet: Add a mutex around puppet runs.  https://review.openstack.org/32016:06
openstackgerritA change was merged to openstack/openstack-ci-puppet: Add support for packages.o.o and 32-bit builds.  https://review.openstack.org/31816:06
znsdolph: yes, I use —trace-calls. It came in handy when I got bug requests over email. I asked for traces and was able to identify the problem. But that might be me and my Python noobness. Is there a better way to do this than tracer/tools?16:07
tr3buchetvishy: happen to know offhand if libvirt supports multiple fixed_ips per vif?16:07
*** bcwaldon has quit IRC16:08
openstackjenkinsProject nova build #1,295: SUCCESS in 3 min 12 sec: https://jenkins.openstack.org/job/nova/1295/16:09
openstackjenkinsTarmac: Servers with metadata will now boot on xenserver with flat_injected==False16:09
*** mattray has joined #openstack-dev16:09
openstackjenkinsProject nova-tarball-bzr-delta build #522: FAILURE in 13 sec: https://jenkins.openstack.org/job/nova-tarball-bzr-delta/522/16:15
openstackjenkins* Tarmac: Added:16:15
openstackjenkins- volume metadata16:15
openstackjenkins- volume types16:15
openstackjenkins- volume types extra_specs16:15
openstackjenkinsVolume create API receives volume types & metadata.16:15
openstackjenkinsWork in progress on Openstack APIs for volume types & extra_specs.16:15
openstackjenkins* Tarmac: Servers with metadata will now boot on xenserver with flat_injected==False16:15
*** zaitcev has joined #openstack-dev16:16
*** bcwaldon has joined #openstack-dev16:16
openstackjenkinsProject nova build #1,296: SUCCESS in 3 min 32 sec: https://jenkins.openstack.org/job/nova/1296/16:20
openstackjenkins* Tarmac: Added:16:20
openstackjenkins- volume metadata16:20
openstackjenkins- volume types16:20
openstackjenkins- volume types extra_specs16:20
openstackjenkinsVolume create API receives volume types & metadata.16:20
openstackjenkinsWork in progress on Openstack APIs for volume types & extra_specs.16:20
openstackjenkins* Tarmac: Servers with metadata will now boot on xenserver with flat_injected==False16:20
openstackjenkins* Tarmac: Once a network is associated with project, I can���t delete this network with ���nova-manage network delete���.16:20
openstackjenkinsAs you know, I can delete network by scrubbing the project with ���nova-manage project scrub���.16:20
openstackjenkinsHowever it is too much.16:20
openstackjenkinsThe cause of this problem is there is no modify command of network attribute.16:20
openstackjenkinsThis branch adds 'nova-manage network modify' command.16:20
openstackjenkinsIn this branch, we only support project and host value modifications.16:20
openstackjenkinsAnother attributes are in future work.16:20
openstackjenkins* Tarmac: Update paste config so that EC2 admin API defaults to noauth.16:20
openstackjenkins* Tarmac: The notifiers API was changed to take a list of notifiers. Some people might want to use more than one notifier so hopefully this will be accepted into trunk.16:20
openstackjenkins* Tarmac: Fixes iscsiadm commands to run properly.16:20
openstackjenkins* Tarmac: Fix default hostname generator so that it won't use underscores, and use minus signs instead.16:20
openstackjenkins* Tarmac: - rebuilds are functional again16:20
ttxbcwaldon: you can prepare a milestone-proposed merge for that bug you mentioned, so that vishy only has to ack it ?16:27
bcwaldonttx: yessir16:27
dolphmzns: yogi likes to debug in Komodo (not sure if he's gotten aptana's debugger attached to keystone yet), while I pretty much depend on automated tests and occasionally inline assertions16:29
ttxbcwaldon: cool.16:29
znsdolphm: Fine for my own debugging needs, I can always just throw in an import for tracer. How about getting traces from customer deployments when they have issues?16:31
