Perl 6 - the future is here, just unevenly distributed

IRC log for #puppet-openstack, 2016-12-13

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:01 ducttape_ joined #puppet-openstack
00:03 EmilienM mwhahaha: yes, we might want to set it to false first. So we can backport it and then we switch poi to use it and then change to true eventually
00:04 EmilienM we'll also need tripleo & fuel with latest packages eventually
00:04 mwhahaha yea
00:04 mwhahaha it's going to be one of those weeks
00:04 mwhahaha also it's release week :D
00:05 * mwhahaha wanders off
00:05 EmilienM oh that
00:45 openstackgerrit zhongshengping proposed openstack/puppet-glare: Added retries for db_sync  https://review.openstack.org/409611
00:59 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
01:11 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
01:22 ducttape_ joined #puppet-openstack
01:22 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
01:43 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
02:19 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Nova Placement API support  https://review.openstack.org/406300
02:19 EmilienM mwhahaha: ready for review again ^
02:19 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: nova: activate placement API service  https://review.openstack.org/406301
02:23 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
02:48 ilbot3 joined #puppet-openstack
02:48 Topic for #puppet-openstack is now CI status: green - Puppet OpenStack modules http://docs.openstack.org/deve​loper/puppet-openstack-guide/
02:56 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
03:23 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
03:26 ducttape_ joined #puppet-openstack
03:50 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
03:50 openstackgerrit Merged openstack/puppet-nova: Add the enable_new_services option to nova::conductor  https://review.openstack.org/409954
04:42 ducttape_ joined #puppet-openstack
04:51 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
05:02 snarwade joined #puppet-openstack
05:08 agrebennikov_ joined #puppet-openstack
05:17 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
05:39 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
05:45 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
05:57 nmathew joined #puppet-openstack
05:58 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
06:00 ducttape_ joined #puppet-openstack
06:07 aruns joined #puppet-openstack
06:19 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
06:22 zhongshengping joined #puppet-openstack
06:42 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
07:09 fxpester joined #puppet-openstack
07:09 imcsk8_ joined #puppet-openstack
07:10 tdb joined #puppet-openstack
07:20 rasca joined #puppet-openstack
07:31 ducttape_ joined #puppet-openstack
07:37 openstackgerrit OpenStack Proposal Bot proposed openstack/puppet-openstack-integration: Updated from Puppet OpenStack modules constraints  https://review.openstack.org/410070
07:50 openstackgerrit zhongshengping proposed openstack/puppet-cloudkitty: Add the acceptance test to deploy the cloudkitty service  https://review.openstack.org/409693
07:54 mkoderer joined #puppet-openstack
07:59 ccamacho joined #puppet-openstack
07:59 jaosorior joined #puppet-openstack
08:05 rcernin joined #puppet-openstack
08:09 basilAB joined #puppet-openstack
08:12 jaosorior joined #puppet-openstack
08:13 openstackgerrit Merged openstack/puppet-murano: Generate separate db for murano service broker  https://review.openstack.org/405065
08:25 rcernin joined #puppet-openstack
08:33 aruns joined #puppet-openstack
08:55 astupnikov joined #puppet-openstack
08:57 jpich joined #puppet-openstack
08:57 rmart04 joined #puppet-openstack
08:57 chem joined #puppet-openstack
09:01 ducttape_ joined #puppet-openstack
09:04 shardy joined #puppet-openstack
09:09 tdb joined #puppet-openstack
09:14 fxpester joined #puppet-openstack
09:21 chem joined #puppet-openstack
09:29 mkarpin joined #puppet-openstack
09:34 gfidente joined #puppet-openstack
09:41 derekh joined #puppet-openstack
09:59 paramite joined #puppet-openstack
10:10 openstackgerrit Merged openstack/puppet-openstack-integration: Pin older openstack-selinux  https://review.openstack.org/409008
10:32 ducttape_ joined #puppet-openstack
11:05 pester joined #puppet-openstack
11:08 openstackgerrit Xing Zhou proposed openstack/puppet-openstacklib: Add support when creating multiple databases with same db username  with puppet-openstacklib  https://review.openstack.org/410166
11:14 richm joined #puppet-openstack
11:31 openstackgerrit Juan Antonio Osorio Robles proposed openstack/puppet-heat: Add manifests to deploy APIs over httpd  https://review.openstack.org/394835
11:39 ntpttr joined #puppet-openstack
12:02 ducttape_ joined #puppet-openstack
12:07 nmathew joined #puppet-openstack
12:21 dgurtner joined #puppet-openstack
12:23 dprince joined #puppet-openstack
12:36 ducttape_ joined #puppet-openstack
13:11 johnmilton joined #puppet-openstack
13:33 pradk joined #puppet-openstack
13:47 openstackgerrit Flavio Percoco proposed openstack/puppet-glance: Don't pass glance-registry.conf in every sync_db  https://review.openstack.org/408554
