| *** mcarden has joined #openstack-ansible | 00:14 | |
| *** gyee has quit IRC | 00:17 | |
| *** mmercer has quit IRC | 00:25 | |
| *** spatel has joined #openstack-ansible | 00:31 | |
| *** spatel has quit IRC | 00:31 | |
| *** nsmeds has quit IRC | 00:36 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Add one test case to the TestMergeDictUnit for same key dict merge https://review.openstack.org/609745 | 00:54 |
|---|---|---|
| *** lbragstad-503 is now known as lbragstad | 00:58 | |
| *** spatel has joined #openstack-ansible | 01:07 | |
| *** dave-mccowan has joined #openstack-ansible | 01:20 | |
| *** logan- has joined #openstack-ansible | 01:24 | |
| *** vnogin has joined #openstack-ansible | 01:47 | |
| *** spatel has quit IRC | 01:50 | |
| *** vnogin has quit IRC | 01:51 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: tests: Do not use ansible_host to pass container IP https://review.openstack.org/582329 | 01:53 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: Allow container detection when host matches container_name https://review.openstack.org/582330 | 01:53 |
| cloudnull | hey mattoliverau :) | 01:56 |
| mattoliverau | cloudnull: o/ | 01:56 |
| cloudnull | I'm not sure of a way to do ipv6 only clouds, however, I do know that if you have IPv6 enabled you can get IPv6 within the containers | 01:57 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: More cleanup of the osquery role https://review.openstack.org/611099 | 01:57 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Add osquery pack setup https://review.openstack.org/611195 | 01:57 |
| prometheanfire | I've done v6 only networks, and I think nothing infra side would REQUIRE v4, but tooling probably needs to change | 01:57 |
| *** cshen has joined #openstack-ansible | 01:58 | |
| spotz | mattoliverau: Hey!!! | 02:00 |
| mattoliverau | spotz: o/ | 02:00 |
| spotz | mattoliverau: Congrats | 02:01 |
| mattoliverau | cloudnull, prometheanfire: I'll just have a play with chaning the aio code/tooling a bit and see what happens :). If i manage it I'll be sure to push something :) | 02:02 |
| cloudnull | ++ cool! | 02:02 |
| prometheanfire | mattoliverau: :D | 02:02 |
| mattoliverau | spotz: thanks! good thing the baby is cute, cause finding time to sleep is the hard part :P | 02:02 |
| *** cshen has quit IRC | 02:03 | |
| spotz | mattoliverau: ehehe. Will we see you in Berlin? | 02:03 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Fix missing logstash device error https://review.openstack.org/610705 | 02:03 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-plugins master: Allow container detection when host matches container_name https://review.openstack.org/582330 | 02:06 |
| mattoliverau | spotz: nope :( I would love too but there happens to be noone (friends or family) available as some backup support for the wife that week (any other week would've been fine). And not quite ready to leave the wife alone with a new born and a monster^Wtoddler. But should be at the next. | 02:06 |
| spotz | mattoliverau: We'll miss you but very understandable | 02:06 |
| mattoliverau | so a nice quiet stay at home for the rest of the year. though maybe quiet isn't the word for it :P | 02:07 |
| *** ram5391 has joined #openstack-ansible | 02:09 | |
| ram5391 | Is there a flag that is supposed to be set in the config to tell ansible to test against https keystone, and not http? | 02:14 |
| ram5391 | I tried setting: keystone_service_publicuri_proto: http | 02:15 |
| ram5391 | openstack_service_publicuri_proto: http | 02:15 |
| ram5391 | But my setup-openstack.yml is failing because it can't resolve the http keystone endpoint. (https is working fine) | 02:16 |
| *** kstev has joined #openstack-ansible | 02:20 | |
| *** kstev has quit IRC | 02:39 | |
| *** kstev has joined #openstack-ansible | 02:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/queens: common-tasks: nodepool: Set openSUSE OBS mirror for OpenStack CI https://review.openstack.org/607505 | 02:46 |
| ram5391 | If anyone there could lend a hand, I just don't know if its worth combing through all the playbooks to change the value of the test string, or if its something I'm doing wrong | 02:47 |
| *** kstev has quit IRC | 02:53 | |
| openstackgerrit | melanie witt proposed openstack/openstack-ansible-os_cinder master: Fix Cinder backup to use full class path https://review.openstack.org/605846 | 03:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Update messaging notification configuration https://review.openstack.org/605112 | 03:05 |
| *** dave-mccowan has quit IRC | 03:33 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 03:40 |
| *** ram5391 has quit IRC | 03:42 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 03:48 |
| *** cshen has joined #openstack-ansible | 03:58 | |
| *** cshen has quit IRC | 04:03 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 04:34 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 04:40 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 04:43 |
| *** pcaruana has quit IRC | 04:46 | |
| *** shyamb has joined #openstack-ansible | 04:58 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 05:05 |
| *** chhagarw has joined #openstack-ansible | 05:06 | |
| *** jbadiapa has quit IRC | 05:10 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 05:36 |
| *** vnogin has joined #openstack-ansible | 05:47 | |
| *** vnogin has quit IRC | 05:51 | |
| *** DanyC has joined #openstack-ansible | 06:07 | |
| *** cshen has joined #openstack-ansible | 06:15 | |
| *** pcaruana has joined #openstack-ansible | 06:34 | |
| *** jbadiapa has joined #openstack-ansible | 06:34 | |
| evrardjp | good morning | 06:44 |
| jrosser | Morning evrardjp | 06:44 |
| evrardjp | morning jrosser :) | 06:45 |
| evrardjp | mattoliverau: did you get help on the ipv6? | 06:45 |
| evrardjp | ipv6 is so long that people tend to use dns with ipv6 behind the scenes. In that case it should be fairly straightforward. The only part is I guess the lxc containers IPs. If you want full ipv6, you can probably use a bare metal deployment, which will not really care about generation of IPs. Or better, improve our nspawn/lxc to handle ipv6 (or just accepting IPs from dhcpv6/slaac) | 06:47 |
| evrardjp | maybe jrosser has an idea there? | 06:47 |
| jrosser | Given that the container macs are generated I’d always thought a slaac-alike approach was viable there | 06:48 |
| jrosser | Which may be osa generating static ipv6 from the generated Mac | 06:49 |
| evrardjp | yes so a way of disabling privacy extensions and just re-using the slaac would be fine to determine the IP of the container | 06:52 |
| jrosser | Perhaps my point really is that all of the tooling exists already to assign unique and unchanging ipv6 to the container interfaces, it would be a relatively small change somewhere to do that | 06:52 |
| *** shyamb has quit IRC | 06:52 | |
| jrosser | evrardjp: can we always rely on there being a RA for the prefix? | 06:55 |
| *** deployer2 has quit IRC | 06:58 | |
| evrardjp | I was first tempted to say yes, but I can think of a few cases where no | 07:02 |
| evrardjp | I would not see a problem to use fe80: in itself | 07:02 |
| evrardjp | or private networks | 07:02 |
| evrardjp | fc00... | 07:03 |
| evrardjp | in that cases there is a slight chance that the user would not have assigned a router | 07:03 |
| evrardjp | but I see many problems happening if there is no proper ipv6 router though | 07:06 |
| evrardjp | so it's maybe a wise assumption | 07:06 |
| jrosser | You’d want it to work in any AIO really | 07:06 |
