| *** DanyC has quit IRC | 00:01 | |
| *** shyamb has quit IRC | 00:28 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_tempest stable/rocky: Set tempest concurrency to int only https://review.openstack.org/643789 | 00:52 |
|---|---|---|
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible stable/rocky: Add support to deploy mistral https://review.openstack.org/642614 | 00:54 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible stable/rocky: nova: restart instead of reloading services https://review.openstack.org/642842 | 00:54 |
| *** marst has joined #openstack-ansible | 01:31 | |
| *** DanyC has joined #openstack-ansible | 02:02 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_tempest stable/rocky: Set tempest concurrency to int only https://review.openstack.org/643789 | 02:03 |
| *** DanyC has quit IRC | 02:07 | |
| *** gkadam_ has joined #openstack-ansible | 02:11 | |
| *** gkadam_ has quit IRC | 02:17 | |
| *** BjoernT has joined #openstack-ansible | 02:19 | |
| *** irclogbot_0 has quit IRC | 02:23 | |
| openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-galera_server master: defaults: bump deployed version to 10.3.13 https://review.openstack.org/643798 | 02:25 |
| *** marst has quit IRC | 02:40 | |
| *** BjoernT has quit IRC | 02:48 | |
| *** BjoernT has joined #openstack-ansible | 02:49 | |
| *** nicolasbock has quit IRC | 02:58 | |
| *** shyamb has joined #openstack-ansible | 03:05 | |
| *** BjoernT has quit IRC | 03:06 | |
| *** shyam89 has joined #openstack-ansible | 03:11 | |
| *** shyam89 has quit IRC | 03:14 | |
| *** shyamb has quit IRC | 03:14 | |
| *** BjoernT has joined #openstack-ansible | 03:16 | |
| *** bgmccollum has quit IRC | 03:56 | |
| *** bgmccollum has joined #openstack-ansible | 04:00 | |
| guilhermesp | chandankumar: https://review.openstack.org/#/c/643789/ | 04:05 |
| guilhermesp | backported but seems we have two tempest tests failing. Have you seen this http://logs.openstack.org/89/643789/2/check/openstack-ansible-functional-opensuse-423/3873aa6/logs/ara-report/result/1b1e5b56-23b9-494c-817d-8a6be46b2e45/? | 04:07 |
| *** udesale has joined #openstack-ansible | 04:08 | |
| *** BjoernT has quit IRC | 04:20 | |
| *** chhagarw has joined #openstack-ansible | 04:45 | |
| *** ivve has quit IRC | 05:22 | |
| *** bgmccollum has quit IRC | 05:28 | |
| *** bgmccollum has joined #openstack-ansible | 05:31 | |
| *** mathlin has quit IRC | 05:34 | |
| *** mathlin has joined #openstack-ansible | 05:36 | |
| *** jbadiapa has joined #openstack-ansible | 06:06 | |
| *** miloa has joined #openstack-ansible | 06:10 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Bump SHAs for master https://review.openstack.org/642208 | 06:18 |
| *** ivve has joined #openstack-ansible | 06:30 | |
| *** ChosSimbaOne has quit IRC | 06:48 | |
| *** ChosSimbaOne has joined #openstack-ansible | 06:48 | |
| *** mkuf_ has joined #openstack-ansible | 06:51 | |
| *** mkuf_ is now known as mkuf | 06:51 | |
| *** phasespace has quit IRC | 06:54 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible master: Imported Translations from Zanata https://review.openstack.org/643864 | 06:55 |
| *** kopecmartin|off is now known as kopecmartin | 07:03 | |
| *** ivve has quit IRC | 07:07 | |
| *** luksky has joined #openstack-ansible | 07:08 | |
| *** DanyC has joined #openstack-ansible | 07:08 | |
| *** luksky has quit IRC | 07:18 | |
| *** pcaruana has joined #openstack-ansible | 07:19 | |
| *** DanyC has quit IRC | 07:24 | |
| *** cshen has joined #openstack-ansible | 07:29 | |
| jrosser | Morning | 07:34 |
| *** ivve has joined #openstack-ansible | 07:38 | |
| fnpanic | good morning | 07:44 |
| *** hamzaachi has joined #openstack-ansible | 07:57 | |
| *** hamzaachi has quit IRC | 07:59 | |
| *** DanyC has joined #openstack-ansible | 08:00 | |
| *** luksky has joined #openstack-ansible | 08:01 | |
| *** mkuf has quit IRC | 08:04 | |
| *** tosky has joined #openstack-ansible | 08:05 | |
| *** fnpanic has quit IRC | 08:06 | |
| *** tosky has quit IRC | 08:07 | |
| *** hamzaachi has joined #openstack-ansible | 08:08 | |
| *** phasespace has joined #openstack-ansible | 08:10 | |
| jrosser | chandankumar: has anything changed in tempest/os_tempest with the amount of concurrency? We see jobs failing tempest in the last couple of weeks with lots of load on keystone.... | 08:15 |
| *** tosky has joined #openstack-ansible | 08:25 | |
| *** priteau has joined #openstack-ansible | 08:25 | |