13:55 aruns joined #puppet-openstack
13:56 iurygregory mwhahaha, when you update https://review.openstack.org/#/c/409476/ can you add Closes-bug: #1604463 ? =)
13:56 openstack bug 1604463 in puppet-mistral "Modules should use keystone::resource::authtoken to configure keystone_authtoken settings" [Medium,In progress] https://launchpad.net/bugs/1604463 - Assigned to Iury Gregory Melo Ferreira (iurygregory)
13:58 ayoung joined #puppet-openstack
14:02 dprince joined #puppet-openstack
14:06 morazi joined #puppet-openstack
14:08 ducttape_ joined #puppet-openstack
14:13 openstackgerrit Giulio Fidente proposed openstack/puppet-ceph: Improves support for ceph-mds  https://review.openstack.org/402385
14:21 dmsimard joined #puppet-openstack
14:22 agrebennikov_ joined #puppet-openstack
14:23 johnmilton joined #puppet-openstack
14:24 mwhahaha iurygregory: k
14:26 openstackgerrit Alex Schultz proposed openstack/puppet-mistral: Move Mistral to authtoken  https://review.openstack.org/409476
14:27 pester joined #puppet-openstack
14:27 morazi joined #puppet-openstack
14:28 dmsimard joined #puppet-openstack
14:30 dmsimard joined #puppet-openstack
14:42 paramite joined #puppet-openstack
14:50 ducttape_ joined #puppet-openstack
14:58 zhongshengping joined #puppet-openstack
15:03 alee joined #puppet-openstack
15:10 nmathew- joined #puppet-openstack
15:25 openstackgerrit Iury Gregory Melo Ferreira proposed openstack/puppet-openstack-guide: Add meeting #101  https://review.openstack.org/410270
15:33 sanagikoki joined #puppet-openstack
15:34 zhongshengping joined #puppet-openstack
15:43 zhongshengping joined #puppet-openstack
15:52 EmilienM iurygregory: do you have a script?
15:54 openstackgerrit Merged openstack/puppet-openstack-guide: Add meeting #101  https://review.openstack.org/410270
15:57 mfisch iurygregory: you around?
15:59 rajinir joined #puppet-openstack
16:00 iurygregory EmilienM, script for what?
16:00 iurygregory meetings?
16:00 iurygregory mfisch, i'm
16:00 _ducttape_ joined #puppet-openstack
16:00 iurygregory sorry i was in lunch time :D
16:01 EmilienM iurygregory: yeah
16:01 iurygregory EmilienM, nope haha
16:01 iurygregory i do ctrl c + v  in the last and just update the dates haha
16:05 mfisch iurygregory: I cant seem to get into gerrit but wondering if you had time to look at authtoken for mistral
16:05 EmilienM we need a puppet module for that
16:06 iurygregory mfisch, oh we had the patch already
16:06 mwhahaha https://review.openstack.org/#/c/409476/
16:06 mfisch thanks
16:06 iurygregory but now is failing in triple own
16:06 mfisch I dont know why I cant get in
16:06 mfisch it says the auth provider is not configured
16:06 mwhahaha clear them cookies :D
16:07 iurygregory ctrl + f5 haha
16:12 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
16:13 jaosorior joined #puppet-openstack
16:16 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
16:19 mwhahaha iurygregory: https://review.openstack.org/#/c/409945/ if you have a second
16:19 iurygregory mwhahaha, sure
16:24 ducttape_ joined #puppet-openstack
16:26 openstackstatus NOTICE: Launchpad SSO is not currently working, so logins to our services like review.openstack.org and wiki.openstack.org are failing; the admins at Canonical are looking into the issue but there is no estimated time for a fix yet.
16:27 ducttape_ joined #puppet-openstack
16:27 Topic for #puppet-openstack is now Launchpad SSO is not currently working, so logins to our services like review.openstack.org and wiki.openstack.org are failing; the admins at Canonical are looking into the issue but there is no estimated time for a fix yet.
16:27 iurygregory mfisch, ^haha
16:27 iurygregory maybe this is your problem XD
16:27 mfisch yep
16:27 mfisch I got 99 problems and Canonical auth is one
16:28 iurygregory omg =(
16:28 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
16:53 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
16:54 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Fix fix region name release note  https://review.openstack.org/410317
16:58 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
17:00 rcernin joined #puppet-openstack
17:02 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Add cell_v2 simple_cell_setup  https://review.openstack.org/409948
17:02 ducttape_ joined #puppet-openstack
17:03 Topic for #puppet-openstack is now CI status: green - Puppet OpenStack modules http://docs.openstack.org/deve​loper/puppet-openstack-guide/
17:03 openstackstatus NOTICE: Canonical admins have resolved the issue with login.launchpad.net, so authentication should be restored now.
17:11 mwhahaha EmilienM: so the cell_v2 error only currently shows up on upgrades
17:11 mwhahaha EmilienM: which is why packages are fine
17:13 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Enable cell_v2 setup by default  https://review.openstack.org/410324
17:21 dmsimard EmilienM: nova placement
17:21 dmsimard you're rechecking a lot and gloriously ignoring my comment ;(
17:21 EmilienM dmsimard: I didn't see it
17:22 markvoelker joined #puppet-openstack
17:22 EmilienM dmsimard: it's a bug in nova, that cdent is working on
17:22 EmilienM placement API works well
17:22 dmsimard ok, do we have any way of testingit ?
