Tuesday, 2016-03-22

openstackgerritMerged openstack/tripleo-heat-templates: Enable heat-manage purge_deleted cron job  https://review.openstack.org/29448700:20
*** rwsu has quit IRC00:28
*** dmacpher has joined #tripleo00:57
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: IPv6: dual-stack support for Keystone  https://review.openstack.org/28634401:01
*** thrash is now known as thrash|pt001:13
openstackgerritMatthew Thode proposed openstack/diskimage-builder: add pkg-map to pip-and-virtualenv element  https://review.openstack.org/29497301:13
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: WIP: Add a OVB te-broker server  https://review.openstack.org/29524301:16
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Add mirror server  https://review.openstack.org/23841401:16
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Use the mirror server for the centos image and git repos  https://review.openstack.org/28525701:16
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Use instack-qcow2 from cache if available and appropriate  https://review.openstack.org/28525901:16
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Upload the instack qcow image after periodic jobs  https://review.openstack.org/28525801:16
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: [NO MERGY] Test a fake periodic job  https://review.openstack.org/22978901:16
*** ccamacho|afk has quit IRC01:25
*** masco has joined #tripleo01:28
*** ccamacho has joined #tripleo01:28
*** lblanchard has joined #tripleo01:28
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: Split out controller manifest steps  https://review.openstack.org/23624301:30
*** lblanchard has quit IRC01:30
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: Configure ControllerServices via resource chains  https://review.openstack.org/25956801:30
openstackgerritDan Prince proposed openstack/tripleo-heat-templates: composable keystone services  https://review.openstack.org/29558801:30
*** yamahata has quit IRC01:43
*** lucas-dinner has quit IRC01:55
*** Marga_ has quit IRC01:55
*** lucasagomes has joined #tripleo02:00
*** panda has quit IRC02:05
*** panda has joined #tripleo02:06
openstackgerritJames Slagle proposed openstack/instack-undercloud: Secure haproxy stats endpoint  https://review.openstack.org/29544402:06
*** shivrao has quit IRC02:09
openstackgerritMerged openstack/tripleo-heat-templates: Allow option to disable Swift ring management  https://review.openstack.org/29542602:11
openstackgerritMerged openstack/python-tripleoclient: Add password for HAProxy stats  https://review.openstack.org/29544102:15
*** masco has quit IRC02:25
*** Marga_ has joined #tripleo02:28
openstackgerritStephanie Miller proposed openstack/diskimage-builder: ironic-agent postinst fails on systemd with no iptables  https://review.openstack.org/29559702:30
*** rhallisey has quit IRC02:33
*** Marga_ has quit IRC02:33
*** shivrao has joined #tripleo02:45
*** lazy_prince has joined #tripleo02:58
*** Marga_ has joined #tripleo03:10
*** Marga_ has quit IRC03:14
openstackgerritMichael Chapman proposed openstack/puppet-tripleo: Add keystone and db sync profiles  https://review.openstack.org/29271203:18
*** links has joined #tripleo03:19
*** shivrao has quit IRC03:23
*** yuanying has quit IRC03:28
*** ayoung has quit IRC03:29
*** Marga_ has joined #tripleo03:31
*** Marga_ has quit IRC03:35
*** yuanying has joined #tripleo03:38
*** yamahata has joined #tripleo03:43
*** masco has joined #tripleo03:49
*** Marga_ has joined #tripleo03:52
*** Marga_ has quit IRC04:01
*** Marga_ has joined #tripleo04:01
*** masco has quit IRC04:04
*** yuanying has quit IRC04:04
*** yuanying has joined #tripleo04:06
lazy_princeHi.. Can someone please review https://review.openstack.org/#/c/287784 ..04:09
*** chlong|wfh has quit IRC04:12
*** rlandy has quit IRC04:20
*** lazy_prince has quit IRC04:21
*** chlong has joined #tripleo04:25
*** shivrao has joined #tripleo04:30
*** qasims_ has joined #tripleo04:40
*** tzumainn has quit IRC05:01
openstackgerritMichael Chapman proposed openstack/puppet-tripleo: Add keystone and db sync profiles  https://review.openstack.org/29271205:14
*** shivrao has quit IRC05:17
*** akuznetsov has joined #tripleo05:26
*** rwsu has joined #tripleo05:29
*** qasims_ has quit IRC05:33
*** saneax_AFK is now known as saneax05:41
*** shivrao has joined #tripleo05:49
*** qasims_ has joined #tripleo05:59
*** masco has joined #tripleo06:04
*** ohamada has joined #tripleo06:11
*** TSCHAK has quit IRC06:20
*** ohamada has quit IRC06:25
*** social has joined #tripleo06:32
*** jprovazn has joined #tripleo06:34
*** jprovazn has quit IRC06:38
*** jprovazn_ has joined #tripleo06:38
*** jprovazn has joined #tripleo06:39
*** dshulyak has joined #tripleo06:44
*** shivrao_ has joined #tripleo06:47
*** shivrao has quit IRC06:51
*** shivrao_ is now known as shivrao06:51
*** akuznetsov has quit IRC06:57
*** davidlenwell has quit IRC07:01
*** shivrao has quit IRC07:07
*** davidlenwell has joined #tripleo07:09
*** ChanServ sets mode: +v davidlenwell07:09
*** mikelk has joined #tripleo07:11
*** oshvartz has joined #tripleo07:13
*** larstobi has quit IRC07:19
*** oshvartz has quit IRC07:20
*** larstobi has joined #tripleo07:20
*** mcornea has joined #tripleo07:21
*** mkovacik__ has quit IRC07:21
*** ohamada has joined #tripleo07:22
*** liverpooler has joined #tripleo07:22
*** tesseract has joined #tripleo07:28
*** tesseract is now known as Guest9125107:28
*** bvandenh has joined #tripleo07:32
*** akuznetsov has joined #tripleo07:32
*** rcernin has joined #tripleo07:39
*** jtomasek has joined #tripleo07:39
openstackgerritMerged openstack/diskimage-builder: Note requirement for parted on gentoo hosts  https://review.openstack.org/29497207:41
*** ebalduf has quit IRC07:48
*** paramite has joined #tripleo07:49
*** paramite is now known as paramite|afk07:52
*** dshulyak has quit IRC07:57
*** paramite|afk is now known as paramite08:00
*** ccamacho has quit IRC08:07
*** olap has joined #tripleo08:07
*** ccamacho has joined #tripleo08:10
*** dshulyak has joined #tripleo08:20
*** paramite is now known as paramite|afk08:22
*** akuznetsov has quit IRC08:23
*** athomas has joined #tripleo08:23
*** ifarkas has joined #tripleo08:27
*** aufi has joined #tripleo08:28
*** hjensas has joined #tripleo08:29
*** dmacpher has quit IRC08:29
*** paramite|afk is now known as paramite08:32
*** openstackgerrit has quit IRC08:33
*** openstackgerrit has joined #tripleo08:33
*** oshvartz has joined #tripleo08:36
*** pcaruana has joined #tripleo08:40
*** apetrich has quit IRC08:50
*** lazy_prince has joined #tripleo08:51
*** ramishra has quit IRC08:56
*** ramishra has joined #tripleo08:59
openstackgerritMerged openstack/tripleo-heat-templates: Wire in HAProxy stats user and password  https://review.openstack.org/29544209:00
*** mbound has joined #tripleo09:02
openstackgerritMichele Baldessari proposed openstack/os-cloud-config: Add AMT support  https://review.openstack.org/29568009:04
*** gfidente has joined #tripleo09:05
*** paramite is now known as paramite|afk09:05
*** rdopiera has joined #tripleo09:06
d0ugalDo folk think it is okay for me to +A https://review.openstack.org/#/c/291723/2?09:16
d0ugalIt was previously, but failed to merge. Last update just resolved that.09:16
*** shardy has joined #tripleo09:17
*** derekh has joined #tripleo09:17
openstackgerritJuan Antonio Osorio Robles proposed openstack/puppet-tripleo: Make cipher suite and SSL options configurable  https://review.openstack.org/29443809:18