| chandankumar | jrosser yes | 08:38 |
| *** DanyC has quit IRC | 08:39 | |
| chandankumar | jrosser: https://review.openstack.org/#/c/643423/ | 08:39 |
| chandankumar | jrosser: what was the default concurrency set in the jobs? | 08:39 |
| chandankumar | default is 0 | 08:39 |
| jrosser | well that is a good question :) | 08:39 |
| jrosser | i think the answer is that it was never set http://codesearch.openstack.org/?q=tempest_run_concurrency&i=nope&files=&repos= | 08:40 |
| chandankumar | jrosser: from here I am seeing basically two for 4 http://codesearch.openstack.org/?q=tempest_concurrency&i=nope&files=&repos= | 08:40 |
| chandankumar | let me propose a pathc to set it to 2 | 08:41 |
| jrosser | what happens when it is 0? N-CPU or something? | 08:41 |
| chandankumar | jrosser: https://github.com/mtreinish/stestr/blob/master/stestr/commands/run.py#L75 -> no. of cpus count | 08:43 |
| jrosser | chandankumar: hmmm right yes it would certainly be good to define it as a small number by default then | 08:44 |
| chandankumar | in os_tempest itself or in openstack-ansible-tests? | 08:44 |
| jrosser | there should be some sensible default behaviour in os_tempest imho | 08:46 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Set tempest_run_concurrency defaults to 2 https://review.openstack.org/643880 | 08:48 |
| chandankumar | jrosser: I need some help on this review https://review.openstack.org/#/c/643573/ | 08:48 |
| chandankumar | jrosser: currently giving pip_install_upper_constraints is undefined http://logs.openstack.org/73/643573/1/check/openstack-ansible-deploy-aio_distro_lxc-opensuse-423/c015fad/job-output.txt.gz#_2019-03-15_14_17_53_508660 | 08:49 |
| *** luksky has quit IRC | 08:55 | |
| *** luksky has joined #openstack-ansible | 08:57 | |
| jrosser | chandankumar: i think that pip_install_upper_constraints is normally only defined for source installs | 08:57 |
| chandankumar | jrosser: but there tempest is installed from source? | 08:58 |
| chandankumar | *. | 08:58 |
| chandankumar | for rocky distro support is not there | 08:59 |
| chandankumar | let me make a single modification there | 08:59 |
| chandankumar | by setting tempest_install_method == source But not sure it will work | 08:59 |
| jrosser | hmmmm ok we had some discussion about this before and it's a but ugly | 09:01 |
| jrosser | *bit | 09:01 |
| *** simon-AS559 has joined #openstack-ansible | 09:02 | |
| * chandankumar waits for odyssey4me to come. | 09:04 | |
| *** mkuf has joined #openstack-ansible | 09:11 | |
| *** DanyC has joined #openstack-ansible | 09:18 | |
| *** fnpanic has joined #openstack-ansible | 09:27 | |
| *** luksky has quit IRC | 09:31 | |
| *** rgogunskiy has joined #openstack-ansible | 09:34 | |
| *** CeeMac has quit IRC | 09:34 | |
| *** CeeMac has joined #openstack-ansible | 09:40 | |
| CeeMac | morning | 09:40 |
| *** shyamb has joined #openstack-ansible | 09:41 | |
| *** tsi has joined #openstack-ansible | 09:51 | |
| tsi | Hello . I have an AIO for testing ( release 18.1.4) and i would like to add some of the openstack services such as telemetry . From https://docs.openstack.org/openstack-ansible/latest/user/aio/quickstart.html#bootstrap-the-aio-configuration as I understand it all i have to do is copy the correct yml.aio to /etc/openstack_deploy/conf.d/ and then do i have to re run bootstrap-aio and the rest of the openstack-ansible setup comma | 09:59 |
| *** shyamb has quit IRC | 10:04 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Pass ceph_mons as a list of inventory hosts https://review.openstack.org/643695 | 10:13 |
| *** shyamb has joined #openstack-ansible | 10:14 | |
| *** luksky has joined #openstack-ansible | 10:21 | |
| *** udesale has quit IRC | 10:28 | |
| *** shyamb has quit IRC | 10:30 | |
| *** shyamb has joined #openstack-ansible | 10:30 | |
| *** nicolasbock has joined #openstack-ansible | 10:35 | |
| *** jbadiapa has quit IRC | 10:44 | |
| *** simon-AS5591 has joined #openstack-ansible | 11:00 | |
| *** simon-AS559 has quit IRC | 11:04 | |
| *** mkuf has quit IRC | 11:08 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Set tempest_run_concurrency defaults to 2 https://review.openstack.org/643880 | 11:10 |
| *** flaviosr has joined #openstack-ansible | 11:18 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible stable/rocky: Always set pip vars for tempest install https://review.openstack.org/643573 | 11:24 |
