Monday, 2012-05-21

*** andresambrois has quit IRC00:05
*** davidha has quit IRC00:21
*** davidha has joined #openstack-dev00:21
*** novas0x2a|laptop has quit IRC00:24
*** ywu has joined #openstack-dev00:40
*** dtroyer_zzz is now known as dtroyer00:51
*** dtroyer is now known as dtroyer_zzz00:53
*** andresambrois has joined #openstack-dev00:53
*** andresambrois has quit IRC01:03
*** bhuvan_ has joined #openstack-dev01:28
*** nati has quit IRC01:52
*** nati has joined #openstack-dev01:52
*** andresambrois has joined #openstack-dev01:54
*** ywu_ has joined #openstack-dev02:09
*** ywu has quit IRC02:12
*** s0mik has joined #openstack-dev02:20
*** danwent has joined #openstack-dev02:26
*** s0mik has quit IRC02:29
*** s0mik has joined #openstack-dev02:35
*** shang has quit IRC02:36
*** ywu_ has quit IRC02:45
*** lloydde has joined #openstack-dev02:46
*** shang has joined #openstack-dev02:50
*** s0mik has quit IRC02:52
*** ewindisch has joined #openstack-dev03:05
*** bhuvan_ has quit IRC03:05
*** danwent has quit IRC03:34
*** lloydde has quit IRC03:44
*** lloydde has joined #openstack-dev03:44
*** danwent has joined #openstack-dev03:48
*** lloydde has quit IRC03:49
*** writerdavid has quit IRC03:59
*** cp16net has quit IRC03:59
*** sleepsonzzz has quit IRC03:59
*** cp16net has joined #openstack-dev03:59
*** sleepsonzzz has joined #openstack-dev03:59
*** writerdavid has joined #openstack-dev03:59
*** glenc_ has quit IRC04:00
*** kpepple has quit IRC04:00
*** sleepsonzzz is now known as sleepsonthefloor04:00
*** glenc has joined #openstack-dev04:00
*** kpepple has joined #openstack-dev04:00
*** btorch_ has quit IRC04:00
*** jkyle has quit IRC04:00
*** andrewsben_zz has quit IRC04:00
*** n0ano has quit IRC04:00
*** ryant has quit IRC04:01
*** wkelly has quit IRC04:01
*** guaqua has quit IRC04:01
*** ryant has joined #openstack-dev04:02
*** guaqua has joined #openstack-dev04:02
*** ewindisch has quit IRC04:03
*** andrewsben_zz has joined #openstack-dev04:04
*** btorch has joined #openstack-dev04:08
*** jkyle has joined #openstack-dev04:11
*** aswadrangnekar has joined #openstack-dev04:14
*** wkelly has joined #openstack-dev04:15
*** dtroyer_zzz is now known as dtroyer04:17
*** garyk has quit IRC04:40
*** bhuvan_ has joined #openstack-dev04:44
*** danwent has quit IRC05:11
*** garyk has joined #openstack-dev05:40
zykes-garyk: are you working on quantum ?05:47
*** davidha has quit IRC05:58
*** hattwick has quit IRC06:02
garykzykes-: morning. yes i am. how may i be of assistance?06:05
*** GheRivero has joined #openstack-dev06:10
*** Ruetobas has quit IRC06:11
*** nati has quit IRC06:13
*** nati has joined #openstack-dev06:13
*** littleidea has quit IRC06:14
zykes-garyk: would it be possible to make the IPAM kind of driver based? So that you can for example integrate it with existing IPAM solutions in a enterprise context06:15
*** Ruetobas has joined #openstack-dev06:15
zykes-for example Alcatel QIP06:15
*** mrunge has joined #openstack-dev06:15
garykzykes-: at the moment melange is being integrated into Quantum. At the moment I am not familiar with it. If this is done in a similar manner to the way in which Quantum works with networking then there should be no reason not to do what you suggest. Once again I am not familiar with the code and implementations (hopefully will be in a week or two). I suggest that you ask - Jason Kölker or Dan W06:18
*** nati has quit IRC06:24
zykes-;)06:30
ttxnotmyname: swift-core/release PPA is deprecated (mtaylor was supposed to remove it a long time ago)06:35
ttxnotmyname: Ubuntu packages for the latest Swift currently lives in https://launchpad.net/~openstack-ppa/+archive/latest-milestone-test06:35
ttxnotmyname: but with Ubuntu agreeing to backport latest openstack to current LTS... we will certainly just have a PPA maintained by the Ubuntu server team directly.06:37
ttxto avoid useless duplication of work06:37
*** mindpixel has joined #openstack-dev06:41
*** eglynn_ has joined #openstack-dev06:41
*** eglynn has quit IRC06:42
*** Mkenneth has joined #openstack-dev06:48
*** bhuvan_ has quit IRC06:56
*** eglynn_ has quit IRC06:56
*** dachary has joined #openstack-dev07:09
*** reidrac has joined #openstack-dev07:12
*** eglynn_ has joined #openstack-dev07:16
*** pmezard has joined #openstack-dev07:22
*** pixelbeat has joined #openstack-dev07:49
*** Stackops-Jorge has joined #openstack-dev07:53
*** andresambrois has quit IRC07:56
*** hattwick has joined #openstack-dev07:59
*** eglynn__ has joined #openstack-dev08:00
*** eglynn_ has quit IRC08:02
*** derekh has joined #openstack-dev08:06
*** darraghb has joined #openstack-dev08:11
*** mancdaz has joined #openstack-dev08:12
*** mcclurmc_ has quit IRC08:12
*** Mkenneth has left #openstack-dev08:14
*** huats has quit IRC08:16
*** gael has joined #openstack-dev08:46
*** fc__ has joined #openstack-dev08:46
*** markmc has joined #openstack-dev08:52
zykes-I guess when a instance is stuck in rebuilding it's no good ? I tried to rebuild it to a new image and now it's just hanging at rebuild08:54