*** akuznetsov has joined #tripleo09:23
*** paramite|afk is now known as paramite09:23
*** jaosorior has joined #tripleo09:23
openstackgerritMichele Baldessari proposed openstack/python-tripleoclient: Add AMT support  https://review.openstack.org/29569009:25
d0ugaljaosorior: ping09:26
jaosoriord0ugal: pong09:27
*** akrivoka has joined #tripleo09:29
d0ugaljaosorior: Hey, would you be able to upgarde a few reviews you made from +1 to +2? Assuming you are happy to do so :)09:29
jaosoriord0ugal, no problem dude, pass them on09:29
d0ugaljaosorior: https://review.openstack.org/#/c/294150/ and https://review.openstack.org/#/c/294129/09:29
d0ugaljaosorior: I go on PTO tomorrow, trying to clean up a few of my smaller reviews09:29
jaosoriord0ugal: No worries, it makes sense09:30
*** sshnaidm has quit IRC09:30
*** sshnaidm has joined #tripleo09:31
jaosoriord0ugal: This could merge too: https://review.openstack.org/#/c/294150/3 although it would be nice to get those values for the tests also09:32
*** mkovacik__ has joined #tripleo09:32
jaosoriorbuuuut, this is already pretty good, so it's a +2, and lets see what's up with the ha gate09:32
d0ugaljaosorior: oh, sorry I missed that it had failed.09:33
*** apetrich has joined #tripleo09:33
jaosoriord0ugal: Got anything other review floating around?09:33
d0ugaljaosorior: I have a bunch, but those are the only ones with another +2 :)09:34
openstackgerritMerged openstack/tripleo-docs: Update the tox config to build the docs by default  https://review.openstack.org/29412909:34
d0ugaljaosorior: but if you want a really easy one: https://review.openstack.org/29165009:34
jaosoriorYeah, it failed, but it timed out... :/09:34
d0ugaland this: https://review.openstack.org/29164509:34
jaosoriorso it seems quite unrelated09:34
jaosoriord0ugal: anyway, lets see what the next run says. then +A09:36
*** athomas has quit IRC09:36
d0ugaljaosorior: yup, thanks09:36
jaosoriorgfidente: I would REALLY use your input for this patch https://review.openstack.org/#/c/292799/ the reason is that I want to start generating certificates for the internal services. So it would be useful to be able to make assumptions for those.09:36
d0ugaljaosorior: oh, there is also this change - you reviewed previously and it had a +A but the merge failed. https://review.openstack.org/#/c/291723/09:37
d0ugalI almost just +A'ed it again09:37
*** sshnaidm has quit IRC09:37
jaosoriord0ugal: Where are you based at?09:40
d0ugaljaosorior: Scotland, you?09:40
d0ugalcool :)09:41
jaosoriorgot any traces of sprint yet?09:41
*** athomas has joined #tripleo09:41
jaosoriorHere it was starting to get sunny, and slowly getting better, then one day it snowed again :P09:41
jaosoriorbut, that's finland09:41
d0ugaljaosorior: spring? Yup. It's been a bit sunnier and some flowers etc. are appearing.09:42
d0ugaljaosorior: We have a similar problem, but rain instead of snow :)09:42
d0ugalActually, it rains all year, so maybe it isn't the same...09:42
jaosoriorhahaha damn09:42
jaosoriorwell, here we get some change09:42
jaosoriorit snows most of the year, then it rains the rest of it09:43
openstackgerritMerged openstack/tripleo-common: Import code working with Ironic nodes from os_cloud_config  https://review.openstack.org/29172309:43
*** sshnaidm has joined #tripleo09:43
*** electrofelix has joined #tripleo09:44
*** ccamacho has quit IRC09:44
*** ccamacho has joined #tripleo09:45
openstackgerritDougal Matthews proposed openstack/tripleo-docs: Correct a link the basic_deployment document  https://review.openstack.org/29165009:47
openstackgerritMerged openstack-infra/tripleo-ci: Reset the ovs nightly  https://review.openstack.org/29248009:49
*** apetrich_ has joined #tripleo09:51
openstackgerritDan Sneddon proposed openstack/os-net-config: Fix order-of-operations bug in os-net-config restart_interfaces  https://review.openstack.org/29142009:51
*** apetrich has quit IRC09:53
*** sshnaidm has quit IRC09:55
*** beagles has left #tripleo09:56
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: Switch to using node utilities from tripleo-common  https://review.openstack.org/29172610:01
*** mgould has joined #tripleo10:04
*** sshnaidm has joined #tripleo10:06
*** apetrich_things_ has joined #tripleo10:09
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: Use Ironic API v1.11 to support ENROLL state  https://review.openstack.org/27220610:11
*** dshulyak has quit IRC10:11
*** pblaho has joined #tripleo10:12
lazy_princeCan some one spend few minutes reviewing https://review.openstack.org/#/c/287784/10:12
*** apetrich_ has quit IRC10:12
*** jaosorior has quit IRC10:15
*** jaosorior has joined #tripleo10:15
openstackgerritDougal Matthews proposed openstack/python-tripleoclient: Use Ironic API v1.11 to support ENROLL state  https://review.openstack.org/27220610:17
*** tosky has joined #tripleo10:24
*** apetrich_things_ has quit IRC10:24
*** nico_auv has joined #tripleo10:29
*** dshulyak has joined #tripleo10:32
*** akuznetsov has quit IRC10:34
*** apetrich has joined #tripleo10:44
adarazsderekh: hello :) can you please look at the machine that we use for reproducing IPv6... I think there are more VMs running in the background or something. everything is very sluggish, and rabbit fails to start with a timeout... I had this before when it started up too slow.10:51
derekhadarazs: there are other VM's on it, I'll shut them down10:51
derekhadarazs: that should be better now11:03
adarazsderekh: now it's a lot more responsive. rabbit still fails to start up on controller 1 and 2 (it's okay on "0") with "unknown error", nothing in the rabbitmq logs to indicate the problem. :/11:03
adarazsI'm trying to figure out what the heck is wrong, but I really don't know pacemaker.11:04
adarazsor rabbit :)11:04
derekhI'm afraid I'm not vary familiar with them myself either, would it be worth trying again now that the load on the host is gone?11:05
adarazsderekh: yeah, I'll redeploy.11:05
d0ugaljaosorior: This is another easy one :) https://review.openstack.org/#/c/291726/11:08
jaosoriord0ugal: Done11:09
jaosoriorI had seen that one before. Just needs CI results and that's that11:09
d0ugalYup, I'll watch out for that.11:09
d0ugalgfidente: That hasn't passed CI11:10
gfidented0ugal which one?11:10
adarazsderekh: btw is there some guide to explain how to create a reproducer environment like this? I would be happy to use one of my dev boxes to make me able to reproduce on it.11:10
d0ugalgfidente: https://review.openstack.org/#/c/291726/511:11
gfidenteah I checked the depends-on11:11
gfidenteand missed the other wasn't finished yet11:11
d0ugalgfidente: I think it should be okay, but better safe than sorry I guess :)11:11
gfidentesure that wasn't my intention :P11:12
derekhadarazs: There shouldn't be anything special about it, it was setup with devtest_testenv http://git.openstack.org/cgit/openstack/tripleo-incubator/tree/scripts/devtest_testenv.sh11:12
derekhadarazs: but the image was built for it about 2 years ago, v. hard to reproduce now, I can provide you with the image if you want11:13
derekhadarazs: but the main thing I think when compared to people dev envs is that the hosts have multiple environments so can be heavily loaded, and you should match the ram/cpu allocated to each instance11:14
d0ugalgfidente: np11:15