| chandankumar | jrosser: ^^ switched it to source for tempest let's see how it goes! | 11:24 |
| *** mkuf has joined #openstack-ansible | 11:29 | |
| *** shyamb has quit IRC | 11:31 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-openstack_hosts master: Install user supplied CA certificates into system trust store https://review.openstack.org/641445 | 11:36 |
| *** kaiokmo has joined #openstack-ansible | 11:39 | |
| *** kaiokmo has quit IRC | 11:39 | |
| *** kaiokmo has joined #openstack-ansible | 11:40 | |
| *** udesale has joined #openstack-ansible | 11:45 | |
| *** maxbab has joined #openstack-ansible | 11:47 | |
| *** ansmith has quit IRC | 11:49 | |
| *** shyamb has joined #openstack-ansible | 11:56 | |
| *** udesale has quit IRC | 12:01 | |
| *** udesale has joined #openstack-ansible | 12:02 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_tempest stable/rocky: Revert "Pin stestr to 2.3.0 for rocky" https://review.openstack.org/643923 | 12:04 |
| odyssey4me | chandankumar jrosser o/ | 12:07 |
| jrosser | hello!! | 12:07 |
| *** cshen has quit IRC | 12:14 | |
| *** cshen has joined #openstack-ansible | 12:22 | |
| *** markvoelker has quit IRC | 12:22 | |
| jrosser | chandankumar: there is something I don't understand | 12:27 |
| jrosser | previously we didnt define tempest_run_concurrency | 12:27 |
| *** udesale has quit IRC | 12:27 | |
| jrosser | so according to this line http://git.openstack.org/cgit/openstack/openstack-ansible-os_tempest/tree/tasks/tempest_run.yml#n24 we should either have --concurrency <N> or --serial on the tempest run command | 12:28 |
| *** udesale has joined #openstack-ansible | 12:28 | |
| jrosser | but here for example, we have neither http://logs.openstack.org/95/643695/1/gate/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/d7458c2/logs/ara-report/ | 12:28 |
| tsi | Hello . I have an AIO for testing ( release 18.1.4) and i would like to add some of the openstack services such as telemetry . From https://docs.openstack.org/openstack-ansible/latest/user/aio/quickstart.html#bootstrap-the-aio-configuration as I understand it all i have to do is copy the correct yml.aio to /etc/openstack_deploy/conf.d/ rename them,and then do i have to re run bootstrap-aio and the rest of the openstack-ansible | 12:31 |
| DimGR | hi , i have opened this issue but not sure if its a bug or not https://bugs.launchpad.net/openstack-ansible/+bug/1820591 | 12:32 |
| openstack | Launchpad bug 1820591 in openstack-ansible "openstack-ansible AIO , heat role detects wrong IP " [Undecided,New] | 12:32 |
| chandankumar | odyssey4me: \o/ | 12:32 |
| jrosser | tsi: i think you can just drop those new files into conf.d and the run the playbooks for the new components you want to deploy | 12:32 |
| chandankumar | odyssey4me: https://review.openstack.org/#/c/643573/ I have updated this patch, not sure that is the correct fix | 12:33 |
| chandankumar | please have a look@ | 12:33 |
| odyssey4me | tsi no need to re-run bootstrap-aio, just run the playbooks that deploy openstack | 12:33 |
| odyssey4me | tsi ie setup-hosts, setup-infrastructure, setup-openstack | 12:33 |
| tsi | odyssey4me: ok thanks . | 12:33 |
| DimGR | jrosser thanks | 12:33 |
| jrosser | DimGR: that is indeed a bug - sort of | 12:34 |
| odyssey4me | jrosser chandankumar yeah, IIRC we use --serial for tempest runs because we found that when using concurrency we'd end up with differing amounts of tests executing on each run | 12:34 |
| jrosser | but i see --serial is missing | 12:34 |
| jrosser | like here http://logs.openstack.org/95/643695/1/gate/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/d7458c2/logs/ara-report/ | 12:35 |
| chandankumar | jrosser: yes | 12:35 |
| jrosser | sorry here http://logs.openstack.org/95/643695/1/gate/openstack-ansible-deploy-aio_lxc-ubuntu-bionic/d7458c2/logs/ara-report/result/1d75b12f-02d2-404a-bf56-4fa4d5a0d285/ | 12:35 |
| chandankumar | jrosser: something wrong | 12:35 |
| chandankumar | jrosser: https://review.openstack.org/#/c/643423/2/tasks/tempest_run.yml got merged | 12:35 |
| chandankumar | for concurrency | 12:36 |
| jrosser | oh | 12:36 |
| chandankumar | jrosser: http://logs.openstack.org/38/640138/2/check/openstack-ansible-functional-centos-7/44fd417/logs/ara-report/result/6ebae9b6-5ccd-45fd-9949-b361c3f63821/ it is from old | 12:37 |
| chandankumar | here also --serial is missing | 12:37 |
| noonedeadpunk | Hi everyone | 12:37 |
| jrosser | hmmm ok i think we need to get this back to --serial because it pretty much DoS on keystone | 12:37 |