bcwaldonttx: https://code.launchpad.net/~rackspace-titan/nova/build-failure-lp833856-d4/+merge/7292316:32
mtaylorzul: great! was it based on the packaging in lp:~openstack-ubuntu-packageres/keystone/ubuntu ?16:33
zulno it was based on something i was working on...but ill sync them up later16:33
mtaylorgreat!16:34
dolphmzns: logging *should* already take care of that -- back traces should be logged along with relevant sanitized data16:34
mtaylorzul: I still need to change our jenkins tooling to deal with the oneiric split I was bitching about like a little ass yesterday, right?16:34
mtaylorzul: (as in, to start looking for per-release branches?)16:35
zulno worries i was bitching as well16:35
mtaylorzul: :)16:35
mtaylorzul: there's nobody really I can bitch to in order to get that requirement lifted is there?16:35
*** zns has quit IRC16:35
*** zns has joined #openstack-dev16:36
zulmtaylor: well not really since we are already past FF16:36
mtaylorzul: ok. that will teach me to pay attention to policy changes more closely in the future16:37
dprincemtaylor: Hey. would you have a look at merging this: lp:~dan-prince/keystone/ubuntu-keystone-nodoc16:37
dprincemtaylor: I like to use nodocs sometimes and it wasn't working for me.16:38
zulmtaylor: heh ditto :)16:39
zykes-anyone that took a look at my error when building keystone docs?16:39
dolphmzykes-: link?16:40
zykes-http://paste.openstack.org/show/2272/16:40
*** zns has quit IRC16:40
*** amccabe has joined #openstack-dev16:43
*** mattray has quit IRC16:43
dolphmzykes-: found this... https://bitbucket.org/birkenfeld/sphinx/issue/756/the-error-typeerror-unhashable-type-list16:44
zykes-is it a code error or ?16:45
zykes-running Ubuntu 11.0416:45
*** jdurgin has joined #openstack-dev16:49
*** Capashen has quit IRC16:53
*** heckj has joined #openstack-dev16:57
openstackjenkinsProject nova build #1,297: SUCCESS in 3 min 16 sec: https://jenkins.openstack.org/job/nova/1297/17:00
openstackjenkins* Tarmac: added unit tests to instance_types for rainy day paths17:00
openstackjenkins* Tarmac: Added:17:00
openstackjenkins- volume metadata17:00
openstackjenkins- volume types17:00
openstackjenkins- volume types extra_specs17:00
openstackjenkinsVolume create API receives volume types & metadata.17:00
openstackjenkinsWork in progress on Openstack APIs for volume types & extra_specs.17:00
*** Gordonz_ has quit IRC17:01
mtaylordprince: I shall look at it in just a minute, the milestone-proposed ppa needs my attention first17:04
dprincemtaylor: thanks.17:06
*** Adri2000 has joined #openstack-dev17:09
*** mgius has joined #openstack-dev17:29
zykes-dolphm: ?17:31
*** jakedahn has joined #openstack-dev17:36
dolphmzykes-: i just found that bug report, which looks similar.. it's not an error i've seen before and i don't know what would have caused it within keystone17:37
*** hisaharu has joined #openstack-dev17:38
vishyttx, bcwaldon: yes that should go in17:40
*** nati has joined #openstack-dev17:40
vishyapproved17:43
Davieymtaylor: It's not *that* funny that people are looking at deploying Oneiric in datacentres based on repository packages.  It's *wise* that people are looking to do that.  Canonical already have a private/public cloud for openstack running Oneiric as a proof of concept.  People really are deploying Natty clouds.. so test driving Oneiric is a good idea.17:45
Davieymtaylor: "OpenStack for Ubuntu is because Canonical has a (asinine) requirement that only software in main can  run in their data centers" <-- is just wrong.17:46
DavieyIt smells bad to state things which are not true.17:47
ttxvishy, mtaylor: fwiw two jenkins jobs were stucj on teh nova slaves, so the approved from yesterday night only got through the washing machine after I got up.17:48
vishyah17:48
vishyok17:48
