Friday, 2011-10-28

« Thursday, 2011-10-27 Index Saturday, 2011-10-29 »
*** f-yang has left #openstack-dev00:07
*** bencherian has quit IRC00:18
*** jakedahn has quit IRC00:25
*** jakedahn has joined #openstack-dev00:27
*** mwhooker has quit IRC00:28
*** jakedahn has quit IRC00:29
*** jakedahn has joined #openstack-dev00:30
*** bengrue has quit IRC00:30
*** jog0 has quit IRC00:31
*** rkukura has joined #openstack-dev00:42
*** mmetheny_ has joined #openstack-dev00:46
*** mmetheny has quit IRC00:46
*** mmetheny_ is now known as mmetheny00:46
*** lorin1 has joined #openstack-dev01:02
lorin1How do you submit a merge proposal in Gerrit that depends on another proposal? Or does Gerrit just figure it out automatically?01:05
*** ayoung has quit IRC01:10
*** novas0x2a|laptop has quit IRC01:13
*** novas0x2a|laptop has joined #openstack-dev01:14
*** hugokuo has joined #openstack-dev01:18
*** cp16net has quit IRC01:32
openstackgerritLorin Hochstein proposed a change to openstack/nova: Added code to libvirt backend to report state info.  https://review.openstack.org/115201:37
*** mwhooker has joined #openstack-dev01:41
*** jog0 has joined #openstack-dev01:58
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Adding identity api v2.0 tests  https://review.openstack.org/115302:05
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Updating images tests  https://review.openstack.org/114902:05
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Further optimize kong.tests.test_servers  https://review.openstack.org/114602:07
*** jog0 has quit IRC02:08
*** jog0 has joined #openstack-dev02:10
*** jdurgin has quit IRC02:11
*** deshantm has quit IRC02:12
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Adding identity api v2.0 tests  https://review.openstack.org/115302:19
*** jog0 has quit IRC02:28
openstackgerritSumit Naiksatam proposed a change to openstack/quantum: Moved the initialization of the blade state so that the interfaces which are configured outside of Quantum are also initialized in the blade state.  https://review.openstack.org/115402:36
*** mwhooker has quit IRC02:41
*** mwhooker has joined #openstack-dev03:14
*** chomping has quit IRC03:32
mtaylorlorin1: gerrit just figures it out03:35
lorin1mtaylor: thx03:35
*** mjfork has quit IRC03:56
openstackgerritLorin Hochstein proposed a change to openstack/nova: Fix to a documentation generation script.  https://review.openstack.org/115503:59
openstackgerritLorin Hochstein proposed a change to openstack/nova: Removed autogenerated API .rst files.  https://review.openstack.org/115603:59
*** mdomsch has joined #openstack-dev04:01
*** lorin1 has quit IRC04:01
*** troytoman-away is now known as troytoman04:07
*** jakedahn_ has joined #openstack-dev04:19
*** jakedahn_ has joined #openstack-dev04:20
*** jakedahn_ has joined #openstack-dev04:20
*** jakedahn has quit IRC04:22
*** jakedahn_ is now known as jakedahn04:22
*** timr1 has quit IRC04:26
*** chomping has joined #openstack-dev04:37
*** troytoman is now known as troytoman-away04:42
*** timr has joined #openstack-dev04:42
*** mdomsch has quit IRC04:46
*** jakedahn_ has joined #openstack-dev04:49
*** jakedahn__ has joined #openstack-dev04:51
*** jakedahn_ has quit IRC04:51
*** jakedahn has quit IRC04:52
*** jakedahn__ is now known as jakedahn04:52
*** zaitcev has quit IRC05:43
*** reed has quit IRC05:50
*** chomping has quit IRC05:51
*** chomping has joined #openstack-dev05:56
*** _adjohn has joined #openstack-dev06:13
*** adjohn has quit IRC06:13
*** _adjohn is now known as adjohn06:13
*** comstud has quit IRC06:13
*** comstud has joined #openstack-dev06:13
*** novas0x2a|laptop has quit IRC06:25
*** jakedahn has quit IRC06:39
ttxannegentle: I'll replace it by the complete link.06:49
ttxannegentle: you're not the first to notice that the tinyurl actually adds more complexity :)06:50
*** adjohn has quit IRC07:38
ttxsandywalsh: meeting time in gCal should be ok now07:46
ttx(please doublecheck)07:47
*** jeffjapan has quit IRC07:59
*** adiantum has joined #openstack-dev08:10
*** timr has quit IRC08:26
*** chomping has quit IRC08:33
*** pixelbeat has joined #openstack-dev08:43
zykes-Kiall: around?09:02
*** ron-slc has quit IRC09:11
*** darraghb has joined #openstack-dev09:16
*** ron-slc has joined #openstack-dev09:19
*** timr has joined #openstack-dev09:20
*** rods has joined #openstack-dev09:45
*** adiantum has quit IRC09:45
Kiallzykes-: yea09:47
zykes-Kiall: your dashboard package what version is it based on ?09:52
Kialldiablo branch09:53
Kiall(everything i use is diablo, or stable/diablo)09:53
*** hugokuo has quit IRC10:05
*** viraptor has joined #openstack-dev10:10
zykes-hmm10:19
zykes-funky10:19
Kiallwhy?10:20
*** ron-slc has quit IRC10:48
*** ron-slc has joined #openstack-dev10:51
*** timr has quit IRC11:03
*** rkukura has quit IRC11:12
sandywalshttx, thanks, I'll give it a shot11:15
openstackgerritSoren Hansen proposed a change to openstack/nova: Retry failed SQL connections (LP #876663)  https://review.openstack.org/115711:15
uvirtbotLaunchpad bug 876663 in nova "Nova should retry connections if started before MySQL" [Wishlist,Confirmed] https://launchpad.net/bugs/87666311:15
sorenI wish Gerrit would let me set a description separate from the commit message. :-/11:18
viraptor++, needed to use first review message for that sometimes, but it's not perfect solution11:20
Kiallsoren: re that MySQL retry fix, If i remember correctly, I've seen similar issues with RabbitMQ being down.. might be a similar fix (assuming I'm actually remembering correctly)11:25
sorenKiall: ISTR that having been addressed already.11:25
