Thursday, 2010-08-19

*** littleidea has quit IRC00:05
*** e1mer has quit IRC00:07
*** aliguori has quit IRC00:14
*** abecc has quit IRC00:16
*** jc_smith has quit IRC00:20
*** spackest has joined #openstack00:24
*** rossij has joined #openstack00:28
*** aliguori has joined #openstack00:29
*** maplebed has quit IRC00:48
*** gustavomzw has quit IRC00:52
*** sophiap has quit IRC01:05
*** miclorb_ has quit IRC01:08
*** xfrogman5 has joined #openstack01:08
vishyeday: old message it is cruft01:08
vishyeday from when the api_server was listening on rabbit as well01:08
*** littleidea has joined #openstack01:11
edayvishy: ahh, ok. thanks :)01:16
*** miclorb has joined #openstack01:16
*** sophiap has joined #openstack01:18
*** littleidea has quit IRC01:35
*** aliguori has quit IRC02:00
*** justinsheehy has quit IRC02:03
*** rdw has quit IRC02:09
*** maplebed has joined #openstack02:12
*** Xenith has quit IRC02:15
*** adjohn has joined #openstack02:15
*** tr3buchet has quit IRC02:41
*** stewart has quit IRC02:42
*** stewart has joined #openstack02:45
*** pvo has quit IRC02:47
*** spackest has quit IRC02:53
*** rdw has joined #openstack03:15
*** pvo has joined #openstack03:16
*** ChanServ sets mode: +v pvo03:16
*** jdmaturen has joined #openstack03:20
*** jdmaturen has joined #openstack03:21
*** jdmaturen has quit IRC03:21
*** justinsheehy has joined #openstack03:26
*** jdmaturen has joined #openstack03:33
*** pvo has left #openstack03:34
*** jdmaturen has quit IRC03:39
*** sirp1 has quit IRC04:04
*** thedot has joined #openstack04:34
*** littleidea has joined #openstack04:40
*** thedot has quit IRC04:47
*** thedot has joined #openstack04:49
*** f4m8_ is now known as f4m804:50
*** xfrogman5 has quit IRC05:00
*** e1mer has joined #openstack05:03
*** e1mer has joined #openstack05:03
*** zaitcev has quit IRC05:03
*** e1mer has quit IRC05:09
*** tr3buchet has joined #openstack05:21
*** tr3buchet has quit IRC05:22
*** adjohn_ has joined #openstack05:31
*** adjohn has quit IRC05:31
*** adjohn_ is now known as adjohn05:31
*** Xenith has joined #openstack05:39
*** e1mer has joined #openstack05:44
*** thedot has quit IRC05:44
*** ibarrera has joined #openstack05:59
*** adjohn_ has joined #openstack06:11
*** adjohn has quit IRC06:13
*** adjohn_ is now known as adjohn06:13
*** guigui has joined #openstack06:20
*** larsemil_ is now known as larsemil06:24
*** jdmaturen has joined #openstack06:33
*** Nacx has joined #openstack07:02
*** allsystemsarego has joined #openstack07:04
*** ibarrera has quit IRC07:06
*** jdmaturen has quit IRC07:11
*** Podilarius has joined #openstack07:23
*** rnewson has joined #openstack07:23
*** Xenith has quit IRC07:29
*** dgm_077 has joined #openstack07:36
*** rajijoom has joined #openstack07:54
*** zul_ has quit IRC08:01
*** lmcdowell has quit IRC08:01
*** Phacops has quit IRC08:01
*** syah has quit IRC08:01
*** pandemicsyn has quit IRC08:01
*** gcc has quit IRC08:01
*** sureshgv has quit IRC08:01
*** Steve0 has quit IRC08:01
*** alekibango has quit IRC08:01
*** vishy has quit IRC08:01
*** rektide has quit IRC08:01
*** tomo_bot has quit IRC08:01
*** dubs has quit IRC08:01
*** _0x44 has quit IRC08:01
*** justinsb has quit IRC08:01
*** timrc_ has quit IRC08:01
*** comstud has quit IRC08:02
*** Mr_T has quit IRC08:02
*** coredump777 has quit IRC08:02
*** tsuehpsyde has quit IRC08:02
*** pkhamre has quit IRC08:02
*** redbo has quit IRC08:02
*** mjmac has quit IRC08:02
*** jonesy has quit IRC08:02
*** wreese_ has quit IRC08:02
*** h1nchtastic has quit IRC08:02
*** soren has quit IRC08:02
*** eday has quit IRC08:02
*** uvirtbot has quit IRC08:02
*** jeremydei has quit IRC08:02
*** londo_ has quit IRC08:02
*** jpopi has quit IRC08:02
*** kainam has quit IRC08:02
*** f4m8 has quit IRC08:02
*** cloud0 has quit IRC08:02
*** miclorb has quit IRC08:18
*** gasbakid has joined #openstack08:19
*** rajijoom has quit IRC08:31
*** rnewson has quit IRC08:34
*** ambo_ has quit IRC08:34
*** ambo_ has joined #openstack08:40
*** rnewson has joined #openstack08:43
*** rnewson has joined #openstack08:43
*** rajijoom has joined #openstack08:47
*** londo_ has joined #openstack08:48
*** f4m8 has joined #openstack08:48
*** jpopi has joined #openstack08:48
*** kainam has joined #openstack08:48
*** cloud0 has joined #openstack08:48
*** letterj_ has joined #openstack08:57
*** guigui has quit IRC09:02
*** sophiap has quit IRC09:02
*** perestrelka has quit IRC09:02
*** jforester has quit IRC09:03
*** Daviey has quit IRC09:03
*** rbergeron has quit IRC09:03
*** btorch_ has quit IRC09:03
*** letterj has quit IRC09:03
*** cw has quit IRC09:03
*** DogWater has quit IRC09:03
*** PiotrSikora has quit IRC09:03
*** ikk has quit IRC09:03
*** londo_ has quit IRC09:03
*** jpopi has quit IRC09:03
*** kainam has quit IRC09:03
*** f4m8 has quit IRC09:03
*** cloud0 has quit IRC09:03
*** dgm_077 has quit IRC09:03
*** Kami_ has quit IRC09:03
*** creiht has quit IRC09:03
*** jpipes has quit IRC09:03
*** ianweller has quit IRC09:03
*** glange has quit IRC09:03
*** polvi has quit IRC09:03
*** msparks has quit IRC09:03
*** dendrobates has quit IRC09:03
*** ivan has quit IRC09:03
*** imeyer has quit IRC09:03
*** FeliXdk has quit IRC09:03
*** rnewson has quit IRC09:03
*** ambo_ has quit IRC09:03
*** metoikos has quit IRC09:03
*** loglaunch has quit IRC09:03
*** joschi has quit IRC09:03
*** gholt has quit IRC09:03
*** benoitc has quit IRC09:03
*** notmyname has quit IRC09:03
*** rackerhacker has quit IRC09:04
*** thom has quit IRC09:04
*** samppah has quit IRC09:04
*** dgm_077 has joined #openstack09:11
*** Kami_ has joined #openstack09:11
*** creiht has joined #openstack09:11
*** jpipes has joined #openstack09:11
*** ianweller has joined #openstack09:11
*** glange has joined #openstack09:11
*** polvi has joined #openstack09:11
*** msparks has joined #openstack09:11
*** FeliXdk has joined #openstack09:11
*** dendrobates has joined #openstack09:11
*** ivan has joined #openstack09:11
*** imeyer has joined #openstack09:11
*** brown.freenode.net sets mode: +vv creiht dendrobates09:11
*** ambo_ has joined #openstack09:12
*** cw_ has joined #openstack09:12
*** rnewson has joined #openstack09:12
*** joschi has joined #openstack09:12
*** metoikos has joined #openstack09:12
*** loglaunch has joined #openstack09:12
*** gholt has joined #openstack09:12
*** benoitc has joined #openstack09:12
*** notmyname has joined #openstack09:12
*** rackerhacker has joined #openstack09:12
*** samppah has joined #openstack09:12
*** thom has joined #openstack09:12
*** brown.freenode.net sets mode: +v notmyname09:12
*** londo_ has joined #openstack09:14