| jrosser | Even one without any v6 on the outside | 07:07 |
| *** fghaas has joined #openstack-ansible | 07:07 | |
| jrosser | So I think generating static addresses in the inventory in a way which resembles slaac is probably the most universal answer. | 07:09 |
| *** hamzaachi has joined #openstack-ansible | 07:35 | |
| *** DanyC has quit IRC | 07:37 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO ELK Fixes https://review.openstack.org/611167 | 07:40 |
| *** goldenfri has quit IRC | 07:46 | |
| *** shyamb has joined #openstack-ansible | 07:46 | |
| *** tosky has joined #openstack-ansible | 07:53 | |
| *** pvradu has joined #openstack-ansible | 07:56 | |
| *** electrofelix has joined #openstack-ansible | 07:59 | |
| *** goldenfri has joined #openstack-ansible | 08:02 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610603 | 08:04 |
| odyssey4me | o/ folks, sorry was out yesterday organising visa things for berlin | 08:11 |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Set OpenSUSE distro jobs to non-voting https://review.openstack.org/608430 | 08:20 |
| jrosser | o/ odyssey4me ..... this was surprising https://review.openstack.org/#/c/610165/ | 08:21 |
| odyssey4me | jrosser: unfortunately I lost your PM from yesterday because my IRC client crashed :/ | 08:21 |
| jrosser | no worries, i lashed some stuff together here https://review.openstack.org/#/c/611146/ which would be a starting point to talk about | 08:22 |
| odyssey4me | jrosser: honestly I find myself thinking that perhaps ceph for ubuntu should just come from UCA instead of trying to use the ceph repo, but meh - I just using the ceph packages across all distros is something people want | 08:23 |
| jrosser | well theres a number of things | 08:25 |
| jrosser | i was trying to debug that rgw segafault and there didnt seem to be any -dbg packages in UCA | 08:26 |
| *** DanyC has joined #openstack-ansible | 08:26 | |
| jrosser | also the apt-pinning is ambiguous - last week UCA had a higher version string of 13.2.1 than download.ceph.com did, but today it's the other way round as theres 13.2.2 at ceph.com | 08:29 |
| *** thuydang has joined #openstack-ansible | 08:30 | |
| evrardjp | oh great | 08:34 |
| evrardjp | that's good news | 08:34 |
| jrosser | oh, and we apt-pin ceph_client but not the server side parts | 08:35 |
| odyssey4me | oh dear | 08:37 |
| evrardjp | odyssey4me: jrosser should we go on a limb here? | 08:38 |
| evrardjp | I am thinking that we should build nodes in infra with ceph on it. | 08:38 |
| evrardjp | so we can request said nodes and just integrate into it :p | 08:38 |
| evrardjp | so no need to care about that then :p | 08:39 |
| evrardjp | two different lifecycles | 08:39 |
| evrardjp | re-usable by other deployment projects | 08:39 |
| evrardjp | so server side problem is solved:p | 08:40 |
| odyssey4me | evrardjp: I don't see how that solves the problem? | 08:40 |
| evrardjp | it just moves it | 08:40 |
| evrardjp | :p | 08:40 |
| evrardjp | we don't need to care about server anymore that way? | 08:40 |
| evrardjp | we just say for release x we support ceph version a/b/c | 08:41 |
| evrardjp | but I said "go on a limb" | 08:41 |
| evrardjp | I agree that fixing the thing is probably better right now | 08:42 |
| jrosser | as i look at this, on the whole OSA takes a really careful approch to ensuring that the right versions of key components get installed in a repeatable manner | 08:42 |
| jrosser | that just needs improving for the ceph parts, which are currently relying somewhat on luck to get a coherent set of client/server installed | 08:43 |
| jrosser | sorting the pinning for the ceph server components to match the client is one thing | 08:43 |
| evrardjp | jrosser: maybe we should move to only a single role for making it easier, and have different "profiles" there? | 08:43 |
| evrardjp | so only one pinning would apply | 08:44 |
| evrardjp | (as it would be in a common 'profile' ) | 08:44 |
| evrardjp | (wild ideas here, no sacred cows) | 08:44 |
| jrosser | and secondly playbooks/ceph-install.yml and the matching rgw one are quite scrappy | 08:44 |
| jrosser | i feel they are quite low hanging fruit for a refactor | 08:44 |
| odyssey4me | evrardjp: I thought the ceph_client role did exactly that? | 08:45 |
| chandankumar | odyssey4me: Hello | 08:45 |
| odyssey4me | jrosser: sounds good - I see you've started on it already | 08:45 |
| odyssey4me | chandankumar: o/ | 08:45 |
| chandankumar | odyssey4me: regarding this review https://review.openstack.org/#/c/591424/ http://logs.openstack.org/24/591424/30/check/openstack-ansible-functional-distro_install-centos-7/b8236ed/job-output.txt.gz#_2018-10-16_12_37_02_749516 | 08:45 |
| jrosser | odyssey4me: i have, but that part i'd really like input on | 08:46 |
| chandankumar | odyssey4me: I am not sure from where /root/etc is taken | 08:46 |
| chandankumar | odyssey4me: for other jobs tempest is running | 08:46 |
| odyssey4me | chandankumar: looking | 08:47 |
| *** vnogin has joined #openstack-ansible | 08:47 | |
| chandankumar | evrardjp: can we merge this one https://review.openstack.org/609992 | 08:48 |
| odyssey4me | chandankumar: in vars/distro_install there is the following set: tempest_workspace: "{{ ansible_env.HOME }}" | 08:51 |
| odyssey4me | chandankumar: then the task 'Copy tempest config' copies the file into dest: "{{ tempest_workspace }}/etc/tempest.conf" | 08:51 |
| chandankumar | odyssey4me: will i remove that part? | 08:51 |
| odyssey4me | that's why it's trying to copy into /root/etc | 08:51 |
| *** vnogin has quit IRC | 08:52 | |
| odyssey4me | so the path you removed in 'Create tempest directories' should have a new path added to create /etc/tempest, and the destination for the conf file should be changed to /etc/tempest | 08:52 |
| chandankumar | odyssey4me: we donot need to create tempest workspace | 08:53 |
| chandankumar | tmepest init <workspace> -> creates a dir and /etc and /logs folder for us | 08:53 |
| odyssey4me | chandankumar: yes, but we do need to create /etc/tempest do we not? | 08:55 |
| chandankumar | odyssey4me: we donot need | 08:55 |
| evrardjp | I forgot to say that 18.0.0 is finally released! Congrats everyone | 08:55 |
| odyssey4me | if not, then just make sure the right destination path is set for the conf file | 08:55 |
| evrardjp | now I can work on 18.0.1 and all the others | 08:56 |
| odyssey4me | evrardjp: yeah! thanks for handling that - I guess we need a sha bump already for 18.0.1 | 08:56 |
| chandankumar | odyssey4me: http://paste.openstack.org/show/732324/ | 08:56 |
| chandankumar | it creates tempest_lock in workspace as well as in /tmp dir | 08:57 |
| chandankumar | it also creates ~/.tempest/ folder where workspace reference is stored | 08:57 |
| odyssey4me | chandankumar: yes, that's for the workspace - but doesn't tempes.conf need to already be in /etc/tempest at the time of init for the workspace to use it as a base config? | 08:57 |
| chandankumar | odyssey4me: nope, during tempest run, tempest first looks for config in workspace if it not found then it looks for other places | 08:58 |
| chandankumar | /etc/tempest | 08:58 |
| odyssey4me | chandankumar: yes, but it won't exist anywhere else yet because the tasks have been re-ordered to only place the conf down afterwards | 08:58 |
| chandankumar | checking | 08:59 |
| odyssey4me | this is why I'm suggesting that we create /etc/tempest and put the tempest.conf there before we init the workspace | 09:00 |
| chandankumar | odyssey4me: will I introduce a new var tempest_conf_path where default will be /etc/tempest ? | 09:01 |