sorenKiall: Maybe it broke again.11:26
*** mjfork has joined #openstack-dev11:33
ttxsoren: https://bugs.launchpad.net/openstack-ci/+bug/84521111:37
uvirtbotLaunchpad bug 845211 in openstack-ci "Making comments" [Wishlist,Triaged]11:37
sorenttx: Cool.11:39
viraptorsoren: re. testing, you could use mox to return a fake engine which fails N connections, then succeeds11:40
sorenviraptor: Yeah, I guess that would be a decent start.11:41
viraptorthen just get_engine with it... or even give that fake engine directly to ensure_connection - not sure which one would best fit this - maybe one test for ensure_... itself and one for get_engine which fails and counts reconnects?11:42
sorenviraptor: Hey, are you involved in https://review.openstack.org/#change,815 at all?11:42
sorenviraptor: I'd verify ensure_connection separately and perhaps check that get_session at some point called ensure_connection.11:43
viraptorsoren: yes, but the author didn't have time to fix it yet, I guess it will be updated next week11:47
Kiallsoren: possibly .. if it happens again I'll report it ..11:51
*** adiantum has joined #openstack-dev12:01
*** markvoelker has quit IRC12:16
*** cmagina has quit IRC12:21
*** cmagina has joined #openstack-dev12:30
zykes-ttx: do you know if there's any way to get a full stack of openstack working ?12:31
ttxzykes-: on what platform ?12:32
zykes-ubuntu or centos...12:32
zykes-i've tried getting openstack-dashboard from gd to work and the trunk version of dashboard from 4p with either version of keystone12:33
zykes-but everytime i run into an error12:33
ttxdevstack is an option12:33
zykes-i don't want to start installing everything over again...12:33
ttxhttp://devstack.org/12:33
zykes-should it really be that hard ?12:34
ttxzykes-: the complexity is in the variety of deployments that are possible.12:35
ttxWithout knowing what you want to deploy and where you want to deploy it, it's difficult to be more specific12:35
ttxAre you after building something the size of Amazon ? Or just after a all-in-one demo ?12:35
zykes-centos or ubuntu diablo core projects with keystone and dashboard, pretty simple in my mind.12:35
zykes-and i got everything except dashboard to work12:36
ttxzykes-: how many machines do you plan to deploy ?12:36
ttxzykes-: OSAPi or EC2 ? Network model ?12:36
zykes-2 nova nodes atm, vlan backed api i dunno12:37
zykes-doesn't the db use the osapi ?12:37
ttxyes it does, doesn't mean your clients have to.12:37
ttxI guess I should refresh fnords.wordpress.com/2010/12/02/bleeding-edge-openstack-nova-on-maverick/12:38
zykes-ttx: i got everything running as i said except a dashboard12:39
zykes-towards nova12:39
ttxzykes-: there is a reason why soi many deployment tools are built around openstack. Complex multi-server systems are difficult to deploy because the "package" abstraction is no longer enough12:39
zykes-so for a prod env one should use the soi instead ?12:40
ttxdashboard and keystone will have rough edges wrt integration with other core projects -- at least until Essex is out12:40
ttxsoi ?12:40
zykes-ehm, nvm12:41
zykes-wonder how i should run a demo then12:41
ttxzykes-: destack is supposed to give you a quick demo.12:41
ttxdevstack*12:41
ttx(haven't used it yet)12:41
ttxanyway, that's not a development discussion, we should continue this on #openstack12:42
zykes-though i guess if i show it to folks at work and they ask "would you deam this production ready" then what should i ask ?12:42
zykes-ok12:42
*** mmetheny has quit IRC12:46
*** mmetheny_ has joined #openstack-dev12:46
openstackgerritSandy Walsh proposed a change to openstack/nova: services updating  https://review.openstack.org/115812:47
openstackgerritSandy Walsh proposed a change to openstack/nova: fixup  https://review.openstack.org/115912:47
openstackgerritSandy Walsh proposed a change to openstack/nova: tests running  https://review.openstack.org/116012:47
openstackgerritSandy Walsh proposed a change to openstack/nova: pep8  https://review.openstack.org/116112:47
openstackgerritSandy Walsh proposed a change to openstack/nova: removed commented out code  https://review.openstack.org/116212:47
openstackgerritSandy Walsh proposed a change to openstack/nova: removed debugging messages  https://review.openstack.org/116312:47
openstackgerritSandy Walsh proposed a change to openstack/nova: Moved host list fetch into ZoneManager and fixed tests  https://review.openstack.org/116412:47
openstackgerritSandy Walsh proposed a change to openstack/nova: joinload and remove double counting  https://review.openstack.org/116512:47
openstackgerritSandy Walsh proposed a change to openstack/nova: working on least_cost  https://review.openstack.org/116612:47
openstackgerritSandy Walsh proposed a change to openstack/nova: working on zones support  https://review.openstack.org/116712:47
openstackgerritSandy Walsh proposed a change to openstack/nova: renamed abstractscheduler to distributedscheduler  https://review.openstack.org/116812:47
openstackgerritSandy Walsh proposed a change to openstack/nova: removed weight_hosts  https://review.openstack.org/116912:47
openstackgerritSandy Walsh proposed a change to openstack/nova: pep8  https://review.openstack.org/117012:47
openstackgerritSandy Walsh proposed a change to openstack/nova: child zone refactoring complete  https://review.openstack.org/117112:47
openstackgerritSandy Walsh proposed a change to openstack/nova: ick  https://review.openstack.org/117212:47
openstackgerritSandy Walsh proposed a change to openstack/nova: syntax errors and related branch merges  https://review.openstack.org/117312:47