| noonedeadpunk | Have a question - do I need to add tag to every task inside included play, or it's enough to place tag on the include? | 12:38 |
| chandankumar | jrosser: yes , looking in the jinja | 12:38 |
| jamesdenton | mornin folks | 12:39 |
| odyssey4me | noonedeadpunk the answer is, it depends - I believe there are some bugs in ansible depending on whether import or include are used, and whether they are included/imported plays/tasks - best is to try what you think is good and to test it | 12:41 |
| *** aedc has joined #openstack-ansible | 12:41 | |
| nicolasbock | noonedeadpunk: Depending on how you include, you can apply tags to he included tasks: https://docs.ansible.com/ansible/latest/modules/include_tasks_module.html#include-tasks-module | 12:43 |
| noonedeadpunk | odyssey4me nicolasbock: I see. Actually the question is regarding https://review.openstack.org/#/c/639960/2 - tasks/horizon_translations_update.yml is included like any other task in this role. And Frank says, that horizon-translations is not working to him. | 12:44 |
| noonedeadpunk | So for me it looks like that I need to test every tag for the role | 12:44 |
| odyssey4me | noonedeadpunk I can confirm that using import_tasks and tags like https://github.com/openstack/openstack-ansible-os_glance/blob/master/tasks/main.yml#L51-L68 definitely works | 12:44 |
| jamesdenton | If it help, the changes here allowed existing tags to work again: https://review.openstack.org/#/c/637026/ | 12:45 |
| noonedeadpunk | jamesdenton: Yep, I think I need to do the same for horizon | 12:47 |
| jrosser | DimGR: by making the changes in your bug report you will have made the basic heat cli work, but unless your created servers/vm can contact your internal VIP (possibly for a lab, highly unlikely for production) then heat (and magnum) wont work properly | 12:47 |
| noonedeadpunk | So the question now - should I do it before mentioned patch, or right after it? | 12:48 |
| *** ansmith has joined #openstack-ansible | 12:48 | |
| *** shyamb has quit IRC | 12:48 | |
| noonedeadpunk | Thanks for your answers:) | 12:53 |
| noonedeadpunk | I mean, for pointing me to import/include | 12:53 |
| *** pcaruana|afk| has joined #openstack-ansible | 12:54 | |
| *** pcaruana has quit IRC | 12:56 | |
| jamesdenton | Trying to rerun os-neutron-install.yml and getting hung up on the python_venv_build tasks: https://pasted.tech/pastes/88f0b29f9df256c74cd78a6b32a71b29a4235d03. Haven't OSA'd in a while. Any known issue here, or should I keep digging? | 13:09 |
| guilhermesp | mornings! | 13:12 |
| jamesdenton | hello! | 13:12 |
| guilhermesp | jamesdenton: o/ | 13:13 |
| *** BjoernT has joined #openstack-ansible | 13:13 | |
| odyssey4me | jamesdenton I'm guessing that's a stale fact issue - force the regathering of facts, or just wipe the fact cache. | 13:13 |
| guilhermesp | hum ok chandankumar so that case just wait the new tempest release to be able to unblock some of the rocky jobs? these are some examples https://review.openstack.org/#/c/642614/7 https://review.openstack.org/#/c/642842/ | 13:13 |
| chandankumar | guilhermesp: for rocky release we have a work in progress patch https://review.openstack.org/#/c/643573/ | 13:14 |
| guilhermesp | huuum ok chandankumar watching it | 13:24 |
| chandankumar | odyssey4me: http://logs.openstack.org/73/643573/2/check/openstack-ansible-deploy-aio_distro_lxc-opensuse-423/f78a1a5/job-output.txt.gz#_2019-03-18_12_35_37_745883 still failing with same error | 13:26 |
| chandankumar | odyssey4me: 'pip_install_upper_constraints' is undefined | 13:26 |
| odyssey4me | chandankumar yep, will look into it a bit later | 13:26 |
| *** marst has joined #openstack-ansible | 13:26 | |
| odyssey4me | I have a call in a bit which I need to prep for. | 13:27 |
| chandankumar | odyssey4me: sure | 13:27 |
| jamesdenton | odyssey4me Turns out I had to rebuild facts for the repo container. Of course I purged them beforehand. Would have been nice to compare. | 13:30 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: setup: allow custom conductor host https://review.openstack.org/628743 | 13:33 |
| *** hwoarang has quit IRC | 13:37 | |
| *** hwoarang has joined #openstack-ansible | 13:38 | |
| *** marst has quit IRC | 13:41 | |
| *** jwitko has joined #openstack-ansible | 13:47 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: [DNM] serial testing Syntex is correct, need to dig why serial flag is not showing https://review.openstack.org/643947 | 13:48 |
