Wednesday, 2012-06-20

*** kindaopsdevy has quit IRC00:00
jgriffithclayg: sorry... something else came up00:01
jgriffithclayg: Ok... so should be updated now00:01
*** danwent has joined #openstack-dev00:01
*** JStoker has quit IRC00:03
*** Mandell has quit IRC00:05
*** danwent has quit IRC00:07
*** andrewsmedina has joined #openstack-dev00:08
claygjgriffith: cool thanks for the email - i'll give that a try too00:08
jgriffithclayg: coolio00:09
jgriffithI'm headed out... be back later, or shoot me an email if you run into problems.00:09
*** adjohn has quit IRC00:10
jeblairGerrit should automatically upgrade to 2.4.1 in 15-20 minutes00:14
*** JStoker has joined #openstack-dev00:14
jeblair(This fixes the problem with missing emails)00:14
*** sleepsonzzz is now known as sleepsonthefloor00:14
bcwaldonjeblair: thanks for pushing back release window00:15
bcwaldonjeblair: whats typical downtime?00:16
jeblairbcwaldon: i did it just for you!00:16
bcwaldonjeblair: aww, how sweet00:16
jeblairbcwaldon: 15 seconds?00:16
bcwaldonjeblair: I still tried to look at it while it as down00:16
bcwaldoncurse you!00:16
adam_gdoes nova.tests.rpc.test_kombu:RpcKombuTestCase.test_iterconsume_errors_will_reconnect spin indefinitely for everyone else, or am i just a special case?00:17
jeblair(and sorry about the short notice, i managed to accidentally hit the earlier puppet run, but meant to hit the next one)00:18
bcwaldonjeblair: no worries00:18
jog0adam_g: I  saw that problem on one of the Jenkins tests one of my commints00:18
jeblairhttps://jenkins.openstack.org/job/gate-nova-python27/754/testReport/junit/nova.tests.rpc.test_kombu/RpcKombuTestCase/test_publishing_errors_will_reconnect/00:18
*** sandywalsh has joined #openstack-dev00:19
*** sandywalsh_ has quit IRC00:19
adam_gjog0: jeblair: thanks00:20
adam_gjeblair: actually im having problems with another test there, test_iterconsume_errors_will_reconnect. tested across 4 different systems (devstack included) hmm, test_publishing_errors_will_reconnect seems okay for me00:24
*** torgomatic has quit IRC00:25
*** andrewbogott has joined #openstack-dev00:28
*** andrewbogott has joined #openstack-dev00:28
*** sleepsonthefloor is now known as sleepsonzzz00:28
*** Aaton is now known as Aaton_off00:29
*** sleepsonzzz is now known as sleepsonthefloor00:29
*** sleepsonthefloor is now known as sleepsonzzz00:37
*** s0mik has quit IRC00:43
*** PotHix has quit IRC00:52
*** andrewbogott has quit IRC00:53
*** anderstj has joined #openstack-dev00:54
*** anderstj has quit IRC01:06
*** novas0x2a|laptop has quit IRC01:07
*** matwood has joined #openstack-dev01:10
*** rods has quit IRC01:13
*** bencherian has quit IRC01:14
*** bencherian has joined #openstack-dev01:17
*** dtroyer_zzz is now known as dtroyer01:18
*** adjohn has joined #openstack-dev01:19
*** rods has joined #openstack-dev01:19
*** littleidea has quit IRC01:19
*** winston-d has joined #openstack-dev01:21
*** littleidea has joined #openstack-dev01:21
winston-djgriffith, around?01:22
*** andrewbogott has joined #openstack-dev01:24
*** andrewbogott has joined #openstack-dev01:24
*** jog0 has left #openstack-dev01:26
*** datsun180b has quit IRC01:26
*** jog0_ has joined #openstack-dev01:27
*** johnpostlethwait has quit IRC01:29
*** bencherian has quit IRC01:31
*** jog0_ has quit IRC01:32
*** kmulvey has quit IRC01:32
*** rods has quit IRC01:33
*** edygarcia has joined #openstack-dev01:57
*** jdurgin has quit IRC02:05
*** littleidea has quit IRC02:17
*** cooper has quit IRC02:17
*** littleidea has joined #openstack-dev02:19
*** markmcclain has joined #openstack-dev02:20
*** adalbas has quit IRC02:20
*** littleidea has quit IRC02:25
*** s0mik has joined #openstack-dev02:26
*** littleidea has joined #openstack-dev02:26
*** zaitcev has quit IRC02:27
*** andrewsmedina has quit IRC02:27
*** agonella has left #openstack-dev02:28
winston-dsmoser, any chance still around?02:28
*** galstrom has joined #openstack-dev02:28
*** littleidea has joined #openstack-dev02:29
*** galstrom has quit IRC02:31
*** dtroyer is now known as dtroyer_zzz02:32
*** johnpur has quit IRC02:32
*** koolhead17|afk has quit IRC02:32
*** Mandell has joined #openstack-dev02:39
*** littleidea has quit IRC02:40
*** salgado-afk has quit IRC02:53
*** blamar has joined #openstack-dev02:57
*** blamar has joined #openstack-dev02:58
jgriffithwinston-d: what's up?02:59
jgriffithclayg: ahaha.... I see there's something wrong with the clone operation02:59
*** matwood has quit IRC03:05
*** andrewbogott has quit IRC03:08
winston-djgriffith, the euca test you mentioned, do you want me to test it against a original nova-vol?03:08
*** mnaser has joined #openstack-dev03:09
jgriffithwinston-d: Actually no need, I already checked it out, also got the reason for the slowness03:09
winston-dgood.03:09
*** s0mik has quit IRC03:10
*** littleidea has joined #openstack-dev03:10
jgriffith:) So vishy pointed out that it's most likely an issue because of the dd being done before delete status is updated03:10
*** anderstj has joined #openstack-dev03:10
jgriffithAt some point I want to background that and change it, but not on the top of my list right now03:10
winston-di see.  then I'll try out cinder + nova.03:11
jgriffithwinston-d: :)03:11
*** anderstj has quit IRC03:12
jgriffithwinston-d: excellent!  It's sort of in a funky state right now, but should work.03:12
jgriffithI promise it will be easier/more stable in the next day or two03:13
winston-dyou've been working so hard to make it happen.03:13
jgriffithwinston-d: It's not just me... sleepsonthefloor, anotherjesse, vishy, rnirmal....03:14
*** s0mik has joined #openstack-dev03:14
jgriffithwinston-d: I'm excited, it's going to be a good project once it's past this intial phase03:15
winston-dsure it will.  i hope i can help more in next couple weeks, once I finished my current job.03:16
*** edygarcia has quit IRC03:16
jgriffithwinston-d: Sure!  The fun stuff will start in a week or so :)  New features!!!03:17
winston-dnice!03:17
*** johnpur has joined #openstack-dev03:19
*** ChanServ sets mode: +v johnpur03:19
*** johnpur has quit IRC03:19
*** matwood has joined #openstack-dev03:20
*** harlowja has quit IRC03:21
hugokuogood morning ~03:37
hugokuowinston-d, long time no see lol03:37
*** dtroyer_zzz is now known as dtroyer03:39
*** nati_ueno has quit IRC03:40
*** sandywalsh has quit IRC03:46
*** ewindisch has quit IRC03:47
*** littleidea has quit IRC03:51
winston-dhugokuo, hi~03:55
winston-dhugokuo, time to have some lunch03:55
*** sleepsonzzz is now known as sleepsonthefloor03:56
*** anderstj has joined #openstack-dev04:08
*** markmcclain has quit IRC04:22
*** dolphm has joined #openstack-dev04:26
*** anderstj has quit IRC04:27
*** novas0x2a|laptop has joined #openstack-dev04:27
*** metral has joined #openstack-dev04:27
*** metral_ has joined #openstack-dev04:30
*** metral has quit IRC04:31
*** metral has joined #openstack-dev04:31
*** metral_ has quit IRC04:35
*** garyk has quit IRC04:38
*** dolphm has quit IRC04:53
*** Mandell_ has joined #openstack-dev04:55
*** Mandell has quit IRC04:55
*** shang has quit IRC04:56
*** s0mik has quit IRC05:00
*** steveb_ has quit IRC05:00
*** zigo has joined #openstack-dev05:02
*** andrewsben is now known as andrewsben_zz05:07
*** oneiroi has quit IRC05:10
*** trapni has quit IRC05:11
*** zykes- has quit IRC05:11
*** oneiroi has joined #openstack-dev05:12
*** trapni has joined #openstack-dev05:12
*** zykes- has joined #openstack-dev05:12
*** winston-d has quit IRC05:17
*** adjohn has quit IRC05:25
*** anderstj has joined #openstack-dev05:30
*** adjohn has joined #openstack-dev05:40
*** dtroyer is now known as dtroyer_zzz05:41
*** dachary has joined #openstack-dev05:53
*** erikzaadi has joined #openstack-dev05:56
*** adjohn has quit IRC05:57
*** adjohn has joined #openstack-dev05:57
*** garyk has joined #openstack-dev05:59
*** hattwick has quit IRC06:01
*** zigo has quit IRC06:02
*** adjohn has joined #openstack-dev06:05
*** JStoker has quit IRC06:09
*** JStoker has joined #openstack-dev06:12
*** Mandell_ has quit IRC06:13
*** blamar has quit IRC06:16
*** anderstj has quit IRC06:19
*** matwood has quit IRC06:24
*** mindpixel has joined #openstack-dev06:25
*** danwent has joined #openstack-dev06:26
*** erikzaadi has quit IRC06:28
*** adjohn has quit IRC06:49
*** dachary has quit IRC06:53
*** metral has quit IRC07:09
*** erikzaadi has joined #openstack-dev07:14
*** reidrac has joined #openstack-dev07:18
*** erikzaadi has quit IRC07:19
*** erikzaadi has joined #openstack-dev07:19
*** mnaser has quit IRC07:30
*** jakedahn is now known as jakedahn_zz07:32
vuntzttx: hey, just wondering, do you think it's worthwile to fix nova-rootwrap to search binaries in path also in essex, or do you feel we should just do it in folsom once we land your rootwrap.d changes?07:41