openstackgerritSandy Walsh proposed a change to openstack/nova: local provisioning working again  https://review.openstack.org/117412:47
openstackgerritSandy Walsh proposed a change to openstack/nova: pep8  https://review.openstack.org/117512:47
openstackgerritSandy Walsh proposed a change to openstack/nova: child zone boot working again  https://review.openstack.org/117612:47
openstackgerritSandy Walsh proposed a change to openstack/nova: unit tests getting in shape  https://review.openstack.org/117712:47
openstackgerritSandy Walsh proposed a change to openstack/nova: distributed scheduler tests working  https://review.openstack.org/117812:47
openstackgerritSandy Walsh proposed a change to openstack/nova: json filter fixed and all host filter tests working  https://review.openstack.org/117912:47
openstackgerritSandy Walsh proposed a change to openstack/nova: json filter fixed and all host filter tests working  https://review.openstack.org/118012:47
openstackgerritSandy Walsh proposed a change to openstack/nova: removed debugging  https://review.openstack.org/118112:47
openstackgerritSandy Walsh proposed a change to openstack/nova: pep8  https://review.openstack.org/118212:47
openstackgerritSandy Walsh proposed a change to openstack/nova: least cost tests working again  https://review.openstack.org/118312:47
openstackgerritSandy Walsh proposed a change to openstack/nova: remove debugging  https://review.openstack.org/118412:47
openstackgerritSandy Walsh proposed a change to openstack/nova: pep8  https://review.openstack.org/118512:47
openstackgerritSandy Walsh proposed a change to openstack/nova: mailmap  https://review.openstack.org/118612:47
sandywalshsorry12:48
sandywalshmissed the squash12:48
sandywalshI assume I need to "abandon" each of these?12:49
*** lts has joined #openstack-dev12:49
ttxow12:51
ttxsandywalsh: I suppose -- you could probably whip up a script that makes the corresponding gerrit calls12:52
sandywalshI'm midway through abandoning each12:52
ttxthat's probably faster, since there doesn't seem to be a gerrit abandon command12:53
*** dprince has joined #openstack-dev12:53
sandywalshhmm, how can I squash when there have been trunk merges in between my commits?12:57
*** mjfork has quit IRC12:58
viraptorsandywalsh: I'd make new branch, rebase -i your old branch to it and pick (squash actually) only own commits...13:12
viraptornot sure if that solves your issue13:13
*** uncleofthestick has joined #openstack-dev13:17
sandywalshviraptor, will that work around the trunk merges?13:19
sandywalshviraptor, I had done a 'rebase -i master' before starting the process, but forgot the squash13:19
viraptorif you drop the merges themselves and select only your commits... it should I think13:20
*** mattray has joined #openstack-dev13:23
*** Gordonz has joined #openstack-dev13:35
*** joesavak has joined #openstack-dev13:42
openstackgerritBrian Waldon proposed a change to openstack/nova: Creating uuid -> id mapping for S3 Image Service  https://review.openstack.org/98813:43
openstackgerritJames E. Blair proposed a change to openstack/horizon: Add .gitreview and rfc.sh.  https://review.openstack.org/118713:50
*** cp16net has joined #openstack-dev13:52
openstackgerritVerification of a change to openstack/horizon failed: Add .gitreview and rfc.sh.  https://review.openstack.org/118713:57
openstackgerritVerification of a change to openstack/horizon failed: Add .gitreview and rfc.sh.  https://review.openstack.org/118714:01
soren*chuckle*14:03
*** rkukura has joined #openstack-dev14:13
*** code_franco has joined #openstack-dev14:16
*** jaypipes has joined #openstack-dev14:17
mtaylorsoren: yes?14:17
mtaylorttx, sandywalsh: you can totally abandon them all with a script, just so you know for next time14:18
*** uncleofthestick_ has joined #openstack-dev14:19
*** uncleofthestick has quit IRC14:20
sandywalshmtaylor, what's the incantation?14:20
*** uncleofthestick_ is now known as uncleofthestick14:20
mtaylorsandywalsh: one sec14:20
jaypipesdprince: thx for taking care of that glance packaging issue (cleaner/pruner/reaper.conf)14:23
dprincejaypipes: No problem.14:24
*** reed has joined #openstack-dev14:25
*** bcwaldon has joined #openstack-dev14:25
openstackgerritDolph Mathews proposed a change to openstack/keystone: Moving contributor docs into rst (bug #843056)  https://review.openstack.org/118814:26
uvirtbotLaunchpad bug 843056 in keystone "Design Documentation" [High,Fix committed] https://launchpad.net/bugs/84305614:26
mtaylorsandywalsh: ssh -p 29418 review.openstack.org gerrit query status:open project:openstack/nova owner:sandy.walsh@rackspace.com | grep ^change | awk '{print $2}' | xargs -n1 ssh -p 29418 review.openstack.org gerrit review --abandon14:27
mtaylorsandywalsh: something like that14:27
mtaylor(untested, for obvious reasons)14:27
sandywalshheh, thanks14:29
sandywalshmtaylor, and ... you won't be getting anything from me for x-mas this year :p14:29
*** adiantum has quit IRC14:29
mtaylorsandywalsh: awww14:29
mtaylorwe've been considering adding something to git review that would warn you if you're about to submit more than one review14:30
sandywalshmtaylor, my spellbook is overflowing with gerrit/git hacks14:30
mtaylorsandywalsh: oh yeah? sounds like a fun spellbook...14:30
openstackgerritVerification of a change to openstack/horizon failed: Add .gitreview and rfc.sh.  https://review.openstack.org/118714:30
*** adiantum has joined #openstack-dev14:32
openstackgerritVerification of a change to openstack/horizon failed: Add .gitreview and rfc.sh.  https://review.openstack.org/118714:37