| *** pcaruana|afk| has quit IRC | 13:48 | |
| mnaser | i am not sure what to do about opensuse in rocky | 13:51 |
| odyssey4me | mnaser it is a little odd that it's only opensuse hitting the issue, but I'll spend some time after my call looking into it | 13:59 |
| mnaser | thanks odyssey4me and good luck :) | 14:01 |
| nicolasbock | I am pulling up an env to test the rocky/opensuse issues | 14:03 |
| nicolasbock | I am seeing a different failure though | 14:04 |
| jrosser | mnaser: i thought this may be some positive movement becasue it looks like an opensuse/rocky test has worked there https://review.openstack.org/#/c/643550/ | 14:06 |
| jrosser | but not in the integrated repo though | 14:07 |
| DimGR | jrosser in other words the changes i did were not enough for heat to work properly ? this is a test lab and the created instances can contact the internal vip | 14:07 |
| *** marst has joined #openstack-ansible | 14:07 | |
| jrosser | DimGR: ok, that should be OK then I think | 14:08 |
| DimGR | i have had instaces working fine for few weeks , its only now that i needed heat and found this bug | 14:09 |
| DimGR | hope it gets resolved with a newer release | 14:09 |
| jrosser | DimGR: there is a bit of ambiguity in the heat code with how keystone endpoints are handled, see the details here http://lists.openstack.org/pipermail/openstack-discuss/2019-February/002925.html | 14:12 |
| jrosser | so it's not really an OSA problem as such, we can configure either internal or external endpoints in heat.conf right now, just different things break depending on which you pick, sadly | 14:13 |
| * DimGR checking | 14:13 | |
| *** mgariepy has quit IRC | 14:18 | |
| mkuf | Hey there, I'd like to serve all my keystone endpoints (public/internal/admin) via https. Setting 'keystone_service_proto: https' in user_variables.yaml and running haproxy-install.yml and os-keystone-install.yml works fine and all Keystone-Endpoints are available via https. Running another playbook (e.g. os-glance-install) fails at the 'ensure glance service' stage, as it is still trying to | 14:24 |
| mkuf | connect via http. http://paste.openstack.org/show/747967/ Is this behaviour intended or am I missing something essential here? | 14:24 |
| mkuf | ^ this occures in an AIO setup on Queens 17.1.7 | 14:26 |
| evrardjp | guilhermesp: jrosser can you vote on https://review.openstack.org/#/c/642738/1 ? | 14:31 |
| evrardjp | the freeze has merged I can finally release | 14:31 |
| *** nurdie has joined #openstack-ansible | 14:31 | |
| jrosser | evrardjp: done | 14:31 |
| evrardjp | thanks | 14:32 |
| jrosser | mkuf: it looks to me like something is making an https request but finding a certificate that is not trusted, but there is not much context in your error message really | 14:33 |
| *** pcaruana has joined #openstack-ansible | 14:33 | |
| *** BjoernT has quit IRC | 14:35 | |
| *** maxbab has quit IRC | 14:37 | |
| *** BjoernT has joined #openstack-ansible | 14:40 | |
| *** shyamb has joined #openstack-ansible | 14:41 | |
| cloudnull | mornings | 14:42 |
| evrardjp | morning | 14:43 |
| *** xenefix has joined #openstack-ansible | 14:50 | |
| cloudnull | hows it on this fine day of days :) | 14:52 |
| mkuf | jrosser: would a full output of the playbook run be more helpful or are there any other switches I can add to assist troubleshooting except -vvv? | 14:52 |
| mkuf | jrosser: the url at the end of line 8 uses http, if this is a certificate error shuldn't there be any ssl exception? | 14:52 |
| spotz | hey cloudnull evrardjp | 14:53 |
| evrardjp | hey | 14:53 |
| jrosser | mkuf: oh right, missed that :/ | 14:53 |
| cloudnull | hola | 14:54 |
| cloudnull | evrardjp have you messed much with the nova_api db things? | 14:54 |
| cloudnull | or anyone for that matter? | 14:54 |
| spotz | cloudnull: Beyond manual setups I have not | 14:55 |
| cloudnull | with cell rabbit connection information stored in the db, i'm wondering what the official steps to update the db are, besides running an `update` query | 14:56 |
| cloudnull | IE: https://pasted.tech/pastes/09b13478ca0dddd56b0779ee42f35a6a318a4d33 | 15:01 |
| cloudnull | and a service restart does not update the fields | 15:01 |
| cloudnull | kinda of a bummer when trying to move/scale rabbitmq nodes. | 15:02 |
| *** BjoernT_ has joined #openstack-ansible | 15:02 | |
| cloudnull | same story for db connections | 15:02 |