gfidenteguys these two https://review.openstack.org/#/c/289812/ and https://review.openstack.org/#/c/293051/ have either passed all CI jobs collectively or making changes unrelated to the failed jobs11:19
jaosoriorgfidente: I'm tempted in merging this https://review.openstack.org/#/c/260466/1311:20
gfidenteI'd like to merge them .. rechecks are not so useful11:20
gfidentejaosorior yeah that one too11:20
gfidentedepends on one of the other two I mentioned above11:20
jaosoriorThis depends on those two? https://review.openstack.org/#/c/260466/1311:20
gfidenteon https://review.openstack.org/#/c/293051/11:21
jaosoriorI see11:22
jaosoriorgfidente: Not sure where it actually fails, it even seems to pass the pingtest11:23
gfidentejaosorior https://review.openstack.org/#/c/294518/ as well yeah11:24
gfidenteI mean, we're just slowing down things on rechecks11:24
jaosoriorgfidente: That is the case, the only reason I haven't merged it is because I'm unsure what's gonna happen; The overcloud ssl recently changed, and it didn't have any merge conflict, which is weird to me11:26
jaosoriorgfidente: Yep, there will be a merge conflict in that file11:26
jaosoriorit needs to be rebased11:26
jaosoriorthis cr https://review.openstack.org/#/c/293051/311:27
jaosoriorgfidente: Should I rebase https://review.openstack.org/#/c/293051/3 ?11:30
gfidentejaosorior actually regarding tls/ssl can you check this https://review.openstack.org/#/c/294071/11:34
gfidenteand see if we need to add columns and where?11:34
*** weshay has joined #tripleo11:35
jaosoriorgfidente: I'll comment on that patch11:36
*** oshvartz has quit IRC11:37
jaosoriorgfidente: Commented11:38
gfidentejaosorior great, thanks! :)11:39
jaosoriorgfidente: Sorry for pinging you so much, but I would really use some input here https://review.openstack.org/#/c/292799/ can you check it out when you have time?11:40
gfidentejaosorior yeah I saw that, sorry, will do11:41
gfidentewe don't use those names yet so it should be safe11:41
gfidentewe just need to figure what's best way to do it11:41
jaosoriorgfidente: What do you mean?11:42
gfidentethe hostnames naming11:43
openstackgerritGiulio Fidente proposed openstack-infra/tripleo-ci: Deploy ceph in the nonha job  https://review.openstack.org/29305111:44
*** mburned_out is now known as mburned11:44
openstackgerritMerged openstack/tripleo-heat-templates: Remove CephStorageCount  https://review.openstack.org/29451811:44
*** dprince has joined #tripleo11:45
jaosoriorgfidente: well, the way I had set it up there was to add sort of sub-domains for the different endpoints. such as internalapi.<controller 1>.<domain>11:46
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Remove CephStorageCount  https://review.openstack.org/29574011:47
gfidentederekh jaosorior  ^^ I ported this one to liberty as we need it for the stable jobs11:48
gfidentejaosorior subdomains yep, it's probably nicer indeed11:49
*** ramishra has quit IRC11:51
*** ramishra has joined #tripleo11:51
*** jcoufal has joined #tripleo11:52
bandinihohum am I the only one seeing this: http://buildlogs.centos.org/centos/7/cloud/x86_64/openstack-liberty/repodata/repomd.xml: [Errno 12] Timeout11:56
*** rhallisey has joined #tripleo11:57
derekhbnemec: when your in, could I get you to take another look at the mirror server patch, https://review.openstack.org/#/c/238414/15  , I'll then rebuild a new server to represent what got merged and then we can recheck the other patches12:00
mariosgfidente: sanity check https://review.openstack.org/#/c/289812/2 pls12:03
*** trown|outtypewww is now known as trown12:05
EmilienMmorning o/12:06
*** apetrich has quit IRC12:07
adarazsEmilienM: good morning12:08
EmilienMadarazs: I updated your CI patch yesterday to have one more Depends-On12:08
EmilienMdid it help?12:08
*** masco has quit IRC12:08
adarazsEmilienM: I saw. all the gate jobs timed out on deployment. not sure if it was infra issue or something else. looking at your patch, it was probably unrelated.12:09
adarazsEmilienM: I did a recheck but the queue is super long.12:09
*** apetrich has joined #tripleo12:09
adarazsEmilienM: I saw you added TLS stuff to the nonha job, why was that needed?12:10
EmilienMadarazs: where?12:11
adarazsEmilienM: https://review.openstack.org/#/c/289445/20..21/toci_gate_test.sh line 73 in the patchset you pushed.12:11
adarazsEmilienM: same here: https://review.openstack.org/#/c/289445/20..21/toci_instack.sh12:12
adarazsmaybe you mixed two tripleo-ci changes?12:12
EmilienMadarazs: I think I was rebasing12:13
EmilienMadarazs: you can fix it if  needed12:13
jaosorioradarazs: TLS is now being tested in CI. That should actually be there12:14
adarazsoh, right, it must be the rebase then :)12:15
jaosorioradarazs, EmilienM: That came from this cr https://review.openstack.org/#/c/281988/12:15
adarazssorry for the confusion, I only checked patchset 20 vs 21.12:15
adarazsjaosorior: I'll probably remove temporarily on this WIP patch to test IPv6 separately. I also hijacked the container job :)12:15
*** apetrich has quit IRC12:15
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the container job to run again  https://review.openstack.org/28891512:17
*** bvandenh has quit IRC12:18
*** apetrich has joined #tripleo12:19
*** jayg|g0n3 is now known as jayg12:20
jaosoriorEmilienM: Do you think this pattern is alright? https://github.com/openstack/tripleo-heat-templates/blob/master/puppet/hieradata/controller.yaml#L54 I just realised that if I want to add another value to the configuration in another file (such as all-nodes-config) hiera will do a priority search and only fetch the top most value it find and not go deeper into the hierarchy. Would you think is fine if I move12:21
jaosoriorthose config-only related values to not use the <class>::<parameter> syntax? and use heira_merge to set those up in the manifests?12:21
*** lazy_prince has quit IRC12:21
*** pradk_ has joined #tripleo12:21
EmilienMjaosorior: we shoud really avoid using <...>::config::<...> and patch puppet-<...> instead12:22
EmilienMbecause if puppet-<...> adds the support for the parameter, it will break tripleo, for dupplicated parameter12:22
jaosoriorEmilienM: alright, will follow that then12:23
gfidentejaosorior also we can change the way hiera does the merging, see https://review.openstack.org/#/c/281822/12:25
gfidentebut it only works for explicit hiera_* calls, not for the automatic data binding to classes12:26
gfidenteso I w-1 that for now :(12:26
openstackgerritOded Shvartz proposed openstack/tripleo-common: overcloud-odl : add new image file definition  https://review.openstack.org/26688112:27
jaosoriorgfidente: Yeah, I checked on that too. So to add the deep merging I'm trying to push this package for the overcloud images: https://review.openstack.org/#/c/295410/12:28
*** morazi has joined #tripleo12:30
*** pradk has quit IRC12:31
*** pradk_ is now known as pradk12:34
*** paramite is now known as paramite|afk12:34
*** lazy_prince has joined #tripleo12:35
gfidentejaosorior ok we might be getting somewhere, left a question inline12:36
pradkshardy, slagle, can i get some help on why the ha job is failing https://review.openstack.org/#/c/289435/5 .. as i noted in the review, this patch was passing ci on Mar 11th and since then we only updated the commit message.12:36
pradklooking at the logs mongo comes up fine and db sync finishes successfully12:38
openstackgerritMarios Andreou proposed openstack-infra/tripleo-ci: Adds the new '-y' required for stack-deletion when tty  https://review.openstack.org/29576112:38