| odyssey4me | chandankumar: I don't think it needs a var - would TripleO like it to be configurable? | 09:02 |
| chandankumar | odyssey4me: in Tripleo, we use the default workspace/etc path for config | 09:02 |
| chandankumar | let me do that | 09:02 |
| chandankumar | and see how it goes, in future we can change if needed | 09:03 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Added support for installing tempest from distro https://review.openstack.org/591424 | 09:04 |
| evrardjp | odyssey4me: yes indeed -- it should have happened on friday, so I am late :p | 09:04 |
| evrardjp | odyssey4me: but if you have noticed, I am slowly catching up said lag. | 09:05 |
| evrardjp | I expect by next release no lag should appear. | 09:05 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Switch to using file-backed VM's only https://review.openstack.org/609110 | 09:05 |
| evrardjp | chandankumar: done | 09:06 |
| chandankumar | evrardjp: thanks :-) | 09:06 |
| odyssey4me | evrardjp: thanks - appreciate the time you're taking on this :) let us know if we can help carry the load | 09:07 |
| noonedeadpunk | morning everyone. | 09:12 |
| noonedeadpunk | do you know if opensuse 15 was either fixed or disabled? Unfortunatelly I didn't catch on yestarday meeting what exactly decided regarding this broken template | 09:13 |
| odyssey4me | noonedeadpunk: the job is now non-voting | 09:13 |
| *** shyamb has quit IRC | 09:14 | |
| noonedeadpunk | odyssey4me: so may I set it as non-voting for distro-install task as well here https://review.openstack.org/#/c/610274/ ? | 09:15 |
| odyssey4me | jrosser: nice work on http://grafana.openstack.org/d/4-pSHcImz/openstack-ansible?orgId=1&from=now-30d&to=now - I think it might be possible to use some wildcards to have all the jobs show up there... and the run time graphs would look better if the lines are joined I think, and perhaps we could also put them all on the same graph | 09:15 |
| odyssey4me | alternatively we could have each job graphed with its data, and show the moving average or something like that | 09:16 |
| odyssey4me | perhaps I should fire up a test grafana host to play around with some things, given that graphite.openstack.org is openly accessible | 09:16 |
| *** hw_wutianwei_ has joined #openstack-ansible | 09:17 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: All distro-install tests are voting now. https://review.openstack.org/610274 | 09:21 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: All distro-install tests are voting now. https://review.openstack.org/610274 | 09:23 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Distro-install tests for Ubuntu and Centos are voting now. https://review.openstack.org/610274 | 09:24 |
| evrardjp | odyssey4me: not much load in itself. I have yet to find time basically on my last round of cleanups. | 09:25 |
| evrardjp | then it can be properly documented | 09:25 |
| evrardjp | and moved away from the current method, and branch independant, and simpler :) | 09:25 |
| odyssey4me | awesome :) | 09:25 |
| *** hamzaachi has quit IRC | 09:26 | |
| evrardjp | I mean it's already branch independant and more modular to me | 09:26 |
| evrardjp | but anyway I am closer. | 09:26 |
| *** hamzaachi has joined #openstack-ansible | 09:26 | |
| *** shyamb has joined #openstack-ansible | 09:38 | |
| *** Emine has joined #openstack-ansible | 09:40 | |
| mattoliverau | evrardjp, jrosser: it's too late to read through and grok all this now :P. ATM I'm just changing the aio tooling and keeping as simple as possible for now.. use static ipv6 addressing. It's just a step 1. Thanks for the comments, they will be very useful and something I will need to think about in the future. | 09:42 |
| mattoliverau | maybe morning matt when he has more brain power | 09:43 |
| *** shyamb has quit IRC | 09:48 | |
| *** shyamb has joined #openstack-ansible | 09:48 | |
| odyssey4me | noonedeadpunk: reviewed with some feedback | 09:52 |
| *** thuydang has quit IRC | 09:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon master: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/611277 | 09:54 |
| noonedeadpunk | odyssey4me: and is the list of zuul templates public available? Can you please point me to it, in case it is available | 09:57 |
| odyssey4me | noonedeadpunk: they come from https://github.com/openstack/openstack-ansible-tests/blob/master/zuul.d/project-templates.yaml | 09:57 |
| noonedeadpunk | got it, thanks | 09:58 |
| jrosser | odyssey4me: on grafana i was thinking that having graphs like those i did per distro family would be good | 09:58 |
| jrosser | then an aggregate one showing overall failure rate, because at the end of the day that is our measure of how easy/hard it is to land a patch | 09:59 |
| jrosser | the per-distro family ones would let us see why things were failing easily | 09:59 |
| jrosser | i only discovered too late that the graphite data was open, so didnt get change to play locally and make the charts look nicer | 10:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Use the OSA inventory and plugins where possible https://review.openstack.org/611279 | 10:01 |
| odyssey4me | jrosser: it's a good start, and I think a worthwhile initiative | 10:02 |
| jrosser | yeah, it'll provide some really good insight | 10:03 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Distro-install tests for Ubuntu and Centos are voting now. https://review.openstack.org/610274 | 10:05 |
| noonedeadpunk | odyssey4me: according to the list of templates, it's time we made bionic voting for upgrade task, or it's still not ready for prod? | 10:05 |
| odyssey4me | noonedeadpunk: we've removed all the upgrade tests, I'm working on using the integrated repo for all role tests and when that's in, we'll use that for upgrade tests instead | 10:06 |
| odyssey4me | so yeah, those comments are out of date | 10:06 |
| noonedeadpunk | oh, sorry then. | 10:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove verbose output for bootstrap https://review.openstack.org/611283 | 10:08 |
| *** shyam89 has joined #openstack-ansible | 10:09 | |
| *** Emine has quit IRC | 10:10 | |
| *** shyamb has quit IRC | 10:13 | |
| *** Emine has joined #openstack-ansible | 10:13 | |
| *** rgogunskiy has joined #openstack-ansible | 10:14 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Switch to using file-backed VM's only https://review.openstack.org/609110 | 10:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Use the OSA inventory and plugins where possible https://review.openstack.org/611279 | 10:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove verbose output for bootstrap https://review.openstack.org/611283 | 10:16 |
| *** Emine has quit IRC | 10:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove duplicated vm_servers group creation https://review.openstack.org/611288 | 10:31 |
| *** tosky has quit IRC | 10:34 | |
| *** tosky has joined #openstack-ansible | 10:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 10:41 |
| *** hamzaachi has quit IRC | 10:43 | |
| *** hamzaachi has joined #openstack-ansible | 10:44 | |
| *** Emine has joined #openstack-ansible | 10:50 | |
| *** cshen has quit IRC | 11:03 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Switch to using file-backed VM's only https://review.openstack.org/609110 | 11:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Use the OSA inventory and plugins where possible https://review.openstack.org/611279 | 11:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove verbose output for bootstrap https://review.openstack.org/611283 | 11:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove duplicated vm_servers group creation https://review.openstack.org/611288 | 11:04 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 11:04 |