17:22 EmilienM and there is no need of tempest test for now, compute works fine. But they are working on adding tempest tests
17:22 dmsimard it*
17:23 dmsimard okay, I saw some functional tests in the tree but it didn't seem like it was using tempest
17:23 EmilienM nope
17:23 EmilienM but that's fine
17:24 EmilienM dmsimard: https://bugs.launchpad.net/nova/+bug/1647316
17:24 openstack Launchpad bug 1647316 in OpenStack Compute (nova) "scheduler report client sends allocations with value of zero, violating min_unit" [Medium,In progress] - Assigned to Ed Leafe (ed-leafe)
17:25 EmilienM you picked a trace which is a bug
17:25 EmilienM but it doesn't block you to spawn a VM
17:25 EmilienM it's on nova side
17:26 EmilienM dmsimard: I actually reported this trace to cdent :D
17:27 mwhahaha is scenario003 broken on centos?
17:27 mwhahaha https://review.openstack.org/#/c/409476/ keeps failing tempest on compute
17:27 mwhahaha for a mistral change
17:34 dmsimard maybe another designate-ish issue ?
17:34 dmsimard where in p-o-i it passes but beaker it doesn't
17:34 * dmsimard looks
17:34 mwhahaha but that's p-o-i
17:35 mwhahaha i'm rechecking a change in p-o-i to see if it's happening there as well
17:35 rcernin joined #puppet-openstack
17:35 dmsimard what do you mean that's p-o-i
17:36 dmsimard oh, yeah it's p-o-i but in puppet-mistral
17:36 dmsimard let me see
17:37 dmsimard mwhahaha: mistral isn't even enabled in p-o-i
17:37 dmsimard https://github.com/openstack/puppe​t-openstack-integration/blob/maste​r/fixtures/scenario003.pp#L17-L34
17:37 mwhahaha i think there's a latent bug in nova somewhere because the qa folks had reported the same error on a tempest change but that was scenario001
17:37 dmsimard yeah it looks like a nova issue from looking at the tempest failures
17:38 dmsimard http://logs.openstack.org/76/409476/4/check​/gate-puppet-openstack-integration-4-scenar​io003-tempest-centos-7/8881991/logs/nova/no​va-compute.txt.gz#_2016-12-13_16_39_50_833
17:38 dmsimard http://logs.openstack.org/76/409476/4/ch​eck/gate-puppet-openstack-integration-4-​scenario003-tempest-centos-7/8881991/log​s/libvirt/qemu/instance-00000001.txt.gz
17:39 dmsimard I think that's the qemu host model issue thing
17:39 dmsimard amoralej: ^ ?
17:39 amoralej yeah
17:39 mwhahaha so what's the fix?
17:39 amoralej you need qemu-kvm-ev 2.6.0
17:39 dmsimard amoralej: so that's fixed when kvm 2.6 lands in 7.3 ?
17:39 amoralej yes
17:40 mwhahaha is this a new thing
17:40 mwhahaha like why is this starting now?
17:40 dmsimard which is signed and pushed so we're just waiting for CDN sync ?
17:40 amoralej only affects to amd, i'd say
17:40 dmsimard mwhahaha: because 7.3 is officially out since yesterday
17:40 dmsimard mwhahaha: and there's a yum update that was added somewhere recently ?
17:40 amoralej nop
17:40 mwhahaha yea we merged that yum update
17:41 amoralej i meant, yes, not only to AMD, :)
17:41 amoralej any processor with arat extension
17:41 dmsimard basically qemu-kvm-ev 2.6.0 was meant to ship at the same time as 7.3 but it looks like it's lagging a bit behind
17:41 amoralej you can, whether wait to get it synced in centos official repos
17:41 amoralej what should happen very soon
17:42 amoralej or enable virtsig testing repo
17:42 amoralej which is what we are doing in rdo-ci
17:56 openstackgerrit David Moreau Simard proposed openstack/puppet-openstack-integration: Pass a generic cpu to qemu instead of trying to pass host-model  https://review.openstack.org/410344
17:57 dmsimard amoralej, mwhahaha: ^
17:57 dmsimard I'll need to add an override on weirdo to maintain nested virt capabilities
18:00 amoralej dmsimard, maybe we could fix it in puppet-nova and set cpu_mode to none as soon as virt_type is qemu,
18:00 dmsimard maybe
18:01 dmsimard mwhahaha: wdyt ? Should we change the puppet-nova default to match devstack config ?
18:01 dmsimard i.e, when qemu, default to 'none' instead of 'host-model'
18:02 dmsimard EmilienM: does tripleo jobs use qemu or kvm ?
18:03 EmilienM dmsimard: qemu
18:03 EmilienM for multinode
18:03 EmilienM for ovb I'm not sure kvm maybe
18:03 dmsimard what about libvirt_cpu_mode ?