*** crobinso has joined #openstack-dev14:38
*** cmagina has quit IRC14:40
*** cmagina has joined #openstack-dev14:41
*** mdomsch has joined #openstack-dev14:51
*** adjohn has joined #openstack-dev14:52
*** bsza has joined #openstack-dev14:52
*** jdg has joined #openstack-dev14:58
*** uncleofthestick has quit IRC15:00
*** darraghb has quit IRC15:03
openstackgerritStuart McLaren proposed a change to openstack/glance: Adds option to encrypt 'location' metadata.  https://review.openstack.org/118915:06
openstackgerritBrian Waldon proposed a change to openstack/nova: Restructuring OSAPI code for Admin API endpoint  https://review.openstack.org/112415:10
*** negronjl_ has quit IRC15:12
*** dragondm has joined #openstack-dev15:30
*** joesavak has quit IRC15:31
*** joesavak has joined #openstack-dev15:31
*** mjfork has joined #openstack-dev15:34
openstackgerritAnne Gentle proposed a change to openstack/openstack-manuals: Fixes bug 882781  https://review.openstack.org/119015:38
uvirtbotLaunchpad bug 882781 in openstack-manuals "Flag for multi_host flag is missing from docs" [Medium,New] https://launchpad.net/bugs/88278115:38
*** adjohn has quit IRC15:45
*** bencherian has joined #openstack-dev15:46
openstackgerritJay Pipes proposed a change to openstack/glance: Fixes LP Bug#882585 - Backend storage disconnect  https://review.openstack.org/119115:48
openstackgerritSandy Walsh proposed a change to openstack/nova: flatten distributed scheduler  https://review.openstack.org/119215:50
*** yogirackspace has joined #openstack-dev15:50
*** yogirackspace has quit IRC15:54
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Further optimize kong.tests.test_servers  https://review.openstack.org/114615:54
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Stop kong/run_tests.sh from building venv each run  https://review.openstack.org/113715:54
*** bencherian has quit IRC15:54
openstackgerritSandy Walsh proposed a change to openstack/nova: flatten distributed scheduler  https://review.openstack.org/119215:56
*** blamar has joined #openstack-dev15:56
*** sdake has quit IRC15:56
openstackgerritBrian Waldon proposed a change to openstack/openstack-integration-tests: Further optimize kong.tests.test_servers  https://review.openstack.org/114615:57
openstackgerritSandy Walsh proposed a change to openstack/nova: flatten distributed scheduler  https://review.openstack.org/119216:09
*** sdake has joined #openstack-dev16:09
*** jsavak has joined #openstack-dev16:10
*** joesavak has quit IRC16:13
sandywalshf-yang https://review.openstack.org/#change,119216:29
*** zaitcev has joined #openstack-dev16:30
*** f-yang has joined #openstack-dev16:31
*** fred_ has joined #openstack-dev16:32
*** fred_ has quit IRC16:32
openstackgerritAlex Meade proposed a change to openstack/nova: Updated NoAuth to account for requests ending in /  https://review.openstack.org/114716:33
*** negronjl has joined #openstack-dev16:36
*** anotherjesse has joined #openstack-dev16:38
anotherjessejeblair: ever seen jenkins barf on startup with *** glibc detected *** java: malloc(): memory corruption: 0x00007fe554595ac0 ***16:38
*** ewindisch has joined #openstack-dev16:43
jeblairanotherjesse: no :(16:43
*** adiantum has quit IRC16:44
anotherjesseam I just really unlucky - every other day jenkins explodes on me.  I think I don't trust this box anymore16:44
anotherjessethe fact that I can kernel panic the host by doing stuff in the guests doesn't help with my happyness with this cluster16:44
zykes-anotherjesse: "Unexpected error: list indices must be integers, not str" < what's that in connection to dashboard ?16:53
mtayloranotherjesse: wow16:57
mtayloranotherjesse: yeah - that's amongst the reasons I've been sticking with only running stuff in slaves rather than in vms on the master16:58
openstackgerritBrad Hall proposed a change to openstack/quantum: Change the ovs plugin create_*() calls to take the kwargs param  https://review.openstack.org/119316:58
anotherjessemtaylor: everything I run is in lxc container or kvm instance16:58
anotherjessemtaylor: oh - nevermind - misread16:58
*** rkukura has quit IRC16:59
zykes-is there any dev that can answer on that ?17:00
*** adjohn has joined #openstack-dev17:00
*** jdurgin has joined #openstack-dev17:00
*** cp16net has quit IRC17:01
anotherjessezykes-: when does that occur?17:08
zykes-when logging onto dash17:08
anotherjessedo CLI tools work?17:09
zykes-yeah17:09
anotherjesseusing keystone with novaclient or talking directly to nova?17:10
zykes-keystone correct17:10
anotherjessecan you paste the result of authing to keystone:17:11
anotherjessecurl -s -d  "{\"auth\":{\"passwordCredentials\": {\"username\": \"$NOVA_USERNAME\", \"password\": \"$NOVA_API_KEY\"}}}" -H "Content-type: application/json" http://$KEYSTONE_IP:$KEYSTONE_PORT/v2.0/tokens17:11
anotherjesseNOVA_API_KEY is password not api key - sorry about that17:12
openstackgerritBrian Waldon proposed a change to openstack/nova: Creating uuid -> id mapping for S3 Image Service  https://review.openstack.org/98817:12
zykes-http://paste.pocoo.org/show/499645/17:16
zykes-http://pastebin.com/7jE8R17:19
anotherjessenot sure if this is the issue - but you are missing %tenant_id% (or something like that) in your keystone endpoint templates17:21
anotherjesse            "publicURL": "http://nova.stack.lan:8774/v1.1/,"17:21
anotherjesseshould have the tenant id17:21
zykes-tenant id is required for nova public or ?17:22
anotherjessemy understanding is that %tenant_id% is required for many (all?) urls - here is what we use: https://github.com/cloudbuilders/devstack/blob/master/files/keystone_data.sh#L3017:26