*** f4m8 has joined #openstack09:14
*** jpopi has joined #openstack09:14
*** kainam has joined #openstack09:14
*** cloud0 has joined #openstack09:14
*** larissa` has joined #openstack09:14
*** sophiap has joined #openstack09:16
*** cloud0 has joined #openstack09:16
*** kainam has joined #openstack09:16
*** jpopi has joined #openstack09:16
*** f4m8 has joined #openstack09:16
*** londo_ has joined #openstack09:16
*** thom has joined #openstack09:16
*** samppah has joined #openstack09:16
*** rackerhacker has joined #openstack09:16
*** notmyname has joined #openstack09:16
*** benoitc has joined #openstack09:16
*** gholt has joined #openstack09:16
*** loglaunch has joined #openstack09:16
*** metoikos has joined #openstack09:16
*** joschi has joined #openstack09:16
*** cw_ has joined #openstack09:16
*** ambo_ has joined #openstack09:16
*** dgm_077 has joined #openstack09:16
*** Kami_ has joined #openstack09:16
*** creiht has joined #openstack09:16
*** jpipes has joined #openstack09:16
*** ianweller has joined #openstack09:16
*** glange has joined #openstack09:16
*** polvi has joined #openstack09:16
*** msparks has joined #openstack09:16
*** FeliXdk has joined #openstack09:16
*** dendrobates has joined #openstack09:16
*** ivan has joined #openstack09:16
*** imeyer has joined #openstack09:16
*** zelazny.freenode.net sets mode: +vvv notmyname creiht dendrobates09:16
*** larissa` is now known as larissa09:16
*** chilts_ has joined #openstack09:17
*** Daviey has joined #openstack09:18
*** comstud has joined #openstack09:19
*** coredump777 has joined #openstack09:19
*** pkhamre has joined #openstack09:19
*** Mr_T has joined #openstack09:19
*** tsuehpsyde has joined #openstack09:19
*** redbo has joined #openstack09:19
*** mjmac has joined #openstack09:19
*** jonesy has joined #openstack09:19
*** zelazny.freenode.net sets mode: +vv comstud redbo09:19
*** PiotrSikora has joined #openstack09:19
*** justinsb has joined #openstack09:19
*** timrc_ has joined #openstack09:19
*** zheng_li has joined #openstack09:19
*** zul_ has joined #openstack09:19
*** lmcdowell has joined #openstack09:19
*** Phacops has joined #openstack09:19
*** syah has joined #openstack09:19
*** pandemicsyn has joined #openstack09:19
*** gcc has joined #openstack09:19
*** sureshgv has joined #openstack09:19
*** zelazny.freenode.net sets mode: +v pandemicsyn09:19
*** btorch has joined #openstack09:20
*** wreese_ has joined #openstack09:20
*** h1nchtastic has joined #openstack09:20
*** soren has joined #openstack09:20
*** uvirtbot has joined #openstack09:20
*** jeremydei has joined #openstack09:20
*** eday has joined #openstack09:20
*** zelazny.freenode.net sets mode: +v soren09:20
*** alekibango has joined #openstack09:20
*** vishy has joined #openstack09:20
*** rektide has joined #openstack09:20
*** tomo_bot has joined #openstack09:20
*** dubs has joined #openstack09:20
*** _0x44 has joined #openstack09:20
*** zelazny.freenode.net sets mode: +v _0x4409:20
*** Steve0 has joined #openstack09:20
*** dgm_077 has quit IRC09:21
*** Kami_ has quit IRC09:21
*** creiht has quit IRC09:21
*** jpipes has quit IRC09:21
*** ianweller has quit IRC09:21
*** glange has quit IRC09:21
*** polvi has quit IRC09:21
*** msparks has quit IRC09:21
*** dendrobates has quit IRC09:21
*** ivan has quit IRC09:21
*** imeyer has quit IRC09:21
*** FeliXdk has quit IRC09:21
*** alekibango has quit IRC09:21
*** vishy has quit IRC09:21
*** rektide has quit IRC09:21
*** tomo_bot has quit IRC09:21
*** dubs has quit IRC09:21
*** _0x44 has quit IRC09:21
*** btorch has quit IRC09:21
*** wreese_ has quit IRC09:21
*** h1nchtastic has quit IRC09:21
*** soren has quit IRC09:21
*** eday has quit IRC09:21
*** uvirtbot has quit IRC09:21
*** jeremydei has quit IRC09:21
*** Steve0 has quit IRC09:21
*** zheng_li has quit IRC09:21
*** zul_ has quit IRC09:21
*** lmcdowell has quit IRC09:21
*** Phacops has quit IRC09:21
*** syah has quit IRC09:21
*** pandemicsyn has quit IRC09:21
*** gcc has quit IRC09:21
*** sureshgv has quit IRC09:21
*** justinsb has quit IRC09:21
*** timrc_ has quit IRC09:21
*** PiotrSikora has quit IRC09:21
*** comstud has quit IRC09:21
*** Mr_T has quit IRC09:21
*** coredump777 has quit IRC09:21
*** tsuehpsyde has quit IRC09:21
*** pkhamre has quit IRC09:21
*** redbo has quit IRC09:21
*** mjmac has quit IRC09:21
*** jonesy has quit IRC09:21
*** Daviey has quit IRC09:21
*** sophiap has quit IRC09:21
*** londo_ has quit IRC09:22
*** jpopi has quit IRC09:22
*** kainam has quit IRC09:22
*** f4m8 has quit IRC09:22
*** cloud0 has quit IRC09:22
*** ambo_ has quit IRC09:22
*** cw_ has quit IRC09:22
*** metoikos has quit IRC09:22
*** loglaunch has quit IRC09:22
*** joschi has quit IRC09:22
*** gholt has quit IRC09:22
*** benoitc has quit IRC09:22
*** notmyname has quit IRC09:22
*** rackerhacker has quit IRC09:22
*** thom has quit IRC09:22
*** samppah has quit IRC09:22
*** maggiolo00 has joined #openstack09:23
*** chilts has quit IRC09:25
*** letterj_ has quit IRC09:28
*** stewart has quit IRC09:32
*** calavera has joined #openstack09:34
*** PiotrSikora has joined #openstack09:34
*** letterj has joined #openstack09:34
*** Steve0 has joined #openstack09:34
*** _0x44 has joined #openstack09:34
*** dubs has joined #openstack09:34
*** tomo_bot has joined #openstack09:34
*** rektide has joined #openstack09:34
*** vishy has joined #openstack09:34
*** alekibango has joined #openstack09:34
*** eday has joined #openstack09:34
*** jeremydei has joined #openstack09:34
*** uvirtbot has joined #openstack09:34
*** soren has joined #openstack09:34
*** h1nchtastic has joined #openstack09:34
*** wreese_ has joined #openstack09:34
*** btorch has joined #openstack09:34
*** sureshgv has joined #openstack09:34
*** gcc has joined #openstack09:34
*** pandemicsyn has joined #openstack09:34
*** syah has joined #openstack09:34
*** Phacops has joined #openstack09:34
*** lmcdowell has joined #openstack09:34
*** zul_ has joined #openstack09:34
*** zheng_li has joined #openstack09:34
*** timrc_ has joined #openstack09:34
*** justinsb has joined #openstack09:34
*** jonesy has joined #openstack09:34
*** mjmac has joined #openstack09:34
*** redbo has joined #openstack09:34
*** zelazny.freenode.net sets mode: +vvvv _0x44 soren pandemicsyn redbo09:34
*** tsuehpsyde has joined #openstack09:34
*** Mr_T has joined #openstack09:34
*** pkhamre has joined #openstack09:34
*** coredump777 has joined #openstack09:34
*** comstud has joined #openstack09:34
*** Daviey has joined #openstack09:34
*** sophiap has joined #openstack09:34
*** cloud0 has joined #openstack09:34
*** kainam has joined #openstack09:34