18:04 dmsimard EmilienM: for context, this can also break tripleo: https://review.openstack.org/#/c/410344/ + http://logs.openstack.org/76/409476/4/ch​eck/gate-puppet-openstack-integration-4-​scenario003-tempest-centos-7/8881991/log​s/libvirt/qemu/instance-00000001.txt.gz
18:05 EmilienM dmsimard: can you submit a patch in THT/ci/environments please?
18:06 dmsimard Sure but ihavenoideawhatimdoing.jpg
18:06 EmilienM dmsimard: learn :)
18:26 openstackgerrit Emilien Macchi proposed openstack/puppet-nova: Nova Placement API support  https://review.openstack.org/406300
18:26 EmilienM ok cdent did a last review, should be fine now
18:27 EmilienM dmsimard: so back to the qemu problem. Is it only when qemu is used and not kvm?
18:28 EmilienM if yes, you might just need to patch https://github.com/openstack/tripleo-h​eat-templates/blob/master/ci/environme​nts/scenario001-multinode.yaml#L61-L62
18:28 EmilienM and add the hiera params to configure that
18:28 EmilienM you'll need to do it for all environments in https://github.com/openstack/tripleo-hea​t-templates/blob/master/ci/environments
18:28 EmilienM and also in https://github.com/openstack-infra/tripleo-ci​/blob/master/test-environments/multinode.yaml
18:37 EmilienM dmsimard: what are the symptoms again? I see a lot of tripleo failures, might be related
18:38 dmsimard EmilienM: tempest/nova failing with http://logs.openstack.org/76/409476/4/check​/gate-puppet-openstack-integration-4-scenar​io003-tempest-centos-7/8881991/logs/nova/no​va-compute.txt.gz#_2016-12-13_16_39_50_833 and http://logs.openstack.org/76/409476/4/ch​eck/gate-puppet-openstack-integration-4-​scenario003-tempest-centos-7/8881991/log​s/libvirt/qemu/instance-00000001.txt.gz
18:38 dmsimard with centos 7.3, kvm 2.6, qemu and host-model
18:38 EmilienM http://logs.openstack.org/92/409692/4/chec​k-tripleo/gate-tripleo-ci-centos-7-ovb-ha/​21f0f1e/logs/undercloud/var/log/nova/nova-​compute.txt.gz#_2016-12-13_16_35_36_622
18:38 EmilienM nevermind
18:38 EmilienM my problem is still the same as we had in tripleo CI and we fixed it but patch is still in gate
18:38 dmsimard EmilienM: I'll send patches to modify those lines you sent me.
18:39 EmilienM dmsimard: merci
18:39 * EmilienM lunch
18:55 dmsimard EmilienM: https://review.openstack.org/#/q/topic:kvm-2.6
18:55 EmilienM dmsimard: nice
18:55 dmsimard A question amoralej asked is if it would make sense to change the default provided by puppet-nova
18:55 EmilienM dmsimard: what is the default in Nova?
18:55 dmsimard I don't have a strong opinion
18:56 dmsimard host-model when using qemu
18:56 EmilienM ok so default is broken, right?
18:56 EmilienM as it should be 'none' iiuc
18:56 dmsimard https://github.com/openstack/puppet-nova/bl​ob/master/manifests/compute/libvirt.pp#L138
18:56 dmsimard well, this is a regression in 2.6.0
18:56 dmsimard at least how I understand the current state of things
18:57 dmsimard There's an internal red hat thread about the issue.
18:57 amoralej this is a bug in qemu 2.6.0, but it's probably uncovering a wrong default, IMHO
18:57 EmilienM does it affect newton/mitaka CI too?
18:57 amoralej yes
18:57 dmsimard newton yes, mitaka no
18:58 amoralej why not dmsimard?
18:58 dmsimard we're not forcing qemu >= 2.3 in mitaka
18:58 dmsimard that merged in newton
18:58 dmsimard I guess it would manifest in mitaka if you're using the virt sig, which was not bundled in rdo release in Mitaka.
18:59 amoralej so, we are not enabling virsig repo in p-o-i for mitaka?
18:59 dmsimard not afaik
18:59 amoralej it'd affect for no p-o-i anyway
18:59 dmsimard what ?
19:00 amoralej packstack and tripleo if it uses qemu
19:00 amoralej i mean
19:00 amoralej if we fix it in puppet-nova, we'd fix to all
19:01 amoralej or even in upstream nova
19:01 dmsimard Unless people are actually putting host-model or until it's merged tested and released
19:01 dmsimard released and people actually update their packages*
19:01 dmsimard and in mitaka opm was monolithic
19:01 dmsimard this is a nightmare
19:01 mwhahaha https://github.com/openstack/puppet-nova/blob/​master/manifests/compute/libvirt.pp#L140-L142
19:01 mwhahaha that's the fix
19:01 mwhahaha remove qemu from setting host-model
19:02 EmilienM dmsimard: so we need to backport https://review.openstack.org/#/c/410344/ ?
19:02 dmsimard EmilienM: like amoralej said, there's probably more value in doing the fix in both puppet-nova and elsewhere
19:02 EmilienM mwhahaha: yes indeed, it sounds like it, though behavior might differ on ubuntu
19:03 EmilienM dmsimard: I agree
19:03 dmsimard EmilienM: yeah that's why I'm hesitating a bit, if it's a regression, it's bound to be fixed
19:03 EmilienM can we try https://github.com/openstack/puppet-nova/blob/​master/manifests/compute/libvirt.pp#L140-L142 ?