*** dachary has quit IRC09:03
*** eglynn has joined #openstack-dev09:05
*** eglynn__ has quit IRC09:07
*** longhudou has joined #openstack-dev09:10
*** apevec has joined #openstack-dev09:12
*** dachary has joined #openstack-dev09:13
*** huats has joined #openstack-dev09:25
*** danpb has joined #openstack-dev09:29
*** byeager has quit IRC09:32
*** byeager has joined #openstack-dev09:55
*** maploin has joined #openstack-dev09:58
*** maploin has quit IRC09:58
*** maploin has joined #openstack-dev09:58
*** mrunge has quit IRC10:40
*** mrunge has joined #openstack-dev10:52
*** flaviamissi has quit IRC10:55
*** rkukura has quit IRC10:55
*** Mkenneth has joined #openstack-dev11:04
*** uvg has joined #openstack-dev11:05
uvgI am newbie to Swift; not able to get the '-A' parameter to swift command11:06
uvgI tried providing the Keystone URL as the auth URL but Swift returned me an 'account not found' error11:08
uvgi am able to list the containers for my account using RESTAPI tool and curl11:13
*** byeager has quit IRC11:14
uvgbut not using the 'swift'  CLI11:14
*** bsza has joined #openstack-dev11:17
*** mjfork has joined #openstack-dev11:22
*** rbasak has joined #openstack-dev11:26
*** wiliam has joined #openstack-dev11:33
*** markvoelker has joined #openstack-dev11:36
*** byeager has joined #openstack-dev11:38
*** shang has quit IRC11:40
*** wiliam is now known as waa11:43
*** Mkenneth has left #openstack-dev11:53
*** byeager has quit IRC11:53
*** mrunge has quit IRC11:54
*** byeager has joined #openstack-dev12:08
*** adalbas has joined #openstack-dev12:11
*** uvg has quit IRC12:13
*** dtroyer is now known as dtroyer_zzz12:14
garykrussellb: ping12:15
russellbgaryk: hi there12:15
garykrussellb: how are you? i am having a few issues with the rpc implementation.12:15
russellbok, have a paste of the problem?12:16
garykrussellb: ok, prior i'll try and explain.12:16
garykrussellb: in the quantum agent once a second i perform a rpc.call for a health check. the health check is meant to return an id.12:17
garykrussellb: i see that the id is returned twice.12:17
russellbwhat do you mean by returned twice?12:17
garykrussellb: please look at: http://fpaste.org/GqMs/12:19
garykrussellb: on the agent side I have NEXT 1051 reset False12:20
garykID - 105112:20
garykNEXT 1051 reset None12:20
garykID - 105112:20
*** flaviamissi has joined #openstack-dev12:20
garykrussellb: questions: 1. on the agent I guess that I need the method health_check to return the ID's returned from the plugin? I implemented http://fpaste.org/i8vO/12:21
russellbthat code looks fine12:22
russellbis there more than one plugin?12:22
garykon the agent side when i print the results, one iteration they are OK, the next one of the parameters is not set.12:22
garykNo, there is only one plugin.12:23
russellbok.12:23
garykthere can be more than one agents, but at the moment there is only one in the test bed.12:23
russellbok, just so you know, when there is more than one, when you call using a topic of 'quantum', you're not going to know which plugin it goes to12:24
russellbit's going to go to one of them, round-robin12:24
russellbjust something to be aware of12:24
garykok.12:25
russellbback to the problem though ... i'm still not clear on what's wrong12:25
russellb        health = self.health_check(id)12:25
russellb        print health12:25
russellbi see that code ... is that printing something unexpected?12:25
garykPlease look at http://fpaste.org/b1qA/12:27
garykThe point of the health check is to keep the configurations in sync. The plugin knows if the agent is not in sync (via the ID) and it informs the agent to resent its devices...12:29
russellbi see two different health_check() implementations that do different things12:33
russellbi suspect you're calling one and then the other, alternating12:33
garyksorry for the confusion. the first on the the agent does the rpc.call. the second on the agent is the one that hadles the reply fro the plugin. this is the one that returns the dictionary. originally i was not sure if i needed this, but there would be an exception if it did not exists12:35
garyki'll change the names and then paste the code12:36
russellbno, i see 2 that return a dict, not do rpc.call()12:37
russellbbottom of http://fpaste.org/GqMs/ ... and on http://fpaste.org/i8vO/12:37
zykes-andrewsben_zz: ping12:37
russellbtry to add some logging and make sure the code you think is running is running on every health check (every second or whatever the time is)12:37
russellbbecause i think you have 2 different bits of code running, alternating each time12:37
russellband when this runs: http://fpaste.org/i8vO/ , is when you're getting reset None12:38
*** alaski has joined #openstack-dev12:39
garykhttp://fpaste.org/qVvW/12:39
garykrussellb: when the code  http://fpaste.org/i8vO/ does not exists I get an exception. I am making sense?12:40
russellbyes.  HostReceiver.health_check() is always going to return reset: None12:41
garykrussellb: when an rpc.call taked place from A to B, the does the function need to be implemented on A and B or just B?12:41
russellbbecause rpc.call() isn't passing reset as an argument12:41
russellbjust B, but in that case, you don't need to do create_consumer() on A12:41
russellbI think that's your issue ...12:41
garyk:)12:41
russellbyou are alternating between A calling B and A calling A12:41
garykGreat. I'll check12:42
garykrussellb ++12:42
russellb:)12:42