*** zns has joined #openstack-dev17:48
ttxvishy: i don't know what you tested exactly when you greenlighted me... trunk ?17:48
jaypipesDaviey: evening :)17:48
Davieyjaypipes: Hello!17:48
jaypipesDaviey: sorry never sent the calming email ;)17:48
ttxvishy: anyway, I waited for Glance --now the plan is to release tomorrow early in my morning.17:49
vishyok17:49
Davieyjaypipes: Oh, time is best spent with online arguments than doing real work.  Everyone knows that.17:49
ttxthat should give mtaylor a chance to fix the PPAs before I push packages to them17:49
vishyi tested proposed last night, so sounds like i should run it again17:49
jaypipesDaviey: :)17:49
mtaylorttx: working on it - apparently kombu also needs pymongo :)17:53
*** dprince has quit IRC17:53
zuluh....kombu is being instead of carrot now?17:54
mtaylorDaviey: sorry - you're right on the second point - I may have been in a bad mood and carpy.17:54
mtaylorDaviey: (although the no-universe on machines in canonical data centers is an actual restriction and I do find it a bit far - it is completely out of scope here)17:54
mtaylorDaviey: on the first point though - I'm not talking about deploying servers on oneiric being crazy. I AM saying that deploying 1000's of nodes of cloud fabric servers on a non-LTS release is rather unlikely to happen17:56
mtaylorDaviey: which does NOT mean that I think it's silly to have oneiric packages, or that I think it's silly to do test deployments on oneiric17:57
*** ohnoimdead has joined #openstack-dev17:57
mtaylorDaviey: merely that I personally believe that it's highly unlikely that people wanting to deploy major installations of diablo in production settings are going to do so on oneiric - and thus it is important for me to support them on lucid17:58
mtaylorand on squeeze17:58
mtaylorzul: glance/keystone use kombu17:59
*** dprince has joined #openstack-dev18:01
medberrymtaylor, which kombu. 1.0 or 1.2?18:01
*** heckj has quit IRC18:02
*** heckj has joined #openstack-dev18:02
*** zns has quit IRC18:05
*** mszilagyi has joined #openstack-dev18:12
*** mnour has joined #openstack-dev18:12
zykes-anyone else know why i can't build ks docs ?18:14
*** hisaharu has quit IRC18:15
sorenjeblair: Is there anything in writing about this deployment? Not how the deployment magic works, but rather how it's configured (one API server, one glance server, network nodes here and there, compute nodes so and so).18:15
mtaylormedberry: 1.0.418:17
medberrynod.18:17
jeblairsoren: not yet, lemme see if i can get you the vital information18:18
*** hisaharu has joined #openstack-dev18:19
jeblairsoren: http://etherpad.openstack.org/sg0YLui6DK18:24
*** heckj has quit IRC18:26
annegentlezykes-: taking a look18:27
*** heckj has joined #openstack-dev18:27
Davieymedberry: You raise a good queston, I don't know if 1.2 has been tested against glance or keystone.18:29
*** joesavak has quit IRC18:35
*** mattray has joined #openstack-dev18:36
openstackgerritJay Pipes proposed a change to openstack/glance: Update rfc.sh to use 'true'.  https://review.openstack.org/34318:37
openstackgerritJay Pipes proposed a change to openstack/glance: Bug lp:829654  https://review.openstack.org/34418:39
mtaylorDaviey, medberry: iirc, kombu 1.2 requires an update to python-amqplib18:40
openstackgerritJay Pipes proposed a change to openstack/glance: Bug lp:829064  https://review.openstack.org/34518:40
mtaylorDaviey, medberry: and since nova uses that with carrot, upgrading glance/keystone to kombu 1.2 would lead us in to bad places18:40
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug #827660 - Swift driver fail 5G upload  https://review.openstack.org/34618:40
uvirtbotLaunchpad bug 827660 in glance "Cannot store files greater than 5GB in Swift" [Critical,Fix committed] https://launchpad.net/bugs/82766018:40