*** jpopi has joined #openstack09:34
*** f4m8 has joined #openstack09:34
*** londo_ has joined #openstack09:34
*** thom has joined #openstack09:34
*** samppah has joined #openstack09:34
*** rackerhacker has joined #openstack09:34
*** notmyname has joined #openstack09:34
*** benoitc has joined #openstack09:34
*** zelazny.freenode.net sets mode: +vv comstud notmyname09:34
*** gholt has joined #openstack09:34
*** loglaunch has joined #openstack09:34
*** metoikos has joined #openstack09:34
*** joschi has joined #openstack09:34
*** cw_ has joined #openstack09:34
*** ambo_ has joined #openstack09:34
*** dgm_077 has joined #openstack09:34
*** Kami_ has joined #openstack09:34
*** creiht has joined #openstack09:34
*** jpipes has joined #openstack09:34
*** ianweller has joined #openstack09:34
*** glange has joined #openstack09:34
*** polvi has joined #openstack09:34
*** msparks has joined #openstack09:34
*** FeliXdk has joined #openstack09:34
*** dendrobates has joined #openstack09:34
*** ivan has joined #openstack09:34
*** imeyer has joined #openstack09:34
*** zelazny.freenode.net sets mode: +vv creiht dendrobates09:34
*** londo_ has quit IRC09:37
*** PiotrSikora has quit IRC09:37
*** PiotrSikora has joined #openstack09:39
*** rbergeron has joined #openstack09:39
*** PiotrSikora is now known as Guest9782109:40
*** londo_ has joined #openstack09:41
*** ambo_ has quit IRC09:43
*** ambo_ has joined #openstack09:43
*** Guest97821 is now known as PiotrSikora09:43
*** PiotrSikora is now known as Guest1277409:44
*** PiotrSikora_ has joined #openstack09:44
*** taihen has joined #openstack09:46
*** justinsheehy has quit IRC09:50
*** gasbakid has quit IRC09:54
*** justinsheehy has joined #openstack09:56
*** gasbakid has joined #openstack09:56
*** stewart has joined #openstack10:00
*** rnewson has joined #openstack10:00
*** ptremblett has joined #openstack10:02
*** avsm has joined #openstack10:08
*** michalis has joined #openstack10:08
*** PiotrSikora_ has quit IRC10:18
*** rbergeron has quit IRC10:18
*** PiotrSikora_ has joined #openstack10:18
*** rbergeron has joined #openstack10:18
sorencreiht:10:27
sorenWhoops10:27
sorenmtaylor: I've filed bug #620368 and bug #620367, grabbed python-eventlet from the swift ppa, updated it to 0.9.10, and fiddled a bit with debian/control to stop pretending it's already in Debian :)10:28
uvirtbotLaunchpad bug 620368 in ubuntu "[FFe] new package: python-eventlet" [Undecided,New] https://launchpad.net/bugs/62036810:28
uvirtbotLaunchpad bug 620367 in ubuntu "[FFe] new package: swift" [Undecided,New] https://launchpad.net/bugs/62036710:28
sorenmtaylor: It still has your name on it (as the Original-Maintainer), though, so I'd like your approval before I actually upload it.10:29
sorenmtaylor: As for Swift, how ready is that for upload? Is there anything you want to do to it, before I review and sponsor the upload?10:30
sorendendrobates: ^ fyi.10:31
maggiolo00hi10:39
maggiolo00which ide do you use for python?10:40
sorenvi10:41
maggiolo00thanks10:42
*** michalis has quit IRC10:48
*** PiotrSikora_ is now known as PiotrSikora10:58
*** sandywalsh has joined #openstack11:17
*** adjohn has quit IRC11:18
*** ctennis has quit IRC11:21
*** dgm_077 has quit IRC11:30
*** justinsheehy has quit IRC11:33
*** ctennis has joined #openstack11:37
justinsbmaggiolo00: eclipse has really good python integration also, if you're already familiar with eclipse: http://pydev.org/11:41
*** michalis has joined #openstack11:41
dendrobatessoren: thanks11:43
*** miclorb has joined #openstack11:44
*** sophiap has quit IRC11:45
*** metoikos has quit IRC11:50
*** metoikos has joined #openstack11:52
*** sophiap has joined #openstack11:56
*** justinsb has quit IRC12:05
*** justinsb has joined #openstack12:07
*** tobym has joined #openstack12:09
*** justinsheehy has joined #openstack12:10
*** gustavomzw has joined #openstack12:12
*** dendrobates is now known as dendro-afk12:20
*** zul_ is now known as zul12:21
*** ChanServ sets mode: +v letterj12:21
*** pvo has joined #openstack12:26
*** ChanServ sets mode: +v pvo12:26
*** dendro-afk is now known as dendrobates12:27
*** kashyapc has joined #openstack12:34
* soren is putting finishing touches on a patch to libvirt to support hotplugging disks for UML.12:43
*** guigui has quit IRC12:43
sorenI think we're pretty close to UML supporting the full feature set needed by Nova. Yay for testing!12:43
justinsbsoren: That will be a great feature... e.g. if we want to fill up capacity with super-low priority tasks (e.g. Hadoop)12:48
sorenjustinsb: I guess. I wasn't really expecting to use it for production. I just don't have much hardware to test with, so being able to test Nova *in the cloud* would make my life much simpler.12:49
*** ppetraki has joined #openstack12:50
* soren contemplates nested uml, but gets confused by the mere thought12:50
justinsbsoren: I'm looking at VirtualBox for the same reasons...12:51
justinsbsoren: I just think your UML feature might have more wide-spread usefulness down the road...12:51
sorenjustinsb: It's possible. I haven't benchmarked it in recent time (last time was probably... wow, 5-7 years ago), but back in the day, it was quite slow.12:57
*** ikk has joined #openstack12:59
justinsbsoren: Though maybe you should just ask your manager for the hardware you need :-)13:00
*** miclorb has quit IRC13:04
*** abecc has joined #openstack13:04
jerouml? is that still developped?13:07
*** pvo has quit IRC13:10
dendrobatessoren: are you pushing you libvirt patches to the ubuntu libvirt package and to upstream13:10
dendrobatessoren: are you testing with Mavrick?13:11
*** jbd has quit IRC13:17
jeronevermind13:17
*** dendrobates is now known as dendro-afk13:29
*** codejunkie has joined #openstack13:30
*** pcrews has joined #openstack13:31
*** gasbakid has quit IRC13:34
sorendendro-afk: Upstream for now. Once reviewed, I'll push to Maverick.13:37
sorendendro-afk: I'm testing with maverick, yes.13:37
*** justinsheehy has quit IRC13:38
*** justinsheehy has joined #openstack13:38
sorenjustinsb: I've considered it :)13:40
* soren takes a break13:40
zulsoren: rackspace must have some spare hardware ;)13:40
sorenzul: I'm sure they do. In San Antonio :)13:40
*** lmcdowell has quit IRC13:40
zulsoren: heh time to move ;)13:41
sorenzul: I think asking San Antonio to move to Aalborg is a bit of a stretch.13:41
soren;)13:41
* soren really takes a break13:42