zykes-ok, fixed but still the same17:27
anotherjessecan you copy/paste the settings you use for verifying via the cli - since without tenant_id in the url I'm not sure how nova would work17:28
zykes-http://paste.pocoo.org/show/499651/17:28
zykes-i've added %tenant_id% now17:28
anotherjessehow about your env variables you use for novaclient17:29
zykes-http://paste.pocoo.org/show/499653/17:30
*** jog0 has joined #openstack-dev17:30
*** novas0x2a|laptop has joined #openstack-dev17:31
anotherjessezykes-: can you post the full traceback from dash?17:31
zykes-http://paste.openstack.org/show/2953/17:33
*** cp16net has joined #openstack-dev17:35
anotherjessezykes - is your novaclient up to date?17:36
zykes-anotherjesse: i got the one that's from openstack packages in oneiric...17:37
zykes-should it be never ?17:38
anotherjessezykes-: not sure - still on natty17:39
*** code_franco has quit IRC17:39
*** bencherian has joined #openstack-dev17:40
anotherjesselooks like this is a known issue - fixed in https://github.com/4P/horizon/pull/167 - reviewing now17:40
anotherjessezykes-: you can try to update your dash to https://github.com/gabrielhurley/openstack-dashboard/tree/keystoneclient17:41
anotherjesseif you want to test in the mean time17:41
zykes-uhm k17:42
*** novas0x2a|laptop has quit IRC17:43
anotherjesseanother casualty of subtle changes to how keystone works ...17:44
anotherjesse:(17:44
zykes-or just projects not being updated :p17:44
*** troytoman-away is now known as troytoman17:44
zykes-i've used keystone diablo > trunk for your info so :p17:44
*** anotherjesse has quit IRC17:45
zykes-pardon myself.17:46
zykes-went a bit over the line there17:47
*** novas0x2a|laptop has joined #openstack-dev17:56
*** cp16net has quit IRC18:03
*** cp16net has joined #openstack-dev18:03
*** liemmn_ has joined #openstack-dev18:05
*** dragondm has quit IRC18:06
*** code_franco has joined #openstack-dev18:06
liemmn_jsavak, do we still use the nova_auth_token for lazy Nova provisioning?18:08
*** jakedahn has joined #openstack-dev18:08
*** jakedahn has quit IRC18:09
*** liemmn_ has quit IRC18:14
*** markwash has quit IRC18:19
*** mwhooker has quit IRC18:20
jaypipes_cerberus_: around?18:23
_cerberus_Ish18:23
jaypipes_cerberus_: so the section on Deploying Glance is wrong on http://wiki.openstack.org/OpenstackDeployability18:23
jaypipes_cerberus_: Glance's API has been versioned now for 4 months or so.18:23
_cerberus_hah, ok18:24
jaypipes_cerberus_: as has the Glance client...18:24
jaypipes_cerberus_: i.e. a Glance client and server communicate on an agreed-upon lowest common denominator API version.18:24
*** tcampbell has joined #openstack-dev18:24
_cerberus_Alright, I think that helps18:25
_cerberus_Sounds like I can just remove the majority of that section18:25
_cerberus_It's a WIP anyway18:25
jaypipes_cerberus_: now, don't get me wrong, Glance has *other* issues with continuous deployment (like the db_sync crap not working all the time), but the API version isn't one of those reasons :)18:25
_cerberus_I just wanted some place to aggregate ideas18:25
jaypipes_cerberus_: yep. lemme give you some bug #'s to put in there... one sec18:25
jaypipeshttps://bugs.launchpad.net/bugs/82479418:25
uvirtbotLaunchpad bug 824794 in glance "Tables are generated outside of migration process" [Low,In progress]18:25
jaypipeshttps://bugs.launchpad.net/bugs/86165318:26
uvirtbotLaunchpad bug 861653 in glance "Glance client-only package should be available on PyPI" [Low,Confirmed]18:26
jaypipeshttps://bugs.launchpad.net/bugs/86165018:26
uvirtbotLaunchpad bug 861650 in glance "Glance client packaging uses non-required dependencies" [Medium,Confirmed]18:26
openstackgerritSandy Walsh proposed a change to openstack/nova: flatten distributed scheduler  https://review.openstack.org/119218:26
jaypipes_cerberus_: there may be more, but those are def relevant.18:26
jaypipes_cerberus_: that first one (824794) is the big one IMHO.18:27
_cerberus_Updated18:27
openstackgerritBrian Lamar proposed a change to openstack/nova: Too much information is returned from POST /servers  https://review.openstack.org/119418:27
jaypipes_cerberus_: feel free to link those bugs to https://blueprints.launchpad.net/nova/+spec/deployability-improvements if you want...18:27
_cerberus_Good idea18:28
*** dubsquared has joined #openstack-dev18:40
*** bencherian has quit IRC18:43
*** bencherian has joined #openstack-dev18:44
*** adjohn has quit IRC18:47
*** rjh has joined #openstack-dev18:48
*** liemmn has joined #openstack-dev18:49
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Fixes LP882760.Changes to return TenantId properly as part of roles.Additional tests to support the same.  https://review.openstack.org/119518:50
*** bencherian has quit IRC18:50
openstackgerritBrian Waldon proposed a change to openstack/nova: Creating uuid -> id mapping for S3 Image Service  https://review.openstack.org/98818:53
*** mwhooker has joined #openstack-dev19:05
*** mwhooker has quit IRC19:06
*** blamar has quit IRC19:06
*** mwhooker has joined #openstack-dev19:07
*** dprince has quit IRC19:07
smoserjaypipes, around ?19:10
jaypipessmoser: yuppers19:10
smoseri interested in making it dead easy for someone to populate their cloud with a build from http://cloud-images.ubuntu.com19:10
mtaylorsmoser: ++19:11
smosermy thought on how to do that would be to have something consume http://uec-images.ubuntu.com/query/ (or some other well formed data we have there).19:11
smoseri have a very bad understanding of glance, but it seems to me that you might think i should run a public glance server or something.19:11