russellbso what you had was both sides set up to receive messages on the 'quantum' topic12:43
garykrussellb - you have put a smile on my face. thank you!12:43
russellband the AMQP broker was alternating between them when you sent a message (rpc.call)12:43
garykok, now I understand12:43
russellbgreat12:43
garykone more question. I need A to perform the health checks as described above, but it also needs to receive cast messages from B. I guess that I need a consumer for this.12:49
*** katell has joined #openstack-dev12:52
*** lts has joined #openstack-dev12:52
*** roge has joined #openstack-dev12:56
*** s0mik has joined #openstack-dev13:00
*** mattray has joined #openstack-dev13:00
russellbgaryk: correct.  You need to set them up to use different topic names13:01
russellbyou had them both set up to use 'quantum'13:01
*** longhudou has quit IRC13:02
garykthanks13:02
russellbnp13:03
*** lorin1 has joined #openstack-dev13:05
*** crobinso has joined #openstack-dev13:08
*** renier has joined #openstack-dev13:09
*** salgado has joined #openstack-dev13:11
*** salgado has joined #openstack-dev13:11
*** littleidea has joined #openstack-dev13:11
*** s0mik has quit IRC13:12
*** davidha has joined #openstack-dev13:13
*** nunosantos has joined #openstack-dev13:16
*** kbringard has joined #openstack-dev13:17
mtaylorchmouel: hey! python-swiftclient added to gerrit ... working on adding the jenkins jobs now13:20
chmouelmtaylor: sweet, thanks13:21
chmouelmtaylor: seems like the other libraries binding have their own launchpad project  any chances to set one up as well?13:23
mtaylorchmouel: actually, that's a historical aberration...13:24
mtaylorchmouel: the current thinking is that the client libs should be managed in the parent project ... so you'd file swiftclient bugs in swift...13:24
mtaylorttx: right? ^^^13:24
*** timello has joined #openstack-dev13:24
ttxmtaylor: that's the current thinking13:25
*** aswadrangnekar has left #openstack-dev13:25
chmouelcool whatever is standard is fine by me :)13:25
ttxmtaylor: they are considered separate deliverables from the same project (i.e. they have the same PTL)13:25
ttxso a single LP project.13:25
ttxchmouel: you can use the "python-swiftclient" tag to mark clieent-specific bugs and find them13:25
chmouelttx: cool thanks.13:26
* ttx goes to create that "official tag" to get autocompletion goodness13:26
ttxdone13:27
*** dtroyer_zzz is now known as dtroyer13:28
*** dprince has joined #openstack-dev13:30
*** lorin1 has quit IRC13:30
*** vanchester12416 has joined #openstack-dev13:32
vanchester12416Hello everyone. Can someone explain me some object auditor's behaviour?13:32
vanchester12416from swift i mean13:33
*** dtroyer is now known as dtroyer_zzz13:33
*** rkukura has joined #openstack-dev13:40
*** sdague has quit IRC13:44
*** bsza has quit IRC13:44
*** sdague has joined #openstack-dev13:44
*** jgriffith has joined #openstack-dev13:48
*** markmcclain has joined #openstack-dev13:49
*** sstent has quit IRC13:50
*** sdague has quit IRC13:50
*** sstent has joined #openstack-dev13:51
*** edygarcia has joined #openstack-dev13:51
*** sdague has joined #openstack-dev13:51
notmynamevanchester12416: I can try13:53
*** deshantm has joined #openstack-dev13:55
*** gael has quit IRC13:55
vanchester12416notmyname, am i right that variable self.passes is for counting correct and processed files?13:57
notmynamevanchester12416: don't know off the top of my head. I'd need to go look at the code13:58
notmynamevanchester12416: ya, passes looks like the count of good objects13:59
vanchester12416thank you, i wasn't sured13:59
*** maploin has quit IRC14:00
vanchester12416and also i don't understand at all config option zero_byte_only_at_fps14:00
vanchester12416i found doc where it's binded with XFS errors when it's create empty files, as i understand14:01
notmynamevanchester12416: a "zero byte scan" is a faster version of auditing that only checks to see if the file is now reported as zero bytes long (a common sympton of an xfs error)14:02
vanchester12416aha!14:02
vanchester12416thank you very-very much14:02
*** gael has joined #openstack-dev14:02
notmynamevanchester12416: zero byte scans can be done on startup (eg when a box dies but the drives are still good)14:02
notmynamevanchester12416: so when the problem hardware comes up, you can get a "good enough" check faster than having to md5 hash all of the data on the disk (which takes a long time)14:03
vanchester12416yeah, i see14:04
notmynamevanchester12416: actually, the main struggle with auditing is that since it does so much disk IO, it can impact customer connections if it runs unrestrained. that's why there is the "bytes per second" limit and also the zero byte scan. just different ways to limit what it does14:06
*** ywu has joined #openstack-dev14:08
*** lorin1 has joined #openstack-dev14:09
*** legkodymov has joined #openstack-dev14:09
*** c28 has joined #openstack-dev14:11
vanchester12416it's important info, nice14:11
*** c28 is now known as sla14:12
*** GheRivero_ has joined #openstack-dev14:13
*** sla has quit IRC14:15
*** hub_cap has joined #openstack-dev14:27
*** dhellmann has joined #openstack-dev14:27
garykrussellb: ping14:29