| *** vollman has joined #openstack-ansible | 11:09 | |
| *** ansmith has joined #openstack-ansible | 11:10 | |
| *** cshen has joined #openstack-ansible | 11:15 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Switch to using file-backed VM's only https://review.openstack.org/609110 | 11:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Use the OSA inventory and plugins where possible https://review.openstack.org/611279 | 11:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove verbose output for bootstrap https://review.openstack.org/611283 | 11:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Remove duplicated vm_servers group creation https://review.openstack.org/611288 | 11:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 11:15 |
| *** shyam89 has quit IRC | 11:16 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 11:18 |
| *** ansmith has quit IRC | 11:18 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610605 | 11:18 |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610601 | 11:18 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610597 | 11:21 |
| openstackgerrit | Merged openstack/openstack-ansible-os_sahara master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610609 | 11:21 |
| *** rgogunskiy has quit IRC | 11:26 | |
| *** pcaruana has quit IRC | 11:28 | |
| *** mbi has joined #openstack-ansible | 11:28 | |
| *** shyam89 has joined #openstack-ansible | 11:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 11:37 |
| *** strobelight has joined #openstack-ansible | 11:39 | |
| *** kaiokmo has joined #openstack-ansible | 11:41 | |
| *** strobelight_ has joined #openstack-ansible | 11:41 | |
| *** strobelight is now known as Guest52277 | 11:41 | |
| *** strobelight_ is now known as strobelight | 11:42 | |
| *** Guest52277 has quit IRC | 11:45 | |
| noonedeadpunk | Folks, can you please review this https://review.openstack.org/#/c/610274/ ? | 11:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/rocky: use include_tasks instead of include https://review.openstack.org/608105 | 11:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon master: Add ability for HTTP access to horizon https://review.openstack.org/607969 | 11:56 |
| *** vollman has quit IRC | 11:57 | |
| *** shyam89 has quit IRC | 12:11 | |
| *** shyam89 has joined #openstack-ansible | 12:11 | |
| *** shyam89 has quit IRC | 12:13 | |
| *** kstev has joined #openstack-ansible | 12:13 | |
| *** shyam89 has joined #openstack-ansible | 12:13 | |
| *** dave-mccowan has joined #openstack-ansible | 12:14 | |
| *** shyam89 has quit IRC | 12:18 | |
| *** pcaruana has joined #openstack-ansible | 12:24 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Zuul: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/609284 | 12:36 |
| openstackgerrit | Marc Gariépy (mgariepy) proposed openstack/openstack-ansible-os_nova master: Update the pci config for nova. https://review.openstack.org/605789 | 12:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Zuul: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/609287 | 12:38 |
| *** hamzaachi has quit IRC | 12:39 | |
| nicolasbock | Good morning | 12:40 |
| *** strattao has joined #openstack-ansible | 12:41 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Zuul: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/609288 | 12:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Zuul: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/609286 | 12:42 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: [WIP] Refactor ceph server playbooks https://review.openstack.org/611146 | 12:43 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible master: Fix rgw playbook pre_tasks https://review.openstack.org/611304 | 12:43 |
| jrosser | well there appears to be a certain amount of whoooooops about 611304 | 12:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Zuul: Use openstack-ansible-role-distro_install-jobs template https://review.openstack.org/609285 | 12:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: Add ability to deploy Octavia load balancer service https://review.openstack.org/600155 | 12:53 |
| *** kstev has quit IRC | 12:53 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Distro-install tests for Ubuntu and Centos are voting now. https://review.openstack.org/610274 | 12:54 |
| strattao | odyssey4me or someone, I’ve got the ec2-api up and running. When I use it to describe regions, it responds and works great. However, when I describe images, or describe instances, it tries to use the public endpoints for both glance and nova respectively and gives me connection errors, that I believe are due to certs not being installed properly. I can look into that workaround, but I’m trying to figure out why it doesn’t use the admin or | 12:59 |
| strattao | internal endpoints first. Would that be something I’ve setup incorrectly with the ec2api user, a missing role or something? | 12:59 |
| odyssey4me | strattao: typically you have to tell it which endpoint to use, and we tell all services in the conf file to use the 'internal' endpoint | 13:00 |
| odyssey4me | so it's likely just some missing config | 13:00 |
| strattao | thanks. I’ll keep playing around with the config settings then | 13:01 |
| strattao | just to confirm, the endpoint setting you are referring to is by specifying the auth_url (like in /etc/glance/glance-api.conf)? | 13:02 |
| *** chandankumar is now known as chkumar|off | 13:03 | |
| odyssey4me | strattao: it may be something like these: https://github.com/openstack/openstack-ansible-os_nova/blob/master/templates/nova.conf.j2#L244-L246 | 13:04 |
| strattao | yup, sounds good. thx odyssey4me | 13:04 |
| *** ansmith has joined #openstack-ansible | 13:14 | |
| *** SimAloo has joined #openstack-ansible | 13:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610602 | 13:34 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610604 | 13:35 |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible master: Add qdrouterd role for oslo.messaging backend deployment https://review.openstack.org/570261 | 13:35 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610606 | 13:36 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610608 | 13:37 |
| hwoarang | does anyone know what may have changed in stable/rocky and broke suse like this http://logs.openstack.org/02/610902/1/check/openstack-ansible-deploy-aio_distro_lxc-opensuse-423/00253bd/job-output.txt.gz#_2018-10-17_12_41_28_672737 ? | 13:38 |
| hwoarang | sounds like requests/cert.py is missing or something | 13:38 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610607 | 13:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610610 | 13:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Add ability for HTTP access to horizon https://review.openstack.org/607969 | 13:45 |
| *** munimeha1 has joined #openstack-ansible | 13:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon stable/rocky: Add ability for HTTP access to horizon https://review.openstack.org/611322 | 14:01 |
| *** cshen has quit IRC | 14:04 | |
| *** rgogunskiy has joined #openstack-ansible | 14:07 | |
| *** chhagarw has quit IRC | 14:12 | |
| *** spatel has joined #openstack-ansible | 14:20 | |
| spatel | jamesdenton: morning | 14:20 |
| jamesdenton | hi | 14:20 |
| spatel | Quick question, how do you guys manage mix hardware compute node | 14:21 |
| spatel | ex. i have 10G and 1G and more memory box all kind of hardware | 14:21 |