ttxvuntz: we don't push behavior changes in stable branches, so that should probably be folsom-only07:48
vuntzok, thanks07:49
*** sleepsonthefloor is now known as sleepsonzzz07:52
*** pixelbeat has joined #openstack-dev07:55
*** darraghb has joined #openstack-dev07:56
*** hattwick has joined #openstack-dev07:58
GheRiveromorning people07:59
garykgongys: ping08:01
*** dachary has joined #openstack-dev08:05
*** comstud has quit IRC08:13
*** comstud has joined #openstack-dev08:13
hugokuodoes anyone know about how to export bugs list/contents from launchpad ....?08:16
*** apevec has joined #openstack-dev08:21
*** zigo has joined #openstack-dev08:24
*** jgruber has joined #openstack-dev08:30
*** popux has joined #openstack-dev08:32
*** danwent_ has joined #openstack-dev08:33
*** danwent has quit IRC08:33
*** danwent_ is now known as danwent08:33
trapniHey. nova-novncproxy needs just to be installed on the controller node (not on each compute node), right? it seems I can only connect to VNCs on instances that is also hosting the controller node services.08:33
mikalDoes anyone here have experience with adding foreign keys in migrations they want to share with me?08:34
*** trapni has quit IRC08:34
*** trapni has joined #openstack-dev08:34
*** jgruber has quit IRC08:37
*** sulochan has joined #openstack-dev08:41
*** danpb has joined #openstack-dev08:56
*** dachary has quit IRC09:14
*** danwent has quit IRC09:15
*** Ryan_Lane has quit IRC09:25
*** danwent has joined #openstack-dev09:29
*** danwent_ has joined #openstack-dev09:40
*** danwent has quit IRC09:40
*** danwent_ is now known as danwent09:40
*** popux has quit IRC09:41
*** m4xmr has joined #openstack-dev09:55
*** matwood has joined #openstack-dev09:59
*** GheRivero has quit IRC10:13
*** GheRivero has joined #openstack-dev10:13
*** ywu has quit IRC10:19
*** Ryan_Lane has joined #openstack-dev10:22
*** zigo has quit IRC10:30
*** littleidea has joined #openstack-dev10:48
*** popux has joined #openstack-dev10:50
*** camm has quit IRC10:50
*** camm has joined #openstack-dev10:51
*** danwent has quit IRC11:02
*** m4xmr has quit IRC11:14
*** vanchester12416 has joined #openstack-dev11:14
*** wiliam has joined #openstack-dev11:21
*** zul has quit IRC11:23
sorendanpb: Do you have ip set support in the Fedora kernels?11:27
*** shang has joined #openstack-dev11:28
danpbsoren: not sure what you mean by that ?11:28
danpbis there a kconfig variable i can check for you ?11:28
sorendanpb: CONFIG_IP_SET11:28
*** jgruber has joined #openstack-dev11:28
danpbin fact here is a config from current F17 kernels http://fpaste.org/OUmW/11:29
danpbyep, that's there as a module11:29
soren\o/11:29
sorenAwesome, thanks.11:29
sorendanpb: IP sets are a construct that allows you to create an ip set in the kernel that you can reference from iptables. There are various kinds for various needs, but the basic idea is to avoid having to traverse through 1000 rules in iptables and instead doing a quick hash lookup in an ip set and have a single rule referencing the ipset in iptables.11:30
danpbyeah makes sense11:31
sorenThey weren't mainstream back when I first implemented security groups in nova, but I think the time is ripe to make it more efficient this way.11:31
danpbin fact now that you mention it, i see Stefan Berger added this ipset  capability to libvirt's nwfilter code11:32
sorenWicked.11:32
danpbi was actually wondering what scenarios we still need the nova  iptables code for ?11:33
danpbif nova's min libvirt version is sufficient, we could just rely on the nwfilter code11:33
*** zul has joined #openstack-dev11:33
sorendanpb: Has the problems with nwfilter been fixed now? I.e. not being able to filter.. err.. incoming packages, but only outgoing ones?11:34
* soren looks for the bug reference11:34
danpbsoren: i don't know - you're the first person to mention any problems11:34
sorenhttps://bugzilla.redhat.com/show_bug.cgi?id=64217111:34
uvirtbotsoren: Error: Could not parse XML returned by bugzilla.redhat.com: HTTP Error 404: Not Found11:34
* soren glances at uvirtbot11:35
danpbuvirtbot: lamo11:35
uvirtbotdanpb: Error: "lamo" is not a valid command.11:35
sorenSilly thing.11:35
*** shang has quit IRC11:35
sorenIt's because of that bug we stopped using nwfilter by default.11:36
sorenI don't think anyone has kept it up to date. There hasn't been much point, seeing as it couldn't actually filter things properly anyway.11:36
sorenBut Stefan suggests it might be fixed now.11:36
sorenThat would be great.11:36
danpbyeah, i guess we have to keep both around for a while, since i doubt the fix is in 0.9.7 versions11:39
*** ewindisch has joined #openstack-dev11:40
sorendanpb: Oh, I don't think this was fixed in libvirt. I think it was fixed in the kernel.11:49
*** alaski has joined #openstack-dev11:50
*** ewindisch has quit IRC11:53
*** zigo has joined #openstack-dev11:55
*** m4xmr has joined #openstack-dev11:57
*** littleidea has quit IRC12:00
*** agonella has joined #openstack-dev12:01
*** apevec has quit IRC12:02
*** popux has quit IRC12:07
*** dolphm has joined #openstack-dev12:11
*** salgado has joined #openstack-dev12:11
*** zul has quit IRC12:16
*** drewlander has joined #openstack-dev12:18
*** markmc has joined #openstack-dev12:20
*** eglynn__ has quit IRC12:22
*** danwent has joined #openstack-dev12:25
*** eglynn__ has joined #openstack-dev12:25
*** dtroyer_zzz is now known as dtroyer12:27
*** dachary has joined #openstack-dev12:34
*** mikal has quit IRC12:35
*** dachary has quit IRC12:35
*** dachary has joined #openstack-dev12:36
*** dolphm has quit IRC12:36
*** danwent has quit IRC12:36
*** danwent has joined #openstack-dev12:37
*** mikal has joined #openstack-dev12:37
*** markvoelker has joined #openstack-dev12:37
agonellaHi, Im running a compute-controller + compute+node, and I already launched one instance based on linux iso. When I tried to launch a second instance, it can not start, it says that instance status is ¨error¨and task status ¨scheduling¨. I m running over QEMU. Any idea?12:43