*** mancdaz has quit IRC14:35
*** mancdaz has joined #openstack-dev14:35
*** ewindisch has joined #openstack-dev14:35
*** byeager has quit IRC14:37
zykes-andrewbogott_: what's the diff of your nova plugins vs extensions ?14:38
andrewbogott_zykes-:  I tried to explain that with a not-especially-clear ven diagram on this page:  http://wiki.openstack.org/novaplugin14:39
andrewbogott_zykes-:  Basically, the extension system provides a well-defined way to add API extensions, but it's fuzzy around the edges; there's not an clear system for actually implementing those extensions.14:39
andrewbogott_zykes-:  Plugins also support actions initiated by nova rather than via a client call.14:40
BLZbubbagood morning14:40
andrewbogott_zykes-:  If that page/diagram are unclear, I welcome suggestions.14:41
zykes-so you'll have extensions, drivers and plugins ?14:41
andrewbogott_zykes-:  A plugin will be a packagable unit that may contain one or more drivers or extensions.14:42
andrewbogott_extensions and drivers will continue to exist in their current form... I'm just trying to whittle down the scope of what those terms mean a bit.14:42
jgriffithandrewbogot: I like it, nice work on the wiki IMO14:43
andrewbogott_thanks14:43
zykes-andrewbogott_: python namespace packages?14:43
sorenvishy: Do you have any recollection of why we decided to disable STP on Nova's bridges?14:44
andrewbogott_zykes-:  Yes, probably.  Right now my code implements a couple of deployment methods but I expect to standardize on pip/setup/namespace/entrypoint14:44
zykes-what is an example plugin ?14:45
andrewbogott_zykes-:  The broadest use-case that I'm working on now is this:  http://wiki.openstack.org/SharedFS.  There's a slideshow that shows the interactions w/nova here:  http://prezi.com/2nrdx_b1k8cc/shared-filesystems-in-openstack-nova/14:48
*** rnirmal has joined #openstack-dev14:48
andrewbogott_It extends the API and responds to notifications and contains a fair bit of support code14:48
*** mindpixel has quit IRC14:48
*** renier has quit IRC14:49
*** rgoodwin_away is now known as rgoodwin14:50
*** datsun180b has joined #openstack-dev14:51
*** devanand1 has joined #openstack-dev14:52
andrewbogott_zykes-:  Both of those sharedfs docs demonstrate adding the functionality before the plugin framework existed.  The goal of the plugin framework is largely to simplify and stabilize the interface that external features use to interact with nova.14:53
*** devananda has quit IRC14:53
mtaylorchmouel: ok. jenkins jobs all there ... I have submitted a small cleanup patch14:59
mtaylorchmouel: https://review.openstack.org/762714:59
chmouelmtaylor: cool i just +2 directly and let's see if that works14:59
vishysoren: I don't15:00
*** lloydde has joined #openstack-dev15:00
*** byeager has joined #openstack-dev15:01
*** adjohn has joined #openstack-dev15:01
chmouelmtaylor: awesome that worked!15:02
*** reidrac has quit IRC15:05
*** belliott has joined #openstack-dev15:09
*** vincentricci has joined #openstack-dev15:13
*** byeager has quit IRC15:22
*** thingee has quit IRC15:23
*** andrewsben_zz is now known as andrewsben15:25
*** n0ano has joined #openstack-dev15:33
*** danwent has joined #openstack-dev15:34
*** matwood has joined #openstack-dev15:38
*** byeager has joined #openstack-dev15:39
*** davidha has quit IRC15:41
*** gyee has joined #openstack-dev15:43
*** lloydde has quit IRC15:44
vishycomstud, pvo: are you guys aware that there is an issue with 3.2 Kernel guests on XenServer?15:44
pvovishy: I am not.15:45
pvovishy: details anywhere?15:45
vishypvo: http://invalidlogic.com/2012/04/28/ubuntu-precise-on-xenserver-disk-errors/15:45
pvoI don't see which version of xenserver its referencing.15:46
vishypvo: I haven't tried adding barrier=0 yet, but if you don't do it on a pvgrubbed cloudserver you get a read-only fs with kernel 3.215:46
pvothey mention XS6, but not that they'r eon that ver15:46
vishypvo: i found a bug report somewhere that it is supposed to be fixed in XS 6.115:47
pvok, will follow up with Citrix.15:47
pvothanks15:47
vishypvo: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/82408915:47
uvirtbotLaunchpad bug 824089 in linux "Xen blkfront i/o errors prevent boot in domU" [Undecided,Invalid]15:47
vishypvo: anyway looks like you might be able to work around it by adding barrier=0 in kernel options when you guys create your precise images15:49
vishypvo: I will let you know if it works, retrying now15:50
pvoyep, I'm making our internal ticket now.15:50
*** gyee has quit IRC15:51
*** byeager has quit IRC15:52
*** Ryan_Lane has joined #openstack-dev15:53
*** bhuvan_ has joined #openstack-dev15:57
*** zaitcev has joined #openstack-dev15:58
*** heckj has joined #openstack-dev15:58
*** eglynn_ has joined #openstack-dev16:00
*** eglynn has quit IRC16:00
*** katell has quit IRC16:01
*** lloydde has joined #openstack-dev16:02
*** mancdaz has quit IRC16:05
mnaservishy: is there any standard way of accumulating performance data from nova at the moment?  doesnt seem like it as far as I read so implementing something that relies on nova server diagnostics to get that info.. any better suggestions?16:09
mnaser(xenserver backend)16:10
*** mancdaz has joined #openstack-dev16:14
*** byeager has joined #openstack-dev16:14
jeblairsleepsonthefloor: can you take a look at https://review.openstack.org/763016:17