*** killer_prince has joined #tripleo12:38
*** Goneri has joined #tripleo12:39
*** killer_prince has quit IRC12:40
*** apetrich has quit IRC12:43
*** rlandy has joined #tripleo12:44
*** trozet has quit IRC12:46
openstackgerritGonĂ©ri Le Bouder proposed openstack/instack-undercloud: ironic: turn on ipxe timout  https://review.openstack.org/29478712:47
*** paramite|afk is now known as paramite12:48
trownis the meeting in 10 min or 70 min?12:50
trownsilly google calendar and its lack of UTC support12:50
mariostrown: in 6 mins (/me also has a call at same time :/)12:54
trownthanks marios12:57
EmilienMtrown: hum, in my calendar it's in 70 min12:57
slaglei do hate the way google calendar has no respect for utc12:58
trownstupid DST12:58
d0ugalI am glad I seen this, I assumed it was in 7012:59
EmilienMhum, meeting is in one hour guys13:00
mariostrown: EmilienM so i'm not so sure anymore... there is *usually* a clash for me... but since US changed times and EU hasn't yet, it may be wrong13:00
EmilienMit's only 1pm UTC now, and meeting is at 213:00
trownEmilienM: yep, it is in an hour13:00
d0ugalEmilienM: Yup, thanks - I thought I was going mad for a moment :)13:00
*** links has quit IRC13:01
*** MaxPC has joined #tripleo13:04
*** lazy_prince has quit IRC13:07
*** lucasagomes is now known as lucas-hungry13:07
*** bswartz has quit IRC13:08
*** lazy_prince has joined #tripleo13:08
*** ayoung has joined #tripleo13:11
ayoungshardy, If I want to update the policy.json file for the overcloud Keystone and Nova instances after deployment, how would I go about doing that.13:12
shardyayoung: I'd check in puppet-nova and puppet-keystone first to see if they support managing that file13:14
*** bswartz has joined #tripleo13:14
shardyif so then you can use this:13:14
shardyotherwise you'd have to modify the config in via of these interfaces:13:15
adarazsslagle, trown: there's a way to have UTC in google calendar :)13:15
*** sshnaidm has quit IRC13:16
shardyprobably OS::TripleO::ControllerExtraConfigPre assuming the sampe file is deployed via the package that's in the image13:16
*** paramite is now known as paramite|afk13:16
adarazsslagle, trown: 1) go to the calendar settings, 2) on the general tab, in the "Your current time zone" section, select the  'Display all time zones' checkbox and then click on "Show an additional  time zone" 3) scroll through the list to find (GMT+00:00) GMT (no daylight saving) and then click 'Save'13:16
ayoungshardy, so the basic flow is to make the change as Heat understands it then re-run openstack overcloud deploy13:16
shardypradk: sure, will check it out - I know the HA job was failing yesterday but I was working on a heat bug so not sure on the latest status13:16
shardyayoung: yup13:16
ayoungshardy, what about setting the FQDN for the overcloud?13:17
shardyayoung: obviously for development you can hack in changes manually, but then a subsequent stack update is liable to stomp over those changes13:17
ayoungshardy, right.  I am trying to document some things for production13:17
ayoungso need to do it "right"13:17
*** tzumainn has joined #tripleo13:18
ayoungshardy, would it make sense to keep the  compute_params.yaml  file in a git repo?13:18
shardyayoung: for FQDN/DNS see here:13:19
shardyayoung: if it's a generally useful example config, you can propose it to tripleo-heat-templates13:19
shardyayoung: most sample environments go in the environments/ directory13:20
trownadarazs: that shows 2PM UTC as 11AM EST... which is also not right13:20
ayoungshardy, I mean that using tripleo quickstart, it generates a config file.  We could keep that in a local git repo and then people could make changes there and redeploy according to how they want to customerize the cluster. Git would allow a roll back if a cahnge was wrong.13:20
adarazstrown: hm, I have a meeting set to UTC 2pm and it shows it as 3pm for CET which is correct, so not sure why that doesn't work for EST :/13:20
trownadarazs: nevermind, user error :P13:21
trownadarazs++ great tip, thanks!13:21
shardyayoung: sure, that's certainly possible, but we may not want to bake that into quickstart, because it overlaps with work going into the new TripleO API13:21
*** jdob has quit IRC13:21
adarazstrown: glad to help, it irked me that I can't set UTC as a meeting timezone before I found this too :)13:21
shardywhere "plans" including environment files/parameters get stored into swift13:21
shardyhaving git be an alternative store there would also be good13:22
*** jdob has joined #tripleo13:22
ayoungshardy, so the compute_params.yaml  would somehow become a maanged rsource?13:22
openstackgerritEmilien Macchi proposed openstack/puppet-pacemaker: Switch to ubuntu-trusty DIBS  https://review.openstack.org/29581213:22
adarazsif anybody has a HA tripleo deployment... rabbitmq-clone should be started on all controllers, right?13:22
shardyayoung: I'm not familiar with compute_params.yaml, what does it contain?13:23
adarazsshadower: I guess your tripleo deployment you have now is not HA...?13:23
*** sshnaidm has joined #tripleo13:24
ayoungshardy, heh...it was in the link you posted13:26
ayounghttp://docs.openstack.org/developer/tripleo-docs/advanced_deployment/node_config.html   " openstack overcloud deploy -e compute_params.yaml "13:26
pandaadarazs: clone resources usually are started on every node, or they shouldn't be clone from the beginning13:27
*** lblanchard has joined #tripleo13:27
shardyayoung: Ah, sorry I thought you were referring to an output from quickstart13:27
ayoungjust an extra params file.  Quickstart does something similar:  It generates...13:27
shardyayoung: yeah, the workflow may change slightly but basically that data ends up in swift13:27
shardythen passed to heat13:27
adarazspanda: oh, okay, "clone"... makes sense. shows how much I know about pacemaker. :)13:27
shardyvs just passed disk->CLI->Heat13:27
shardyayoung: right now for the CLI flow, checking all that into a local git repo makes perfect sense13:28
ayoungshardy, having the config options browsable, and modifiable prior to a redeploy would be very valuable.  "Here is what you wil get next time"  to include some way of saying "this is what is different"  which is why I was thinking git13:29
shardyand probably having a little wrapper script which manages the exact deploy command13:29
shardyin future, that may not be needed tho (although it'll remain possible)13:29
shardyayoung: Yeah, we actually have "preview" features added into heat to enable that13:30
shardywe just havent wired in to them yet13:30
ayoungshardy, thanks.  This is the stuff I need to learn.  I am going to try to use this way of updating the config to do what I was proposing with Dynamic Policy last year13:30
shardyayoung: kk, sounds good - be interested to hear how it goes13:30
ayoungone of the big ones was refferred to in the Heat email thread:  the ability to tag a project as an admin project is there in Mitaka, but we can't change the policy out from under the deployment13:31
shardyayoung: Yeah, it'd be good to revive that discussion and figure out how to (optionally) enable a consistent "admin" policy accross all services13:31
ayoungshardy, the idea is that you tag on project as the admin project, and then the token has an additional field "is_admin_project"=True13:31
ayoungbut the policy files in the remote systems need to be able to honor that.13:32
shardyayoung: is that wired in via oslo context so we can just consume that flag?13:32