*** zigo has quit IRC12:46
*** vanchester12416 has quit IRC12:46
*** m4xmr has quit IRC12:52
*** andrewsmedina has joined #openstack-dev12:53
*** vanchester has joined #openstack-dev12:54
*** m4xmr has joined #openstack-dev12:56
*** sandywalsh has joined #openstack-dev12:56
*** sulochan has quit IRC13:01
*** mindpixel has quit IRC13:01
*** andrewsmedina has quit IRC13:03
*** iccha has joined #openstack-dev13:03
*** mindpixel has joined #openstack-dev13:03
*** sulochan has joined #openstack-dev13:05
*** sulochan_ has joined #openstack-dev13:05
*** adalbas has joined #openstack-dev13:07
*** markmcclain has joined #openstack-dev13:07
*** dachary has quit IRC13:08
*** sulochan has quit IRC13:09
*** sulochan_ is now known as sulochan13:09
*** mindpixel has quit IRC13:10
*** joesavak has joined #openstack-dev13:12
*** dachary has joined #openstack-dev13:12
*** sandywalsh_ has joined #openstack-dev13:13
*** sandywalsh has quit IRC13:13
*** rods has joined #openstack-dev13:14
*** stuntmachine has joined #openstack-dev13:15
*** roge has joined #openstack-dev13:16
*** zul has joined #openstack-dev13:20
*** lts has joined #openstack-dev13:25
*** sandywalsh_ has quit IRC13:27
*** vanchester has quit IRC13:29
*** markmcclain has quit IRC13:31
*** markmcclain has joined #openstack-dev13:31
*** andrewsmedina has joined #openstack-dev13:31
*** sandywalsh has joined #openstack-dev13:31
*** edygarcia has joined #openstack-dev13:32
*** japage has joined #openstack-dev13:32
*** dachary has quit IRC13:35
*** davidkranz has quit IRC13:35
*** mnaser has joined #openstack-dev13:36
*** dtroyer is now known as dtroyer_zzz13:37
*** ewindisch has joined #openstack-dev13:38
*** davidkranz has joined #openstack-dev13:39
garykdanwent: ping13:41
danwentgaryk: hello13:42
danwentconfig patch?13:42
garykhi, do you have a few minutes to discuss ip address allocations13:42
*** dubsquared has joined #openstack-dev13:42
danwentsure13:43
*** dolphm has joined #openstack-dev13:43
*** dubsquared has quit IRC13:43
danwentoh good, looks like main config comon changes got in?13:43
*** dolphm_ has joined #openstack-dev13:43
*** davidkranz has quit IRC13:44
*** davidkranz has joined #openstack-dev13:44
*** shang has joined #openstack-dev13:45
*** shang has quit IRC13:47
*** shang has joined #openstack-dev13:48
*** dprince has joined #openstack-dev13:48
*** dolphm has quit IRC13:48
*** rbasak has quit IRC13:49
*** dtroyer_zzz is now known as dtroyer13:51
*** dubsquared has joined #openstack-dev13:52
*** apevec has joined #openstack-dev14:05
*** Gordonz has joined #openstack-dev14:06
*** uksysadmin has joined #openstack-dev14:07
*** uksysadmin has left #openstack-dev14:08
sstentanyone know if there is away to get pat the current devstack lesscss bug?  https://bugs.launchpad.net/devstack/+bug/101473514:08
uvirtbotLaunchpad bug 1014735 in devstack "Horizon now needs LESS installed..." [High,Confirmed]14:08
*** utlemming has quit IRC14:12
*** zigo has joined #openstack-dev14:13
*** utlemming has joined #openstack-dev14:13
*** Mandell has joined #openstack-dev14:14
*** segfault923 has quit IRC14:15
*** Gordonz has quit IRC14:20
*** Gordonz has joined #openstack-dev14:20
*** ccorrigan has quit IRC14:23
*** dachary has joined #openstack-dev14:27
*** chuck_ has joined #openstack-dev14:28
*** zul has quit IRC14:29
*** chuck_ has quit IRC14:29
*** AlanClark has joined #openstack-dev14:30
*** maoy has joined #openstack-dev14:31
trapnihow do I recover a properly running VM from its fixed "REBOOT" state w/o breaking anything? (I do not want to terminate the instance, just because `nova reboot $N` failed - I forced a (re)boot via virsh and it works now. but the nova state says REBOOT still.14:32
*** drewlander has quit IRC14:32
*** rbasak has joined #openstack-dev14:32
*** AlanClark has quit IRC14:33
trapnihttp://paste.openstack.org/show/18651/ is the trace I found in its nova-compute.log14:33
*** zul has joined #openstack-dev14:35
*** shang has quit IRC14:41
*** nunosantos has joined #openstack-dev14:42
*** shang has joined #openstack-dev14:45
*** segfault923 has joined #openstack-dev14:47
*** davidkranz_ has joined #openstack-dev14:49
*** davidkranz has quit IRC14:49
*** markmc has quit IRC14:50
*** littleidea has joined #openstack-dev14:52
*** sulochan has quit IRC14:52
*** blamar has joined #openstack-dev14:53
*** datsun180b has joined #openstack-dev14:56
*** sleepsonzzz is now known as sleepsonthefloor14:58
*** bencherian has joined #openstack-dev14:58
maoytrapni: I'm not aware of anything you can do to reset the state unless you modify the db for the moment. but i'd like to see patches that allow you to stop during reboot state14:58
*** sstent has quit IRC15:00
*** littleidea has quit IRC15:00
*** sstent has joined #openstack-dev15:01
*** reidrac has quit IRC15:01
notmynamettx: who has permissions to update blueprint details?15:04
*** dubsquared has quit IRC15:04
*** rnirmal has joined #openstack-dev15:05
trapnimaoy: hey. thanks. do you know that it's safe to set this flag in the db (i.e. did you do it yourself already)? or is there anything else I should do in addition?15:06
garykdanwent:ping. have a question regarding the fixed ips.15:06
maoytrapni: it's safe. just make sure you update both vm_state and task_state15:07
*** anderstj has joined #openstack-dev15:07
*** sulochan has joined #openstack-dev15:08
maoytrapni: it's a hack.. but should work. if we allow you to stop the VM in reboot state, does that make you happier?15:08
*** anderstj has quit IRC15:08
*** blamar has quit IRC15:09
trapnimaoy: ah. yeah. task_state to NULL (instead of "rebooting", ya)15:11
trapnimaoy: vm_state is "active" in any case, right?15:11
*** mattray has joined #openstack-dev15:11
maoytrapni: yes15:12
trapniyay. worked (so far) :-)15:12
trapnithanks ;)15:12
*** dachary has quit IRC15:12
*** galstrom has joined #openstack-dev15:13
trapnimaoy: do you maybe also know what "Invalid Token" means (said by nova-novncproxy when the web browser connects and fails)15:13
maoytrapni: no idea.. :)15:13
*** kbringard has joined #openstack-dev15:13
trapnitoo bad15:13
*** dachary has joined #openstack-dev15:17
eglynn__bcwaldon: if you've time to review https://review.openstack.org/8701 & https://review.openstack.org/8722 ... it would be great to get them in & start getting some data on the glance test failures15:19
*** shang has quit IRC15:19
*** maoy has quit IRC15:19
eglynn__(strace won't be so useful in Jenkins pending https://lists.launchpad.net/openstack/msg13381.html but should get something from the Fedora workers running smokestack)15:19
*** dachary has quit IRC15:21
*** dachary has joined #openstack-dev15:21
*** Mandell has quit IRC15:21
*** dachary has quit IRC15:21
*** andrewbogott has joined #openstack-dev15:22
ttxnotmyname: drivers, assignee, drafter...15:22
notmynamettx: but not core devs?15:22
ttxnotmyname: no. Launchpad just doesn't give special rights to core devs. It's a Gerrit thing.15:22
notmynameok15:22
notmynamettx: so if I make someone the assignee, then they can edit stuff on it?15:23
ttx(well, core devs are usually also bug supervisors)15:23
*** dachary has joined #openstack-dev15:23
ttxnotmyname: yes15:23
notmynamettx: cool, thanks15:23
ttx(or drafter)15:23
*** blamar has joined #openstack-dev15:23
*** heckj has joined #openstack-dev15:25
*** dachary has quit IRC15:27
*** maoy has joined #openstack-dev15:30
*** dubsquared has joined #openstack-dev15:30
*** shang has joined #openstack-dev15:36
*** shang has quit IRC15:40
bcwaldoneglynn__: what changed in the latest two patchsets of 870115:47
eglynn__bcwaldon: just a rebase15:47
bcwaldoneglynn__: kk15:47
bcwaldoneglynn__: ok, so with this patch that captures logs, what do I have to do differently now?15:49
bcwaldoneglynn__: or what *can* I now do?15:49
*** shang has joined #openstack-dev15:50
eglynn__bcwaldon: just 'export NOSE_GLANCELOGCAPTURE=true' in your environment (same as before, difference is now that the log is captured even if the tests run via tox)15:51
*** shang has quit IRC15:51
bcwaldonok, and where does the output go?15:51
bcwaldoneglynn__: sorry, I never used it before :/15:51
eglynn__bcwaldon: now its embedded in the failing assertion msg15:51
bcwaldonaha15:51