*** cdub has joined #openstack-dev16:18
jeblairsleepsonthefloor: that fixes a problem that has kept the devstack-gate test image from updating for the past few days16:19
jeblairvishy: ^ if you have a second to look at it too, i'd appreciate it16:19
*** davidha has joined #openstack-dev16:20
*** dolphm has joined #openstack-dev16:23
*** Robbin has joined #openstack-dev16:25
openstackgerritVerification of a change to openstack/nova failed: Remove deprecated quota code.  https://review.openstack.org/736516:27
*** issackel_ has joined #openstack-dev16:29
*** edygarcia has quit IRC16:33
*** gael has quit IRC16:33
BLZbubbais max_cores a per-host option?  i am adding it to nova.conf on everything but it doesn't appear to be working16:34
*** edygarcia has joined #openstack-dev16:37
*** maplebed has joined #openstack-dev16:37
jgriffithclayg: ping16:40
*** thingee has joined #openstack-dev16:41
*** anderstj has joined #openstack-dev16:42
zykes-anyone here running Force10 ?16:44
*** reed has joined #openstack-dev16:45
*** eglynn__ has joined #openstack-dev16:47
BLZbubbai set check_cores=false, and quadrupled the "vcpus" column in compute_nodes; yet still I get:16:47
BLZbubba"Not enough allocatable CPU cores remaining"16:47
*** Stackops-Jorge has quit IRC16:48
*** eglynn_ has quit IRC16:48
BLZbubbahacking simple.py  seems to have done the trick; i'll see if i can figure out why it is so buggy16:51
*** markmc has quit IRC16:53
*** byeager has quit IRC16:54
*** jdurgin has joined #openstack-dev16:54
*** dprince has quit IRC16:57
*** vincentricci has quit IRC16:57
*** bhuvan has quit IRC16:59
*** utlemming has quit IRC17:05
*** katell has joined #openstack-dev17:05
*** PotHix has joined #openstack-dev17:06
*** deshantm has quit IRC17:07
*** DiamRem has joined #openstack-dev17:10
*** byeager has joined #openstack-dev17:13
*** derekh has quit IRC17:16
*** utlemming has joined #openstack-dev17:19
*** jakedahn_zz is now known as jakedahn17:19
*** utlemming has quit IRC17:19
*** darraghb has quit IRC17:20
*** utlemming has joined #openstack-dev17:20
*** andrewsben is now known as andrewsben_zz17:21
*** torgomatic has joined #openstack-dev17:24
*** mdomsch has joined #openstack-dev17:24
*** Ryan_Lane has quit IRC17:24
*** berendt has joined #openstack-dev17:25
*** vanchester12416 has left #openstack-dev17:26
*** andrewsben_zz is now known as andrewsben17:27
claygjgriffith: pong17:28
*** harlowja has joined #openstack-dev17:30
*** epim has joined #openstack-dev17:30
*** n0ano has quit IRC17:31
*** adalbas has quit IRC17:31
*** adalbas has joined #openstack-dev17:32
*** dachary has quit IRC17:34
*** renier has joined #openstack-dev17:35
*** andresambrois has joined #openstack-dev17:38
jgriffithclayg: pm sent17:39
*** garyk has quit IRC17:40
vishyjeblair: sleepsonthefloor is on vaca17:40
*** vincentricci has joined #openstack-dev17:44
jeblairvishy: thanks, i guess he forgot to change his nick to sleepsonthefloor_vacation.  :)17:45
*** koolhead17 has joined #openstack-dev17:47
*** dprince has joined #openstack-dev17:48
*** adjohn has quit IRC17:52
*** troytoman-away is now known as troytoman17:53
*** andresambrois has quit IRC17:54
*** matwood has quit IRC17:56
*** legkodymov has quit IRC17:57
*** apevec has quit IRC17:58
*** johnpostlethwait has joined #openstack-dev18:01
*** andresambrois has joined #openstack-dev18:01
*** Robbin has quit IRC18:03
*** utlemming has quit IRC18:04
sdaguevishy: did you have a chance to look at my follow up email on the driver loader?18:04
*** danwent has quit IRC18:04
*** garyk has joined #openstack-dev18:05
*** dachary has joined #openstack-dev18:05
*** maplebed_ has joined #openstack-dev18:06
*** maplebed_ is now known as Guest7676718:07
*** maplebed has quit IRC18:07
*** Guest76767 is now known as maplebed18:07
*** maplebed has joined #openstack-dev18:07
jgriffith/nick jgriffith_away18:08
*** jgriffith is now known as jgriffith_away18:08
*** Ryan_Lane has joined #openstack-dev18:08
*** Ryan_Lane has joined #openstack-dev18:08
*** kiall has joined #openstack-dev18:11
*** jakedahn is now known as jakedahn_zz18:13
*** dachary has quit IRC18:14
*** dwalleck has joined #openstack-dev18:20
*** danpb has quit IRC18:21
*** koolhead17 is now known as koolhead17|afk18:27
openstackgerritVerification of a change to openstack/nova failed: Add zeromq driver. Implements blueprint zeromq-rpc-driver partially implements blueprint host-topic-matchmaking.  https://review.openstack.org/763318:32
*** koolhead17|afk has quit IRC18:32
comstudvishy: Ya, I ran into that a while back... barrier=0 does fix it18:36
comstudpvo: ^18:36
comstudvishy, pvo: ran into it when building wheezy image18:39
*** Adri2000 has quit IRC18:40
*** jgriffith_away has quit IRC18:40
*** Adri2000 has joined #openstack-dev18:40
* ewindisch has fixed the smokestack complaint.18:44
*** mattray has quit IRC18:44
*** dwalleck has quit IRC18:47
*** mattray has joined #openstack-dev18:48
*** adjohn has joined #openstack-dev18:53
*** jakedahn_zz is now known as jakedahn18:54
*** utlemming has joined #openstack-dev18:58
davidhanotmyname: hi18:59
notmynamedavidha: hi18:59
davidhanotmyname: I am having some problem with teh COPY option - For some reason PUT with the X-Copy-From header duplicates the file body (delivers the file content twice)19:02