*** gasbakid has joined #openstack13:48
*** dagger has joined #openstack13:49
*** f4m8 is now known as f4m8_13:50
*** aliguori has joined #openstack13:52
*** pvo has joined #openstack13:54
*** ChanServ sets mode: +v pvo13:54
*** binaryWarrior has joined #openstack13:56
*** kashyapc has quit IRC14:03
*** binaryWarrior has quit IRC14:11
*** rajijoom has quit IRC14:13
*** ppetraki has quit IRC14:17
*** lmcdowell has joined #openstack14:28
*** mrayzenoss has joined #openstack14:28
*** mrayzenoss is now known as mray14:28
*** mray has joined #openstack14:28
*** lmcdowell has quit IRC14:36
*** lmcdowell has joined #openstack14:40
* soren returns14:43
*** sirp1 has joined #openstack14:43
notmyname^ sounds like a sequel to a movie14:43
uvirtbotnotmyname: Error: "sounds" is not a valid command.14:43
jero^ shutdown14:44
uvirtbotjero: Error: "shutdown" is not a valid command.14:44
*** lmcdowell has quit IRC14:45
*** lmcdowell has joined #openstack14:46
jero^ whoami14:46
uvirtbotjero: I don't recognize you.14:46
*** jforester has joined #openstack14:48
*** gundlach has joined #openstack14:49
*** jforester has quit IRC14:50
*** rbergeron has quit IRC15:00
*** rbergeron has joined #openstack15:00
*** rackerhacker has quit IRC15:00
*** rackerhacker has joined #openstack15:00
*** ppetraki has joined #openstack15:02
sorenjero: Rude. 14:45 < jero> ^ shutdown15:08
* soren heads dinnerwards.. be back in a few hours15:09
*** gasbakid has quit IRC15:10
jero:-)15:11
*** lmcdowell has quit IRC15:12
mtaylorsoren: swift is pretty ready for upload... although it might be a good idea to cut a new tarball release and release that15:14
*** calavera has quit IRC15:26
*** jdmaturen has joined #openstack15:26
*** jdmaturen has quit IRC15:30
*** ppetraki has quit IRC15:34
*** ppetraki has joined #openstack15:34
*** littleidea has quit IRC15:40
*** zaitcev has joined #openstack15:40
*** littleidea has joined #openstack15:41
*** rektide has quit IRC15:48
*** jdmaturen has joined #openstack15:48
*** lmcdowell has joined #openstack15:49
*** Nacx is now known as ibarrera15:51
*** jdmaturen has quit IRC15:52
edaysoren: is there any reason why regular qemu wasn't working for cloud in the cloud?16:00
*** rektide has joined #openstack16:05
*** pcrews has quit IRC16:11
*** pcrews has joined #openstack16:16
*** zheng_li has quit IRC16:42
*** ptremblett has quit IRC16:43
*** ptremblett has joined #openstack16:43
*** ibarrera has quit IRC16:54
*** avsm has left #openstack16:54
*** littleidea has quit IRC16:56
*** aliguori has quit IRC16:58
*** pcrews has quit IRC16:58
*** jamiew has joined #openstack17:00
*** maggiolo00 has quit IRC17:03
*** jamiew has left #openstack17:03
*** pcrews has joined #openstack17:04
*** michalis has quit IRC17:07
*** zheng_li has joined #openstack17:13
*** thedot has joined #openstack17:15
*** pcrews is now known as pcrews_bbl17:19
*** littleidea has joined #openstack17:20
*** mtaylor has quit IRC17:37
*** holoway has quit IRC17:44
*** dendro-afk is now known as dendrobates17:45
cw_wrt to swift, when the ring is updated ... how do you encorce all object stores see the update and import it in a timely manner?17:46
*** holoway has joined #openstack17:46
*** rajijoom has joined #openstack17:47
gholtOne option is to just scp the new ring to all the servers. They will auto-reload when they see the modtime change, ~15seconds.17:49
gholtAnother option is to place the ring on an interval web server and have each store server wget it if the last-modified header has changed. -N I think.17:52
gholts/interval/internal/17:52
*** tobym has quit IRC17:54
gholtThat'd be good to add to the admin guide for swift. I'll make a bug to do that.17:55
eday_cerberus_, gundlach: any chance I can get an Approve (or reasons not) on the api-tests branch? would like to land that today if possible. thanks!17:56
gundlachi'll leave that to _cerberus_ as he's way more familiar with testing17:56
gundlach(unless you're slammed, _cerberus_)17:56
_cerberus_I've been looking at it. It looks basically the same from yesterday17:56
_cerberus_i.e. good17:57
eday_cerberus_: yeah, nothing changed but a few pylintrc updates17:57
_cerberus_reviwed as approved :-)17:58
*** ptremblett has quit IRC17:59
edaythanks!18:01
uvirtbotNew bug: #620586 in swift "Add how to push out new rings to the admin guide" [Undecided,New] https://launchpad.net/bugs/62058618:05
cw_gholt: if hw is down ... it won't see the update18:13
cw_it seems the protocol should have a ring generate count18:13
cw_so when talking you check you have the updated ring ... and if not you pull from a peer that does18:13
*** gundlach is now known as cerberus18:13
*** cerberus is now known as gundlach18:13
gholtThose are possibilities. Right now ring changes are done manually and verifying they are pushed out is done manually. Automating that stuff is on the to do list. :)18:15
*** mtaylor has joined #openstack18:21
*** ChanServ sets mode: +v mtaylor18:21
*** tobym has joined #openstack18:21
*** rajijoom has quit IRC18:29
sorenmtaylor: Cool beans. Can you take care of that today? The FFe bugs were approved.18:30
soreneday: Too darn slow.18:30
soreneday: Uselessly slow.18:30
edaysoren: ahh, ok18:30
soreneday: uml has a way smaller footprint both in terms of memory and CPU, so I can actually run a number of them even on small rackspace cloud instances.18:31
mtaylorcreiht: you ok with cutting a new tarball from what's in trunk?18:31
sorenmtaylor: Are you cool with the eventlet package?18:31
mtaylorsoren: yeah18:31
sorenmtaylor: Enough that I can keep your name on it, that is.18:31
mtaylorsoren: I feel like I pulled that from someone else...18:32
sorenmtaylor: Otherwise, I can just put my own.18:32
sorenmtaylor: Very likely. The ITP bug linked to an almost-done package. Maybe that's where you started.18:32
mtaylorsoren: oh - no, I think I did that one from scratch actually18:33
mtaylorsoren: yeah - that's all me18:33
sorenmtaylor: Great. I'll upload it now.18:33
mtaylorsoren: cool. where did you make your changes? I suppose I should push my packaging branch...18:34
mtayloror we can just make one once you've uploaded18:34
sorenmtaylor: I made a few changes, and pushed the fresh version to my PPA.18:34
sorenmtaylor: If you want to have a glance first, that's cool.18:35
sorenmtaylor: I fiddled a bit with debian/rules.18:35
sorenmtaylor: Most of the rules in there seemed superfluous.18:35
sorenmtaylor: and I changed the maintainer stuff in debian/control.18:36
sorenmtaylor: ..and pruned the changelog a bit.18:36
sorenmtaylor: Yup, I think that's it.18:36
mtaylorcool18:38
*** ppetraki has quit IRC18:38
sorenmtaylor: So you want to look at it first, or am I ok to upload?18:38
*** aliguori has joined #openstack18:42
*** ianw has joined #openstack18:43
mtaylorsoren: nope. I think you're fine to go18:44