| *** shyamb has quit IRC | 15:03 | |
| *** phasespace has quit IRC | 15:03 | |
| spotz | cloudnull: I think it's changed with a restart of the nova api, etc. Least when doing installs if you need to change that's the proceedure. I do not believe there is a reload function | 15:03 |
| cloudnull | on another note, anyone want to give this a shove through https://review.openstack.org/#/c/643612 ? | 15:03 |
| spotz | looking | 15:04 |
| *** BjoernT has quit IRC | 15:05 | |
| cloudnull | spotz I restarted the services a few times, in fact went through a little hair pulling event last weekend trying to figure out why my cluster was so unstable: http://eavesdrop.openstack.org/irclogs/%23openstack-ansible/%23openstack-ansible.2019-03-17.log.html#t2019-03-17T04:42:11 | 15:06 |
| cloudnull | I removed the venvs, rebuilt all the things, and it wasn't until I started hacking on oslo messaging and logging all kinds of random things that I caught the bad rabbitmq connection string. | 15:07 |
| *** BjoernT has joined #openstack-ansible | 15:07 | |
| DimGR | jrosser i read the pipermail list about it . Its the mother of all bugs this heat bug | 15:07 |
| spotz | cloudnull: oh oh.... I think there's a second setting if you're having issues let me think@ | 15:07 |
| cloudnull | ++ would greatly appreciate it :) | 15:07 |
| spotz | cloudnull: Bringing up old notes give me a few | 15:09 |
| *** BjoernT_ has quit IRC | 15:10 | |
| chandankumar | jrosser: hello | 15:13 |
| chandankumar | jrosser: the serial syntex was correct | 15:13 |
| jrosser | mnaser: i dug a bit at your galera upgrade patch and it can't find wsrep_sst_xtrabackup-v2 | 15:13 |
| chandankumar | jrosser: but on this patch https://review.openstack.org/#/c/643947/ -> serial is alive now http://logs.openstack.org/47/643947/1/check/openstack-ansible-functional-centos-7/aa78849/logs/ara-report/result/ba5c68f4-2c97-4835-998f-30bcb61707fe/ | 15:13 |
| jrosser | i'm not sure where that is supposed to be installed from | 15:14 |
| mnaser | jrosser: i looked so much and i couldn't find it either | 15:15 |
| mnaser | :\ | 15:15 |
| jrosser | idk which package that is supposed to be in | 15:15 |
| mnaser | i looked and couldn't find it :\ | 15:15 |
| jrosser | there are a ton of other percona-* packages and i was going to just throw some of those on to see what they contained | 15:15 |
| jrosser | i've also seen those galera "this is probably a bug" error messages | 15:16 |
| spotz | cloudnull: What port is rabbit? Not finding what I'm looking for but I may be misremembering what I need to look for:) | 15:16 |
| *** miloa has quit IRC | 15:17 | |
| *** tsi has quit IRC | 15:17 | |
| cloudnull | spotz: 5671 | 15:18 |
| cloudnull | though could also be on 5672 | 15:18 |
| cloudnull | ssl vs non-ssl | 15:18 |
| chandankumar | jrosser: I am modifying the dnm patch | 15:18 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Bring back the --serial flag alive https://review.openstack.org/643947 | 15:21 |
| spotz | cloudnull: this is going to drive me insane as I've told people this setting before and can't find it! | 15:24 |
| *** BjoernT has quit IRC | 15:37 | |
| *** BjoernT_ has joined #openstack-ansible | 15:37 | |
| *** mgariepy has joined #openstack-ansible | 15:38 | |
| odyssey4me | jrosser mnaser looks like the heat discussion on the ML has yielded some good advice and a few actionable options - one is a workaround we can implement (adding '[clients_keystone]/auth_uri') in rocky, and the other is a more suitable fix for the future | 15:46 |
| jrosser | odyssey4me: we got a patch already https://review.openstack.org/#/c/642812/ | 15:47 |
| jrosser | i think there are two parts, a new config variable which shouldn't change any behaviour unless set | 15:47 |
| jrosser | then a later revisit of the whole keystone endpoint handling at some point later | 15:47 |
| odyssey4me | boom gshippey is rocking it! | 15:48 |
| jrosser | i don't think anything can be done right now with the existing config options | 15:48 |
| chandankumar | odyssey4me: jrosser https://review.openstack.org/#/c/643947/ will bring serial flag alive | 15:49 |
| chandankumar | odyssey4me: jrosser I am thinking of doing the same for all long shell scripts, if it is ok? | 15:49 |
| odyssey4me | jrosser it seems that adding '[clients_keystone]/auth_uri' should work? | 15:49 |
| jrosser | adjusting any of the existing vars away from internal endpoint breaks our deployment here, in some way | 15:50 |