notmynamedavidha: so you end up with a "abcabc" when the original only had "abc"?19:03
davidhaYep19:04
davidhanotmyname: is there anyone who may hypothise why?19:04
notmynamedavidha: do you have a reproducible case?19:05
*** troytoman is now known as troytoman-away19:06
davidhaYes it si reproducable. It is under the mod_wsgi work. CONTENT-LENGTH is your example would be 3 in the PUT sent to the Object Server while the object server will actually read 3 and then when the object server try to read more to see if tehr eis more, it would read another 3...19:06
davidhaCome to think of it, if teh Proxy would have sent 6, the object servfer would have read 6 in teh first round19:08
notmynamedavidha: that sounds like a bug in the object server, but it's not something we are seeing with eventlet wsgi19:08
davidhaI agree - I will look deaper into the object server19:09
notmynamedavidha: there are unit tests for copy that ensure that the copied object is the same as the original object. that makes me think it's not actually a problem with the swift part but perhaps the wsgi integration19:10
notmynamedavidha: I'm sure it's possible we could be unintentionally relying on some edge case behavior in eventlet wsgi that shows up when you swap in a different wsgi server. or, there could be a bug in your mod_wsgi integration (aren't distributed, layered systems fun? ;-)19:11
davidhanotmyname: :) yep - lots and lots of fun! Sometimes too much fun.19:13
*** maplebed has quit IRC19:13
davidhanotmyname: I have more issues...19:13
notmynamedavidha: do tell :-)19:14
*** maplebed has joined #openstack-dev19:14
davidhanotmyname: test.functional.tests:TestAccount.testInvalidPath test adds an extra / and expects a 40419:14
zykes-anyone here good at switches and so on ?19:14
davidhaSo GET //v1/AUTH_test HTTP/1.1 is expected to produce a 40419:14
*** maplebed has quit IRC19:14
davidhazykes-   I can try19:15
*** joearnold has joined #openstack-dev19:15
*** johnpostlethwait has quit IRC19:15
*** maplebed has joined #openstack-dev19:15
notmynamedavidha: ok, that seems right to me19:15
*** joearnold has left #openstack-dev19:15
zykes-davidha: does one use like 1 ToR switch pr rack for nova or 2 ?19:15
zykes-for management stuff19:15
davidhazykes- This is not a networking Q -sorry (an experianced IT person - maybe at #openstack a better chance)19:16
davidhanotmyname: ry using an extra / in the internet19:16
zykes-it networking "gear" :)19:16
davidhanotmyname: try using an extra / in the internet - the 'standrad' defacto is that the extra slashes are eliminated and ignored19:17
davidhanotmyname: in fact apache simply ignores the extra slash and delivers it without it to the application who than returns a 200 for success - which I find perfrectly good behaviour19:17
*** dwalleck has joined #openstack-dev19:18
notmynamedavidha: internet in this case being the web (which implies web browsers as the client)? I don't think that assumption holds for an API19:18
*** jgriffith has joined #openstack-dev19:18
*** adjohn has quit IRC19:19
davidhawell I do not think it is being removed by te client, but by the server side19:19
davidhanotmyname: I do not know what the standard (if any) say, but as far as I can see pepole tent to implement allowing this and apache is infact hiding this error19:20
*** johnpostlethwait has joined #openstack-dev19:20
notmynamedavidha: the problem is that "/" has a particular meaning in the swift API. it delineates the different parts of the URI19:20
davidhatent=tend19:20
notmynamedavidha: simply squashing and duplicated "/" at the beginning may be something worth considering for swift, but I don't think that's a given19:21
davidhanotmyname: still it is a choice if to consider: //x/y as: an empty account name, a container of x and an objetc of y or as a account of X and container of y19:22
davidhaoh - this is not only in the beginning19:22
davidhanotmyname: it is anywhere19:22
*** timello has quit IRC19:23
*** troytoman-away is now known as troytoman19:23
davidhanotmyname: as an exmple:  //////x/////////y///////z   = /x/y/z    -again I do not know what the standard have to say about this - and I wonder why apache mod_wsgi does this (or is it an earlier component of apache) - I will inquire with Graham19:24
notmynamedavidha: I'd love to hear the answer too19:24
davidhanotmyname: ok - last Q for now :)19:24
davidhanotmyname: chunked19:25
openstackgerritVerification of a change to openstack/nova failed: Add zeromq driver. Implements blueprint zeromq-rpc-driver  https://review.openstack.org/763319:25
davidhanotmyname: PUT with chunked does not seem to include CONTENT-LENGTH in Swift19:25
notmynamedavidha: right. that is correct behavior from the rfc19:26
davidhanotmyname: apache seem to dislike this behaviour - can you highlight what is standard in this case?19:26
notmynamedavidha: ya, let me look it up19:26
davidhanotmyname: note that we are discusing an upload , not a download (PUT rather than GET)19:26
davidhadiscussing19:26
*** dwalleck has quit IRC19:26
*** n0ano has joined #openstack-dev19:27
notmynamedavidha: http://www.w3.org/Protocols/rfc2616/rfc2616-sec4.html#sec4.4 number 319:30
notmynamedavidha: don't have both a transfer-encoding: chunked and a content-length. if you do, ignore content-length19:30