*** cw_ is now known as cw18:54
sorenmtaylor: it's in the NEW queue.18:54
mtaylorsoren: excellent18:55
sorenmtaylor: I really would like to get swift uploaded today, too. I'm going to go over the packaging now.18:57
mtaylork.18:59
creihtmtaylor: hrm18:59
mtaylorcreiht: or should we just release 1.0.2 into ubuntu?18:59
creihtI'm close to having some changes19:00
*** mray has quit IRC19:00
creihtbut it may not until first of next week19:00
creihtThere are also some changes in the current stuff19:00
creihtie. gholt's container metadata19:01
creihtWhich has hardly been tested19:01
mtaylorok19:01
*** mrayzenoss has joined #openstack19:01
*** mrayzenoss is now known as mray19:01
*** mray has joined #openstack19:01
creihtNot sure if I'm really comfortable with making it an official release19:01
creihtUnless there was a real compelling reason to do so19:01
mtaylorcreiht: alright - we'll just get 1.0.2 in then and then do an update whenever it's ready19:01
creihtk19:02
creihtmtaylor: With the amount of changes we are doing, it is likely we are going to want to bump to 1.1 anyways19:04
creihton the next release19:04
mtaylorexcellent19:05
mtaylorsoren: grab lp:swift/debian up through rev 4519:05
mtaylorsoren: I hadn't set up the other branches there properly yet ... this seems like a good time to do so19:05
sorenmtaylor: mkay19:06
*** jhenry1 has joined #openstack19:07
*** pcrews_bbl is now known as pcrews19:11
sorenmtaylor: Why the --no-start to dh_installinit?19:15
*** perestrelka has quit IRC19:15
mtaylorsoren: I didn't do that bit ... that's the way they had it set up19:16
sorenok19:16
*** perestrelka has joined #openstack19:17
*** thedot has quit IRC19:17
edaymtaylor, soren: Do either of you know why LP will show conflicts in a merge proposal when they don't actually exist with trunk? What else would it be diffing against?19:23
edayI've seen it a couple times, always wondered why19:24
zaitcevWhat am I supposed to do to know where .unittests fails?19:24
zaitcevA bunch of 'E' is seen and then a hang forever.19:25
soreneday: link?19:25
uvirtbotNew bug: #620628 in swift "Add public container facility" [Undecided,In progress] https://launchpad.net/bugs/62062819:25
edaysoren: https://code.edge.launchpad.net/~eday/nova/api-tests/+merge/3296019:26
edaysoren: just merged into trunk with no issues19:26
edayzaitcev: run with verbose enabled to get more details, and if it's legit, report a bug19:27
zaitcevthanks, on it19:28
*** ptremblett has joined #openstack19:28
mtayloreday: no. it shouldn't do that19:28
mtayloreday: I'll ask thumper when he wakes up19:29
edayall I can think is it is diffing against an old version of trunk, possibly when it was proposed rather than latest19:29
edaygundlach: ^^19:29
mtayloreday:  it _should_ update the merge prop diff when trunk is updated19:30
gundlacheither way, it makes the diff not reviewable, so if you didn't actually commit files with >>>>>>>>> TREE in them, then it's an LP bug19:30
edaygundlach: in that case, we can just merge into our own copy of trunk and review diff (I do that sometimes anyways since it's easier to read than the web page diff)19:31
soreneday: Good question. I have a hypothesis that I'm testing right now.19:32
*** ianw has quit IRC19:32
gundlacheday: so i'd need to 'bzr branch lp:nova temp; cd temp; bzr merge lp:~eday/somebranch; bzr diff'?19:33
edaygundlach: or just use the trunk copy you probably already have locallly. ie: cd nova/trunk; bzr merge ...; bzr diff; bzr revert  (revert gets trunk back to state before merge)19:34
soreneday: Nope, no clue.19:35
gundlacheday: got it.  so, that's a workaround, but i'd still rather not have to work around the bug.  i'll review your code via the workaround, but should i be filing a bug report or something, mtaylor?19:35
mtaylorgundlach: I'll file one - you want me to subscribe you to it?19:36
gundlachmtaylor: no thanks -- just knowing that it'll get taken care of is plenty :)19:36
_cerberus_eday: Crap, I missed those merge lines as well. I must have scrolled right by them19:37
mtayloreday: it shows that it's been merged already19:39
edaymtaylor: it has, I just marked it as approved to see if it would still go through19:39
edaymtaylor: so, despite the diff showing a conflict, it landed just fine :)19:40
mtaylordamn. I should have told you not to do that ... I was trying to point lifeless at it real quick19:40
mtayloroh well19:40
edayoh, I did that a bit ago.. not just now19:41
mtayloreday: k19:41
vishymtaylor: there were a couple of times when a branch was reported as rejected: no diff"19:41
vishybut it was actually merged19:41
vishyi assume that was a tarmac bug?19:41
mtaylorvishy: that's actually probably more likely correct- it's possible that a branch got merged in as part of someone elses branch? point me at it next time it happens and I'll doublecheck for you19:42
*** pvo_ has joined #openstack19:42
*** ChanServ sets mode: +v pvo_19:42
mtayloreday: I see different approved rev than merged-at rev19:42
edaymtaylor: hmm19:43
mtayloreday: I have no idea what you've done here :)19:43
sorenTarmac merges the rev of the branch at the time it was set to "needs review", doesn't it?19:45
edaymtaylor: oh, that's just cause I was up to 253 in my branch, and trunk was at 248 (and 249 was the mrege)19:45
mtayloreday: ah. gotcha19:45
mtaylorsoren: yes19:45
mtaylorsoren: sorry - I was trying to match two different unrelated bits of data in my brain19:46
*** pvo has quit IRC19:46
*** pvo_ has quit IRC19:47
ctennishttp://cloudscaling.com/blog/cloud-computing/technology/openstack-swift-barn-raising19:47
edaymtaylor: could it be due to there being a prerequisite branch? almost looks like it diffed against trunk without the pre-req19:48
vishymyaylor: it happened here19:48
vishyhttps://code.launchpad.net/~vishvananda/nova/fix-zipfile/+merge/3262719:48
vishystatus was changed back to Needs Review19:49
vishybut the changes showed up in trunk19:49
edaymtaylor: This and my 'api-port' branch (the prereq) was done with bzr pipelines too, if that's doing anything funky. I did have to pump some changes from api-port into api-tests19:49
edaymtaylor: although that shouldn't matter as they just appear as later merges19:50
mtayloreday: ok - spoke with lifeless - diff only updates when the source branch updates, not when the target does19:51
mtayloreday: so there may have been a conflict when you pushed which subsequently got fixed but the diff didn't get re-genned19:51
mtayloreday: there is an outstanding bug about this19:51
edaymtaylor: ahh, ok. thanks!19:52