Davieymtaylor: it does18:40
mtaylorDaviey, medberry: once we get nova from carrot to kombu, then we should be able to talk about updating all three18:41
Davieymtaylor: Are you following the carrot -> kombu issue?18:41
mtaylorDaviey: I am not18:41
Davieymtaylor: bug 798876, being developed against kombu > 1.018:42
uvirtbotLaunchpad bug 798876 in nova "Consider Switching from Carrot to Kombu for AMQP" [Low,Confirmed] https://launchpad.net/bugs/79887618:42
mtaylorDaviey: cool18:42
openstackgerritA change was merged to openstack/glance: Update rfc.sh to use 'true'.  https://review.openstack.org/34318:43
*** zns has joined #openstack-dev18:44
*** johnpur has joined #openstack-dev18:44
*** ChanServ sets mode: +v johnpur18:44
*** lorin1 has left #openstack-dev18:49
zykes-annegentle: finding anytihng ?18:51
*** fly-macbook has joined #openstack-dev18:51
zykes-why should keystone be on amqp ?18:51
openstackgerritVerification of a change to openstack/glance failed: Bug lp:829654  https://review.openstack.org/34418:52
vishyDaviey: is it a big priority for you to have the kombu version in the next ubuntu release?18:54
*** mfer is now known as mfer-food18:55
Davieyvishy: not really.. the reason for the newer kombu version was that when i sniffed migrating nova carrot->kombu .. 1.0 seemed to be missing things.18:55
vishyDaviey: so if we switch to kombu in essex it isn't a huge issue?18:56
DavieyI might be wrong, comstud is more involved than me.18:56
Davieyvishy: i'd welcome it18:56
*** nati has quit IRC18:56
Davieyvishy: Ah, sorry - i misread what you said.18:57
Davieyvishy: So deferring that for D?18:58
vishyDaviey: yes it is a big feature and we're past freeze18:59
mtaylorzul: how did you get keystone into ubuntu with python-unittest2 ?18:59
vishyso the question is, does that cause problems for you18:59
DavieyEither way is OK from my POV18:59
vishyok18:59
zulmtaylor: because its already there?18:59
Davieyvishy: We'll make progress on carrot then :).. thanks.19:01
mtaylorzul: yeah?19:01
zulyeah19:02
annegentlezykes-: so I have Sphinx 1.0.4 installed and don't see that error.19:02
mtaylorzul: my bad - I must have fat-fingered the search19:02
zulmtaylor: python-dtest also got accepted as wel19:02
zykes-1.0.1-1 here annegentle19:02
mtaylorzul: great!19:03
*** annegentle has left #openstack-dev19:07
openstackjenkinsProject nova-milestone build #24: SUCCESS in 18 sec: https://jenkins.openstack.org/job/nova-milestone/24/19:07
openstackjenkinsTarmac: porting fix for rebuild failure bug lp:83385619:07
*** annegentle has joined #openstack-dev19:08
annegentlezykes-: also I'm running python setup.py build_sphinx, not sphinx-build19:09
annegentlebut that just also allows the man page to build (I think)19:09
mtaylorzykes-: pretty sure python setup.py build_sphinx sets some additional options19:10
openstackgerritJames E. Blair proposed a change to openstack/openstack-ci-puppet: Revert "Add a mutex around puppet runs."  https://review.openstack.org/34719:12
openstackgerritA change was merged to openstack/openstack-ci-puppet: Revert "Add a mutex around puppet runs."  https://review.openstack.org/34719:13
*** darraghb has quit IRC19:13
*** fly-macbook has quit IRC19:15
*** nati has joined #openstack-dev19:18
openstackgerritA change was merged to openstack/keystone: Move tools/tracer into the keystone code. Fixes ImportError's when running keystone as a .deb package.  https://review.openstack.org/33919:20
openstackgerritA change was merged to openstack/glance: Switch file based logging to WatchedFileHandler for logrotate  https://review.openstack.org/34019:21
*** mwhooker has joined #openstack-dev19:23
ttxvishy: milestone-proposed rev1152 good for you ?19:23
ttxjaypipes: still struggling with backporting ?19:24
ttxmtaylor: how is it looking ?19:25
