Thursday, 2011-06-09

*** Binbin has joined #openstack-dev00:25
*** vladimir3p has quit IRC00:47
*** ohnoimdead has quit IRC00:49
*** Tv has quit IRC00:59
*** Tv has joined #openstack-dev01:15
*** RobertLaptop has quit IRC01:16
*** pyhole has quit IRC01:24
*** pyhole has joined #openstack-dev01:24
*** jdurgin has quit IRC01:33
*** Tv has quit IRC02:25
*** RobertLaptop has joined #openstack-dev02:33
*** RobertLaptop has quit IRC02:36
*** RobertLaptop_ has joined #openstack-dev02:36
*** cloudgroups has joined #openstack-dev03:02
*** cloudgroups has left #openstack-dev03:03
*** RobertLaptop_ has quit IRC03:55
*** RobertLaptop has joined #openstack-dev03:55
*** Zangetsue has joined #openstack-dev03:59
*** jdurgin has joined #openstack-dev05:39
*** jdurgin has quit IRC05:42
*** deepakn has joined #openstack-dev05:46
*** RobertLaptop has quit IRC06:02
*** RobertLaptop has joined #openstack-dev06:17
*** reidrac has joined #openstack-dev07:00
*** reidrac has quit IRC07:02
*** reidrac has joined #openstack-dev07:02
*** zaitcev has quit IRC07:09
*** heden_ has joined #openstack-dev07:10
*** heden_ has quit IRC07:10
*** heden has joined #openstack-dev07:11
*** openpercept_ has joined #openstack-dev07:49
*** openpercept_ has quit IRC08:18
*** openpercept_ has joined #openstack-dev08:20
*** openpercept_ has quit IRC08:20
*** openpercept_ has joined #openstack-dev08:20
*** vinkeshb has joined #openstack-dev08:41
*** openpercept_ has quit IRC08:49
*** openpercept_ has joined #openstack-dev08:51
*** openpercept_ is now known as Guest1224208:52
*** cloudgroups has joined #openstack-dev09:38
*** cloudgroups has quit IRC09:49
*** cloudgroups has joined #openstack-dev09:51
*** Daviey is now known as Da10:18
*** Da is now known as Daviey10:18
*** openpercept_ has joined #openstack-dev10:19
*** openpercept_ has quit IRC10:19
*** openpercept_ has joined #openstack-dev10:19
*** openpercept_ is now known as Guest5043710:19
*** Binbin has quit IRC10:20
*** Guest12242 has quit IRC10:21
*** Zangetsue_ has joined #openstack-dev10:32
*** Zangetsue_ has quit IRC10:34
*** Zangetsue__ has joined #openstack-dev10:34
*** Zangetsue has quit IRC10:35
*** Zangetsue__ is now known as Zangetsue10:35
*** cloudgroups has quit IRC10:48
*** cloudgroups has joined #openstack-dev10:49
*** cloudgroups has left #openstack-dev10:54
*** markvoelker has joined #openstack-dev11:00
*** markvoelker has joined #openstack-dev11:01
*** Zangetsue has quit IRC11:56
*** Zangetsue has joined #openstack-dev12:04
*** ameade has joined #openstack-dev12:43
*** bcwaldon has joined #openstack-dev12:53
*** lorin1 has joined #openstack-dev12:54
*** pmullaney has joined #openstack-dev13:03
*** RobertLaptop has quit IRC13:16
*** Zangetsue has quit IRC13:31
*** mattray has joined #openstack-dev13:36
*** Zangetsue has joined #openstack-dev13:37
*** markvoelker has quit IRC13:50
*** Zangetsue has quit IRC14:02
*** RobertLaptop has joined #openstack-dev14:10
*** dragondm has joined #openstack-dev14:18
annegentlegah, I hadn't sourced the nova-venv correctly. I am now fine, building docs as always. Double gah.14:28
*** deepakn has quit IRC14:32
*** vladimir3p has joined #openstack-dev14:33
*** jkoelker has joined #openstack-dev14:44
*** vinkeshb has quit IRC14:46
*** Guest50437 has left #openstack-dev14:55
*** pyhole has quit IRC15:08
*** pyhole has joined #openstack-dev15:08
notmynamemtaylor: can you give gholt access to https://launchpad.net/swauth, please?15:18
notmyname(or myself)15:18
*** rnirmal has joined #openstack-dev15:19
*** vinkeshb has joined #openstack-dev15:21
*** reidrac has quit IRC15:23
notmynamemtaylor: by "access" I mean "has access to the 'Configure code hosting' link on http://code.launchpad.net/swauth/")15:26
*** Rubberneck has quit IRC15:56
*** markvoelker has joined #openstack-dev16:06
*** bcwaldon has quit IRC16:08
*** Tv has joined #openstack-dev16:32
jaypipesnotmyname: gholt is the driver of the project. He should be able to configure it... https://code.launchpad.net/swauth/trunk/+setbranch16:33