mtayloreday: I'm going to me-too the bug and add it to our bucket19:53
*** viewlogic has quit IRC19:59
*** mrayzenoss has joined #openstack20:04
*** mrayzenoss has joined #openstack20:04
*** tobym has quit IRC20:06
*** mray has quit IRC20:07
*** pvo has joined #openstack20:08
*** ChanServ sets mode: +v pvo20:08
*** gundlach has quit IRC20:13
*** codejunkie has quit IRC20:16
*** tobym has joined #openstack20:23
*** littleidea_ has joined #openstack20:25
*** ctennis has quit IRC20:27
*** littleidea has quit IRC20:28
*** littleidea has joined #openstack20:32
letterjsoren: I see have looked at the packaging.  What is the proper way to have the /etc/swift directory created?20:33
*** littleidea_ has quit IRC20:35
sorenletterj: Just shipping files in there will create it.20:38
sorenletterj: Hang on, I have a number of patches for the packaging. I'll push them to launchpad.n20:38
sorenletterj: Before uploading, I'll also trim the changelog down to just one entry saying "initial upload to ubuntu." and a reference to the FFe bug.20:39
letterjsoren:  So should the *.conf-sample files be moved there?20:39
sorenletterj: That's how it's usually done.20:39
sorenletterj: lp:~soren/swift/packaging20:39
sorenletterj: Also, usually we start the daemons we install.20:40
gholtHehe, the samples used to be in /etc/swift, but were then moved to /usr/share/doc for some reason.20:40
*** allsystemsarego has quit IRC20:41
gholtFor swift, I'm pretty sure starting the daemons up automatically would be odd. They almost certainly will need some configuration first.20:41
letterjWe had a request from our OPS guys not to start the daemons on install20:41
letterjand reboot20:42
*** justinsheehy has quit IRC20:42
gholtAh, that's true. If a server reboots, they'd prefer to inspect it manually before starting things up again.20:43
*** joearnold has joined #openstack20:43
mtaylorsoren: I moved them out of /etc/swift because they were *.conf-sample and I didn't know enough to install them as *.conf20:45
mtaylorsoren: with sample settings in there - seemed weird to install them in a conf file location and not actually have them be conf files, you know?20:45
jerojust move all that to /bin20:46
gholtWe could change the .conf-sample files to .conf files and just comment everything in them out [and document them a little bit better].20:48
mtaylorgholt: I think that might solve my general beef with them20:48
gholtWe would need to tweak the SAIO instructions a bit in such a case. swift-init has conditional code based on the location of conf files. /etc/swift/object-server.conf vs. /etc/swift/object-server/[1-4].conf for example. But not a big deal.20:50
*** novas0x2a has quit IRC20:51
sorenmtaylor: Hm..20:51
*** ppetraki has joined #openstack20:51
sorenNot starting on boot wouldn't be against policy, but it would be very, very out of the ordinary.20:51
_0x44termie, vish: Do you guys have a fast-import branch that fixes the BTreeBuilder bug?20:51
letterjgholt: Everything is commented out of the .conf-sample files20:52
gholtWell, I know rsync is that way. But they have a RSYNC_ENABLE setting in /etc/default/rsync20:52
sorenI don't have the mental capacity to attack this right now, I'm afraid. It'll have to wait until tomorrow.20:52
letterjI'm talking about the swift *.conf-sample files20:54
gholtletterj: That's almost true with a couple not very notable exceptions. But I mean we could even comment out the section headers. Some things run based on the header existing.20:54
mtayloryeah - the normal way to handle a package not booting on start is to put in a setting in /etc/default/foo that disables booting20:56
mtaylorand/or a debconf question that can set it20:56
*** aliguori has quit IRC20:57
cwi don't see a debian/ dir in the repo21:03
cwis that something you can push out?21:03
cwcertainly if you have pkg'ing stuff that would be useful21:03
gholtIt's a separate swift/debian branch.21:04
cwactually, i do have some of it21:04
cwhmm21:04
letterjthe debian dir is in a different branch lp:~swift-core/swift/debian21:05
cwhow in bzr do i21:05
cwgah21:05
cwhow in bzr do i switch to that?  just reclone?21:05
gholtI just do it separately yeah. But I'm no bzr expert.21:05
letterjbzr branch lp:~swift-core/swift/debian21:06
letterjI just merge code in locally that I want to build packages for21:06
letterjif that code is different than what is in trunk21:07
* mtaylor needs to push another debian branch and write up some expanded instructions21:08
mtaylorsince we're updating lp:~swift-core/swift/debian from hudson - we need another branch for actual releases based on tarball releases ... must document for people who aren't me21:09
letterjmtaylor: if soren is going to clean up the packaging then I guess we don't need to merge my branch21:11
_cerberus_eday: you around?21:11
*** e1mer has quit IRC21:16
gholtmtaylor: Here's an idea. What if the packages all installed as single replica stupid simple Swift? It'd work for say 3rd party developers that want something to test against. Everybody else would have to reconfigure things, but then again, they always would have.21:16
mtaylorgholt: yes. I think that makes the most sense to me21:17
*** ctennis has joined #openstack21:17
mtaylorgholt: I really prefer when installing a package gets me a basic working thing, personally21:17
gholtThe /etc/default thing doesn't make sense the more I looked at it. It seems that controls boot-up and manual starts. For instance, rsync still won't run manually until you enable it in /etc/default21:17
*** jhenry1 has left #openstack21:18
gholtI'll play around with a SwiftOne install. A single replica isn't something we've ever really done, heh.21:18
mtaylorsweet21:19
*** gasbakid has joined #openstack21:21
*** justinsheehy has joined #openstack21:22
eday_cerberus_: yeah21:22
_cerberus_If you have some time today, would you mind looking at https://code.launchpad.net/~cerberus/nova/api_tests ?21:23
edaylooking now21:23
_cerberus_I started moving some stuff around and I'd like to get your opinion21:23
edayahh, moving test/wsgi_test into nova/tests21:25
*** novas0x2a has joined #openstack21:25
edayso, for those you should use bzr mv, not a manual move21:25
_cerberus_eday: Yeah, I know. There's more to it than that. I ran into issues with the git-bzr bridge so I had to do some wizardry with a patchfile and it forced a manual move :-/21:26
*** spackest has joined #openstack21:26
edayheh21:27
spackestso, I want to write something from java that interacts with swift.  is the best bet to just form the http calls myself?21:27
spackestno java bindings yet, right?21:27
spackestbest there is a restful web service, right?21:27