mtaylorttx: had to backport unittest2 in as well - it just build, so I re-triggered the kombu build. hopefully _THAT_ will finish soon19:25
mtaylorat which point we can start copying packages places19:25
ttxmtaylor: cool19:26
*** mfer-food is now known as mfer19:28
jaypipesttx: no.. https://review.openstack.org/#q,status:open+project:openstack/glance+branch:milestone-proposed,n,z19:29
zykes-in glance what's meant with a non pub image?19:29
jaypipesttx: I'm going to try and get a final review on one other bug branch I'd like to see in D4.19:30
ttxjaypipes: sure, will check back in ~1 hour19:30
jaypipesbcwaldon, blamar: hey Brians, mind doing a review on this one please? https://review.openstack.org/#change,33319:30
jaypipesttx: ok. ty!19:30
bcwaldonjaypipes: sure thing19:31
jaypipesbcwaldon: appreciated!19:32
*** markmc has quit IRC19:33
mwhooker /join #lpthw19:36
mwhookeroops19:37
bcwaldonjaypipes: its not obvious why TestPrivateImagesApi is skipped19:38
bcwaldonjaypipes: do you know?19:38
smoseranyone have a nova install up that they could test an image on for me really quickly?19:49
smoserat http://smoser.brickies.net/ubuntu/ttylinux-uec/ , on the nova i have access to, the dhcp is failing, and i wanted to see if its my nova .19:49
smoserhttp://smoser.brickies.net/ubuntu/ttylinux-uec/ttylinux-uec-amd64-12.1_2.6.35-22_1.tar.gz19:49
kbringardI'll try it out, gimme like 5 or so mins19:50
kbringardI'm running  2011.3~rc~20110825.1492-0ubuntu0ppa1~natty1, btw19:50
smoserthanks. mine is dying like: http://paste.ubuntu.com/674764/19:52
kbringardsmoser: just uploaded it, launch it now19:52
smoserand other (Ubuntu) images running fine.19:52
kbringardlooks like it worked for me...19:53
smosercan you pastebin output ?19:53
kbringardhttp://paste.openstack.org/show/2279/19:53
kbringardyea, done19:53
smoser:)19:53
smoserthanks19:53
kbringardI cut out all the ACPI stuff19:53
kbringardlet me know if you want the entire thing19:53
kbringardhttp://paste.openstack.org/show/2280/19:54
smoserthats fine. but just in case you were unaware of pastebinit: euca-get-console-output | pastebinit19:54
kbringardthat's the whole thing19:54
kbringardoh, did not know19:54
kbringardthanks :-)19:54
smoseryeah, it rocks19:54
smoserwould you mind quickly trying http://smoser.brickies.net/ubuntu/ttylinux-uec/dev/snapshot-20110825.0.tar.gz19:55
kbringardsure, one moment19:55
smoserfor me it dies the same way, but thats my development ttylinux and i've not verified on nova outside of lxc19:55
kbringardlaunching now19:57
kbringardhttp://paste.ubuntu.com/674771/19:58
kbringardlooks good to me19:59
*** dprince has quit IRC20:02
jaypipesbcwaldon: so, skip_is_disabled is used there to skip running the test if keystone isn't configured on the system... it's the same technique we use in the Swift and S3 functional tests20:06
bcwaldonjaypipes: I was more interested in the explanation in the skip output20:07
bcwaldonjaypipes: the S3 tests tell you what variable to set in the skip message20:07
jaypipesbcwaldon: ah, I see! one sec...20:07
openstackgerritJay Pipes proposed a change to openstack/glance: Switch file based logging to WatchedFileHandler for logrotate  https://review.openstack.org/34820:08
bcwaldonjaypipes: any way I can help you get the milestone-proposed reviews done, or is that just for you?20:09
jaypipesbcwaldon: oh, that's just for me...20:10
jaypipesbcwaldon: they've all been previously approve...20:10
bcwaldonjaypipes: yeah, just making sure20:10
jaypipesbcwaldon: just cherry-picking them from master into milestone-proposed, which is the branch that the release gets cut from20:10
bcwaldoncool, I've just been getting emails and wanted to make sure I could ignore them20:11
jaypipesVek: ping20:11
jaypipesbcwaldon: yeah, ignore :)20:11