| jamesdenton | in my experience, most environments tend to leverage the same hardware/capabilities | 14:21 |
| jamesdenton | but there is some flexibility | 14:21 |
| mgariepy | having different hw can cause some issues | 14:22 |
| jamesdenton | ^^ can definitely impact migrations | 14:22 |
| spatel | That is what i am struggling with | 14:22 |
| jamesdenton | what in particular? | 14:22 |
| mgariepy | if you don't need live migration it might be ok-ish. but that really depend on your usecase | 14:23 |
| spatel | We have private cloud so its not for customer and trying to figure out how should i manage mix hardware | 14:23 |
| spatel | I don't need Live migration because we hare using most of time SR-IOV compute node | 14:23 |
| jamesdenton | well maybe you're better off making multiple aggregates based on groupings of similar hardware | 14:24 |
| spatel | Trying to understand how should i layout flavor and groups to target them... | 14:24 |
| spatel | Currently i have created AZ-10G & AZ-1G and AZ-Mem Host aggregates | 14:25 |
| spatel | I am using availability zone for grouping in short... ;) | 14:25 |
| spatel | i don't know what is the best practice out there which mostly people follow | 14:26 |
| jamesdenton | hard to say, really. i bet if you asked 20 people you'd get 20 answers | 14:27 |
| spatel | also i have create small 30 compute node AZ-Live for only live migration instance (whoever need more reliability ) | 14:27 |
| jamesdenton | if what you're doing works for you, then it's OK | 14:27 |
| spatel | Hmm! i totally understand there is no hare written answer for this kind of question | 14:28 |
| spatel | I think i continue what i am doing and someday i may find holy grill to make it easy | 14:28 |
| spatel | I have notice block migration also not bad.. its slow but it does work | 14:30 |
| *** openstackgerrit has quit IRC | 14:36 | |
| *** noonedeadpunk[m] has joined #openstack-ansible | 14:40 | |
| *** logan_ has joined #openstack-ansible | 14:40 | |
| *** hamzy has quit IRC | 14:41 | |
| *** noonedeadpunk_ has joined #openstack-ansible | 14:43 | |
| *** noonedeadpunk_ has left #openstack-ansible | 14:43 | |
| *** noonedeadpunk_ has joined #openstack-ansible | 14:43 | |
| *** openstackgerrit has joined #openstack-ansible | 14:44 | |
| openstackgerrit | Alex Redinger proposed openstack/openstack-ansible-os_keystone master: Add memcache flushing handler on db migrations https://review.openstack.org/608066 | 14:44 |
| *** noonedeadpunk[m] has quit IRC | 14:45 | |
| *** kstev has joined #openstack-ansible | 14:47 | |
| *** munimeha1 has quit IRC | 14:49 | |
| *** nicolasbock has quit IRC | 14:49 | |
| *** logan- has quit IRC | 14:49 | |
| *** bgmccollum has quit IRC | 14:49 | |
| *** noonedeadpunk has quit IRC | 14:49 | |
| *** logan_ is now known as logan- | 14:49 | |
| *** cshen has joined #openstack-ansible | 14:50 | |
| *** jbadiapa has quit IRC | 14:52 | |
| *** jbadiapa has joined #openstack-ansible | 14:53 | |
| *** weezS has joined #openstack-ansible | 14:56 | |
| *** noonedeadpunk_ has quit IRC | 14:57 | |
| *** noonedeadpunk_ has joined #openstack-ansible | 14:57 | |
| noonedeadpunk_ | folks, I've got disconnected from freenode with my main nickname. As it was still appearing as online, I've tried to RELEASE it. Now freenode says, that my nickname is just unavailable. Probably someone faced with such problems and may point me to some doc or some direction | 14:59 |
| *** ansmith has quit IRC | 15:01 | |
| *** noonedeadpunk_ is now known as noonedeadpunk | 15:02 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: releasenotes: oslo-messaging-separate-backends add project name https://review.openstack.org/610611 | 15:03 |
| noonedeadpunk | ok, it seems, that I've resolved the problem) | 15:04 |
| *** noonedeadpunk has quit IRC | 15:05 | |
| *** larsks has joined #openstack-ansible | 15:05 | |
| *** noonedeadpunk has joined #openstack-ansible | 15:05 | |
| *** larsks has left #openstack-ansible | 15:05 | |
| *** rgogunskiy has quit IRC | 15:07 | |
| jrosser | odyssey4me: bionic/ceph passes https://review.openstack.org/#/c/611146/ | 15:10 |
| mgariepy | after many many recheck it seems to passes , can we merge this one please ? https://review.openstack.org/#/c/605789/ | 15:10 |
| noonedeadpunk | Folks, and what do you think about this thing on the summit? https://www.openstack.org/summit/berlin-2018/summit-schedule/events/22736/rsvp-required-from-day-1-to-day-2-to-day-1000-practical-openstack-deployment-and-operations-with-openstack-ansible Is it worth wisiting? | 15:11 |
| *** hamzy has joined #openstack-ansible | 15:12 | |
| *** kaiokmo has quit IRC | 15:12 | |
| *** pvradu has quit IRC | 15:12 | |
| odyssey4me | noonedeadpunk: You'll meet cloudnull and I there - we're helping fghaas for the day. :) | 15:13 |
| *** ansmith has joined #openstack-ansible | 15:15 | |
| noonedeadpunk | odyssey4me: so is it designed for newbies or it will be interesting even for folks with some basics (not the strongest ones, but like mine)? As it will take full day, with tons of interesting activities... | 15:17 |
| odyssey4me | noonedeadpunk: I've never attended one, but Florian is very knowledgeable. However, it is a whole day so you might want to rather focus on spreading yourself around - but popping in in between things to listen in the back. | 15:18 |
| noonedeadpunk | odyssey4me: got you, thanks:) | 15:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server master: Clean up role tests https://review.openstack.org/611359 | 15:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/rocky: Clean up role tests https://review.openstack.org/611361 | 15:25 |
| *** blinkiz has quit IRC | 15:26 | |
| *** cshen has quit IRC | 15:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/queens: Clean up role tests https://review.openstack.org/611362 | 15:28 |
| *** mardim has quit IRC | 15:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Clean up role tests https://review.openstack.org/611364 | 15:30 |
| *** ansmith has quit IRC | 15:31 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Remove wsrep-new-cluster from cluster.cnf https://review.openstack.org/610992 | 15:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Enable the xinetd service for reboot https://review.openstack.org/605778 | 15:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/pike: Allow configuration of startup timeout https://review.openstack.org/604999 | 15:34 |
| *** fghaas has quit IRC | 15:34 | |
| *** cshen has joined #openstack-ansible | 15:37 | |
| *** hamzaachi has joined #openstack-ansible | 15:38 | |
| *** fghaas has joined #openstack-ansible | 15:39 | |
| *** macza has joined #openstack-ansible | 15:39 | |
| fghaas | noonedeadpunk: I get far more praise from odyssey4me there than I deserve :) | 15:40 |
| fghaas | but yes, worth visiting if you'll accept that I'm clearly biased | 15:40 |
| *** smoke87 has quit IRC | 15:41 | |
| *** cshen has quit IRC | 15:43 | |
| *** ansmith has joined #openstack-ansible | 15:43 | |
| *** vnogin has joined #openstack-ansible | 15:43 | |
| noonedeadpunk | fghaas: I believe that it's all true:) So then I have double check the schedule for to free up the first day | 15:44 |
| noonedeadpunk | s/first/last/ | 15:46 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 15:48 |