| odyssey4me | chandankumar anything to make things readable is good - assuming that works... not sure if you might hit some whitespace shenanigans there | 15:50 |
| odyssey4me | jrosser ah ok, then this might get a bit complicated because it'll involve a stable backport... but let's get something into master first and worry about that later | 15:51 |
| jrosser | yes, we've tried to keep that patch super super minimal and not change any existing (maybe wrong) behaviour so that it will backport | 15:51 |
| mnaser | woot. Yeah, maybe might be good | 15:54 |
| mnaser | To update the ML | 15:54 |
| *** cshen has quit IRC | 15:56 | |
| trident | cloudnull: Just a heads up if you have messaging related instabilities: https://bugs.launchpad.net/oslo.messaging/+bug/1800957 | 15:57 |
| openstack | Launchpad bug 1800957 in oslo.messaging (Ubuntu) "Upgrading to pike version causes rabbit timeouts with ssl" [Undecided,New] | 15:57 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_heat master: [DNM] heat tempest tests finding https://review.openstack.org/630694 | 15:57 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_heat master: [DNM] heat tempest tests finding https://review.openstack.org/630694 | 15:57 |
| cloudnull | trident: oslo.messaging is working great. Sadly the issue I have at this time is related to the rabbit connection string being stored in the nova-api db and not being updated when I change things. | 15:59 |
| cloudnull | I guess maybe that is something that should be handled in oslo? | 15:59 |
| *** weshay|rover is now known as weshay | 16:02 | |
| trident | cloudnull: Yeah, that's not dynamic at all. But might make sense to make it so... | 16:02 |
| trident | As far as I have seen that is. Have always had to change it manually... | 16:02 |
| trident | Or with nova-manage | 16:02 |
| trident | cloudnull: I am not sure if it is supposed to be dynamically updated though. Might make more sense to have osa update it if rabbit settings have changed? | 16:05 |
| *** mkuf has quit IRC | 16:11 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/rocky: Pass ceph_mons as a list of inventory hosts https://review.openstack.org/644265 | 16:12 |
| *** cmart has joined #openstack-ansible | 16:12 | |
| jrosser | mnaser: magnum related, if anyone ever needs to use magnum.conf/openstack_ca_file=<blah> I did this https://review.openstack.org/#/c/641445/ | 16:13 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/queens: Pass ceph_mons as a list of inventory hosts https://review.openstack.org/644267 | 16:14 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible stable/pike: Pass ceph_mons as a list of inventory hosts https://review.openstack.org/644268 | 16:15 |
| odyssey4me | cloudnull trident I dunno why it's in the DB, but it makes sense to me that if it changes in the .conf file, then on the next service load it should update it in the DB... or it should just not be in the DB. :p Maybe this is another one of those things that really should go to the ML for discussion and to figure out a solution. Either that, or perhaps a discussion in #openstack-nova at least. | 16:15 |
| trident | cloudnull: The case here I think is that you might want to use different settings for each cell. The transport_url there is set when creating the cell with nova-manage cell__v2 create_cell | 16:19 |
| odyssey4me | trident if that is the case, then perhaps we should use nova-manage to always set the transport URL to whatever's current as a task in the nova role? That way it would always be the current one and we could forget about the hassle | 16:21 |
| *** udesale has quit IRC | 16:21 | |
| *** BjoernT_ has quit IRC | 16:21 | |
| odyssey4me | ideally we should only set it again if it is changing, but then we'd have to figure out a way to determine the change - it might just be simpler to force it every time | 16:22 |
| trident | cloudnull: Hm, the behaviour seem to have changed in OSA though. It doesn't seem to set transport-url when creating cells any longer. IIRC if you don't specify transport_url in cell db, it will use the default from config. | 16:22 |
| trident | https://github.com/openstack/openstack-ansible-os_nova/blob/ab9642f95ca870f2616dfab3f9e22937646cc648/tasks/nova_db_setup.yml#L121 | 16:22 |
| chandankumar | odyssey4me: I will propose a seperate patch to fix all long shell script for os_tempest tomorrow! | 16:22 |
| *** cmart has quit IRC | 16:23 | |