openstackgerritKevin L. Mitchell proposed a change to openstack/glance: Add functional tests  https://review.openstack.org/33320:11
bcwaldonjaypipes: I'll look again in a bit, have to step out for a few20:12
*** mattray has quit IRC20:13
kbringardsmoser: that all cool and the gang? need anything else?20:13
jaypipesbcwaldon: looks like Kevin fixed your issue :)20:14
bcwaldonawesome20:14
bcwaldonI'll start running tests before i step out20:14
smoserkbringard, can you pastebin that too ?20:15
smoserthank you very much for trying20:15
smoseroh, you did.20:15
smoserthanks.20:15
kbringardnp20:15
kbringardnow that I have pastebinit installed, I'm going to pastebin everything! :-p20:15
smoseryou can, if you like, go even further and use pastebins as a general file transfer mechanism20:16
smoserhttp://blog.dustinkirkland.com/2010/10/bikeshed-pbput-and-pbget.html20:16
smosersee 'pbput' and 'pbget' (including encryption support)20:16
kbringardthat's awesome20:17
*** mattray has joined #openstack-dev20:19
ttxjaypipes: Jenkins reported fail on https://review.openstack.org/#change,34420:26
jaypipesttx: hmm, looks like a cache thing.. I'll retry20:27
ttxjaypipes: let me know if you need me to push them through20:27
*** novas0x2a|laptop has joined #openstack-dev20:32
openstackgerritA change was merged to openstack/glance: Bug lp:829064  https://review.openstack.org/34520:32
openstackgerritA change was merged to openstack/glance: Fixes LP Bug #827660 - Swift driver fail 5G upload  https://review.openstack.org/34620:32
openstackgerritA change was merged to openstack/glance: Bug lp:829654  https://review.openstack.org/34420:32
uvirtbotLaunchpad bug 827660 in glance "Cannot store files greater than 5GB in Swift" [Critical,Fix committed] https://launchpad.net/bugs/82766020:32
*** blakeyeager has joined #openstack-dev20:33
ttxjeblair: mental note: would be better if the above messages mentioned milestone-proposed in some way20:35
bcwaldonjaypipes: no experience with keystone. What's the quickest way to get the glance/keystone functional tests passing?20:35
*** AhmedSoliman has joined #openstack-dev20:36
jaypipesbcwaldon: email Vek and/or _cerberus_ about that... they have a good grasp.20:38
bcwaldonjaypipes: thanks20:39
*** bcwaldon has quit IRC20:40
openstackgerritA change was merged to openstack/glance: Switch file based logging to WatchedFileHandler for logrotate  https://review.openstack.org/34820:45
jeblairttx: thanks, added to todo list20:46
sorenttx: Still around?20:47
ttxsoren: yes20:48
sorenttx: Great.20:48
sorenttx: I expect to file a lot more like this: https://code.launchpad.net/~soren/nova/virt-layer-cleanup2/+merge/7269420:48
sorenttx: I understand your desire for a bug reference (given where we are in the release cycle), but...20:48
sorenttx: Should I file a bug when I start working on adding tests for a particular area?20:49
ttxsoren: I just noticed that this review was down on the reviewlist because it wasn't attached to anything20:49
sorenttx: Oh.20:49
sorenGood point.20:50
sorenMaybe attach it to a blueprint?20:50
ttxattaching it to a BP or bug makes it easier to up the prio20:50
sorenI think there's a blueprint for "improving test coverage".20:50
ttxsoren: the issue with a bp saying that is that it's never completed20:50
sorenFound it.20:50
sorenttx: I know.20:51
sorenttx: Suggestions?20:51
sorenhttps://blueprints.launchpad.net/nova/+spec/unittest-coverage for reference.20:51
ttxthat's why I suggested a wishlist bug - easier to target anc omplete20:52
ttxand complete20:52
sorenSo file a bug whenever I start working on a new set of tests?20:52
sorenI can do that.20:52
ttxThat would be perfect.20:53
sorenI guess we've had this discussion before.. :)20:53
sorenNot sure why I didn't apply the same logic here. My bad :)20:53
ttxbecause the logic is sunoptimal20:56