ayoungAnd deploying that has to be a deliberate decision right now, or we break things13:32
shardyor do we need to explicitly extract if from what auttoken gives us?13:32
ayoungshardy, uh...good question13:32
ayoungI think it works as is13:32
ayounglet me confirm...13:32
shardyayoung: Ok, I won't get to the heat part of integrating with that until Newton now, but will add it back on my todo list if it's all ready to go13:33
ayoungshardy, ++  that is the deal with all the services.  We just got the feature into Mitaka on the Keystone side.13:33
ayoungshardy, does oslo context expand all of the token data?  I thought it did13:34
shardyayoung: nice - if you felt like brain-dumping a quick summary of how it's supposed to work into this bug, that would be super-helful when I revive the heat work:13:34
openstackLaunchpad bug 1466694 in heat "admin role can't delete other tenants stacks" [Medium,Triaged] - Assigned to Steven Hardy (shardy)13:34
*** jdob has quit IRC13:34
*** jdob has joined #tripleo13:34
*** beagles has joined #tripleo13:35
ayoungshardy, actually, we should sit down and talk policy needs for heat at some point.  How about we schedule something video-wise after the release but before the summit?13:35
shardyayoung: sure, that would be good13:35
shardyayoung: I don't see anything in context.RequestContext for is_admin_project, only is_admin13:36
shardyso we may need to wire that in13:36
ayoungshardy, cool.  I want to test out the policy redeploy for Nova anyway, so this will help flesh it out and flush out the bugs\13:36
ayoungshardy,  'tenant': self.tenant,  really?  REALLY?13:37
openstackgerritMerged openstack/tripleo-common: Use a common constant for the Heat timeout value  https://review.openstack.org/29415013:37
*** paramite|afk is now known as paramite13:39
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: IPv6: dual-stack support for Keystone  https://review.openstack.org/28634413:40
*** saneax is now known as saneax_AFK13:40
*** pradk_ has joined #tripleo13:44
*** pradk has quit IRC13:44
*** pradk_ is now known as pradk13:44
adarazscan somebody help me debug some elaborate looking IPv6+pacemaker+rabbitmq issue? not sure who is the resident rabbitmq magician. chem? I remember you were playing with new erlang builds for the backport stuff. can you maybe take a look? I'm a bit stuck.13:44
adarazsI have an environment where I managed to reproduce the problem consistently and it's up currently. very recent trunk.13:45
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Use 'deeper' hiera_hash merge behaviour for all roles  https://review.openstack.org/28182213:46
gfidentejaosorior ^^13:46
jaosoriorgfidente: Cool, added a comment13:49
*** trozet has joined #tripleo13:49
gfidentejaosorior wait I don't get it13:51
gfidenteoh behavior13:52
*** liverpooler has quit IRC13:52
gfidentenot behaviour13:52
dmsimardtrown: +8 now, I think we're going to break some record13:52
gfidentesorry !13:52
*** qasims_ has quit IRC13:52
*** qasims_ has joined #tripleo13:53
*** dustins has joined #tripleo13:55
*** ohamada has quit IRC13:56
dprincemeeting time in 1 minute13:59
*** akuznetsov has joined #tripleo14:00
*** ohamada has joined #tripleo14:00
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Use 'deeper' hiera_hash merge behavior for all roles  https://review.openstack.org/28182214:01
chemadarazs: ah, good, where it is ?14:01
adarazschem: I think I figured it out! so the problem is this: Mar 22 13:59:09 localhost crmd[11961]:  notice: overcloud-controller-1-rabbitmq_start_0:746 [ Error: unable to connect to nodes ['rabbit@overcloud-controller-0']: nodedown\n\nDIAGNOSTICS\n===========\n\nattempted to contact: ['rabbit@overcloud-controller-0']\n\nrabbit@overcloud-controller-0:\n  * unable to connect to epmd (port 4369) on overcloud-controller-0: nxdomain (non-existing dom14:02
adarazsnode name: 'rabbitmqctl10213@overcloud-controller-1'\n- home dir: /var/lib/rabbitmq\n- cookie hash: Vl7t+6LKdH65klK+3LpaKg==\14:02
*** apetrich has joined #tripleo14:02
adarazschem: it seems like we switched to use hostnames...14:02
adarazsbut there's no host defined as overcloud-controller-0 in /etc/hosts14:03
chemadarazs: for what it's worth, rabbitmq is *very* sensitive to hostname issue14:03
chemadarazs: hostname and hostname -f  should spill out coherent result14:03
chemadarazs: then it's a core to restart the rabbimq properly (remove cookie and kill epmd)14:04
openstackgerritSagi Shnaidman proposed openstack-infra/tripleo-ci: Tempest for tripleo CI  https://review.openstack.org/29584414:04
adarazschem: that's fine. but "overcloud-controller-1" doesn't know who "overcloud-controller-0" is.14:04
adarazslet me try to add it to the hosts file and see if it helps...14:05
chemadarazs: ah, oki, I get it14:05
*** links has joined #tripleo14:05
chemadarazs: out of curiosity how did you set up your env, did you follow a particular documentation/script ?14:05
EmilienMchem: it's meeting time, don't expeect much activity14:06
*** apetrich has quit IRC14:06
adarazschem: it's a full reproducing of the new IPv6 job with HA and pacemaker. https://review.openstack.org/#/c/289445/ -- this patch.14:06
*** apetrich has joined #tripleo14:06
adarazschem: so it's the latest trunk deployed with the args on line 68; https://review.openstack.org/#/c/289445/21/toci_gate_test.sh14:07
adarazsi'll try to figure out what's wrong and send a patch :)14:07
chemadarazs: humm .. tried to deploy this beast in using instack-virt-setup but what missing the added interfaces (nic1 to nic5), do you use instack ?14:09
adarazschem: umm, I'm using some environment set up with devtest_testenv.sh -- provided by derekh.14:10
chemadarazs: can you paste it ? or is it available somewhere ?14:10
adarazschem: you can to log in and look around?14:11
adarazs*you want to14:11
*** akuznetsov has quit IRC14:11
*** ebalduf has joined #tripleo14:12
chemadarazs: oki, whichever, the script or log somewhere and get the script or anything that bootstrap the proper requirement to have the ci script working properly would be awesome :)14:12
adarazschem: so, I don't completely understand yet how this works. Derek pointed me to this script: http://git.openstack.org/cgit/openstack/tripleo-incubator/tree/scripts/devtest_testenv.sh14:13
chemadarazs: awesome, thanks :)14:13
adarazschem: I want to figure out how to recreate the environment, I'll write a blog post about it as soon as I'm able to :)14:14
chemadarazs: ack :)14:14
*** akuznetsov has joined #tripleo14:17
*** dmacpher has joined #tripleo14:20
*** lucas-hungry is now known as lucasagomes14:21
*** mcornea has quit IRC14:24
*** mcornea has joined #tripleo14:25
*** mgould has quit IRC14:28
*** TSCHAK has joined #tripleo14:29
*** olap has quit IRC14:31
*** openstackgerrit has quit IRC14:33
*** openstackgerrit has joined #tripleo14:34
*** mgould has joined #tripleo14:35
*** ohamada has quit IRC14:35
*** rook has joined #tripleo14:41
rookHey all - I had a Mitaka deploy fail using TripleO, curious if someone can tell me what needs to be updated with :14:41
rook2016-03-22 14:20:12 [overcloud]: CREATE_FAILED Resource CREATE failed: resources.NetworkConfig: resources.Compute.resources[0].Property error: resources.OsNetConfigImpl.properties.config: The Parameter (ControlPlaneDefaultRoute) was not provided.14:41
beaglesrook: just a heads up that the weekly tripleo meeting is on the go right now, so might take a second to get a response14:42