notmynamejaypipes: ya, we saw that he's the driver. just weren't able to see that link. we'll try the direct link...16:34
jaypipesnotmyname: if not, lemme know. I'm in Openstack Admins and can configure it...16:34
*** zaitcev has joined #openstack-dev16:34
notmynamethanks16:35
jaypipesany time.16:35
jaypipesnotmyname: my way of saying thx for letting me know about the webob thing yesterday :)16:35
notmyname:-)16:35
jaypipesnotmyname: what a pita...16:36
*** deepakn has joined #openstack-dev16:42
jaypipesBK_man: ping16:50
jaypipesnotmyname: hey, swauth is going away once keystone integration is done, right?16:51
notmynamebefore that, actually16:52
notmynamewell, those 2 things are independent16:52
*** jdurgin has joined #openstack-dev16:57
BK_manjaypipes: pong16:59
*** bcwaldon has joined #openstack-dev16:59
jaypipesBK_man: heya. Just wanted to give you a quick heads up. Because of a bug in webob < 1.0.7, Glance is going to have to require 1.0.7+. Will this be any issue for you in RPM packaging?17:00
jaypipesbcwaldon: yo, you in the office yet? is dprince around?17:01
*** ohnoimdead has joined #openstack-dev17:01
BK_manjaypipes: thanks! will update python-glance RPM17:01
jaypipesBK_man: cheers. not yet in trunk, but will be shorly.17:01
jaypipesnotmyname, gholt: you ever seen this error before? https://answers.launchpad.net/glance/+question/16006017:03
*** RobertLaptop has quit IRC17:14
*** vladimir3p has quit IRC17:16
bcwaldonjaypipes: I'm here17:20
bcwaldonjaypipes: dprince is, too17:20
bcwaldonjaypipes: we were in a meeting17:20
notmynamejaypipes: no. I don't know why swift.common is missing17:21
jaypipesbcwaldon: ok, coolio. I need to talk with dan about the packaging dependency on webob 1.0.7 after the S3 bug stuff lands in trunk.17:21
jaypipesnotmyname: Error uploading image: [Errno 1] _ssl.c:490: error:140770FC:SSL routines:SSL23_GET_SERVER_HELLO:unknown protocol17:22
jaypipesnotmyname: not the swift.common thing.. wondering about the above.17:22
notmynameah17:22
jaypipesnotmyname: sorry, should have specified :)17:22
notmynamessl error then17:22
notmynameso either you are talking ssl and the proxy isn't listening with ssl or vice versa17:23
notmynameat least, that's my first guess :-)17:23
*** jk0 has quit IRC17:24
*** jk0 has joined #openstack-dev17:25
*** ChanServ sets mode: +v jk017:25
*** RobertLaptop has joined #openstack-dev17:26
*** cloudnod has joined #openstack-dev17:30
*** cloudnod has left #openstack-dev17:32
*** scottsanchez has joined #openstack-dev17:32
jaypipesnotmyname: hmm..17:35
*** dragondm has quit IRC17:41
*** deepakn has quit IRC17:42
jaypipesblamar: tell me, do you feel strongly about the Glance API load balancing thing? ;P17:43
blamarjaypipes,: Yes.17:43
jaypipesblamar: :)17:44
*** dragondm has joined #openstack-dev17:45
tr3buchetcould someone suggest a way to add a column with a foreignkey to a pre-existing table in sqlalchemy which works in sqlite for the tests?17:54
*** vladimir3p has joined #openstack-dev17:56
*** rods has quit IRC18:05
*** vinkeshb has quit IRC18:12
*** rods has joined #openstack-dev18:18
*** Tv has quit IRC18:19
*** troytoman-away has quit IRC18:19
*** troytoman-away has joined #openstack-dev18:19
*** Tv has joined #openstack-dev18:19
jaypipestr3buchet: are you asking how to write the migrate script?18:46
jaypipestr3buchet: or something else?18:46
tr3buchetjaypipes: migrate18:46
jaypipestr3buchet: are you running into the lovely "Cannot alter non-existing index" issue?18:47
tr3buchetnope18:47
tr3buchetNotSupportedError: SQLite does not support ALTER TABLE ADD CONSTRAINT18:47
jaypipestr3buchet: ah... ok.18:47
jaypipestr3buchet: so, you will need to write a 0XX_add_fk_sqlite.sql file to the migrate/versions directory.18:48
tr3buchetwell i've got a whole migration already written18:48
jaypipestr3buchet: adding a .sql file will override the migrate Python script just for the specific RDBMS.18:48
tr3bucheti'm dropping columns, creating tables, moving data around18:49
jaypipestr3buchet: k. you could also do this:18:49