mtaylorsmoser, jaypipes: off-topic - but would also be kick-ass if cobbler could speak glance api for its images. just saying19:12
jaypipessmoser: Yep, sounds like an excellent case for a public Glance server.19:13
jaypipessmoser: shall we chat next week at UDS?19:13
smoseri was afraid you'd say that.19:13
smoseri'm really hoping to have something "just work" for ubuntu users in 12.04.19:14
smosermtaylor, for you personally, (as opposed to end-user-y) you can use cloud-publish-ubuntu that just went into oneiric from cloud-utils19:15
smoserhttp://pb.daviey.com/2ppE/19:15
*** adjohn has joined #openstack-dev19:15
smoserif you have the euca2ools funcitoning (which you probably dont) then you just do : cloud-publish-ubuntu --disk i386 natty my-bucket19:16
jaypipessmoser: afraid I'd say what?19:17
smoserwell, afraid you'd suggest the glance sort of.19:17
smoseri'm not completely opposed to it. i just probably ahve to read more and understand what it is.19:17
smoseror what that would entail.19:17
jaypipessmoser: you'd have to be a bit more descriptive when you say "would be to have something consume"...19:18
jaypipeswhat is "something"19:18
*** markwash has joined #openstack-dev19:18
smoserwell 2 ideas at the moment:19:18
smoser * openstack dashboard19:18
smoser * something that runs on installation of openstack that just downloads and populates glance19:19
*** troytoman is now known as troytoman-away19:19
smoserso a definite goal of 12.04 will be to have some tool that i can run that basically mirrors cloud-images.ubuntu.com, and populates my local cloud with that stuff. (configurable to keep the latest N releases or something)19:20
*** tcampbell has quit IRC19:22
openstackgerritjoe-gordon0 proposed a change to openstack/nova: fix typo in scheduler/driver.py assert_compute_node_has_enough_memory. fixes bug 883233.  https://review.openstack.org/119619:22
uvirtbotLaunchpad bug 883233 in nova "typo in nova/scheduler/driver.py  assert_compute_node_has_enough_memory()" [Undecided,New] https://launchpad.net/bugs/88323319:22
smoseris anyone doing work on the dashboard to have a frontend to a public glance server ? jaypipes ?19:22
smoseror am i just completly off base as to if such a thing is reasonable.19:23
jaypipessmoser: dashboard already includes an images tab. though I'm not sure if authentication is required (probably)19:23
smoserright. and it can surely view images that exist in your cloud19:24
jaypipessmoser: yeah, auth is required for dashboard AFAICT :(19:24
smoserbut i want to view images that do not exist in your cloud19:24
jaypipesright19:24
jaypipessmoser: https://blueprints.launchpad.net/glance/+spec/glance-dashboard19:25
jaypipessmoser: I marked it Obselete because of dasboard, but sounds like it may be reopened?19:25
smoserwell, it seems like a reasonable thing to have, no?19:27
smoserthe first thing a user has to do is get an image.19:27
smoserand clearly, i want ubuntu to be the first thing that they find.19:28
jsavakjaypipes - i think i asked you this once, but forgot - who can tag keystone?19:29
jaypipesjsavak: hmm, good question. not sure at this point. jeblair or mtaylor? ^^19:33
*** dragondm has joined #openstack-dev19:33
jaypipessmoser: yup, agreed. might be a little pet project for me to tackle on the plane to Orlando ;0\)19:33
jaypipesugh, that smiley did not come out right...19:33
smoseroh. you're going to orlando. good.19:34
smoserthen i can bother you there.19:34
smoser:)19:34
openstackgerritYogeshwar Srikrishnan proposed a change to openstack/keystone: Fix tenant_id for Role  https://review.openstack.org/104419:34
smoserjaypipes, uds, right?19:34
jaypipessmoser: ya19:35
jsavakjaypipes: thanks19:35
*** dolphm has joined #openstack-dev19:35
*** dolphm has quit IRC19:50
*** joesavak has joined #openstack-dev19:53
mtaylorjaypipes, jsavak: ttx19:55
*** jsavak has quit IRC19:56
*** troytoman-away is now known as troytoman19:57
*** mszilagyi has joined #openstack-dev19:57
*** dolphm has joined #openstack-dev20:07
*** joesavak has quit IRC20:08
*** dolphm has quit IRC20:10
*** dolphm has joined #openstack-dev20:11
*** Gordonz has quit IRC20:18
*** heckj has joined #openstack-dev20:22
heckjjeblair or monty - either of you around?20:26
*** dubsquared has quit IRC20:29
zykes-hey heckj, got a clue on endpoints for nova?20:29
*** dubsquared has joined #openstack-dev20:29
*** dubsquared has quit IRC20:30
heckjzykes-: maybe - what do you need? I can tell you how it's working now and how to configure them.20:30
*** dubsquared has joined #openstack-dev20:30
zykes-oooh happy days!20:31
*** dubsquared has quit IRC20:31
zykes-heckj: i think i got it to work!20:31
heckjexcellent!20:31
*** dubsquared has joined #openstack-dev20:31
mtaylorheckj: I am here20:31
*** zns has joined #openstack-dev20:31
mtaylorheckj: (monty doesn't ping my screen nearly as much as mtaylor does, btw. :) )20:32
heckjmtaylor: ah - just writing you. Devin said there was some problem with the tests running on Horizon in the new gerrit setup?20:32
heckjmtaylor: wanted to know if I could help resolve, although it looks like it's working now20:32
mtaylorheckj: yes!20:32
mtaylorheckj: nope, we just disabled them20:32
mtaylorheckj: how far down the rabbit hole would you like to go here at first?20:33
heckjmtaylor: this the relevant one: https://jenkins.openstack.org/view/Horizon/job/gate-horizon-unittests/20:33
mtaylorheckj: yes20:33
mtaylorheckj: so here's the thing...20:33
mtaylorheckj: (we'll ignore for just right now that I don't like repos with more than one project in them ... that's actually not the issue at the moment)20:34