*** links has quit IRC14:42
beagless/second/few minutes/14:43
*** bvandenh has joined #tripleo14:43
rookbeagles: roger.14:45
*** absubram has joined #tripleo14:46
beaglesrook: out of curiousity - what did your deployment command look like?14:46
*** lazy_prince has quit IRC14:48
*** pkovar has joined #tripleo14:48
beaglesrook: I figured that particular value actually came with defaults, so unless something else was going on with it in your deployment, where else the value might have "disappeared"14:49
*** ohamada has joined #tripleo14:49
sshnaidmdoes anybody know what the "noop" from Jenkins means here? https://review.openstack.org/#/c/295844/14:52
pradkshardy, slagle, seems like this could be related to ha failure we're seeing? https://bugs.launchpad.net/tripleo/+bug/155896214:53
openstackLaunchpad bug 1558962 in tripleo "Error: Could not find dependent Service[redis] for Exec[cp -p /etc/redis.conf.puppet /etc/redis.conf] at /etc/puppet/modules/redis/manifests/config.pp" [Undecided,Triaged]14:53
trownsshnaidm: that just lets you know that your change is queued up in CI14:54
sshnaidmtrown, oh, thanks14:54
rookbeagles:  https://gist.githubusercontent.com/jtaleric/8764f3c00995f5438d59/raw/8d3db8707a29da6fb1085bf87a50d40b09397732/gistfile1.txt14:55
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Environment Configuration error handling  https://review.openstack.org/29588014:58
shardypradk: it does look possibly related - lets recheck against it14:59
*** lazy_prince has joined #tripleo14:59
openstackgerritJiri Tomasek proposed openstack/tripleo-ui: Plans Dropdown in Deployment Plan page  https://review.openstack.org/29588215:00
openstackgerritMerged openstack/tripleo-heat-templates: Remove GlanceRegistry from EndpointMap  https://review.openstack.org/28981215:00
openstackgerritMarios Andreou proposed openstack-infra/tripleo-ci: Adds the new '-y' required for stack-deletion when tty  https://review.openstack.org/29576115:11
mariosbnemec: updated thanks! ^15:11
mariosbnemec: so i see https://review.openstack.org/#/c/295047/ is older - am happy to abandon mine if it is preferable to land /#/c/295047/ (assming it gets updated though)15:14
bnemecmarios: Yours works, that one doesn't right now. :-)15:16
mariosbnemec: k wfm15:17
adarazschem: any idea why this might be happening/15:19
adarazschem: http://fpaste.org/343762/58659961/ ?15:19
chemadarazs: I'm in another meeting please give me ~5/10min15:20
adarazschem: actually the rabbitmq is running on controller-2.15:20
adarazschem: sure.15:20
*** mcornea_ has joined #tripleo15:22
*** mcornea has quit IRC15:25
*** mcornea_ is now known as mcornea15:26
*** rodrigods has quit IRC15:27
*** rodrigods has joined #tripleo15:27
openstackgerritBen Nemec proposed openstack/os-collect-config: Switch to oslo.log again  https://review.openstack.org/29590315:31
*** bvandenh has quit IRC15:35
*** yamahata has quit IRC15:38
openstackgerritBen Nemec proposed openstack/os-collect-config: Switch to oslo.log again  https://review.openstack.org/29590315:40
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Fix the path to get_host_info.sh  https://review.openstack.org/29591315:44
derekhIf anybody has notice host_info.txt missing on the overcloud nodes ^^15:45
derekhbnemec: if you have a sec, can you take another look at https://review.openstack.org/#/c/238414/ please15:47
beaglesrook: sorry for the noise and I admit it isn't too likely that I'll have you answer but maybe the other guys will jump in - but am I right in guessing that what you have there worked in liberty and you've tried against Mitaka and it no longer works. Judging from your initial question you suspect something has changed that you need to accomodate to get the deploy to work. Is this pretty much it?15:47
beaglesrook: fwict a plausible scenario is that AFAICT there was a potential source of a default value for the ControlPlaneDefaultRoute in liberty that no longer appears to exist in mitaka. If you aren't setting it yourself in the 'network-environment.yaml' environment file you are passing in the command line, than maybe that's what is going on.15:48
openstackgerritMerged openstack-infra/tripleo-ci: Add mirror server  https://review.openstack.org/23841415:51
*** qasims_ has quit IRC15:52
derekhbnemec: thanks, I'll redeploy a new mirror server with that now, and makes sure everything else still works before removing my WIP off the next patch15:52
bnemecderekh: Sounds good, thanks15:53
pradkshardy, slagle, did we recently add redis auth to tripleo?15:55
slaglepradk: i think so15:57
pradkshardy, slagle, yea looks like its not setup correctly  .. https://bugzilla.redhat.com/show_bug.cgi?id=1319916 and https://bugzilla.redhat.com/show_bug.cgi?id=1319916 .. i think this could be why ha is failing15:57
openstackbugzilla.redhat.com bug 1319916 in rhel-osp-director "All Redis HAProxy backends are marked as down" [Urgent,New] - Assigned to mandreou15:57
slagleit was added in I373f97ada4e4101700a12b42dfb8ee4b2ff701f215:58
openstackgerritJason E. Rist proposed openstack/tripleo-ui: TripleO Logo for Login and Navbar  https://review.openstack.org/29592115:58
gfidenteoh the haproxy check!15:59
jristwoot woot15:59
pradkslagle, i think we might have to revert or fix it quickly.. none of the coordination works currently16:01
*** ohamada has quit IRC16:01
slaglepradk: i think gfidente is working on it16:03
shardystrangely that did pass the HA job before it landed16:03
slagleperhaps the error is only triggered by aodh16:04
pradkyea as aodh needs coordination16:04
openstackgerritPradeep Kilambi proposed openstack/tripleo-heat-templates: Fix redis coordination url  https://review.openstack.org/29592816:05
pradk^^ that fixes the coordination url16:05
pradkgfidente, are you looking into fixing haproxy config?16:05
gfidentepradk yes I think that's the only issue16:06
*** apetrich has quit IRC16:06
gfidenteyou think the actual url is wrong?16:06
hewbroccaoh hmm I wonder if this is the ceilo bug we just assigned to marios16:06
pradkyea the format is incorrect16:06
pradkhewbrocca, yep it came up in our ceilo product call16:06
pradkgfidente, seems like there are two issues.. haproxy config is not correct either16:07
hewbroccaRHBZ 1320036, 1319916 ?16:07
pradkgfidente, and the url is incorrectly formatted16:07
pradkhewbrocca, yes sir16:07
hewbroccamarios: looks like these guys are doing your work for you :D16:07
adarazschem: sorry to bother (ignore it if you're still busy)... I looked into it a bit more and  it's interesting.16:07
adarazsthe other rabbitmq nodes are complaining when they are trying to connect to the first one that its domain doesn't exist. but I'm not sure if the problem is that or if the problem is that epmd is not listening on ipv6. though rabbitmq is configured in /etc/rabbitmq/rabbitmq.config to listen on the IPv6 address.16:07
chemadarazs: no problem, meeting just ended 5min ago ... was a long one16:08
openstackgerritMerged openstack/diskimage-builder: Remove all interface configs for simple-init  https://review.openstack.org/29487116:08
lazy_princeshardy: Hi.. Were you able to test https://review.openstack.org/#/c/287784 ..?16:08
chemadarazs: can I log somewhere, to look around ?16:08
adarazschem: can you send me your pubkey16:09
adarazsderekh: I hope you don't mind if I add his key to that machine so that chem can help me debug a rabbitmq issue.16:09
derekhadarazs: go for it16:10