tr3buchetso do i have to do all of that in the .sql version?18:49
jaypipestr3buchet: one sec...18:49
tr3buchetkk18:49
jaypipestr3buchet: this is how I handled it:18:50
jaypipes    dialect = migrate_engine.url.get_dialect().name18:50
jaypipes    if not dialect.startswith('sqlite'):18:50
jaypipestr3buchet: for reference: http://bazaar.launchpad.net/~hudson-openstack/glance/trunk/view/head:/glance/registry/db/migrate_repo/versions/005_size_big_integer.py18:51
tr3buchetjaypipes: so i just don't add the foreginkey at all in sqlite?18:52
tr3buchetas soon as a test depends on the foreignkey wouldn't it fail?18:52
*** mattray has quit IRC18:53
jaypipestr3buchet: if you want to change something for just SQLite, you would need to create the .sql file I talked about above.18:53
jaypipestr3buchet: I had to write .sql files just for SQLite when I needed to change indexes. migrate doesn't work properly for SQLite when changing indexes...18:56
jaypipestr3buchet: and I think that's exactly what you might be running into (since a constraint is just a unique index...)18:56
tr3bucheti see18:56
tr3bucheti think i need to do it for both...18:56
tr3buchetif the foreignkey doesn't exist in sqlite it would probably break tests..18:56
gholtjaypipes: No luck accessing https://code.launchpad.net/swauth/trunk/+setbranch18:59
*** johnpur has joined #openstack-dev18:59
*** ChanServ sets mode: +v johnpur18:59
jaypipesgholt: ok, you want me to just set you as the admin (must be the admin not the driver...)19:00
jaypipesgholt: try again.19:00
gholtjaypipes: Gracias, senor!19:01
jaypipestr3buchet: I wouldn't actually make FKs a requirement in the schema and tests, but that's just my opinion...19:01
jaypipestr3buchet: sometimes people remove them for scaling reasons...19:01
tr3bucheti guess i could do that19:02
tr3buchetwould have to remove all resulting references and backrefs19:02
jaypipestr3buchet: up to you :)19:02
jaypipesgholt: np!19:02
tr3buchetso long story short, sqlite is in the way...19:02
tr3buchetthanks jay i'll screw with it some more, probably end up writing two migrations..19:03
jaypipestr3buchet: ok, ping me if you run into more issues. you can use the glance stuff as examples (look for the .sql files in glance/registry/db/migrate_repo/versions19:04
tr3buchetyeah will do, thanks19:05
tr3buchetjaypipes: one last thing, both migrations get run correct?19:07
tr3buchet0xx - blah.py and 0xx - blah.sql ?19:07
*** antonym has quit IRC19:21
*** clayg has quit IRC19:22
*** clayg has joined #openstack-dev19:23
*** antonym has joined #openstack-dev19:26
*** antonym is now known as Guest1385319:27
*** vladimir3p has quit IRC19:39
*** mcclurmc_home has joined #openstack-dev19:40
mtaylorjaypipes: ola hombr?@!19:54
jaypipesmtaylor: yo19:57
jaypipestr3buchet: yes, sorry, both get run.19:58
tr3buchetjaypipes: is there an order they get run in? in need 0xx_blah.py to run before 0xx_upgrade_sqlite.sql20:10
*** vladimir3p has joined #openstack-dev20:10
jaypipestr3buchet: no. migrate will simply run the .sql file instead of the .py file if your RDBMS is the $RDBMS in the filename. So, 00X_blah_sqlite_upgrade.sql will be the sqlite-only upgrade for the 00X migration.20:13
jaypipestr3buchet: make sense?20:13
*** bcwaldon has quit IRC20:14
tr3buchetjaypipes: yes, but i would prefer if they were both run even on sql. i guess i can make two migrations 00n_blah.py and 00n+1_blah.sql20:14
tr3buchet*sqlite20:14
tr3buchethmm that doesn't work20:15
tr3buchetwhen the .sql is the only migration of that number it fails20:15
tr3bucheti guess i'll make an empty .py file to go with it20:15
tr3buchetthis is pure nastiness20:15
jaypipestr3buchet: you still need a .py migration for the other DBs....20:15
*** bcwaldon has joined #openstack-dev20:15
tr3buchetso here's the thing i've got a big nasty migration. and i need a special sqlite migration to run afterwards which adds a foreignkey constraint. i really really dont' want to have to rewrite the whole big nasty migration in the .sql file20:16