*** zul has quit IRC15:52
eglynn__bcwaldon: this is a more limited use-case than before, i.e. capture logs if the test service fails to launch as opposed to on any type of test failure15:53
*** dubsquared has quit IRC15:53
eglynn__bcwaldon: the old logcapture still works for any failure if run via run_tests.sh15:53
bcwaldoneglynn__: ok, we need to get to tox15:53
bcwaldoneglynn__: I want run_tests.sh to go away15:53
andrewbogottI could use some reviews from anyone who cares about openstack-common and/or logging and/or notifications.  Any volunteers?15:54
eglynn__bcwaldon: yep, so this is a first step on that transition (Jenkins switching from run_tests.sh to nosetests broke the old lopcapture)15:54
eglynn__bcwaldon: this will enable log capture in the particular failure case that I want to track down (the intermittent 'Unexpected service launch status' ...)15:55
bcwaldoneglynn__: yep, you've got my +215:55
bcwaldoneglynn__: markwash is looking at it, too15:55
bcwaldoneglynn__: I just posted feedback on your dependent branch15:55
eglynn__bcwaldon: cool, thanks! ... loking now15:56
*** littleidea has joined #openstack-dev15:58
*** agonella has left #openstack-dev15:59
*** novas0x2a|laptop has quit IRC16:00
*** littleidea has quit IRC16:01
ijwI've just had a weird devstack install failure with http_proxy set.  Is it possible that the keystone CLI client respects http_proxy and none of the others do?16:02
*** novas0x2a|laptop has joined #openstack-dev16:02
*** apevec has quit IRC16:04
eglynn__bcwaldon: #8722 updated16:05
*** dubsquared has joined #openstack-dev16:07
bcwaldoneglynn__: +2'd16:07
eglynn__thx!16:07
bcwaldoneglynn__: now if only markwash would look at it16:07
bcwaldoneglynn__: and I've got this one thats been sitting for a while: https://review.openstack.org/#/c/8507/16:08
eglynn__bcwladon: looking now ...16:08
bcwaldoneglynn__: and does this conflict with the work you're doing: https://review.openstack.org/#/c/8704/16:08
*** lloydde has joined #openstack-dev16:09
*** bencherian has quit IRC16:09
eglynn__bcwaldon: #8704 shouldn't conflict, monty's patch is more about controlling whether the tmp test dir is cleaned up or retained (as opposed to embedding the log content in the results)16:11
bcwaldoneglynn__: ok, as you can tell I didnt look at the actual change16:12
bcwaldoneglynn__: but the commit message sounded similar16:12
eglynn__bcwaldon: just keeping the tmp dir around isn't that useful for an ordinary developer without ssh access to the Jenkins slave16:12
jeblaireglynn__: is monty's change no longer necessary if yours goes in?16:12
bcwaldoneglynn__: well, it would be if ou werent capturing logs and dumping them to stdout16:13
bcwaldonjaypipes: around?16:13
mtaylorttx: what change?16:13
mtaylorgah16:13
mtaylorsorry ttx16:13
jaypipesbcwaldon: for you? any time.16:13
mtaylorbcwaldon: what change?16:13
jeblair(i think the end-state for this should be that the appropriate log info is captured by the nose logcapture plugin, and it sounds like that's what eglynn__ is doing)16:14
eglynn__jeblair: well mtaylor's patch would be useful for the CI team in tracking down test failures16:14
bcwaldonjaypipes: awww16:14
bcwaldonjaypipes: https://review.openstack.org/#/c/8713/16:14
bcwaldonjaypipes: super easy16:14
jeblaireglynn__: i don't want to track down failures, i want other people to be able to.  :)16:14
jaypipesbcwaldon: on it.16:14
bcwaldonjaypipes: and I sent you an email about another review16:14
eglynn__jeblair: ;) yep, previously the glance logcapture plugin would dump logs to stdout for failing tests, but this was broken by the switch from run_tests.sh to nosetests16:15
jaypipesjeblair: but that won't be compatible with my new get-jeblair-to-fix-it nose plugin.16:15
bcwaldonmtaylor: https://review.openstack.org/#/c/8746/16:15
*** sstent has quit IRC16:15
jeblairjaypipes: man, if you think things are non-deterministic now...  :)16:15
mtaylorah. I forgot that glance used to have a logcapture plugin16:16
jaypipeslol!16:16
mtaylorbcwaldon: why do we want to revert that? where is the patch that fixes it better? (I'm fine - I'm just curious)16:17
*** garyk has quit IRC16:17
bcwaldonmtaylor: ask markwash16:17
mtaylorgreat.16:18
mtaylormarkwash: what did I break?16:18
bcwaldonmtaylor: watch your tone with me16:18
* mtaylor watches his tone at bcwaldon16:18
bcwaldonmtaylor: he and ameade are seeing disgusting output16:18
bcwaldonmtaylor: and I havent run tests :/16:18
jaypipesbcwaldon: done.16:19
bcwaldonjaypipes: thank you, sir16:19
jaypipesnp16:19
mtaylorbcwaldon: well, it _should_ be being captured by nose, but I'll take their word for it16:19
bcwaldonmtaylor: I'm running tests to see what happens16:19
mtaylorgreat. I will do the same16:19
bcwaldonmtaylor: I can already see the initial db migrations are dumped16:19
bcwaldonmtaylor: if thats it, its not a big deal16:20
mtaylorwell, that's ugly, but probably something that can be fixed16:20
bcwaldonmtaylor: sure, I didnt +2 his revert because I want to figure out whats up first16:21
bcwaldonI already talked to him about it16:21
mtaylorcool. I like investigating things16:21
bcwaldonmtaylor: let's CSI this up16:21
*** nati_ueno has joined #openstack-dev16:21
bcwaldonmtaylor: also, can you drop the run times from SKIPped tests?16:21
bcwaldonmtaylor: holy crap16:22
bcwaldonmtaylor: markwash is not lying16:22
mtaylorbcwaldon: I will look in to it.16:22
mtaylorbcwaldon: yeah?16:22
*** s0mik has joined #openstack-dev16:22
bcwaldonmtaylor: http://stsh.me/2NL16:22
bcwaldonmtaylor: its not every test16:22
mtaylorhrm. weird.16:23
bcwaldonmtaylor: that screenshot is from the v2 tests that use the requests lib16:23
mtaylorI wonder what's different about them. does the requests lib set up logging differently?16:23
jeblaireglynn__, mtaylor: i'm working on a patch to puppet to flip the ptrace flag16:25
mtaylorjeblair: aroo?16:25
eglynn__jeblair: sweet, thanks!16:25
jeblairmtaylor: mailing list16:25
mtaylorthanks16:25
mtaylorbcwaldon: DEBUG Attempting to import store glance.store.filesystem.Store16:27
bcwaldonmtaylor: what about it16:27
mtaylorbcwaldon: that looks like our message - any ideas why that one might be outputting differently (so as to escape the logcapture?)16:27
bcwaldonmtaylor: also, did you impersonate Richard Nixon's head up there?16:27
mtayloryes16:28
bcwaldonmtaylor: bravo16:28
*** zul has joined #openstack-dev16:28
*** metral has joined #openstack-dev16:29
bcwaldonmtaylor: I don't have enough info to tell you why things are happening wrt logcapture16:29
mtaylork16:30
mtaylorI think I've got an easy test case to poke at16:30
*** reed has quit IRC16:31
bcwaldonmtaylor: did you get my request about SKIPped tests up there?16:33
*** salgado is now known as salgado-lunch16:33
*** danwent has quit IRC16:35
*** danwent has joined #openstack-dev16:35
mtaylorbcwaldon: yup16:36
*** anderstj has joined #openstack-dev16:37
*** somik has joined #openstack-dev16:39
*** danwent has quit IRC16:39
*** danwent has joined #openstack-dev16:39
mtaylorbcwaldon: would you try something for me?16:39
*** s0mik has quit IRC16:40
*** somik is now known as s0mik16:40
bcwaldonmtaylor: yes16:42
mtaylorbcwaldon: nm. false lead16:42
mtaylorbcwaldon: BUT - I think I might have a real fix16:43
bcwaldonmtaylor: twould be great16:43
jeblaireglynn__: https://review.openstack.org/#/c/8751/16:44
jeblairmtaylor: ^16:44
bcwaldonmtaylor: in the meantime, can you review this https://review.openstack.org/#/c/8746/16:44
mtaylorbcwaldon: what should I review about it?16:45
bcwaldonmtaylor: either +2 or -216:46
bcwaldonmtaylor: I dont want it to land if you have something else16:46
*** garyk has joined #openstack-dev16:47
mtaylorbcwaldon: I will have something withing the next half hour16:47