| *** vnogin has quit IRC | 15:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Switch to using file-backed VM's only https://review.openstack.org/609110 | 15:49 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Use the OSA inventory and plugins where possible https://review.openstack.org/611279 | 15:50 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Remove verbose output for bootstrap https://review.openstack.org/611283 | 15:52 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Remove duplicated vm_servers group creation https://review.openstack.org/611288 | 15:52 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Copy SSH keys into containers if they are present https://review.openstack.org/611292 | 15:52 |
| *** hamzaachi has quit IRC | 15:55 | |
| *** smoke87 has joined #openstack-ansible | 15:55 | |
| *** shardy has joined #openstack-ansible | 16:01 | |
| *** hamzy has quit IRC | 16:01 | |
| *** hamzy has joined #openstack-ansible | 16:01 | |
| spatel | I have flavor and i want to change vPCU number is it possible to do that ? | 16:03 |
| fghaas | noonedeadpunk: right. You could just be there the morning and then sneak out to go see something else in the afternoon, but doing so wouldn't exactly be nice to a person that does want to spend the whole day and whose seat you might be taking. | 16:06 |
| *** nsmeds__ has joined #openstack-ansible | 16:06 | |
| fghaas | So do choose wisely, and if you have something super interesting to see either half of the day, you might want to look at https://academy.citycloud.com/courses/course-v1:citycloud+cc201+201810/about instead | 16:07 |
| *** cshen has joined #openstack-ansible | 16:07 | |
| fghaas | However, we'll also be very happy to have you in the room, just to be crystal clear about that. :) | 16:08 |
| *** priteau has joined #openstack-ansible | 16:08 | |
| noonedeadpunk | fghaas: yeah, I know it, that's why I decided to ask whether there's a sence to take part for a person, who is familiar with OSA and already deploy with it | 16:08 |
| odyssey4me | cores, I'd appreciate some reviews for https://review.openstack.org/#/q/I7d1f07878dba7d21aecd59fa75e9117b50b886e6 - the broken upgrade tests are preventing other patches frommerging in pike especially | 16:08 |
| noonedeadpunk | As I don't want to take someone's place but really interesnted in the event | 16:09 |
| fghaas | you can also come to my Kubernetes tutorial that I'm doing with Tyler from Rackspace, there's no OSA there but at least you'll be interacting with an OSA-deployed Magnum :) | 16:10 |
| fghaas | s/my/the/ | 16:13 |
| noonedeadpunk | fghaas: to be honest - I was also very interested in it, but there are really interesting sessions both in the morning and in the evening | 16:14 |
| fghaas | classic Summit problem | 16:14 |
| noonedeadpunk | yeah. And the problem starts, when there are monre than 3 cool sessions at the same time... | 16:17 |
| noonedeadpunk | I'd wish such tutorials being held the day before/after main event... | 16:20 |
| goldenfri | spatel no, the the horizon people told me that the nova people removed the ability to edit flavors because it didn't actually work | 16:24 |
| *** munimeha1 has joined #openstack-ansible | 16:24 | |
| *** nicolasbock has joined #openstack-ansible | 16:24 | |
| spatel | goldenfri: thanks for confirmation! also i didn't find any single doc related that to edit directly.. I have deleted flavor and re-created | 16:25 |
| *** fghaas has quit IRC | 16:32 | |
| *** jbadiapa has quit IRC | 16:33 | |
| *** strattao has quit IRC | 16:39 | |
| *** strattao has joined #openstack-ansible | 16:40 | |
| *** strattao has quit IRC | 16:41 | |
| *** shardy has quit IRC | 16:41 | |
| *** strattao has joined #openstack-ansible | 16:43 | |
| *** jbadiapa has joined #openstack-ansible | 16:47 | |
| *** strattao has quit IRC | 16:53 | |
| *** kaiokmo has joined #openstack-ansible | 16:54 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_octavia master: Ensure that the flavor/network facts are available to all targets https://review.openstack.org/611389 | 16:54 |
| *** jbadiapa has quit IRC | 16:55 | |
| openstackgerrit | Nicolas Bock proposed openstack/openstack-ansible-os_octavia master: Fix updating amphora images https://review.openstack.org/611393 | 17:01 |
| nicolasbock | I think I fixed my problem with octavia's image update playbook (https://review.openstack.org/#/c/611393/) | 17:02 |
| nicolasbock | xgerman_ (IRC): Could you have a look? | 17:02 |
| * xgerman_ reading scrollback | 17:03 | |
| openstackgerrit | Michael Vollman proposed openstack/openstack-ansible-ops master: Add vars to configure prometheus module https://review.openstack.org/611394 | 17:04 |
| *** vollman has joined #openstack-ansible | 17:04 | |
| odyssey4me | nicolasbock: yep, that looks right - great catch! | 17:05 |
| odyssey4me | nicolasbock: oh, you might have to use openstack_image[0].id | 17:06 |
| odyssey4me | because the facts module returns a list, I think | 17:06 |
| *** hamzy has quit IRC | 17:07 | |
| *** hamzy has joined #openstack-ansible | 17:08 | |
| xgerman_ | nicolasbock: we use the same name for all images — so the id is in reality the name? | 17:08 |
| odyssey4me | xgerman_: the ID must be used, because we want to delete the previous image of the same name | 17:09 |
| odyssey4me | if we delete by name, we'd be deleting both the old and new one | 17:09 |
| odyssey4me | by deleting the old after uploading the new, octavia can still deploy while that process is happening | 17:09 |
| xgerman_ | my question is more: https://docs.ansible.com/ansible/2.6/modules/os_image_module.html says name is the image name and not id | 17:10 |
| xgerman_ | but you give name the is from before… so wonder if we need to issue both? | 17:10 |
| *** hamzy has quit IRC | 17:12 | |
| odyssey4me | good question - I thnk the shade library will use name or ID - let's take a peek | 17:12 |
| *** hamzy has joined #openstack-ansible | 17:12 | |
| odyssey4me | xgerman_ nicolasbock for ansible 2.5 (rocky/stein), 'name' needs to be provided for the image delete function, but it will accept an id for the name: https://github.com/ansible/ansible/blob/stable-2.5/lib/ansible/modules/cloud/openstack/os_image.py#L195 | 17:15 |
| *** fghaas has joined #openstack-ansible | 17:20 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests master: Convert test network setup to use the networkd role https://review.openstack.org/592596 | 17:20 |
| nicolasbock | Yes, the documentation is a bit confusing | 17:22 |
| xgerman_ | who is submitting that fix? | 17:22 |
| nicolasbock | I verified that it works with the `name` key and the `id` as value | 17:22 |
| odyssey4me | nicolasbock: did it work without using openstack_image[0].id instead of openstack_image.id ? | 17:24 |
| odyssey4me | my env is hanging due to too many processes :/ | 17:24 |
| cloudnull | prometheanfire -> https://review.openstack.org/592596 | 17:24 |
| nicolasbock | xgerman_ (IRC): I have to look at the module source to answer that. I'll do that. | 17:24 |
| nicolasbock | And what fix? | 17:24 |
| xgerman_ | fix to the od_image docs | 17:25 |
| xgerman_ | :-) | 17:25 |
| odyssey4me | I'll be happy to do that if no-one else wants to. | 17:26 |
| odyssey4me | I might be wrong, but it looked to me like 'id' is not actually even used. | 17:26 |
| *** tosky__ has joined #openstack-ansible | 17:26 | |
| macza | hey i know this is out of the openstack ansible scope but hoping i could get some help, in case someones run into this. im running ocata using nova with ceph, a compute hypervisor crashed and now all the instances wont connect to their drives after bringing the hypervisor back up | 17:27 |
| *** tosky has quit IRC | 17:27 | |
| nicolasbock | xgerman_ (IRC): I am on it :) | 17:27 |
| xgerman_ | :- | 17:27 |
| xgerman_ | 0 | 17:27 |
| *** kstev has quit IRC | 17:28 | |