ttxsuboptimal20:56
ttxjaypipes: jenkins still digesting your pushes, apparently it can only digest one of them at a time20:57
sorenttx: Bug filed, branch updated to include --fixes lp:<bugno>. I'm curious how far up the lists it'll go. :)20:57
ttxjaypipes: and runs every 15 min :)20:57
ttxsoren: if someone targets it to diablo-rbp, it will go near the top as soon as d4 is released20:58
sorenttx: Sorry, I forget what "rbp" stands for?20:59
mtaylorsoren, ttx: in a few weeks, post switch, you could just have a branch for working on unittest-coverage blueprint and gerrit will auto-associate each change with that blueprint ... so it might get easier :)21:00
ttxrelease branch point21:01
ttxa.k.a. poening of essex21:01
ttxopening21:01
ttxmtaylor: doesn't solve the "never completed" issue21:02
sorenmtaylor: I can do that now, too?21:02
mtaylorsoren: you can?21:02
sorenmtaylor: Sure.21:03
sorenmtaylor: Link a branch to a blueprint, hack on stuff, propose it for merge, keep adding more stuff to it. When approved and merged, just file another merge proposal for the same branch.21:04
sorenand keep hacking on the same branch.21:04
ttxjaypipes: looks like it caught up -- please confirm that glance-2011.3~d4~20110825.r996.tar.gz is good to ship21:08
ttx20f09f6bd589613043e5b1eba70aa38bb63746b421:08
ttxvishy: around ?21:08
*** nmistry has joined #openstack-dev21:09
ttxvishy: please confirm that milestone-proposed rev1152 is good for you21:10
* ttx is going to bed -- jaypipes, vishy: will release abovementioned versions in ~8 hours unless your redflag me on IRC or email21:10
jaypipesttx: just making a final check. will do.21:11
ttxmtaylor: do your best fixing the PPA by then :)21:11
*** lts has quit IRC21:11
ttxjaypipes: thx!21:11
jaypipesttx: no, thank you!21:11
*** novas0x2a|laptop has quit IRC21:14
jaypipesttx: milestone-proposed is good to go.21:16
jaypipesttx: moved remaining in progress bug to RBP21:16
*** novas0x2a|laptop has joined #openstack-dev21:20
*** martine has quit IRC21:20
*** amccabe has quit IRC21:36
*** mfer has quit IRC21:40
*** jsavak has quit IRC21:45
*** markvoelker has quit IRC21:48
*** nmistry has quit IRC21:56
*** nmistry has joined #openstack-dev21:56
*** bsza has quit IRC22:07
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Redefining credential types.  https://review.openstack.org/33422:11
vishyreviewers needed: https://code.launchpad.net/~vishvananda/nova/fix-iscsiadm/+merge/7296522:13
*** kbringard has quit IRC22:14
*** zns has quit IRC22:22
*** bengrue has joined #openstack-dev22:22
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Redefining credential types.  https://review.openstack.org/33422:24
*** ameade has quit IRC22:32
*** mattray has quit IRC22:49
*** code_franco has quit IRC22:55
*** AhmedSoliman has quit IRC22:59
*** mwhooker has quit IRC23:00
*** anotherjesse has joined #openstack-dev23:06
anotherjesseblamar: around?23:07
*** rnirmal has quit IRC23:07
*** hisaharu has quit IRC23:16
openstackjenkinsProject nova build #1,298: SUCCESS in 3 min 19 sec: https://jenkins.openstack.org/job/nova/1298/23:20
openstackjenkinsTarmac: Add a set of generic tests for the virt drivers.23:20
openstackjenkinsUpdate a bit of documentation to match reality.23:20
*** anotherjesse has quit IRC23:24
*** ewindisch has joined #openstack-dev23:24
*** anotherjesse has joined #openstack-dev23:32
*** anotherjesse has quit IRC23:35
openstackjenkinsProject nova-milestone build #25: SUCCESS in 18 sec: https://jenkins.openstack.org/job/nova-milestone/25/23:42
openstackjenkinsTarmac: Backported fix for run_iscsiadm.23:42
*** mgius has quit IRC23:46
*** mgius has joined #openstack-dev23:47
*** anotherjesse has joined #openstack-dev23:58
*** anotherjesse has quit IRC23:58

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