heckjheh, ok20:34
mtaylorheckj: for the other projects, we keep a job that builds the virtualenv20:34
mtaylorheckj: that job rebuilds the virtualenv nightly, or if the pip-requires changes20:34
mtaylorheckj: that way we don't have test runs failing if pypi has network issues (happens often)20:34
mtaylorheckj: then the beginning of the test run itself is copying the pristine venv dir into the source tree, and then running run_tests.sh20:35
mgiusmtaylor: I always assumed I was getting packet loss to my virtual machines...20:35
zykes-heckj: you work on dashboard?20:35
heckjmtaylor: with you so far...20:35
heckjzykes-: among other things, yeah20:35
mtaylormgius: that may also be the case- but pip install certainly fails more often than I would like for it to20:35
mtaylorheckj: so - the issue with horizon is that one of its steps is to cd to openstack-django and then run boostrap.py and buildout20:36
mtaylorheckj: which does not install things into a venv - it installs things into the django dir20:36
heckjyeah, damned icky/messy20:36
mtaylorso I'm not sure what to save in the venv building script so that I can copy it into a fresh new thing20:36
mtayloryup. also - using both pip-requires/pip AND buildout is a little ugh. but I can live with that even :)20:37
mtaylorI just don't have a way to build a venv in one place, then copy it in so that I can run horizon's run_tests without downloading/installing things20:37
mtaylorSO....20:38
mtaylorfirst box of kittens I'd like is just to figure out the pre-building of venvs in such a way that it works for horizon20:38
heckjso other than the .venv, there's the directories "launchpad", "eggs", and "parts" that are the pieces that build-out slaps into place.20:38
zykes-heckj: "2011-10-28 20:36:28,672 DEBUG routes.middleware [-] No route matched for GET /1/extras/servers" seen that before?20:38
mtaylorheckj: and bin20:38
mtaylorheckj: (also, glance no longer lives on launchpad - perhaps not pulling it from bzr is a better idea?)20:39
heckjthose directories are under "django-openstack" and are what build-out creates to set things up.20:39
mtaylorheckj: unrelated and a small bug20:39
heckjerp ":-)20:39
mtaylorheckj: so, if I just tar up django-openstack/{launchpad,eggs,parts,bin} and untar it on a new clone, do you think that would work ?20:40
heckjI think it should - but I also think that buildout will still try and recreate it20:40
mtaylorthat should be fine - run_tests.sh doesn't re-run buildout everytime does it?20:40
mtaylorit's only during venv building, no?20:40
heckjmtaylor: unfortunately, it's not terribly smart - it's fairly niave and does try to create/recreate it each time20:41
*** reed has quit IRC20:41
*** zns has quit IRC20:41
heckjIt optimizes to re-use from whaty's there, but some of the buildout recipes try every time.20:42
mtaylorheckj: YAY20:42
* mtaylor spits on buildout20:42
* mtaylor growls20:42
* mtaylor roars20:42
heckjmtaylor: yeah, although it's the bar recipe that sucks, so we might be able to kill that bird and the glance thing with one stone20:43
mtaylorheckj: sweet20:43
zykes-heckj: do you guys have fully working dashboards ?20:43
heckj(bar == bzr in my crappy typing)20:43
*** dubsquared has quit IRC20:43
heckjzykes-: yeah, definitely - we keep them working to be able to develop on it20:43
mtaylorheckj: so, I'll try tarring the other stuff up - just know that until we fully solve making buildout not trying to do network things, that we could have false negatives on trunk gating20:43
zykes-heckj: horizon or ?20:44
mtaylorheckj: can I just keep throwing beefs your way, or do you want me to file bugs for other things20:44
heckjmtaylor: definitely - I'll see about getting the buildout updated to not use any BZR recipes either - which should help dramatically20:44
mtaylorheckj: (there are a few things that I'd really like to see change, but aren't immediate needs)20:44
heckjmtaylor: definitely throw it my way - feel free to file a bug and assign to me directly too, but i'll catch and cover this stuff20:44
* mtaylor coughs "openstackx"20:44
*** dubsquared has joined #openstack-dev20:44
* mtaylor coughs "openstackx"20:45
mtaylor:)20:45
heckjmtaylor: active work to remove it by the devs - some stuff (minimal) still remaining, but we're hammering on it.20:45
mtayloryeah - I know you guys are working on that one - I just thought I'd be an asshole for a sec20:45
*** zns has joined #openstack-dev20:45
heckjmtaylor: lots of active movement recently to move needed functionality into python-novaclient and cut the cord, just not quite there yet20:45
*** dubsquared has quit IRC20:46
heckjmtaylor: All for it! And I agree :-)20:46
zykes-heckj: do you know if dashboard (horizon) goes out looking for /1/extras/servers ?20:46
mtaylorheckj: the other one is - how much of a battle will I have and who will the battle be with to split dashboard and openstack-django into two independent code trees?20:46
mtaylorheckj: I would REALLY love it if dashboard pip-require'd openstack-django20:46
mtayloralthough I understand we may not be there infrastructure-wise to support that sanely just yet20:46
heckjmtaylor: that would be nice, I'll see what I can do there. There's an added complication that dashboard needs all of glance dependencies due to glance not having a separate client.20:47
mtayloryes. there are several things we need to fix about how all of this hangs together20:48
* mtaylor imagines na?vely that the openstack pypi server will solve all of them20:48
heckjmtaylor: have we advanced glance to the point that we can pip install it (glance) and have all the dependencies flow in?20:48