*** e1mer has quit IRC16:48
*** littleidea has joined #openstack-dev16:48
*** littleidea_ has joined #openstack-dev16:48
markwashmarkwash vindicated once!16:50
markwashtake that, universe!16:50
markwashbut really, I'm sorry I didn't poke the change more before I +2'd16:50
*** littleidea has quit IRC16:52
*** littleidea_ is now known as littleidea16:52
*** pixelbeat has quit IRC16:53
*** jog0 has joined #openstack-dev16:53
mtaylormarkwash, bcwaldon: would you test: https://review.openstack.org/875416:53
mtaylorit works locally for me16:54
mtaylorhowever ... you might want to also do this:16:54
mtaylorhttp://paste.openstack.org/show/18661/16:54
mtaylorto verify that debug logging is captured and output properly during a failing test16:55
*** joesavak has quit IRC16:55
*** PotHix has joined #openstack-dev16:55
mtaylor(that test does have debug logging that happens during the test)16:55
mtaylormarkwash: well, _I_ should have apparently tested that patch better before I submitted it16:56
mtaylormarkwash: it worked in context I tested in , but not for full test runs. Sigh.16:56
bcwaldonmtaylor: tsk tsk16:56
markwashthat time it took three to tango16:56
mtaylormmm. three person tango16:57
bcwaldonuh16:57
bcwaldonguys16:57
mtayloryeah?16:57
bcwaldonlets, uh, lets not do that again16:57
mtaylorhehehe16:57
mtaylorbcwaldon: dude, you live in SF now, not virginia. three-person-all-male-tangos are the new black16:58
markwashberkeley is similar, except its required that you are also high16:58
bcwaldonmtaylor: I did just move to berkeley16:58
bcwaldonmtaylor: jeblair and I tango all the time16:58
*** littleidea has quit IRC16:58
mtaylorbcwaldon: oh, berkeley. you can't do that in berkeley. you are only allowed to grow long grey beards and not bathe there16:59
bcwaldonon it16:59
mtaylorpoint taken, I'll keep you out of it16:59
markwashmtaylor: so, that branch cleans up tests, but then on a failing test I see the same output duplicated in both "captured logging" and "captured stdout"17:00
markwashso I'm wondering, do we even need a logger at all?17:00
markwasherr17:00
markwasha "handler"17:00
bcwaldonmarkwash: we would need the handler for unit tests17:00
bcwaldonmarkwash: where we dont spin up servers17:01
bcwaldonmarkwash: and functional tests should only be giving output through logs17:01
markwashbcwaldon: I'm getting the dup'd output on a failing unittest17:01
bcwaldonmarkwash: I'm not telling you how it is17:01
bcwaldonmarkwash: thats how it *should* be17:01
mtaylormarkwash: oh! so it is. FASCINATED17:02
markwashbcwaldon: it seems like nose is creating its own handler17:02
bcwaldonmarkwash: I bet it is17:02
mtaylorI think it is - although when I don't register the sys.stdout handler the nose one doens't seem to work17:02
mtaylorperhaps I should register a null handler?17:03
bcwaldonmtaylor: is there a null handler?17:03
bcwaldonmtaylor: just dont registrr a handler at all17:03
*** torgomatic has joined #openstack-dev17:03
*** GheRivero_ has joined #openstack-dev17:03
mtaylorbcwaldon: AHA! That works in the new patch17:04
mtaylorit didn't work before17:04
markwashbcwaldon: yeah, I just removed all the logging stuff from glance/tests/__init__.py and I get the log output on a failing (unit) test17:04
mtaylorin the old patch17:04
bcwaldonmarkwash: is that what you expect?17:04
zulim looking at the iscsi.py is there a reason why ietadm is still the default i think most modern linux distros, ubuntu, fedora, rhel, debian use tgtadm by default17:04
markwashbcwaldon: yeah, that's what i want I think17:04
markwashlogs are in the "captured logging" section17:05
mtaylorhrm17:05
markwashstdout is in the "captured stdout" section17:05
mtaylorok. me stands VERY corrected17:05
markwashthis is with ./run_test.sh17:05
markwashI haven't tried out tox, I know that's a factor here17:05
mtaylorI thought that we needed the extra bits to get the logging captured, and for some reason it wasn't working for me without it17:05
mtaylorbut let's just say that I must have tested it right17:05
mtaylormarkwash: Ijust verified under tox17:05
markwashcool17:06
*** andrewbogott has quit IRC17:06
markwashso it sounds like I can abandon my revert patch17:06
mtaylorlemme do a full test run real quick - I only verified single test run17:06
markwashand we can put something in to just *not* add the handler?17:07
mtaylorno - it sounds like your revert patch is the right thing to do17:07
markwashoh?17:07
mtayloryeah. your revert patch essentially just removes setting up the logged17:07
mtaylorlogger17:07
*** danwent has quit IRC17:08
mtaylorthe line in the patch about the STDOUT in tox.ini is not all that important17:08
*** metral_ has joined #openstack-dev17:08
markwashmtaylor: I'd actually like to remove the run_tests.log logging as well17:08
markwashso maybe that is just a separate concern17:08
markwashI've had times in the past where some error with run tests would just get buried17:09
markwashbut hmm17:09
markwashmaybe that was being buried somewhere else17:09
*** metral has quit IRC17:09
*** metral_ is now known as metral17:09
*** andrewbogott has joined #openstack-dev17:09
markwashhmm17:09
markwashokay I'm on board17:12
*** koolhead17|afk has joined #openstack-dev17:14
*** jakedahn_zz is now known as jakedahn17:15
*** dachary has joined #openstack-dev17:15
*** darraghb has quit IRC17:17
*** sulochan has quit IRC17:17
*** dachary has quit IRC17:18
mtaylormarkwash: cool. I'm going to un--2 your patch17:19
markwashmtaylor: kk17:19
*** koolhead17|afk is now known as koolhead1717:22
bcwaldonyou two just tell me what to approve17:22
mtaylorbcwaldon: already done17:23
bcwaldonkk17:23
mtaylorbcwaldon: and I think I've got a patch for your SKIP no times request17:23
bcwaldonawesome17:24
*** salgado-lunch is now known as salgado17:24
mtaylorbcwaldon: https://review.openstack.org/875817:24
mtaylorjkoelker: ^^17:24
*** rohitk has joined #openstack-dev17:26
*** harlowja has joined #openstack-dev17:27
mtaylorbcwaldon: so did eglynn__ have something which conflicted with this: https://review.openstack.org/#/c/8704/ ?17:28
bcwaldonmtaylor: well, I think its more of an alternate approach17:29
mtaylorbcwaldon: ok. so I should go read his thing17:29
bcwaldonmtaylor: he said that your patch makes it so the log files are kept around for collection17:29
bcwaldonmtaylor: but his dumps logs to stdout17:30
bcwaldonmtaylor: so yes17:30
mtaylorbcwaldon: does his dump logs to stdout so that they just get picked up by nose?17:30
bcwaldonmtaylor: idk17:30
mtaylork.17:30
mtaylorneat. well, his is merged, so I'm just going to abandon mine for now17:31
*** rohitk_recon has joined #openstack-dev17:32
*** rohitk has quit IRC17:34
arosenHi garyk you around?17:34
*** Ryan_Lane has quit IRC17:35
*** danpb has quit IRC17:37
eglynn__mtaylor: my original glance logcapture plugin dumped the logs for failing tests into the nose output ...17:38
eglynn__mtaylor: but the alternate approach (#8701) simply embeds the log content for a service that fails to launch properly into the failed assertion message17:39
markwasheglynn__: the change in question focuses on functional tests, correct?17:39
*** rohitk_recon has quit IRC17:39
eglynn__markwash: yep, on the sepecific case that the api or registry server fails to launch in a functional test17:40
eglynn__markwash: i.e. the intermittent unkown issue that's been causing jenkins & smokestack to fail regularly17:40
eglynn__gotta run, bbiab ...17:41