19:03 EmilienM and see how CI works
19:03 mwhahaha it might but this might be a case of where the modules are oppinionated and it's biting us
19:03 dmsimard yeah but you already approved my patch :P
19:03 dmsimard mwhahaha: we'd also be changing a default of stable release
19:04 mwhahaha dmsimard: well we should fix it for ocata forward at least
19:04 mwhahaha brb
19:06 EmilienM brb too, third attempt to have lunch
19:07 dmsimard mwhahaha, EmilienM: to be fair, puppet-nova is simply emulating what nova is already doing... https://github.com/openstack/nova/blob/mast​er/nova/virt/libvirt/driver.py#L3411-L3419
19:08 dmsimard So one could argue that the proper fix would be in nova and then we could set libvirt_cpu_mode to $::os_service_default or something
19:08 dprince joined #puppet-openstack
19:08 dmsimard just thinking out loud
19:14 rcernin joined #puppet-openstack
19:26 mwhahaha Yea it should be os service default
19:50 openstackgerrit Alex Schultz proposed openstack/puppet-openstack-integration: Enable cell_v2 setup for p-o-i  https://review.openstack.org/410377
19:51 openstackgerrit Alex Schultz proposed openstack/puppet-nova: Enable cell_v2 setup by default  https://review.openstack.org/410324
19:51 mwhahaha lining up those patch sets
20:01 mwhahaha http://logs.openstack.org/77/410377/1/​check/gate-puppet-openstack-integratio​n-puppet-syntax-3-centos-7/330baf0/con​sole.html#_2016-12-13_19_52_43_016108
20:02 mwhahaha ‎(ノಥ益ಥ)ノ ┻━┻
20:02 mwhahaha that is all.
20:06 openstackgerrit Alex Schultz proposed openstack/puppet-openstack_spec_helper: Pin fast_gettext  https://review.openstack.org/410380
20:06 mwhahaha EmilienM: iurygregory -^
20:06 iurygregory mwhahaha, done :D
20:06 iurygregory and you have change your emoji lol
20:06 iurygregory (ノಥ益ಥ)ノ ┻━ hahaha
20:06 EmilienM mwhahaha: +A. Don't forget to backport
20:08 mwhahaha ruby 2.1 is now the bane of my existance and will probably be so for the next several years
20:10 iurygregory #fail
20:10 iurygregory https://review.openstack.org/#/c/410380/
20:11 mwhahaha wat
20:11 mwhahaha guess we need 1.1.0
20:12 openstackgerrit Alex Schultz proposed openstack/puppet-openstack_spec_helper: Pin fast_gettext  https://review.openstack.org/410380
20:12 mwhahaha that's odd, wonder who we skipped from 1.1.0 1.3.0
20:12 mwhahaha someone must have added 1.3.0 as a dep somewhere
20:13 mwhahaha i wish bundler/ruby took ruby version into consideration
20:19 iurygregory gate-puppet-openstack-librarie​s-puppet-beaker-rspec-centos-7 #failure
20:19 mwhahaha because of libvirt?
20:20 mwhahaha nope couldn't install r10k
20:20 mwhahaha of course
20:20 mwhahaha why not
20:21 mwhahaha 2016-12-13 20:17:32.434358 |   ERROR:  Error installing r10k:
20:21 mwhahaha 2016-12-13 20:17:32.434619 |   fast_gettext requires Ruby version >= 2.1.0.
20:21 mwhahaha ok so CI is completely hosed
20:21 * mwhahaha gives up
20:21 iurygregory NOOOOOOOOOO <O>
20:21 mwhahaha if you need me i'll be under my desk crying
20:22 * iurygregory time to leave the office ha
20:24 mwhahaha gettext-setup released today
20:24 mwhahaha so it's probably that
20:24 mwhahaha Hunner: is that a puppet thing?
20:32 mwhahaha EmilienM: http://logs.openstack.org/80/410380/2/c​heck/gate-puppet-openstack-libraries-pu​ppet-beaker-rspec-centos-7/a7fa8bb/cons​ole.html#_2016-12-13_20_17_25_799715 where does this come from?