mtaylorheckj: probably not20:49
mtaylorjaypipes: ^^ do you know?20:49
*** dolphm has quit IRC20:49
mtaylorheckj: I'm imagining these are likely slightly longer-term issues we should just solve over the course of the cycle20:50
heckjmtaylor: Okay - probably stuck with buildout for a while then.20:50
mtaylorok. fair - we'll work around it for now then20:50
heckjmtaylor: yeah, the whole build-out thing is to test django-openstack separately, which requires django and other bits (req) for even testing, but is used inversely (where glance, django, etc) already exist.20:50
mtaylorheckj: that's why I'm thinking _eventually_ it would be sweet if django-openstack can just have a lifecycle and be uploaded/released to pip like a normal lib20:52
mtaylorheckj: and then horizon could just consume it20:52
mtaylorbut I live in a world of idealism20:52
heckjmtaylor: That would be darned nice. I'd like to get away from the buildout bits myself, but that's because I don't want to have to maintain two different mechanisms like pip to do a similar thing. Haven't yet looked deeply at how to replace it effectively.20:53
*** markwash has quit IRC20:53
zykes-heckj: are you running horizon?20:53
mtaylorheckj: sweet. well, as long as we're in agreement that it would be a nice to have ... we can solve the actual thing next time we're drunk20:53
*** mattray1 has joined #openstack-dev20:53
*** mattray has quit IRC20:53
heckjmtaylor: next week, I think - yeah :-)20:54
mtaylorheckj: hell yes.20:54
mtaylorheckj: when are you getting there?20:55
*** lts has quit IRC20:55
heckjzykes-: I'm one the core team for horizon, we run it internally, and I help do a lot of other weird shit for Nebula20:55
heckjmtaylor: Sunday night20:55
mtaylorzykes-: the weird shit he does is all stuff that really shouldn't be discussed in public ;)20:56
mtaylorheckj: awesome. I'm getting in tomorrow, so I'll be around20:56
zykes-heckj: should trunk horizon work with diablo nova ?20:56
*** crobinso has quit IRC20:56
heckjzykes-: yeah, it should - I don't think there's any breaking changes that we've implemented (yet). We will, but right now it should.20:58
*** mattray1 has quit IRC20:58
zykes-when going to instances page then i get errors about not having openstackx and yada leading to this in nova-api.log "2011-10-28 20:42:07,900 DEBUG routes.middleware [-] No route matched for GET /1/extras/servers"21:00
heckjzykes-: I'm sorry, I don't know off the top. Have you opened a Question in the answers section of launchpad? They're watching it closely, but that's deeper in the API than I know off the top of my head.21:02
heckjhave to run - back later21:02
*** markwash has joined #openstack-dev21:06
*** bencherian has joined #openstack-dev21:13
*** cp16net has quit IRC21:15
*** dolphm has joined #openstack-dev21:20
*** markwash has quit IRC21:24
*** liemmn has quit IRC21:29
*** bencherian has quit IRC21:29
*** mdomsch has quit IRC21:30
*** bencherian has joined #openstack-dev21:32
*** benheilers has joined #openstack-dev21:37
openstackgerritGabriel Hurley proposed a change to openstack/horizon: Migrates Horizon to use KeystoneClient  https://review.openstack.org/119721:38
dolphmcool! ^21:38
*** benheilers has quit IRC21:40
openstackgerritMatt Dietz proposed a change to openstack/nova: Fixes lp883279  https://review.openstack.org/119821:42
*** bsza has quit IRC21:45
*** bencherian has quit IRC21:45
*** errr has quit IRC21:46
*** errr has joined #openstack-dev21:47
*** benheilers has joined #openstack-dev21:48
*** rods has quit IRC21:52
*** benheilers has quit IRC21:53
*** nati2 has joined #openstack-dev21:55
*** rods has joined #openstack-dev22:04
*** dolphm has quit IRC22:04
*** dolphm has joined #openstack-dev22:07
*** code_franco has quit IRC22:09
*** reed has joined #openstack-dev22:14
openstackgerritDolph Mathews proposed a change to openstack/keystone: Track post-Diablo database evolution using migrations (BP: database-migrations)  https://review.openstack.org/120022:36
*** dragondm has quit IRC22:41
*** dragondm has joined #openstack-dev22:41
nati2Hey nova.crypt.mkcacert looks not used now.22:44
nati2I greped the code of nova. But I found only definition of the code22:45
*** bencherian has joined #openstack-dev22:46
*** dolphm has quit IRC22:50
*** rbasak has quit IRC22:52
*** heckj has quit IRC22:54
*** bencherian has quit IRC23:02
*** ewindisch has quit IRC23:04
*** jog0 has quit IRC23:08
*** pixelbeat has quit IRC23:12
*** zns has quit IRC23:16
*** negronjl has quit IRC23:23
*** jdg has quit IRC23:26
*** adjohn has quit IRC23:32
*** dragondm has quit IRC23:34
openstackgerritJay Pipes proposed a change to openstack/glance: More cache refactoring - Management Middleware  https://review.openstack.org/120123:38
*** zns has joined #openstack-dev23:43
openstackgerritRazique Mahroua proposed a change to openstack/openstack-manuals: Add a conceptual figure for Keystone Fixes bug 854409  https://review.openstack.org/103323:53
openstackgerritRazique Mahroua proposed a change to openstack/openstack-manuals: Update Swift doc : # Fix bug 791584 # Update formatting  https://review.openstack.org/104023:53
uvirtbotLaunchpad bug 854409 in openstack-manuals "Concepts in Identity Service would benefit from a diagram" [Undecided,Fix committed] https://launchpad.net/bugs/85440923:53
uvirtbotLaunchpad bug 791584 in openstack-manuals "Documentation missing detail" [Low,Fix committed] https://launchpad.net/bugs/79158423:53
« Thursday, 2011-10-27 Index Saturday, 2011-10-29 »

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