*** rohitk has joined #openstack-dev17:42
*** shang has joined #openstack-dev17:42
*** mszilagyi has joined #openstack-dev17:45
*** reed has joined #openstack-dev17:46
*** zul has quit IRC17:46
*** Mandell has joined #openstack-dev17:49
*** jakedahn is now known as jakedahn_zz17:50
*** zul has joined #openstack-dev17:52
*** GheRivero_ has quit IRC17:53
*** matwood has quit IRC17:53
*** sstent has joined #openstack-dev17:55
*** matwood has joined #openstack-dev17:55
sstentFor the node-less stuff in horizon, where exactly should it be installed?17:55
cloudflywherever you want that has a path to the api server?17:56
sstentcloudfly: I isntalled to /usr/lib/node_modules/less/bin/ but dahsboard seems to be looking in /usr/lib/python2.6/site-packages/17:57
cloudflyah path17:58
sstentcloudfly: as in add, /usr/lib/node_modules/less/bin/ to my apache path?17:59
*** novas0x2a|laptop has quit IRC18:02
*** Dr_Who has joined #openstack-dev18:03
*** Dr_Who has quit IRC18:03
*** Dr_Who has joined #openstack-dev18:03
*** m4xmr has quit IRC18:03
*** Dr_Who has quit IRC18:04
markwasheglynn__: glad it got in, sorry I was going to approve it but then lunch came up and jk0 beat me to it :-)18:04
*** erikzaadi has quit IRC18:04
eglynn__markwash: cool, thanks18:05
jeblaireglynn__: your ptrace scope change is active on the jenkins slaves18:05
eglynn__jeblair: excellent ... now we just need a few failures, shouldn't be long ...18:06
mtaylorcan I get someone to look at https://review.openstack.org/#/c/8742/ for python-keystoneclient and the corresponding https://review.openstack.org/#/c/8761/ for openstack-common?18:06
mtaylorheckj, jkoelker: ^^ ?18:06
* heckj looks18:07
*** shang has quit IRC18:07
*** shang has joined #openstack-dev18:08
mtaylorjaypipes: you can review openstack-common, can't you?18:08
*** torgomatic has quit IRC18:08
heckjmtaylor: I don't have approve on openstack-common, but I gave you a +1 there18:08
heckj+2'd in keystoneclient18:08
*** torgomatic has joined #openstack-dev18:09
jaypipesmtaylor: I believe so. link?18:09
*** dhellmann has quit IRC18:10
mtaylorjaypipes: https://review.openstack.org/#/c/8761/ and https://review.openstack.org/#/c/8760/ and https://review.openstack.org/#/c/8284/ (in decreasing order of current importance)18:10
*** dhellmann has joined #openstack-dev18:11
jaypipesmtaylor: will try to get to that shortly.18:11
mtaylorjaypipes: you are a magical pony18:11
mtaylorjaypipes: just be careful about the rhinestones18:11
garykarosen: you there?18:11
jgriffithjaypipes: is a unicorn?18:11
mtaylorjgriffith: is he not?18:12
jgriffithmtaylor: I believe you may be correct18:12
jaypipesjgriffith: a centaur.18:12
jaypipesjgriffith: bejazzled centaur.18:12
jgriffithjaypipes: Now we're talking18:13
mtaylorjaypipes: vajazzled?18:13
arosenhey garyk i hadded a comment on https://review.openstack.org/#/c/817618:13
jgriffithmtaylor: I'm out18:13
jaypipesmtaylor: nope. just bejazzled.18:13
jgriffithmtaylor: ya just hads to go there18:13
mtaylorjaypipes: k. just making sure that there wasn't a speech impediment thing going on ... v is just an unvoiced version of b you know18:14
mtaylorjgriffith: I was checking the linguistics, I don't know where your mind went :)18:14
jgriffithmtaylor: DOH!18:14
*** metral has quit IRC18:14
*** metral has joined #openstack-dev18:14
* mtaylor lied. v is actually voiced f, p is unvoiced b. hopes no one else in the channel studied diction for performance in school...18:15
jaypipeslol18:15
garykarosen: i saw it. you are correct. the quantum change that went through this morning does away with the plugins.ini file. so i think that it should be ok.18:15
*** dubsquared has quit IRC18:19
arosengaryk: I still think that should still be changed. Otherwise I think we should remove:       sudo sed -i -e "s/^provider =.*$/provider = $Q_PLUGIN_CLASS/g" $Q_PLUGIN_INI_FILE18:20
cloudflyby the by I would love some input on this:  https://lists.launchpad.net/openstack/msg13420.html18:21
garykarosen: i agree with you. but please note that from the quantum version from this morning the plugins.ini file no longer exists. the next quantum change to devstack should remove this code.18:23
*** dubsquared has joined #openstack-dev18:24
garykarosen: i understand that you are making changes at the moment. i'll be happy to review and test when you are ready18:24
arosengaryk:  the only change I have up right now is https://review.openstack.org/#/c/8726/ but I'm going to add one that moves policy.json  to /etc/quantum/ shortly. (This only needs to be done for nodes with q-svc enabled right?)18:26
*** rohitk has quit IRC18:26
arosenthanks18:26
garykarosen: yes, that is correct. i'll add in a few other comments. thanks!18:28
*** jog0 has quit IRC18:29
arosennp, excited to be getting my feet wet in quantum :)18:29
cloudflyi'd love to see a quickstart guide for quantum in devstack18:30
garykarosen: welcome aboard18:31
*** ExxonValdeez has joined #openstack-dev18:32
*** danwent has joined #openstack-dev18:33
*** novas0x2a|laptop has joined #openstack-dev18:33
*** wiliam has quit IRC18:34
jgriffithdtroyer:https://review.openstack.org/#/c/7042/ I dunno.. it all passed for me18:34
garykarosen: added my comments - thanks. i am going now. chat tomorrow18:34
*** littleidea has joined #openstack-dev18:35
*** wiliam has joined #openstack-dev18:41
*** nunosantos has quit IRC18:43
*** nunosantos has joined #openstack-dev18:43
*** joesavak has joined #openstack-dev18:43
*** eglynn__ has quit IRC18:44
*** eglynn has joined #openstack-dev18:44
*** jog0 has joined #openstack-dev18:49
*** jog0 has quit IRC18:50
*** jog0 has joined #openstack-dev18:50
*** epim has joined #openstack-dev18:54
*** nati_ueno has quit IRC18:55
*** anotherjesse_zz is now known as anotherjesse18:56
*** danwent_ has joined #openstack-dev18:58
*** danwent has quit IRC18:58
*** danwent_ is now known as danwent18:58
*** markmc has joined #openstack-dev18:58
*** andrewbogott has quit IRC19:00
markmcttx, there?19:05
markmcsilly me, he's marked away19:06
*** utlemming has quit IRC19:09
*** utlemming has joined #openstack-dev19:09
*** m4xmr has joined #openstack-dev19:11
*** jakedahn_zz is now known as jakedahn19:15
*** koolhead17 has quit IRC19:16
*** lts has quit IRC19:23
bcwaldonmtaylor: oy19:26
bcwaldonmtaylor: what do you want me to do with this https://review.openstack.org/#/c/8704/19:26
*** epim has quit IRC19:31
maoyvishy, jerdfelt: please take a look when you have time at this patch: https://review.openstack.org/#/c/8254/19:36
dtroyerjgriffith: are you running with n-vol enabled?  I tried it and get the same failure as Jenkins...19:37
*** rkukura has quit IRC19:39
jgriffithdtroyer: Hmm.. yes, I was19:40
jgriffithdtroyer: guess I better check again... volume.sh failed for you or ecua.sh (or both)19:40
dtroyerjgriffith: both19:41
*** joesavak has quit IRC19:41
jgriffithdtroyer: I was worried about the euca.sh additions becuase the delete snapshot can get hung for some reason.  The other is "new"19:41
*** dolphm_ has quit IRC19:42
vishymaoy: got a link to the resulting graphviz?19:42
maoyvishy: give me a minute to do that. I have it in my mac but it's with apple care now. :(19:43
maoyvishy: now I have a png.. where to upload to?19:47
*** segfault923 has quit IRC19:47
vishymaoy: wherever, I just want to see what it looks like19:47
vishyyou could put it in the wiki if you want19:48
maoyvishy: emailed you..19:48
*** heckj has quit IRC19:49
*** zigo has quit IRC19:50
vishymaoy: ok so looking at the code, you've decided to always default to stopped if powerstate and vmstate don't agree19:53
vishyI think that makes sense, although I'm wondering if some people might prefer some attempt to fix it19:53
vishymaoy: I'm a little concerned about a reboot from inside the guest. Will there be a few moments where the state is something other than running in that case?19:54
jgriffithdtroyer: Not sure what I'm doing differently.... exercises/volumes.sh no problems19:54