20:33 mwhahaha oh it comes from puppet-openstack-integration
20:33 * mwhahaha sighs
20:33 dgurtner joined #puppet-openstack
20:33 dgurtner joined #puppet-openstack
20:33 EmilienM mwhahaha: http://git.openstack.org/cgit/openstack/puppet-o​penstack-integration/tree/install_modules.sh#n25
20:33 EmilienM http://codesearch.openstack.org/?q=gem%20ins​tall%20r10k&amp;i=nope&amp;files=&amp;repos=
20:33 mwhahaha so we can't fix spec_helper because p-o-i and we can't fix p-o-i because of spec_helper
20:35 openstackgerrit Alex Schultz proposed openstack/puppet-openstack-integration: Enable cell_v2 setup for p-o-i  https://review.openstack.org/410377
20:36 iurygregory so we are in a loop LOL
20:37 EmilienM in a mtg, i'll look in a bit
20:37 EmilienM but we can still hack something I think
20:38 mwhahaha yea looking at it
20:38 mwhahaha might be able to force something in p-o-i
20:38 openstackgerrit Alex Schultz proposed openstack/puppet-openstack-integration: Pin fast_gettext  https://review.openstack.org/410388
20:43 openstackgerrit Alex Schultz proposed openstack/puppet-openstack-integration: Pin fast_gettext  https://review.openstack.org/410388
20:44 openstackgerrit Alex Schultz proposed openstack/puppet-openstack_spec_helper: Pin fast_gettext  https://review.openstack.org/410380
20:44 mwhahaha that might do it
20:44 iurygregory ok ^^
20:46 mwhahaha why does CI seem to always fall apart every week
21:02 EmilienM back
21:02 EmilienM this TC meeting never ends :)
21:02 EmilienM mwhahaha: +2
21:02 EmilienM let's revert https://review.openstack.org/#/c/410388/ once it's merged in spec_helper
21:02 mwhahaha we can't
21:03 mwhahaha because r10k can't install
21:03 EmilienM ah
21:03 mwhahaha we can partial revert the Gemfile but the other is here to stay
21:03 EmilienM mwhahaha: let's see how CI works and we'll merge it
21:03 mwhahaha yea it looked green so far
21:03 mwhahaha i'm sure we'll jsut hit the libvirt issue
21:03 mwhahaha so i've pretty much given up on anything actually merging today
21:03 EmilienM do we need to align the stars?
21:04 mwhahaha quite possibly, but lets see
21:04 * EmilienM is invoking stars
21:04 * EmilienM makes special dance
21:04 iurygregory hahahha
21:05 mwhahaha and here i was hoping to actually get some work done today
21:05 mwhahaha so far it's been queuing things up and maybe some day they'll merge
21:05 * mwhahaha goes back to queuing things up while we wait
21:08 EmilienM mwhahaha: if things are broken, squash commits
21:09 mwhahaha sure
21:19 openstackgerrit Alex Schultz proposed openstack/puppet-barbican: Add oslo_messaging_notification transport_url  https://review.openstack.org/410400
21:20 iurygregory mwhahaha, i will review the patches after my dinner o/
21:20 mwhahaha iurygregory: thanks
21:20 iurygregory np ^^
21:21 EmilienM so iiuc we need notification_transport_url and default_transport_url
21:22 EmilienM why don't we set notification_transport_url in poi yet?
21:23 mwhahaha i think we do for ceilometer maybe?
21:23 mwhahaha i see in cinder, cinder::ceilometer does the notification_transport_url which is odd
21:24 mwhahaha i have no idea what notification_transport_url is actually used for
21:24 mwhahaha maybe neutron->nova or vice versa might use it
21:25 EmilienM https://review.openstack.org/#/c/233258/
21:25 mwhahaha oh but it inherits from default_transport_url i assume so thats why we never change it
21:25 mwhahaha because generally people aren't deploying two instances for each
21:26 EmilienM where do you see that?
21:26 mwhahaha well i can only assume that's how it's still working :D let me see if i can validate in code
21:26 mwhahaha because i would assume 'DEFAULT/transport_url' would be the fall back for stuff
21:27 mwhahaha https://github.com/openstack/oslo.messaging/b​lob/afaa4d935dd77760492f36dba622f6a01fe14fd9/​oslo_messaging/notify/notifier.py#L142-L154
21:28 mwhahaha yea it falls back
21:28 mwhahaha which is why this is generally not a problem because tripleo/fuel/etc sets a single rabbitmq endpoint
21:28 mwhahaha we aren't deploying multiple rabbit clusters at the moment
21:30 EmilienM mwhahaha: ok so we are good with this param for now
21:30 EmilienM mwhahaha: the only thing we need to solve is the ssl thing
21:30 mwhahaha EmilienM: switch use_ssl to accept whatever then just hard code the true/false openstacklib
21:36 mnaser is it possible that puppet-openstack_infra_spec_helper is broken under centos 7?  http://logs.openstack.org/02/410402/1/chec​k/gate-openstackci-beaker-centos-7/77bd08e​/console.html#_2016-12-13_21_27_28_999218
21:37 mwhahaha mnaser: probably
21:37 mnaser can i update the gemspec to the older release?
21:37 mwhahaha yea sec we're trying to do that ourselves
21:37 mnaser like pin fast_gettext to 1.20
21:37 mnaser oh okay
21:37 mwhahaha https://review.openstack.org/#/c/410380/
21:37 mwhahaha 1.10
21:37 mwhahaha 1.1.0
21:37 mwhahaha 1.2.0 also requires ruby >= 2.1
21:37 mnaser so fast D:
21:38 EmilienM mwhahaha: how can we see if ssl is enabled in the url?
21:38 mwhahaha EmilienM: ssl can be enabled via a query param in the url
21:38 mwhahaha EmilienM: so i'm not sure what you mean
21:38 EmilienM I'll look in your function
21:39 EmilienM I want to patch your function to accept a boolean
21:39 dprince joined #puppet-openstack
21:39 EmilienM does it make sense?