*** aliguori has joined #openstack21:28
gholtThere is http://github.com/rackspace/java-cloudfiles21:28
gholtMeant for Rackspace's CloudFiles but can be commandeered to work with Swift.21:28
gholtThough, honestly, I prefer direct REST calls as I'm a control freak. :)21:29
*** theron_ has joined #openstack21:30
*** tobym has quit IRC21:31
eday_cerberus_: you can remove the import * directives to group them all together too.. we'll probably want to just run nosetests which finds and picks up any unittests  (ie, nosetests nova/tests/api)21:31
gholtmtaylor: One interesting caveat. We need xfs (or something support xattrs). A loopback device is fine, but can we mod /etc/fstab to automount that without getting into packaging trouble?21:31
edayI've slowly been plotting how to convert the old tests to be self containted and able to be nosetests-ized21:31
_cerberus_eday: will do21:32
*** murteas has joined #openstack21:33
spackestgholt: any guess on how much tweaking we'd have to do to get java-cloudfiles to work with swift?21:33
gholtafaik, you just need to tweak the auth_url in the cloudfiles.properties and don't use any of the CDN features.21:34
eday_cerberus_: so minus those couple things, everything else looks good21:35
*** theron_ has quit IRC21:35
_cerberus_eday: thanks for looking. I'm going to go ahead and write out some of the tests now looking for expected behavior, and then see about merging it in. Then we can work on the controllers and make the tests pass21:36
*** pvo has quit IRC21:36
*** pvo has joined #openstack21:37
*** ChanServ sets mode: +v pvo21:37
eday_cerberus_: sounds good21:37
spackestwe might just try the rest, there a good guide for the swift rest api?21:37
spackestnot finding one in my simple searching21:37
*** littleidea has quit IRC21:38
cwrackspace cloud flies docs21:39
*** zooko has joined #openstack21:39
cwsadly it21:39
gholtUnfortunately, not yet. It's something we're working on though. The ... yeah those docs are your best bet right now. Just ignore the CDN stuff.21:39
cwsadly it's PDFs .... which bugs me when web site link to something fat21:39
gholthttp://www.rackspacecloud.com/files/cf-devguide-20100217.pdf21:39
mtaylorgholt: week. that's gonna be fun.21:39
mtaylorgholt: I do not know the answer to that21:40
*** zul has quit IRC21:40
gholtcw: We will fix that with Swift docs. I swear! :)21:40
*** littleidea has joined #openstack21:41
*** sophiap has quit IRC21:44
letterjgholt: Doesn't building the ring in the install present a few problems as well?21:46
zookoGreetings, people of #openstack! Yay for fully open source cloud services.21:46
gholtletterj: Well, I can ship a premade 127.0.0.1 ring.21:46
gholtzooko: Howdy21:47
letterjI was thinking about zones and devices21:47
gholtletterj: The problem with the fstab thing is: if modify it on package install, how to cleanly unmodify it on package uninstall.21:47
gholtThis is going to be SwiftOne: 1 device, 1 zone21:48
pvozooko: welcome to earth21:48
letterjwill that even work?21:48
gholtProbably not with the current code, but it can work in theory.21:49
gholtWhich is what I was going to work on. :)21:49
cwgholt: what are the 'rules' wrt to # zones and handoff?21:50
* gholt is finding the relevant part in the code... lp is slow :)21:52
letterjI don't see a SwiftOne being very useful.  I would rather see a package install run on a server with multiple drives and xfs (or some appropriate file system)21:52
gholtBut j, you can't guarantee somebody installing the packages has more than the one drive.21:53
letterjTo me package install is for a multi-server installation.21:53
gholtIt's useful for say, a tool builder who just needs something to test against.21:53
gholtThe package install will be for both, but it can't be multi-server out of the box. There's no way to know what the settings should be.21:54
gholtcw: http://bit.ly/b3AdEu at the end, get_more_nodes21:54
letterjI guess I thought that's what the SAIO was for.21:55
gholtBasically, it grabs another node from zone that isn't in the set of zones for the previously given nodes.21:55
zaitcevI dunno if Ubuntu allows it, but on Fedora you can have openstack-swift-core (which is what you install your piles of 2U, 4U servers) and openstack-swift-1node that only contains a few entries in /etc/swift and gets slapped on top.21:55
cwgholt: so basically that implies 4 zones minimum21:56
gholtYeah, or handoff simply won't happen.21:56
gholtI guess more specifically, replicas + 121:56
cwyeah21:57
cwbut only 3 is tested :-)21:57
gholtHehe21:57
cwper container replica control would be nice21:57
cwdont really see an easy way to do that right now21:58
gholtYeah, that's a thing we'd like too. Haven't given a whole lot of though past "Damn, that'd be nice" heh.21:58
redboThat's probably going to be on our roadmap.21:58
cwyou could probably have multiple ses of rings21:58
cwsets21:58
cwfor each level of redundancy21:59
cwand map containers to them21:59
gholtMaybe, the downside to that is changing it on the fly.21:59
redboI don't think that's necessary, you might need a ring with the maximum redundancy and only use like the first n nodes for whatever redundancy.21:59
cwthere doesn't seem like there should be any problem in having large numbres of zones22:00
redboreplica count could be part of the container metadata22:00
cwi've pushed here to do one per rack for now22:00
redbothen you just have to futz with replication to make it know how many replicas an object should have22:01
cwredbo: say you have a max of 10 replicas ... and someone only wants 2 ... you have to know where those bits are22:01
cwand if you get failures, etc ... it gets more complex22:01
redboyou can just use the first 2 servers from the 10 node ring22:01
cwthen it's uneven22:01
cwoh, i see what you mean22:01
cwfirst two listed ... not lowest two22:01
redboThat's all pretty easy, I think.  Mostly replication is more complicated.  I have a few ideas of how to make it work, but nothing solid.22:03
redbofor example, under the object directory, you could have like 1/ 2/ 3/ 4/ 5/ .. and under those are the files with that many replicas.22:04
redbothen you wouldn't have to rework replication too much22:04
redbobut that sucks in some ways, like you can't bump up the replica count of a container after it's created22:05
redboand also replication might have to do n times the work to traverse partitions22:05
zaitcevtabled hardcodes R=322:06
redboRight now swift's replica count is chosen when you build the ring, and we usually do 3.  But it'd be nice if you could choose it per container or even per object.22:08