*** andrewbogott has joined #openstack-dev19:55
*** andrewbogott has joined #openstack-dev19:55
maoyvishy: not always. e.g. when they disagree and vm_state is rescued, resized, suspended, i don't do anything yet. yes people are welcome to fix those..19:55
vishycool diagram19:55
jgriffithdtroyer: FWIW the only change from the last succesful Jenkins run was removal of a comment ???19:55
vishymaoy: i'm specifically referring to the powered off guest vs active/stopped19:55
maoyvishy: yes. that's correct.19:56
vishymaoy: so do you know if a guest reboot will temporarily change power state?19:56
maoyvishy: I haven't seen state being not RUNNING if the guest does reboot yet. but that's libvirt only19:56
maoyvishy: no idea about xenapi.19:56
vishyi'd hate to have someone do a reboot right when the periodic task is running and end up stopping the instance19:56
dtroyerjgriffith: do we have a rebase-type problem?  with 3 people uploading patch sets there could easily be some differences...19:57
vishymaoy: might be good to find out19:57
jgriffithdtroyer: possibly... Looking back through jenkins console, see what I might be missing19:57
vishypvo, comstud, tr3buchet, _cerberus_, Vek, jerdfelt: any idea if a reboot inside the guest will change the power state reported by xen temporarily?19:57
*** dachary has joined #openstack-dev19:58
vishymaoy: the othe thing that I fundamentally dislike is the handling of the state changes in the compute node19:58
vishymaoy: sorry the state syncing19:58
maoyvishy: agreed on that. but have to fix that after russellb's getting-rid-of-db work19:59
vishymaoy: we are running a copy of the api code inside of the compute manager which seems really dangerous19:59
jgriffithdtroyer: I think I know what the deal is....  and the additional euca volume tests are the culprit19:59
jgriffithdtroyer: volumes.sh fails because euca-delete-snapshot failed so we don't have any free space left20:00
maoyvishy: agree. but we already do that for confirm_resize anyway..20:01
vishymaoy: yes I'm thinking if there is any way that we can improve it for the moment, and I'm not coming up with anything20:01
vishymaoy: yes I hate that code too :)20:01
jgriffithdtroyer: the euca-delete-snapshot fail is intermittent it seems, thus the succesful run on my workstaion and on jenkins last night20:02
vishymaoy: ok I guess we have to fix it later. At the very least this is a big improvement20:02
*** rkukura has joined #openstack-dev20:02
maoyvishy: thanks. i'll definitely keep that in mind.20:03
jgriffithdtroyer: I guess for now the best way forward is to remove the euca volume tests I added (for now) agree?20:03
jgriffithdtroyer: perhaps this is why they weren't there to begin with :)20:03
dtroyerjgriffith: probably.20:03
maoyvishy: there are still race conditions but i prefer to get this patch out. seems some tempest work might depend on the vm states20:03
vishymaoy: what did you change between patch set 1 and patch set 2?20:03
dtroyerjgriffith: maybe make them conditional on cinder being enabled?20:04
vishymaoy: er 6 and 720:04
jgriffithdtroyer: Cinder will have the same issue with the snapshots, it's just the snapshots that are the issue becuase of the stupid dd operation20:04
maoyvishy: in patch 7, the sync_power_state is rewritten.20:04
maoyvishy: I also allow to call stop when it's in STOPPED state.20:05
dtroyerjgriffith: roger20:05
jgriffithdtroyer: I'll make the changes and resubmit20:05
vishymaoy: ok I was trying to look at the diff but the rebase made it hard. Jenkins hasn't weight in on the new patchset yet20:05
*** jaustinpage has joined #openstack-dev20:05
*** japage has quit IRC20:05
jgriffithdtroyer: I suspect just removing the snapshot portions will be sufficient so I'll start with that20:06
dtroyerjgriffith: sounds good.  I'd love to get this in once it works reliably for n-vol then we can do any follow-up for cinder operation…it has been a PITA to maintain it while everything changes around us20:07
maoyvishy: i should probably note those in commit message next time before I forget.20:07
jgriffithdtroyer: ++++++++120:07
jgriffithdtroyer: My proposal is get this buttoned up, then add a change later this week when we're ready to switch cinder on as default20:07
*** matwood is now known as matwood-lunch20:07
jgriffithdtroyer: In fact, I'll remove all my test additions for now so you can get this off your plate20:08
* vishy cries20:08
* vishy loves having tests for snapshots20:09
jgriffithvishy: They'll be back!!  I just want to let dtroyer get on with his life20:09
*** dubsquared has quit IRC20:09
dtroyervishy: too much stuff got piled in to this single patch...20:10
* jgriffith turns red in the face20:10
*** datsun180b_ has joined #openstack-dev20:10
jgriffithdtroyer: Ok, see what that does for us.  Then I'll figure out a better way to get the snapshot tests back in20:12
*** datsun180b has quit IRC20:14
*** datsun180b_ is now known as datsun180b20:14
*** dolphm has joined #openstack-dev20:14
*** sdake has quit IRC20:15
*** sdake has joined #openstack-dev20:15
*** sdake has quit IRC20:16
*** sdake has joined #openstack-dev20:16
*** wiliam has quit IRC20:16
*** johnpostlethwait has joined #openstack-dev20:16
jkoelkermtaylor just sent in the nose_plugin stuff, new version will be on pypi momentarily20:18
*** anotherjesse is now known as anotherjesse_zz20:19
*** jakedahn is now known as jakedahn_zz20:19
jkoelkermtaylor, bcwaldon nose_plugin updated on pypi, hulk out mode averted20:22
bcwaldonawesome20:23
*** dubsquared has joined #openstack-dev20:23
jgriffithdtroyer: Coool... I've drug this out long enough, hopefully it'll get reviewed and merged here shortly20:26
jgriffithdtroyer: Thanks for all the work on this20:26
dtroyerjgriffith: np, testing patch16 now...20:27
jgriffithdtroyer: Jenkins like it at least :)20:27
mtaylorjkoelker: awesome20:28
*** datsun180b has quit IRC20:31
*** datsun180b has joined #openstack-dev20:31
*** ncode has joined #openstack-dev20:33
*** segfault923 has joined #openstack-dev20:33
*** zul has quit IRC20:35
*** epim has joined #openstack-dev20:42
*** dprince has quit IRC20:42
vishymtaylor: why on earth would libvirt be a string? https://jenkins.openstack.org/job/gate-nova-python26/815/testReport/junit/nova.tests.test_libvirt/LibvirtDriverTestCase/test_finish_migration/20:44
vishyI can fix the test my mocking it out but i'm really curious how the heck that happened20:44
*** alaski has quit IRC20:44
vishyjeblair: ^^20:45
mtaylorvishy: me looky20:45
*** alaski has joined #openstack-dev20:46
mtaylorvishy: well, first positive: we're finally actually doing libvirt testing on jenkins!20:47
sorenvishy: What's the change being tested there?20:47
vishymtaylor: I should mock out the call regardless so I will fix, but it looks like __import__('libvirt') is probably throwing an errror.20:47
vishysoren: it is my change to the usage of loopingCall() to make it actually wait properly20:48
sorenvishy: The line numbers in test_libvirt.py don't seem to match trunk, afaics.20:48
vishysoren: yes I changed it :)20:48
vishysoren: https://review.openstack.org/#/c/8731/20:49
vishyi believe prior the test finished before the looping call activated so it wasn't actually hitting the code path where it calls get_info20:49
vishyI'm just surprised with the actual failure. It makes me think that libvirt tests aren't really running correctly on the jenkins boxes20:50
*** stuntmachine has quit IRC20:50
mtaylorvishy: connection.libvirt = ""20:51
sorenvishy: Yeah. Interesting.20:51
mtaylorvishy: I tried __import__('libvirt') by hand on that machine ...20:52
mtaylorvishy: lemme try it in a venv real quick20:52
mtaylorvishy: AHA!20:52