21:39 mwhahaha EmilienM: Yea you can do that it should accept string/bool
21:40 mwhahaha https://github.com/openstack/puppet-op​enstacklib/blob/master/lib/puppet/pars​er/functions/os_transport_url.rb#L72
21:40 mwhahaha might have to some tricky logic there to do it hto
21:41 EmilienM mwhahaha: wdyt about carlos comment about string in https://review.openstack.org/#/c/40​9911/6/manifests/profile/base/nova.pp ?
21:41 EmilienM I think my code is ok
21:41 mwhahaha i guess it depends on how THT outputs it
21:41 mwhahaha not sure if int or string
21:41 mwhahaha the function requires a string
21:41 EmilienM ah right
21:44 openstackgerrit Alex Schultz proposed openstack/puppet-congress: Add oslo_messaging_notification transport_url  https://review.openstack.org/410409
21:46 mwhahaha stupid libvirt
21:46 mwhahaha wheres that libvrit p-o-i change
21:46 dmsimard mwhahaha: mine ?
21:46 dmsimard https://review.openstack.org/#/c/410344/
21:46 mwhahaha dmsimard: yea we need to merge our changes together
21:46 mwhahaha https://review.openstack.org/#/c/410388/
21:46 dmsimard pff I don't want your patch
21:47 mwhahaha well youre patch is red
21:47 mwhahaha so if you want my patch,you must accept my patch :D
21:47 dmsimard lol
21:47 dmsimard sure, you want to squash it ?
21:47 mwhahaha er first my = your
21:47 mwhahaha yea i can
21:47 dmsimard I'm busy in bikeshedding in regards to the libvirt issue
21:47 mwhahaha sounds glorious
21:47 johnmilton joined #puppet-openstack
21:49 openstackgerrit Alex Schultz proposed openstack/puppet-openstack-integration: Align stars to fix ci  https://review.openstack.org/410388
22:06 rmart04 joined #puppet-openstack
22:08 rajinir joined #puppet-openstack
22:14 dmsimard EmilienM: I don't understand your comments on my tripleo patches :(
22:14 EmilienM dmsimard: you created a new parameter, just use it
22:14 EmilienM dmsimard: like we do the Ceph FSID later in the file
22:15 dmsimard but libvirt_virt_type is also written as just "qemu" but there's a parameter for it
22:15 dmsimard I looked at what existed for libvirt_virt_type and did the same thing for libvirt_cpu_mode ?
22:15 EmilienM dmsimard: there isn't parameter because nobody created it
22:15 dmsimard sure there is
22:15 EmilienM let me look again
22:16 EmilienM dmsimard: indeed, that's a bug
22:16 dmsimard EmilienM: https://review.openstack.org/#/c/41​0358/1/puppet/services/nova-libvirt.yaml
22:16 EmilienM we should replace this hiera thing
22:16 dmsimard it's right above what I added
22:16 EmilienM and use the actual parameter :)
22:16 EmilienM we should use NovaComputeLibvirtType
22:16 EmilienM i'll propose a patch that change that, or feel free to do it in your patch
22:17 dmsimard ok so in the scenarios, I put {get_param: NovaComputeLibvirtCpuMode} instead ?
22:17 dmsimard but then how do I make sure it's set to none ? we're defaulting to kvm/host-model in the parameters
22:17 EmilienM dmsimard: no, NovaComputeLibvirtCpuMode: 'qemu'
22:17 EmilienM err
22:17 dmsimard oh, okay I set the parameter, not the hiera value
22:17 EmilienM NovaComputeLibvirtCpuMode: none
22:17 EmilienM dmsimard: exactly!
22:18 dmsimard eh, ok
22:18 dmsimard I'll do the cpu_mode thing, just to touch as little as possible -- fixing the other non-parameterized stuff is out of scope
22:19 EmilienM dmsimard: sure, you can also propose a separated patch. using parameters makes things easier to understand
22:20 EmilienM using hiera is kind of hack
22:29 EmilienM mwhahaha: https://review.openstack.org/#/c/410388/ is passing CI jobs, i'm approving it
22:29 EmilienM (looking at zuul)
22:29 mwhahaha k
22:30 dmsimard I +2'd it so it doesn't look like you self approved it :p
22:35 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align stars to fix ci  https://review.openstack.org/410438
22:35 EmilienM the backport to newton ^
22:38 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Align stars to fix ci  https://review.openstack.org/410439
22:38 EmilienM the backport to mitaka ^
22:48 dmsimard mgariepy: FYI there's issues with CentOS 7.3 (released yesterday) and qemu-kvm-ev 2.6.0 (released today) when using virt_type=qemu and cpu_mode=host-model. You need to use cpu_mode=none (like devstack does) or you'll run into something like this:
22:48 dmsimard http://logs.openstack.org/76/409476/4/ch​eck/gate-puppet-openstack-integration-4-​scenario003-tempest-centos-7/8881991/log​s/libvirt/qemu/instance-00000001.txt.gz
22:48 dmsimard damnit wrong channel
22:48 dmsimard mgariepy: FYI there's issues with CentOS 7.3 (released yesterday) and qemu-kvm-ev 2.6.0 (released today) when using virt_type=qemu and cpu_mode=host-model. You need to use cpu_mode=none (like devstack does) or you'll run into something like this:
22:48 dmsimard http://logs.openstack.org/76/409476/4/ch​eck/gate-puppet-openstack-integration-4-​scenario003-tempest-centos-7/8881991/log​s/libvirt/qemu/instance-00000001.txt.gz
22:48 dmsimard wtf
23:05 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Remove testing for Glance Registry  https://review.openstack.org/410448
23:07 EmilienM dmsimard: qemu version: 2.3.0
23:07 EmilienM is it expected?