pradkgfidente, the haproxy config needs to be something like what mehdi suggested https://bugzilla.redhat.com/show_bug.cgi?id=1319916#c316:10
openstackbugzilla.redhat.com bug 1319916 in rhel-osp-director "All Redis HAProxy backends are marked as down" [Urgent,New] - Assigned to mandreou16:10
adarazschem: FYI there's this magic thing that github does that this works: https://github.com/adarazs.keys16:10
adarazschem: not sure what's your username.16:10
pradkgfidente, fixed the coordination url that seems to take care of tooz errors16:10
shardylazy_prince: Sorry, I've not tried it yet, but I had a question - this only works for VM according to building_an_image.rst?16:10
marioshewbrocca: heh, well i am chasing another haproxy related (upgrades) thing atm so no objection ;)16:11
chemadarazs: hold on one second for the key I'm updating it (old ssh-dss... bad) , (my usual username is chem, but whatever you want)16:11
*** yamahata has joined #tripleo16:12
*** mcornea has quit IRC16:12
adarazschem, oh, I meant for username like https://github.com/<username>.keys :) so I guess it's https://github.com/chem.keys ; let me know when you updated it.16:12
lazy_princeshardy: it works with physical servers too.. just that the UEFI Image is in disk image.. not a partition image.. Hence VM element is a requirement..16:13
rookbeagles: right, the templates I used for Liberty worked just fine16:13
chemadarazs: -> https://github.com/sathlan.keys (sathlan on github, I'm becoming schizophrenic)16:14
*** aufi has quit IRC16:14
shardylazy_prince: Ok, I'll try to give it a test tomorrow - I was confused because TripleO only uses the "baremetal" element, not the "vm" one16:14
shardywhich I know because I had to hack stuff to get an image that worked on VMs recently16:15
lazy_princewell.. for partition images, Ironic should setup the boot-loader during provisioning. But Images will not be on par with functionality.. hence we posted the patch to bring them on par with each other..16:17
lazy_princes/But Images/But Disk Images/16:17
*** hjensas has quit IRC16:17
shardylazy_prince: Ok, I hit issues specific to booting a baremetal uefi box deployed via ironic, so it may well be that still doesn't work16:18
*** jaosorior has quit IRC16:18
lazy_princeshardy: with the above patch, the images should work...16:19
shardylazy_prince: Ok, thanks - sorry for the delay getting to it, I'll try to carve out some time to give it a try soon16:20
*** akuznetsov has quit IRC16:20
lazy_princeshardy: no hurry.. just wanted to check if you were able to spend time on it..16:21
*** bswartz has quit IRC16:21
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Allow the redis custom monitor to use authentication  https://review.openstack.org/29593716:24
mgouldhi everyone, is there a known problem with Mongo in the tripleo-ci gate?16:25
*** paramite has quit IRC16:28
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Allow the redis custom monitor to use authentication  https://review.openstack.org/29593716:31
*** mikelk has quit IRC16:35
openstackgerritMerged openstack/os-net-config: Use assertTrue/False instead of assertEqual(T/F)  https://review.openstack.org/26460716:44
pradkEmilienM, can you take a peek at https://review.openstack.org/295944 asap .. we want this in mitaka16:44
EmilienMpradk: yeah it's in my backlog already16:44
EmilienMpradk: it will be in mitaka16:44
openstackgerritMerged openstack/tripleo-docs: remove python 2.6 trove classifier  https://review.openstack.org/26084716:45
*** akshai has joined #tripleo16:45
EmilienMpradk: do we need to drop --skip-storage when doing upgrade on an existing deployment?16:46
*** tosky has quit IRC16:46
pradkEmilienM, if we want the storage up then yea, but we can do that as part of the upgrade migration scripts?16:47
EmilienMpradk: it should be done via puppet and nothing else16:47
*** tosky has joined #tripleo16:47
EmilienMpradk: tripleo does bash scripts but lot of people does upgrades only with puppet16:47
*** ccamacho is now known as ccamacho|afk16:48
EmilienMpradk: imo, your patch is an hack16:48
EmilienMour CI does that: https://github.com/openstack/puppet-openstack-integration/blob/master/manifests/gnocchi.pp#L3816:48
pradkEmilienM, so should be expose an option to include if its an upgrade then?16:48
EmilienMpradk: so ceph cluster is ready when gnocchi run db-upgrade16:48
EmilienMpradk: if you look at my link, Exec['create-gnocchi']  is an Exec resource in puppet-ceph, that creates ceph pool for gnocchi16:49
EmilienMif you need that in tripleo, I suggest we make some ordering16:49
*** qasims has joined #tripleo16:50
EmilienMpradk: what is the impact of disabling storage during dbsync?16:50
pradkEmilienM, right but in tripleo we want swift and it wont work until the users/services are created in keystone which is at the end16:50
openstackgerritMerged openstack/tripleo-heat-templates: Remove CephStorageCount  https://review.openstack.org/29574016:50
pradkEmilienM, dn sync without this option will fail and halt deploy16:50
pradkEmilienM, until puppet manages the keystone resource16:50
pradkEmilienM, which i assume wont get into mitaka16:51
EmilienMpradk: what is the impact of putting this option I mean?16:51
*** pcaruana has quit IRC16:52
*** akshai has quit IRC16:52
*** akshai has joined #tripleo16:52
beaglesrook: since nobody is leaping to the rescue ;) two things, can you check whether this file exists on your undercloud VM: /usr/share/openstack-tripleo-heat-templates/environments/network-environment.yaml and also post version of the openstack-tripleo-heat-templates rpm you are using16:53
pradkEmilienM, the storage wont be upgraded by default, we will need this option during upgrades i think .. this will be a non issue if gnocchi user is created by this time16:53
rookbeagles: i modifed the templates and redeployed16:53
rookill see if I made a good guess :)_16:53
beaglesrook: I have an older undercloud vm that was kicking around from earlier this month and it is missing that file16:53
beaglesrook: ack16:53
pradkEmilienM, what if i expose a param say upgrade_storage = in puppet?16:54
*** akshai_ has joined #tripleo16:54
openstackgerritMerged openstack/tripleo-heat-templates: Add BondInterfaceOvsOptions parameter to net-config-bond.yaml  https://review.openstack.org/24508616:54
EmilienMpradk: if set to true by default, yes16:54
pradkEmilienM, yea we could and in tripleo set to false when we invoke?16:55
pradkcool i'll update the patch, seems like a good compromise16:55
EmilienMpradk: right16:56
*** mgould has quit IRC16:58
*** qasims has quit IRC16:58
openstackgerritHonza Pokorny proposed openstack/tripleo-ui: Add LICENSE  https://review.openstack.org/29595716:58
*** akshai has quit IRC16:58
openstackgerritGiulio Fidente proposed openstack/tripleo-heat-templates: Wire redis_password to the tripleo module  https://review.openstack.org/29595817:00
*** trown is now known as trown|lunch17:01
*** lucasagomes is now known as lucas-afk17:02
*** Guest91251 has quit IRC17:02
*** pkovar has quit IRC17:02
*** mgould has joined #tripleo17:13
openstackgerritGiulio Fidente proposed openstack/puppet-tripleo: Allow the Redis specific monitor to use authentication  https://review.openstack.org/29593717:15
*** dprince has quit IRC17:17
openstackgerritDan Sneddon proposed openstack/os-net-config: Fix order-of-operations bug in os-net-config restart_interfaces  https://review.openstack.org/29142017:18