| nicolasbock | xgerman_ (IRC): The thing with using both `name` and `id` was that the module errors out saying that there are multiple images with that name (I had two amphora images at the time) | 17:28 |
| *** tosky__ is now known as tosky | 17:28 | |
| xgerman_ | ok, makes sense… tose modules really need to be documented | 17:28 |
| odyssey4me | nicolasbock: hmm, perhaps that delete action should loop through the list | 17:28 |
| nicolasbock | Ah, good point | 17:29 |
| nicolasbock | odyssey4me: I think you are right about the modules and its use of `id` and `name`. It's a little confused I think. It sometimes uses the `name` and sometimes the `id`. I am still trying to decipher how exactly though... | 17:30 |
| nicolasbock | Sorry no, it's not using the `id` at all :( | 17:31 |
| odyssey4me | it seems to be to return id in the facts, but it never uses the module parameter | 17:31 |
| nicolasbock | But it's using the `name` in `cloud.name_or_id` which seems to explain why it works with `name = id` | 17:31 |
| odyssey4me | I'm not entirely sure why you'd ever want to use id when you could put id into the name parameter. | 17:31 |
| logan- | macza: check the auth caps for the cephx key that nova is using, ensure that it has the "osd blacklist" capability under mon | 17:32 |
| nicolasbock | xgerman_ (IRC): one fix: https://github.com/ansible/ansible/pull/47218 | 17:32 |
| nicolasbock | odyssey4me: good point | 17:32 |
| odyssey4me | the commit doesn't say why either :/ https://github.com/ansible/ansible/commit/de5fb8d98c40ad806420187bbf87fd3f642d7846 | 17:34 |
| logan- | macza: or better, use the rbd profile so you don't have to end up in this situation next time the required rbd auth caps change :) make your caps look something like: http://paste.openstack.org/show/732352/ | 17:34 |
| odyssey4me | nicolasbock: I guess when uploading an image, an ID can be set. I didn't know that glance allowed that, but meh. That's fine. | 17:35 |
| nicolasbock | Yes, me neither | 17:36 |
| *** DanyC has quit IRC | 17:36 | |
| nicolasbock | Seems kind of wrong to set and ID :) | 17:36 |
| nicolasbock | But the documentation should really explain that the `name` can also be an `id`. | 17:37 |
| macza | logan- thanks a bunch 'll try that | 17:37 |
| spotz | nicolasbock: I think that might be a holdover from when the project clients mainly used UUID vs Names, but ID is now sometimes used interchangeably in the help and docs | 17:38 |
| nicolasbock | Yes, good point spotz (IRC) | 17:39 |
| odyssey4me | nicolasbock: yeah, it should say something like 'The ID of the image when uploading an image' for the 'id' parameter. And for the 'name' parameter it should say something like 'The name of the image when uploading - or the name/ID of the image if deleting' | 17:39 |
| nicolasbock | Thanks odyssey4me I'll use that in my PR :) | 17:39 |
| odyssey4me | nicolasbock: you're welcome :) | 17:40 |
| macza | hey logan- shoudl i just reboot nodes after that change? | 17:42 |
| nicolasbock | https://github.com/ansible/ansible/pull/47222 | 17:42 |
| *** strattao has joined #openstack-ansible | 17:45 | |
| *** macza has quit IRC | 17:52 | |
| *** macza has joined #openstack-ansible | 17:52 | |
| prometheanfire | cloudnull: nice, watching it | 17:55 |
| *** kstev has joined #openstack-ansible | 17:56 | |
| *** kstev has quit IRC | 18:00 | |
| logan- | macza: I don't think you should need to, probably just reboot the instances | 18:03 |
| *** jbadiapa has joined #openstack-ansible | 18:06 | |
| *** electrofelix has quit IRC | 18:14 | |
| odyssey4me | logan-: could you review this if you have a moment? https://review.openstack.org/610992 | 18:16 |
| logan- | yep.. is this only needed on pike? | 18:18 |
| odyssey4me | logan-: yep, it seems so - we're not hitting the same issues any later | 18:18 |
| odyssey4me | logan-: I think it's because we basically rewrote the role in queens | 18:19 |
| logan- | ah ok | 18:20 |
| *** hamzaachi has joined #openstack-ansible | 18:20 | |
| *** jbadiapa has quit IRC | 18:21 | |
| *** fghaas has quit IRC | 18:23 | |
| *** bgmccollum has joined #openstack-ansible | 18:23 | |
| nicolasbock | I am getting lots of errors like "ERROR nova.console.websocketproxy error: [Errno 32] Broken pipe" in an AIO based on master. Is that a known issue in Nova? | 18:25 |
| mgariepy | anyone have a minute to review this one ? odyssey4me cloudnull ? :D https://review.openstack.org/#/c/605789/ | 18:35 |
| noonedeadpunk | folks, what to do, if run-upgrade.sh failed at one step? Is it safe to launch it one more time, or it's better to proceed with manual upgrade since failure step? | 18:40 |
| odyssey4me | noonedeadpunk: you can run it again, it tracks where it was | 18:41 |
| odyssey4me | if it doesn't work, then that's a bug which we should fix | 18:41 |
| noonedeadpunk | oh, that's great | 18:41 |
| odyssey4me | mgariepy: any chance you could review https://review.openstack.org/#/q/I7d1f07878dba7d21aecd59fa75e9117b50b886e6 ? | 18:41 |
| *** vnogin has joined #openstack-ansible | 18:42 | |
| mgariepy | sure give me a sec | 18:42 |
| noonedeadpunk | It failed on nova console setup with http://paste.openstack.org/show/732358/ | 18:42 |
| noonedeadpunk | but this commit exists, so it' strange why it failed. Probably I should clean facts like evrardjp suggested several days before | 18:43 |
| mnaser | i'm sorry for not being able to keep up with everything | 18:47 |
| mnaser | i'll do my best to do a catch up sprint later this week | 18:48 |
| odyssey4me | mnaser: it's not all on you, sir - we're all in this together :) | 18:48 |
| cloudnull | ++ | 18:48 |
| mnaser | i've wanted to check off some of the things we talked about at the ptg | 18:50 |
| mnaser | thank you odyssey4me and cloudnull | 18:50 |
| mnaser | https://review.openstack.org/#/c/610311/ is an increased coverage if everyone is ok with it | 18:50 |
| prometheanfire | https://i.imgur.com/UqbcYt2.jpg | 18:51 |
| odyssey4me | it'd be nice to get eyes on this backport too https://review.openstack.org/610363 | 18:51 |
| mnaser | odyssey4me: got it | 18:52 |
| *** ram5391 has joined #openstack-ansible | 18:56 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add osquery testing https://review.openstack.org/611226 | 18:58 |
| ram5391 | Is there a flag that is supposed to be set in the config to tell ansible to test against https keystone, and not http? | 18:58 |
| ram5391 | I tried setting: keystone_service_publicuri_proto: https and openstack_service_publicuri_proto: https, but my deployment is failing because it can't get to the http endpoint (https) is working fine | 18:59 |
| odyssey4me | ram5391: that was recently fixed in master, and the backport is pending review | 19:04 |
| odyssey4me | https://review.openstack.org/611322 | 19:04 |
| odyssey4me | ram5391: oh, I misread - by default keystone's public endpoint is https and internal endpoint is http | 19:05 |
| odyssey4me | what exactly are you trying to do? | 19:05 |
| ram5391 | odyssey4me I'm just trying to execute setup-hosts.yml. it fails on the "waiting for services to be up" because it's checking http://x.x.x.x:5000 and nothing is answering | 19:10 |
| ram5391 | I haven't mucked with the haproxy or default os-keystone files | 19:10 |
| ram5391 | weirdly then my public endpoint https is working fine, and my private is not. They're both on the same network, and I used the same IP for both | 19:11 |
| noonedeadpunk | odyssey4me: my problem was mostly related with network lag in DC, as git pull and git checkout after it worked. | 19:14 |