*** jgriffith has quit IRC19:31
notmynamedavidha: or slightly lower down: "Messages MUST NOT include both a Content-Length header field and a non-identity transfer-coding."19:32
*** bhuvan_ has quit IRC19:33
davidhanotmyname:  Ok, I will try and figure out what is going on than.19:33
*** dolphm has quit IRC19:34
*** timello has joined #openstack-dev19:36
openstackgerritVerification of a change to openstack/nova failed: Only permit alpha-numerics and ._- for instance type names.  https://review.openstack.org/763619:42
*** legkodymov has joined #openstack-dev19:45
*** adjohn has joined #openstack-dev19:47
ryantany nova-network experts out there?19:47
*** troytoman is now known as troytoman-away19:48
*** mrunge has joined #openstack-dev19:49
*** danwent has joined #openstack-dev19:59
*** bhuvan has joined #openstack-dev20:00
*** spiffxp has joined #openstack-dev20:01
*** ncode has joined #openstack-dev20:04
*** ncode has joined #openstack-dev20:04
*** rkukura has quit IRC20:07
*** davidha has quit IRC20:07
*** davidha has joined #openstack-dev20:08
*** troytoman-away is now known as troytoman20:09
*** azbarcea has joined #openstack-dev20:15
*** berendt has quit IRC20:16
*** jgriffith has joined #openstack-dev20:16
jgriffithmtaylor: python-cinderclient is ready, whenever it's convenient for you20:20
*** torgomatic_ has joined #openstack-dev20:22
mtaylorjgriffith: booyah20:22
mtaylorjgriffith: where is the new repo?20:22
jgriffithmtaylor: It's pushed to openstack/python-cinderclient20:22
jgriffithmtaylor: if it makes it easier I can put it back in my personal github20:22
mtaylorjgriffith: ah, so you just want me to squash it to one commit and re-push, right?20:22
*** anderstj_ has joined #openstack-dev20:22
jgriffithYes please :)20:22
*** novas0x2a|laptop has joined #openstack-dev20:23
mtaylorjgriffith: I don't know that I agree with you ...20:24
mtaylorjgriffith: latest commit I show in master is 2a1b5ee5ed8b5024caebaa4185b003c229f2045120:24
*** anderstj has quit IRC20:24
*** torgomatic has quit IRC20:24
*** torgomatic_ is now known as torgomatic20:24
jgriffithmtaylor: ?20:26
jgriffithmtaylor: lemme check20:26
jgriffithmtaylor: DOH it hasn't merged yet20:27
mtaylorjgriffith: oh, wait - that's it, yeah?20:27
mtaylorjgriffith: https://review.openstack.org/#/c/7537/?20:27
jgriffithmtaylor: Talk about the cyring wolf20:27
jgriffithmtaylor: yes 753720:27
mtaylorjgriffith: /me hangs head20:27
mtayloryeah - ok, on it20:27
jgriffithmtaylor: nope, my fault, I should've waited til it merged :)20:27
jgriffithmtaylor: Thank you much!20:28
mtaylornope. it merged. it's all good20:28
jgriffithmtaylor: yep, just now LOL20:28
*** davidha has quit IRC20:28
jgriffithmtyalor: 1:44, whatever20:28
claygjgriffith: mtaylor: I was trying to look at https://review.openstack.org/#/c/7637/1, but it looks like it has some conflict with the last merge20:28
claygis there something that gets flagged in gerrit to see when a review is in conflict?20:29
jgriffithclayg: not sure what you mean?20:29
*** waa has quit IRC20:29
claygjgriffith: monty's branch that fixes the coverage_xml stuff, when I tried to merge it into master (after you big changeset langed) I got some conflicts (to be expected?)20:30
claygbut I didn't know where to look on the gerrit/review.openstack page that would show me obviously that it doesn't merge cleanly?20:30
jgriffithclayg: not sure, the big changeset was just the client... the stuff from mtaylor is in cinder I thought20:31
mtaylorcorrect...20:31
claygoic20:31
*** davidha has joined #openstack-dev20:31
mtaylorthat might explain why you can't merge it on to cinderclient :)20:31
jgriffithhehe... good thing20:31
*** troytoman is now known as troytoman-away20:32
mtaylorjgriffith: ok. I'm going to push now ... means all python-cinderclient things will probably just going to need to get re-cloned20:33
jgriffithmtaylor: Great!20:33
mtaylorjgriffith: done20:34
jgriffithmtaylor: sweet!!!20:34
*** mrunge has quit IRC20:35
jgriffithmtaylor: That looks much better!!  Thanks20:35
*** legkodymov has quit IRC20:35
jgriffithclayg: And thank you for looking at it!20:36
claygno problem, sry for my confusion about the other merge, rebase/reclone looks good20:37
*** hub_cap has quit IRC20:39
*** lts has quit IRC20:39
jgriffithclayg: No worries20:40
*** jgriffith has quit IRC20:45
*** Stackops-Jorge has joined #openstack-dev20:45
*** jgriffith has joined #openstack-dev20:46
mtaylorjgriffith: I let a thing creep in to https://review.openstack.org/#/c/7637 before that shouldn't have been in there ... can you re-review?20:48
mtaylor(it was causing test run hangs ... which I REALLY want to figure out)20:48
jgriffithmtaylor: Looking at it now20:49
claygno problem, sry for my confusion about the other merge, rebase/reclone looks good20:49
jgriffithmtaylor: Done, don't think anybody would disagree on that one20:51
*** Stackops-Jorge has quit IRC20:52
mtaylorjgriffith: thanks!20:52
jgriffithmtaylor: least I could do20:52
jgriffithmtaylor: BTW, is there an easier way to diff two patch sets, I've been pulling both down and using meld.  Would think there's something built into gerrit?20:53