*** bswartz has joined #tripleo17:20
*** mbound has quit IRC17:27
*** akuznetsov has joined #tripleo17:29
*** shivrao has joined #tripleo17:32
*** lucas-afk is now known as lucasagomes17:34
*** ifarkas has quit IRC17:36
*** beagles has left #tripleo17:38
openstackgerritPradeep Kilambi proposed openstack/tripleo-heat-templates: Fix redis coordination url  https://review.openstack.org/29592817:43
*** pcaruana has joined #tripleo17:44
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: ssh less  https://review.openstack.org/29597317:45
*** derekh has quit IRC17:45
*** akrivoka has quit IRC17:48
*** akrivoka has joined #tripleo17:48
*** beagles has joined #tripleo17:50
*** ebalduf_ has joined #tripleo17:53
*** trown|lunch is now known as trown17:54
*** mcornea has joined #tripleo17:54
*** gfidente has quit IRC17:55
*** apetrich has joined #tripleo17:56
*** klindgren has joined #tripleo17:56
*** lucasagomes is now known as lucas-dinner17:57
*** tosky has quit IRC17:57
*** dprince has joined #tripleo17:57
*** d0ugal has quit IRC17:58
*** d0ugal has joined #tripleo17:59
*** mcornea has quit IRC18:00
*** tosky has joined #tripleo18:01
*** akuznetsov has quit IRC18:03
*** akuznetsov has joined #tripleo18:04
*** panda has quit IRC18:04
*** panda has joined #tripleo18:05
*** d0ugal has quit IRC18:09
*** akuznetsov has quit IRC18:09
*** akrivoka has quit IRC18:11
*** mkovacik__ has quit IRC18:12
*** mbound has joined #tripleo18:20
*** apetrich has quit IRC18:23
*** dmacpher is now known as dmacpher-afk18:26
*** shardy has quit IRC18:32
*** Marga_ has quit IRC18:36
*** jtomasek has quit IRC18:37
*** ccamacho|afk is now known as ccamacho18:38
openstackgerritKeith Schincke proposed openstack/python-tripleoclient: Add support for CephRgwKey generation  https://review.openstack.org/29600218:43
openstackgerritKeith Schincke proposed openstack/tripleo-heat-templates: Add support for deploying RGW on controllers.  https://review.openstack.org/28902718:44
*** mbound has quit IRC18:46
colonwqEmilienM, I am much closer on 289027 but have a question about one issue I know about.18:46
*** athomas has quit IRC18:46
*** mburned is now known as mburned_out18:50
*** mgould has quit IRC18:54
*** tosky has quit IRC18:57
*** Marga_ has joined #tripleo19:02
openstackgerritMerged openstack/instack-undercloud: Secure haproxy stats endpoint  https://review.openstack.org/29544419:02
bnemecEasy +A: https://review.openstack.org/#/c/295913/19:02
bnemecAnd \o/19:02
openstackgerritMerged openstack-infra/tripleo-ci: Fix the path to get_host_info.sh  https://review.openstack.org/29591319:02
bnemecslagle is on a roll :-)19:03
openstackgerritKeith Schincke proposed openstack/python-tripleoclient: Add support for CephRgwKey generation  https://review.openstack.org/29600219:04
openstackgerritAttila Darazs proposed openstack-infra/tripleo-ci: WIP: Use IPv6 on the gate jobs  https://review.openstack.org/28944519:06
mgagneWhen running DIB on Ubuntu Trusty, we found /dev/shm to be broken. It is a symlink to /run/shm which looks to be unavailable in the image. We are trying to run Ansible which relies on semaphore and it fails with this error: http://paste.openstack.org/show/491497/19:07
mgagnehow can this be fixed? should /run be mount --bind too in the image?19:07
dmsimarddprince or slagle might know ^19:09
EmilienMcolonwq: go ahead and ask your question if you need help.19:10
EmilienMno, I think pabelanger worked on it recently19:11
EmilienMbut he's not on this channel19:11
pradkEmilienM, see if this helps https://review.openstack.org/#/c/295944/19:11
*** electrofelix has quit IRC19:12
colonwqEmilienM, (and anyone else) line 80 of ceph_cluster-config.yaml, I tried to set ceph_rgw_key to ceph_rgw_key but got an incorrect value.19:13
mgagnelooks like eglibc is hardcoded to use /dev/shm19:14
*** ebalduf has quit IRC19:14
mgagneso I guess not much can be done unless /run is mount --bind too or host remounts /dev/shm itself19:15
EmilienMpradk: +219:15
*** trozet has quit IRC19:16
pradkEmilienM, ty sir19:17
openstackgerritMerged openstack/tripleo-specs: tripleo-quickstart  https://review.openstack.org/27681019:19
*** mcornea has joined #tripleo19:21
*** trozet has joined #tripleo19:31
openstackgerritRyan Hallisey proposed openstack-infra/tripleo-ci: Allow the container job to run again  https://review.openstack.org/28891519:33
*** mburned_out is now known as mburned19:39
*** mcornea has quit IRC19:40
*** Goneri has quit IRC19:47
*** david-lyle_ has joined #tripleo19:56
*** david-lyle has quit IRC19:57
*** dshulyak has quit IRC19:58
*** lynxman has quit IRC19:58
*** lynxman has joined #tripleo19:59
*** lynxman has quit IRC19:59
*** lynxman has joined #tripleo19:59
*** nico_auv has quit IRC20:00
*** david-lyle_ is now known as david-lyle20:02
*** dprince has quit IRC20:02
*** jayg is now known as jayg|g0n320:23
*** rdopiera has quit IRC20:26
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: ssh less  https://review.openstack.org/29597320:31
*** ccamacho has quit IRC20:44
*** ccamacho has joined #tripleo20:45
*** MaxPC has quit IRC20:53
*** trozet_ has joined #tripleo20:59
*** trozet has quit IRC20:59
*** lblanchard has quit IRC21:01
*** dustins has quit IRC21:03
*** jcoufal has quit IRC21:10
*** trown is now known as trown|outtypewww21:10
EmilienMmichchap_: doing glance split now21:20
michchap_EmilienM: I'm going to try to repro the keystone ha issue once my meetings are done for the day21:21
EmilienMmichchap_: do you have all pacemaker constraints? look at https://review.openstack.org/#/c/236243/21/puppet/manifests/overcloud_controller_pacemaker_5.pp21:23
michchap_EmilienM: I don't have all of them, but I didn't move all of them. I only moved rabbit and haproxy.21:26
openstackgerritJames Slagle proposed openstack/tripleo-heat-templates: Fix satellite registration for http or https  https://review.openstack.org/29607121:28
openstackgerritPradeep Kilambi proposed openstack/tripleo-heat-templates: Fix redis coordination url  https://review.openstack.org/29607421:32
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: Add Glance profiles  https://review.openstack.org/29607621:33
EmilienMmichchap_: ^21:33
openstackgerritMerged openstack/tripleo-heat-templates: Fix redis coordination url  https://review.openstack.org/29592821:35
EmilienMmichchap_: feel free to review it when you have time21:37
michchap_EmilienM: I left a quick note - just to move the hiera() calls to parameters so we have the interface locked down21:37
EmilienMmichchap_: makes sense.21:38
*** morazi has quit IRC21:41
*** ebalduf_ has quit IRC21:48
*** openstackgerrit has quit IRC21:48
*** openstackgerrit has joined #tripleo21:49
openstackgerritEmilien Macchi proposed openstack/puppet-tripleo: Add Glance profiles  https://review.openstack.org/29607621:51
EmilienMmichchap_: should be better now ^21:51
*** jprovazn has quit IRC22:01
*** panda has quit IRC22:04
*** panda has joined #tripleo22:05
*** ebalduf has joined #tripleo22:07
*** trozet_ has quit IRC22:08
openstackgerritDerek Higgins proposed openstack-infra/tripleo-ci: Use the mirror server for the centos image and git repos  https://review.openstack.org/28525722:10
*** ccamacho has quit IRC22:23
*** r-mibu has quit IRC22:30
*** r-mibu has joined #tripleo22:31
*** dmacpher-afk has quit IRC22:41
*** ayoung has quit IRC22:43
*** absubram has quit IRC22:47
*** rlandy has quit IRC22:49
*** dmacpher-afk has joined #tripleo22:53
*** saneax_AFK is now known as saneax23:05
*** Erming_ has quit IRC23:47

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