| odyssey4me | ram5391: then something must have failed earlier in the play, or something is misconfigured | 19:19 |
| odyssey4me | ram5391: also, setup-hosts doesn't do anything that would do a port check for port 5000 - that would be the keystone play which is well beyond setup-hosts | 19:20 |
| *** fghaas has joined #openstack-ansible | 19:21 | |
| *** hamzy has quit IRC | 19:21 | |
| *** hamzy has joined #openstack-ansible | 19:21 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/pike: Clean up role tests https://review.openstack.org/611364 | 19:22 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/pike: Allow configuration of startup timeout https://review.openstack.org/604999 | 19:23 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/pike: Enable the xinetd service for reboot https://review.openstack.org/605778 | 19:23 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/pike: Remove wsrep-new-cluster from cluster.cnf https://review.openstack.org/610992 | 19:23 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/queens: Clean up role tests https://review.openstack.org/611362 | 19:23 |
| ram5391 | odyssey4me sorry, you're correct. I misstyped, its the setup-openstack.yml playbook that fales | 19:23 |
| ram5391 | fails* hosts and infrastructure execute 100% succesfully | 19:24 |
| odyssey4me | ram5391: I would suggest inspecting why the keystone service failed to start rather than trying to add more overrides | 19:24 |
| ram5391 | and I'm following the hosts -> infrastructure -> openstack order | 19:24 |
| ram5391 | odyssey4me thats just it, it doesn't fail to start. I can get to https://x.x.x.x:5000 totally fine. | 19:25 |
| odyssey4me | ram5391: good, but if keystone is failing to respond then everything else will fail - identify why it failed and resolve that | 19:25 |
| odyssey4me | ok, in that case you need to look more carefully at the failed task result, perhaps with debug output | 19:26 |
| odyssey4me | IIRC the task does a localhost port check | 19:26 |
| ram5391 | can't I configure both public and private endpoints to interact with keystone on https? The weird thing is that haproxy and keystone are on the same machine. I looked at haproxy to see if maybe it wasn't listening to http requests, but I'm not familiar enough with haproxy to make sense of it. It _looks_ like the config might be setup to not allow localhost to talk to http? | 19:28 |
| odyssey4me | ram5391: you should be able to, but it's not recommended because the certs used are self-signed - you'd need real certs, and even then it's not a well tested path | 19:29 |
| ram5391 | this is the full log output: http://paste.openstack.org/show/732377/ | 19:29 |
| ram5391 | the machine I'm running the playbooks on, is the same machine that haproxy, keystone and all the other core services are on | 19:30 |
| odyssey4me | ram5391: also, they're not on the same machine - haproxy is acting as a front-end and forwarding to a container which is a wholly separate 'machine' | 19:30 |
| *** hamzy has quit IRC | 19:30 | |
| odyssey4me | it may be on the same host, but they are not the same machine | 19:30 |
| ram5391 | odyssey4me touche, but the containers are all picking up IP's on the same /16 that the host is configured with | 19:31 |
| odyssey4me | ram what release/branch is this? | 19:31 |
| ram5391 | checking now | 19:31 |
| odyssey4me | also, what is the failed task's name - the log you posted doesn't show the name of the task | 19:32 |
| ram5391 | odyssey4me I'm pulling master at: 3e458a8ef2d8b3ba3a94aee8de6427ca203b6c31 | 19:35 |
| ram5391 | and its failing at this task: TASK [os_keystone : Wait for services to be up] | 19:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Update galera_server SHA to include cluster fixes https://review.openstack.org/611419 | 19:36 |
| *** strattao has quit IRC | 19:37 | |
| odyssey4me | ram5391: ok, are you intending to pull master - because that's always going to be in a varying state | 19:37 |
| ram5391 | I tried pulling stable/rocky, but was having issues with that too. Since you said that got fixed in master, I'll try pulling again and see if its still an issue | 19:38 |
| *** strattao has joined #openstack-ansible | 19:38 | |
| *** macza has quit IRC | 19:43 | |
| *** macza has joined #openstack-ansible | 19:44 | |
| odyssey4me | ram5391: switching back and forth is also not a good idea | 19:46 |
| odyssey4me | rocky is released and stable | 19:46 |
| *** fghaas has quit IRC | 19:46 | |
| odyssey4me | some more fixes are trickling in, but it's in a generally good working state | 19:46 |
| odyssey4me | ram5391: are you trying, perhaps, to use the same ip for the internal and external LB addresses? | 19:47 |
| ram5391 | odyssey4me understandable, I am in fact | 19:47 |
| macza | thanks logan- you're a lifesaver | 19:47 |
| odyssey4me | ram5391: just don't do that - it does not work because http and https cannot listen on the same port | 19:47 |
| odyssey4me | ram5391: keep them separate | 19:48 |
| *** fghaas has joined #openstack-ansible | 19:48 | |
| odyssey4me | ram5391: once https://review.openstack.org/#/c/611322/ is merged (whenever that happens) and is available in a release, you might be able to have them both on the same IP - but I wouldn't bet on it... just use different IP's | 19:48 |
| *** strattao has quit IRC | 19:48 | |
| ram5391 | ah, that makes sense. But I can have another vif on the same nic and still be okay right? as long as its another IP? | 19:49 |
| odyssey4me | ram5391: yep, they can just be secondary IP's - if you could have keepalived bind them for you if you like | 19:49 |
| *** vnogin has quit IRC | 19:50 | |
| ram5391 | odyssey4me ah okay awesome, that makes sense then. I'll add another vif to my headnode interface, add the new IP and try again. Thank you! Will report back | 19:51 |
| *** vnogin has joined #openstack-ansible | 19:51 | |
| *** spatel has quit IRC | 19:53 | |
| *** jbadiapa has joined #openstack-ansible | 20:02 | |
| odyssey4me | ram5391: good luck! | 20:04 |
| *** fghaas has quit IRC | 20:06 | |
| *** hamzy has joined #openstack-ansible | 20:17 | |
| *** DanyC has joined #openstack-ansible | 20:33 | |
| *** fghaas has joined #openstack-ansible | 20:46 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible stable/pike: Stop MariaDB nodes to reset quorum during upgrade https://review.openstack.org/611438 | 20:52 |
| *** fghaas has quit IRC | 20:53 | |
| *** vnogin has quit IRC | 20:57 | |
| *** ansmith has quit IRC | 21:02 | |
| *** kaiokmo has quit IRC | 21:24 | |
| *** olivierb has quit IRC | 21:30 | |
| *** cshen has quit IRC | 21:42 | |
| *** priteau has quit IRC | 21:50 | |
| *** SimAloo has quit IRC | 21:53 | |
| *** hamzaachi has quit IRC | 21:58 | |
| logan- | macza: no prob glad it worked! | 22:18 |
| *** munimeha1 has quit IRC | 22:20 | |
| *** DanyC has quit IRC | 22:54 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/rocky: Clean up role tests https://review.openstack.org/611361 | 22:56 |
| *** weezS has quit IRC | 23:01 | |
| *** nsmeds__ has quit IRC | 23:08 | |
| *** rgogunskiy has joined #openstack-ansible | 23:09 | |
| *** rgogunskiy has quit IRC | 23:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Clean up role tests https://review.openstack.org/611359 | 23:13 |
| *** ram5391 has quit IRC | 23:15 | |
| *** macza has quit IRC | 23:17 | |
| *** tosky has quit IRC | 23:17 | |
| *** lbragstad is now known as lbragstad_503 | 23:26 | |
| *** cshen has joined #openstack-ansible | 23:38 | |
| *** cshen has quit IRC | 23:42 | |
| *** DanyC has joined #openstack-ansible | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