jaypipesbcwaldon: looks good on the local image service branch. MP it.20:16
jaypipestr3buchet: yeah, that sucks.20:18
bcwaldonjaypipes: On it20:18
openstackjenkinsProject nova build #993: FAILURE in 6.6 sec: http://jenkins.openstack.org/job/nova/993/20:19
*** bcwaldon has quit IRC20:22
*** bcwaldon has joined #openstack-dev20:23
openstackjenkinsYippie, build fixed!20:23
openstackjenkinsProject nova build #994: FIXED in 2 min 53 sec: http://jenkins.openstack.org/job/nova/994/20:23
openstackjenkinsProject nova-tarball-bzr-delta build #240: STILL FAILING in 11 sec: http://jenkins.openstack.org/job/nova-tarball-bzr-delta/240/20:29
*** RobertLaptop has quit IRC20:30
*** mattray has joined #openstack-dev20:31
jaypipesbcwaldon: you are a bug filing MACHINE.20:34
bcwaldonthat's how I roll20:34
*** RobertLaptop has joined #openstack-dev20:43
*** markvoelker has quit IRC20:55
*** lorin1 has quit IRC21:08
*** zaitcev has quit IRC21:18
*** vladimir3p has quit IRC21:22
*** zaitcev has joined #openstack-dev21:24
openstackjenkinsProject swift build #273: SUCCESS in 29 sec: http://jenkins.openstack.org/job/swift/273/21:31
openstackjenkinsTarmac: In light of the upcoming Keystone auth service, which should become the standard OpenStack auth service, Swauth will need to move to its own project. Because we don't want the Swift project to require Keystone (or Swauth or any other auth service for that matter) we need a "placeholder". In this proposal I'm adding a TempAuth to fill our auth testing needs within Swift and removing Swauth.21:31
openstackjenkinsTo quickly change from Swauth on a standard SAIO install, change swauth in your pipeline to tempauth and add the following section:21:31
openstackjenkins[filter:tempauth]21:31
openstackjenkinsuse = egg:swift#tempauth21:31
openstackjenkinsuser_test_tester = testing .admin21:31
openstackjenkinsuser_test2_tester2 = testing2 .admin21:31
openstackjenkinsuser_test_tester3 = testing321:31
openstackjenkinsuser_system_root = testpass .admin .reseller_admin21:31
notmynamejaypipes: you were asking earlier about swauth. that last commit just removed it21:54
*** ameade has quit IRC21:56
notmynamemtaylor: I'm getting a lot of errors about the swift PPAs building21:58
*** johnpur has quit IRC22:09
mtaylornotmyname: those emails?22:15
notmynameya. sorry. I meant emails :-)22:16
mtaylor(just making sure we were talking about the same thing :) )22:16
notmynameya. looks like they may have stopped. I haven't seen one for a bit now. perhaps it's resolved?22:17
mtaylorcp: cannot stat `debian/tmp/usr/bin/swauth-add-account': No such file or directory22:17
mtaylorno, I just got one22:17
mtaylorit seems to be related to that22:17
mtaylorsoren: ^^^22:17
jaypipesand boom goes the dynamite.22:17
*** mcclurmc_home has quit IRC22:34
gholtsoren: ttx: mtaylor: jaypipes: notmyname: andwhoeverelse:): I'll be sending a change to soren regarding the packaging. It's because my deswauth branch landed.22:35
*** mattray has quit IRC22:35
*** pmullaney has quit IRC22:35
mtaylorgholt: sweet22:36
*** bcwaldon has quit IRC22:41
gholtnotmyname: If you wouldn't mind reviewing and merging: https://github.com/crashsite/swift_debian/pull/322:44
notmynamegholt: done22:45
gholtnotmyname: Coolio, gracias.22:46
notmynamegholt: if anything is weird about this, remember that we have to do it again soon with de-slogging and write it down :-)22:49
gholtThat's what, minus 5k lines of code from the Swift repo? :)22:49
gholtIt was /easy/, except for the intial review process. Heheh.22:50
notmynameheh22:50
notmyname1) highlight 2) delete 3) wait two weeks for approval22:50
notmynameI gotta run to dinner. ping me if you need anything else22:50
gholtThe hardest part is making the new standalone project. But you can steal from swauth I'm sure.22:51
*** RobertLaptop has quit IRC23:01
*** jkoelker has quit IRC23:16
*** agarwalla has joined #openstack-dev23:16
*** agarwalla has quit IRC23:33
*** rnirmal has quit IRC23:45

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