mtaylorjgriffith: there is, in fact (although it gets messed up by rebase, which is a bug which is fixed in the next version of gerrit)20:54
mtaylorjgriffith: if you go to the patch view page, you'll see a header thing Differences -> Patch Sets20:54
mtaylorjgriffith: which will allow you to select what you'd like to view from and to20:54
jgriffithmtaylor: hmmm... sorry, patch view page?20:57
jgriffithmtaylor: Ahh.. nevermind, got it I think20:58
jgriffithmtaylor: Excellent!  Thanks, much better20:58
*** andrewsben is now known as andrewsben_zz20:58
mtaylorjgriffith: w00t!20:58
jgriffithmtaylor: Now if only we could do all files simultaneously :)20:59
jgriffithmtaylor: I have to say between this and the Draft feature I'm really starting to like gerrit quite a bit20:59
mtaylorjgriffith: it's on the roadmap ... promise20:59
mtaylorjgriffith: w00t!20:59
mtaylorjgriffith: just wait til you see the crack we're rolling out this week...20:59
jgriffithmtaylor: Me wants more!!! :)21:00
*** Stackops-Jorge has joined #openstack-dev21:00
*** alaski has quit IRC21:01
*** timello has quit IRC21:02
ewindischlooking for reviews on the zeromq driver: https://review.openstack.org/#/c/7633/21:03
*** timello has joined #openstack-dev21:05
*** jtran has joined #openstack-dev21:06
*** edygarcia_ has joined #openstack-dev21:06
*** timello has quit IRC21:07
*** edygarcia has quit IRC21:09
*** edygarcia_ is now known as edygarcia21:09
*** markvoelker has quit IRC21:10
*** timello has joined #openstack-dev21:10
*** dprince has quit IRC21:14
*** heckj has quit IRC21:17
*** edygarcia_ has joined #openstack-dev21:18
*** bhuvan has quit IRC21:21
*** edygarcia has quit IRC21:21
*** edygarcia_ is now known as edygarcia21:21
*** timello has quit IRC21:23
openstackgerritVerification of a change to openstack/nova failed: Fix XenServer diagnostics to provide correct details  https://review.openstack.org/764321:29
*** lorin1 has quit IRC21:31
*** azbarcea has quit IRC21:36
*** Adri2000 has quit IRC21:37
*** troytoman-away is now known as troytoman21:37
*** Adri2000 has joined #openstack-dev21:37
*** Stackops-Jorge has quit IRC21:38
rmkvishy: Question. Is there a technical reason the instance names use a hex-converted instance.id?  Just want to make sure I'm not going to break anything by changing it.21:45
*** edygarcia_ has joined #openstack-dev21:47
*** timello has joined #openstack-dev21:47
*** edygarcia has quit IRC21:49
*** edygarcia_ is now known as edygarcia21:49
kbringardjaypipes: you about?21:54
*** openstackjenkins has quit IRC21:56
*** openstackjenkins has joined #openstack-dev21:56
*** flaviamissi has quit IRC22:02
*** s0mik has joined #openstack-dev22:04
*** DiamRem has quit IRC22:04
*** ewindisch has quit IRC22:04
*** nati_ueno has joined #openstack-dev22:07
*** kbringard has quit IRC22:09
*** anderstj_ has quit IRC22:09
*** dtroyer_zzz is now known as dtroyer22:16
*** troytoman is now known as troytoman-away22:16
*** pmezard has quit IRC22:23
*** jtran has quit IRC22:24
*** littleidea has quit IRC22:29
*** openstackjenkins has quit IRC22:30
*** openstackjenkins has joined #openstack-dev22:31
*** asalkeld has quit IRC22:34
*** Adri2000 has quit IRC22:35
*** littleidea has joined #openstack-dev22:36
*** katell has quit IRC22:38
*** ewindisch has joined #openstack-dev22:39
*** Adri2000 has joined #openstack-dev22:43
*** Adri2000 has quit IRC22:43
*** Adri2000 has joined #openstack-dev22:43
*** datsun180b has quit IRC22:43
*** salgado has quit IRC22:43
*** spiffxp has quit IRC22:44
*** edygarcia has quit IRC22:44
*** dwalleck has joined #openstack-dev22:45
*** mattray has quit IRC22:45
*** littleidea has quit IRC22:53
*** littleidea has joined #openstack-dev22:56
*** dwalleck has left #openstack-dev22:57
*** GheRivero_ has quit IRC23:01
*** dolphm has joined #openstack-dev23:02
*** spiffxp has joined #openstack-dev23:05
*** agonella has joined #openstack-dev23:06
*** sdake has quit IRC23:11
*** eglynn has joined #openstack-dev23:11
*** rnirmal has quit IRC23:13
*** eglynn__ has quit IRC23:14
*** lloydde has quit IRC23:14
*** flaviamissi has joined #openstack-dev23:15
*** jakedahn is now known as jakedahn_zz23:15
*** Ryan_Lane has quit IRC23:16
*** danwent_ has joined #openstack-dev23:17
*** flaviamissi has quit IRC23:18
*** danwent has quit IRC23:18
*** danwent_ is now known as danwent23:18
*** flaviamissi has joined #openstack-dev23:18
*** sdake has joined #openstack-dev23:19
*** dolphm has quit IRC23:25
*** lloydde has joined #openstack-dev23:35
*** s0mik has quit IRC23:35
*** edygarcia has joined #openstack-dev23:36
*** sdake has quit IRC23:43
*** sdake has joined #openstack-dev23:43
*** Adri2000 has quit IRC23:45
*** edygarcia has quit IRC23:47
*** jakedahn_zz is now known as jakedahn23:49
*** lloydde has quit IRC23:49
*** lloydde has joined #openstack-dev23:51
openstackgerritVerification of a change to openstack/nova failed: Fix XenServer diagnostics to provide correct details  https://review.openstack.org/764323:51
*** jgriffith has quit IRC23:52
*** Adri2000 has joined #openstack-dev23:53
*** Adri2000 has quit IRC23:53
*** Adri2000 has joined #openstack-dev23:53
*** s0mik has joined #openstack-dev23:56
*** lloydde has quit IRC23:57
*** edygarcia has joined #openstack-dev23:59

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