mtaylorvishy: there it is - we're hitting the "libvirt not found" - so apparently I was lying about the libvirt problem on the builders being fixed20:53
*** anotherjesse_zz is now known as anotherjesse20:53
mtaylorvishy: so potentially instead of mocking that test just needs one of the if_libvirt barriers20:53
mtayloror whatever -20:53
mtaylorin the meantime, I will figure out why the **** we're losing libvirt in the venv20:54
*** jakedahn_zz is now known as jakedahn20:54
*** alaski has quit IRC20:55
vishymtaylor: I mocked it out instead20:56
vishymtaylor: :)20:56
mtaylorvishy: kk20:56
*** m4xmr has quit IRC20:57
*** steveb_ has joined #openstack-dev21:02
*** rbasak has quit IRC21:03
jog0vishy:  Is http://api.openstack.org/ the official AP reference?     The host aggregates section doesn't mention an api to list the existing aggregates even though the code already exists. along with nova-client21:07
vishyjog0: sounds like it needs to be added21:08
vishyi21:08
vishyjog0: it is generated from openstack-manuals project iirc21:08
annegentlejog0: yes the source is in openstack-manuals. Can you log a doc bug at http://bugs.launchpad.net/openstack-manuals with details?21:09
annegentlejog0: do you happen to know the call for listing existing aggregates? Is it an os-extension?21:10
*** segfault923 has quit IRC21:12
jog0vishy:  Thanks. annegentle: yes, it is on os-exetension.  I do know the existing call, and I will file a bug and file a patch21:16
*** anotherjesse is now known as anotherjesse_zz21:18
*** salgado is now known as salgado-afk21:20
*** kbringard has quit IRC21:20
annegentlejog0: excellent, thanks, I can help if needed21:20
jog0annegentle: I can do all the aggregates related updates, but  I noticed the list command is missing for flavors as well21:22
annegentlejog0: is that this bug? bug 93354721:23
uvirtbotLaunchpad bug 933547 in openstack-manuals "api-site FlavorRef and ImageRef are not documented as parameters when creating a server" [High,Confirmed] https://launchpad.net/bugs/93354721:23
jog0annegentle: no,  the list command is missing under the "flavor management"  section21:24
annegentlejog0: oh I see, just POST and DELETE, no GET21:25
annegentlejog0: er that was meant to be a question :)21:25
jog0annegentle:  bingo.21:25
annegentlejog0: got it, logging a bug now21:25
*** matwood-lunch is now known as matwood21:27
*** epim has quit IRC21:27
*** sieutruc has joined #openstack-dev21:27
sieutruchello21:28
*** markvoelker has quit IRC21:28
*** jakedahn is now known as jakedahn_zz21:28
*** dolphm has quit IRC21:29
sieutruci would like to edit the data integrity of openstack , but i don't know which part in openstack i do21:29
sieutruccan you give me a hint for that ?21:29
*** markmc has quit IRC21:30
*** dolphm has joined #openstack-dev21:31
*** danwent has quit IRC21:31
*** danwent_ has joined #openstack-dev21:31
*** maoy has quit IRC21:32
*** jgruber has quit IRC21:34
*** sc68cal has joined #openstack-dev21:35
*** jaypipes has quit IRC21:37
*** nati_ueno has joined #openstack-dev21:37
*** sstent has quit IRC21:37
*** sstent has joined #openstack-dev21:39
vishyI suspect resize with attached volumes will epicly fail, although I'm curious to see how :)21:41
*** tgall_foo has quit IRC21:42
mtaylorvishy: thanks for finding that error earlier - I have found a bug in the tox.ini entry for nova, so the system site packages were still not getting included21:44
vishymtaylor: cool21:45
mtaylorvishy: https://review.openstack.org/878121:47
mtaylorvishy: with that - we should be back to having live libvirt unittests21:47
*** Aaton_off is now known as Aaton21:57
*** markmcclain has quit IRC21:59
*** dtroyer is now known as dtroyer_zzz22:03
*** johnpostlethwait has quit IRC22:04
*** nati_uen_ has joined #openstack-dev22:05
*** littleidea has quit IRC22:06
*** johnpostlethwait has joined #openstack-dev22:07
*** jakedahn_zz is now known as jakedahn22:07
arosenis there an easy way to determine which HV vm's are located on?22:07
*** littleidea has joined #openstack-dev22:07
arosennvm using ps -eaf | grep kvm I can match up the uuid between that and nova list.22:08
*** nati_ueno has quit IRC22:09
*** jaustinpage has quit IRC22:14
*** johnpostlethwait has quit IRC22:17
*** andrewsmedina has quit IRC22:17
*** ben_duyujie has joined #openstack-dev22:20
*** eglynn has quit IRC22:21
*** dubsquared1 has joined #openstack-dev22:23
*** dubsquared has quit IRC22:23
*** dolphm has quit IRC22:23
*** eglynn has joined #openstack-dev22:29
*** ben_duyujie has quit IRC22:38
*** e1mer has joined #openstack-dev22:39
*** anotherjesse_zz is now known as anotherjesse22:39
*** shang has quit IRC22:40
*** heckj has joined #openstack-dev22:42
*** mattray has quit IRC22:42
heckjdolphm - you around?22:42
*** nati_uen_ has quit IRC22:44
*** nati_ueno has joined #openstack-dev22:45
*** danwent_ has quit IRC22:45
*** littleidea has quit IRC22:45
*** danwent has joined #openstack-dev22:46
*** ben_duyujie has joined #openstack-dev22:46
*** littleidea has joined #openstack-dev22:46
*** datsun180b has quit IRC22:48
*** johnpostlethwait has joined #openstack-dev22:49
*** eglynn has quit IRC22:51
*** eglynn has joined #openstack-dev22:52
cloudflyarosen admin can see host sysems in a euca-describe-instances i believe22:56
*** heckj has quit IRC22:56
cloudflythe instance ids i thought matched those in libvirt22:56
cloudfly=/22:56
*** nunosantos has quit IRC22:57
*** eglynn has quit IRC22:59
*** galstrom has quit IRC22:59
*** pixelbeat has joined #openstack-dev22:59
*** steveb_ has quit IRC23:00
*** eglynn has joined #openstack-dev23:00
*** danwent has quit IRC23:00
*** dubsquared1 has quit IRC23:01
*** danwent has joined #openstack-dev23:01
*** rnirmal has quit IRC23:04
*** steveb_ has joined #openstack-dev23:05
*** tgall_foo has joined #openstack-dev23:06
*** dtroyer_zzz is now known as dtroyer23:09
*** Ryan_Lane has joined #openstack-dev23:10
*** metral_ has joined #openstack-dev23:10
*** metral_ has quit IRC23:11
*** andrewbogott has quit IRC23:12
*** metral has quit IRC23:14
*** dtroyer is now known as dtroyer_zzz23:14
*** dachary has quit IRC23:16
*** lloydde has quit IRC23:18
*** salgado-afk has quit IRC23:19
*** danwent has quit IRC23:19
*** eglynn has quit IRC23:20
*** ben_duyujie has quit IRC23:20
pballandwhat's the preferred style for wrapping long import lines such as "from really_long_package_path.subpath.subsubpath import NameThatExceeds80Cols as Foo" ?23:25
*** anotherjesse is now known as anotherjesse_zz23:25
*** rbasak has joined #openstack-dev23:25
*** mestery has quit IRC23:26
*** eglynn has joined #openstack-dev23:34
*** dolphm has joined #openstack-dev23:34
*** ncode has quit IRC23:36
*** andrewsmedina has joined #openstack-dev23:37
*** anderstj has quit IRC23:38
*** dolphm has quit IRC23:38
*** nati_uen_ has joined #openstack-dev23:39
steveb_vishy, thanks for approving my membership. Gerrit still isn't letting me push, but I assume there is a delay before it syncs23:40
vishysteveb_: takes about 15 min. Make sure your git email matches your launchpad email23:40
steveb_vishy, OK, thanks. It does AFAIK23:41
*** dtroyer_zzz is now known as dtroyer23:41
*** nati_ueno has quit IRC23:42
*** littleidea has quit IRC23:48
*** matwood has quit IRC23:49
*** pvo has quit IRC23:51
*** pvo has joined #openstack-dev23:52
*** nati_ueno has joined #openstack-dev23:53
*** nati_uen_ has quit IRC23:55
*** mszilagyi has quit IRC23:55
*** nati_ueno has quit IRC23:55
*** nati_ueno has joined #openstack-dev23:55
*** anotherjesse_zz is now known as anotherjesse23:58
*** nati_uen_ has joined #openstack-dev23:59
*** nati_ueno has quit IRC23:59

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