23:07 dmsimard what version of openstack
23:07 dmsimard mitaka ?
23:08 EmilienM in your logs
23:08 EmilienM in https://review.openstack.org/#/c/409476/
23:13 dmsimard 2.3.0 doesn't work with 7.3
23:13 dmsimard and 2.6.0 doesn't work with 7.3 when cpu_mode=host-model
23:14 dmsimard 2.6.0 was release sometime today and is progressively being sync'd to all mirrors
23:14 dmsimard 7.3 was released yesterday
23:14 EmilienM are we ok to land the backport ? https://review.openstack.org/#/c/410438/
23:14 mwhahaha oh noes mitaka ci looks very unhappy
23:15 EmilienM http://logs.openstack.org/39/410439/1/che​ck/gate-puppet-openstack-integration-3-sc​enario001-tempest-ubuntu-trusty/8ec0e86/c​onsole.html#_2016-12-13_22_54_22_191684
23:15 EmilienM sounds like a mirror issue
23:15 mwhahaha and all the centos is red for different reasons :D
23:15 EmilienM ceph.com is always down
23:15 EmilienM let me see if we can use the mirror for ceph in mitaka
23:16 EmilienM I'll let you see the centos issue
23:16 EmilienM http://mirror.regionone.osic-clou​d1.openstack.org/ceph-deb-hammer/
23:16 mwhahaha should be ok on newton if dmsimard wants to +A https://review.openstack.org/#/c/410438/
23:16 EmilienM perfect
23:16 dmsimard I sent an email to rdo-list, openstack-dev and openstack-operators about the issue
23:16 EmilienM dmsimard: lol - please ping the world
23:16 EmilienM all@all.com
23:17 dmsimard http://lists.openstack.org/pipermail/o​penstack-dev/2016-December/108884.html
23:17 dmsimard I prefer to pre-emptively communicate so that people don't lose too much time over this
23:17 EmilienM nice, thanks
23:18 mwhahaha dmsimard: your pro-activeness is appreciated and will be abused later :D
23:18 dmsimard I was an operator at $oldjob :(
23:23 openstackgerrit Emilien Macchi proposed openstack/puppet-ceph: Move enable_sig and release to ceph::params  https://review.openstack.org/410454
23:24 openstackgerrit Emilien Macchi proposed openstack/puppet-ceph: repos: allow to configure Debian mirror  https://review.openstack.org/410455
23:29 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: repos: deploy ceph using OpenStack mirrors  https://review.openstack.org/410459
23:29 EmilienM ok that should fix the ceph mirror problem
23:30 EmilienM don't look at code please just +A :D
23:30 EmilienM https://review.openstack.org/​#/q/topic:ceph/mirror/hammer
23:31 EmilienM now the centos7 thing
23:31 EmilienM scenario001 sounds broken because of ceph
23:32 EmilienM scenario003 seems to fail on 3 tempest tests, probably something with nova/neutron
23:32 EmilienM test_server_basic_ops works, so I guess it's not the qemu thing
23:32 EmilienM the floating ip is not reachable
23:33 EmilienM same for scenario002
23:33 EmilienM which means we have a general regression somewhere
23:34 EmilienM https://review.openstack.org/#/q​/project:%255Eopenstack/puppet-.*+status:merged+branch:stable/mitaka
23:34 EmilienM everything worked fine on December 9th, so 4 days ago
23:34 EmilienM let's diff a bit
23:36 EmilienM https://www.diffchecker.com/Tr1gVPvy
23:36 EmilienM ok it's a mess
23:37 EmilienM I don't see any OpenStack related thing
23:37 EmilienM though I see a lot of libvirt/qemu
23:39 EmilienM dmsimard: in mitaka, I see qemu older
23:40 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Mitaka CI test  https://review.openstack.org/410460
23:42 mwhahaha i'll be back later
23:42 * mwhahaha wanders off
23:42 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Remove testing for Glance Registry  https://review.openstack.org/410448
23:42 EmilienM mwhahaha: happy meetup
23:44 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: Pin fast_gettext to 1.2.0  https://review.openstack.org/410463
23:45 openstackgerrit Emilien Macchi proposed openstack/puppet-openstack-integration: repos: deploy ceph using OpenStack mirrors  https://review.openstack.org/410459

| Channels | #puppet-openstack index | Today | | Search | Google Search | Plain-Text | summary