*** pvo_ has joined #openstack22:09
*** pvo_ has joined #openstack22:09
*** ChanServ sets mode: +v pvo_22:09
mtaylorgholt: does it need to be in fstab for SwiftOne?22:09
gholtNo, just needs to be auto-mounted somehow if we're going to have a auto-swift-is-running install.22:10
mtaylorgholt: perhaps the SwiftOne init script just does the mount -o loop on it22:10
mtaylorgholt: and then unmounts it on shutdown22:10
gholtYeah, problem there is it's messy for "real" installations. Cheezy SwiftOne in my Production.22:10
spackestyou guys see my blog posts about doing saio?  http://blog.spack.net/?p=181 and http://blog.spack.net/?p=16222:11
mtaylortotally- make the mounting behavior something easy to turn off22:11
spackestis that the goal of SwiftOne?22:11
*** dendrobates is now known as dendro-afk22:12
gholtspackest: Yes, somewhat. SwiftOne would actually be just one replica and [possibly] much simpler for a packaging system to handle.22:12
*** pvo has quit IRC22:13
*** pvo_ is now known as pvo22:13
gholtSAIO would remain as a simulated 4 server setup on one machine for Swift dev work. Whereas SwiftOne could be use for dev work against Swift.22:14
hazmatdo folks here use coverage.py for testing code coverage.. in nova.. i get some errors trying to run it under coverage.py22:14
hazmatfwiw, the same works fine with swift22:15
gholtI'm not sure on nova, but on swift we use nosetests --with-coverage which is based on coverage.py iirc.22:16
cwgholt: do you know how 'full' you run the fs's typically?22:21
cwi'm assuming 80% or more once you hit some scale22:21
cwthen you add more hw and rebalance22:21
gholtI think redbo knows this more than I, but yeah I think 75-80% is where things start falling off, performance wise.22:22
cwwhat's the fall off there?22:23
cwdir scanning time?22:23
cwcertainly xfs on 2tb drives won't degrade very much for 5gb and smaller files quite that soon22:23
gholtUhm, unsure to be honest.22:23
gholtredbo: Any thoughts?22:24
cwif your dirs get large, that can hurt a bit ... as those fragment22:24
*** murteas has quit IRC22:24
*** murteas has joined #openstack22:27
*** murteas has left #openstack22:29
*** spackest has quit IRC22:34
hazmatre nova, are there going to be two apis going forward (ec2) and rackspace.. or is the ec2 api in need of a maintainer?22:35
*** silassewell has joined #openstack22:38
termie_0x44: i do, it is literally copying the _find_ancestors from BtreeIndex to Btree Builder22:38
termiein btree_index.py in bzr22:38
termie_0x44: i don't see how it came back, it was gone for a long time22:38
_0x44termie: I'm not having the problem but _cerberus_ is22:39
termie_0x44: i started getting it recently also22:39
_0x44termie: He's been really frustrated with bzr/lp and started working on his own bridge using patch22:40
termie_0x44: a bunch of peopl just reported it on the project's issue trackeralso, i'm not sure what the best way to address it is22:40
termieit is has been raised with bzr22:40
termiebut nobody wants to claim it is a bug22:41
termiei can fork bzr but it seems silly to tell people to install my branch22:41
_0x44Well, now that there's a fix, at least he can patch it on his local machine22:41
_0x44Thanks :D22:41
_cerberus_termie: thanks termie. I'll consider installing your branch, actually22:42
termie_cerberus_: i'll push it to lp22:42
termiebzr branch bzr bzr_ancestors is somehow outrageously slow22:43
*** littleidea has quit IRC22:44
termiefrom a local checkout22:44
_0x44termie: s/branch.*how/is/22:44
_0x44Fixed it for you.22:44
gholtmtaylor, soren, letterj, redbo: Well, running swift with just one replica seems to be just fine. Of course, there's a lot of packaging work that'll go into making it install that way, but it's reasonably possible.22:45
*** littleidea has joined #openstack22:46
gholtBut, I'm done for today. There's always tomorrow.22:46
termie_cerberus_: https://code.launchpad.net/~termie/bzr/bzr_btree_ancestors22:46
termie_cerberus_: added two new features to git-bzr-ng today, btw22:46
_cerberus_Thanks man, much appreciated22:46
*** miclorb_ has joined #openstack22:47
*** mrayzenoss has quit IRC22:50
termiehttp://github.com/termie/git-bzr-ng/commit/f03d109c16761d171fdb2c229ae8e4ed629fb93a22:52
*** viewlogic has joined #openstack22:55
*** rnewson has quit IRC22:56
termie_cerberus_: let me know if that stuff solves the problem, and git bzr clear may help you clear out messed up info so that you can try to push again22:58
zaitcev  File "/usr/lib64/python2.7/httplib.py", line 736, in connect23:00
zaitcev    self.timeout, self.source_address)23:00
zaitcevTypeError: create_connection() takes at most 2 arguments (3 given)23:00
zaitcevBet this stuff works dandy on a box with 2.623:00
*** aliguori has quit IRC23:07
mtaylor_cerberus_: sorry you're having issues with launchpad - anything I can do to help?23:07
mtaylortermie: do you happen to know the bzr bug number for the bug you just worked around?23:07
termiemtaylor: yeah, one sec23:09
termiemtaylor: https://bugs.launchpad.net/bzr-fastimport/+bug/54162623:09
uvirtbotLaunchpad bug 541626 in bzr-fastimport "'BTreeBuilder' object has no attribute '_find_ancestors'" [Undecided,Confirmed]23:09
termiemtaylor: also if you google for bzr and _find_ancestors23:09
mtaylortermie: excellent. thanks ... I'll follow up with them on that23:09
termiemtaylor: it got moved from bzr to bzr-fastimport at some point, just copying the method seems to fix it23:10
mtaylorhaving to have people install your branch of bzr seems... unworkable long term :)23:10
termiemtaylor: agreed23:10
*** ppetraki has quit IRC23:18
littleideaThis is a chef'd install of SAIO we're using for our work. http://github.com/cloudscaling/swift-solo/23:18
*** pvo has quit IRC23:19
littleideaIf it's useful for anyone or you have improvements, let us know.23:19
*** binaryWarrior has joined #openstack23:21
*** zheng_li has quit IRC23:28
*** littleidea has quit IRC23:29
*** zul has joined #openstack23:30
creihtgholt: I had started working on a swift one proof of concept (except all services would work in one process), would you like to see that?23:30
creihtIt didn't quite work the way I wanted, but some of the other pieces may be useful for the approach that you are doing23:31
creiht(unless you already got that going)23:31
*** littleidea has joined #openstack23:35
*** alekibango has quit IRC23:38
*** littleidea has quit IRC23:39
*** binaryWarrior has quit IRC23:44
*** gasbakid has quit IRC23:57
*** novas0x2a has quit IRC23:58
*** novas0x2a has joined #openstack23:59

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