| *** DanyC has quit IRC | 16:26 | |
| trident | cloudnull, odyssey4me: Changed here: https://github.com/openstack/openstack-ansible-os_nova/commit/2448db8064a6ef019f4fa75d700345ca65a905e1#diff-794858b9a821f0ebca63fffea0e0143b | 16:27 |
| *** DanyC has joined #openstack-ansible | 16:27 | |
| *** shyamb has joined #openstack-ansible | 16:28 | |
| trident | So, for transport_url this should no longer be an issue for new installations as transport-url is no longer provided when creating the cells so it will use the default from config which in the majority of cases will make sense. | 16:30 |
| *** BjoernT has joined #openstack-ansible | 16:31 | |
| *** DanyC has quit IRC | 16:31 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Always set pip vars for tempest install https://review.openstack.org/643573 | 16:32 |
| trident | However I'd think for an existing installation the easiest way to go if you just want it to use default is to just clear it out of the db (with the reservation that I have not tested it). | 16:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Always set pip vars for tempest install https://review.openstack.org/643573 | 16:33 |
| odyssey4me | chandankumar mnaser ^ I think that will sort out rocky's current integrated build failure | 16:34 |
| chandankumar | odyssey4me: ah nice thanks guilhermesp ^^ will fix tempest issue! | 16:37 |
| *** shyamb has quit IRC | 16:46 | |
| *** SmearedBeard has joined #openstack-ansible | 16:48 | |
| guilhermesp | cool! thanks chandankumar odyssey4me | 16:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/rocky: Always set pip vars for tempest install https://review.openstack.org/643573 | 16:51 |
| *** mmercer has joined #openstack-ansible | 16:52 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Revert "Freeze for milestone release" https://review.openstack.org/642738 | 16:55 |
| *** hamzaachi has quit IRC | 17:04 | |
| *** ivve has quit IRC | 17:13 | |
| *** DanyC has joined #openstack-ansible | 17:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Update role for new source build process https://review.openstack.org/620340 | 17:13 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client master: Add additional linked libs for cephfs https://review.openstack.org/643612 | 17:17 |
| *** BjoernT has quit IRC | 17:19 | |
| *** DanyC has quit IRC | 17:21 | |
| *** DanyC has joined #openstack-ansible | 17:22 | |
| *** simon-AS5591 has quit IRC | 17:22 | |
| *** kopecmartin is now known as kopecmartin|off | 17:22 | |
| *** jwitko has left #openstack-ansible | 17:24 | |
| evrardjp | jrosser: are you there? | 17:34 |
| evrardjp | well not only jrosser, people that are really old in here, for example cloudnull odyssey4me.... | 17:34 |
| chandankumar | odyssey4me: jrosser mnaser cloudnull https://review.openstack.org/#/c/643947/ can we get this merged, it fixes one of the oldest bug | 17:35 |
| evrardjp | If you were to redo multi os distro support, would you make sure all the distros are configured the same, or like we do it today, embracing the differences per distro? | 17:35 |
| evrardjp | I would do it like we do, but I am not strongly in favour or not | 17:35 |
| chandankumar | bug == never filed | 17:36 |
| odyssey4me | evrardjp https://media.giphy.com/media/K6aZlaZEJjt4s/giphy.gif | 17:36 |
| evrardjp | hahaha | 17:37 |
| evrardjp | you're totally like that now | 17:37 |
| odyssey4me | evrardjp well, we do a weird mix now, which is a bit confusing - and I find that people who try it, and the rest of us get all confused all the time too | 17:37 |
| evrardjp | yeah | 17:37 |
| evrardjp | I am not surprised | 17:37 |
| evrardjp | I wonder what we can do better | 17:38 |
| odyssey4me | some packages assume the setup which they use for their 'native' tooling | 17:38 |
| evrardjp | or more | 17:38 |
| odyssey4me | we try to block that and smash it into what we do | 17:38 |
| evrardjp | "what did we learn" | 17:38 |
| odyssey4me | the trouble is that every distro does things differently | 17:38 |
| odyssey4me | so, quite frankly... just don't do it IMO | 17:38 |
| evrardjp | yeah, but I wonder if we should say "no let's do it OUR way, and disregard ALL distro" or the other way around | 17:39 |
| odyssey4me | either pick one distro and only support it on that, or have each distro setup in the way each distro packager expects things setup | 17:39 |
| cloudnull | evrardjp I'd probably keep things using the current pattern of embracing the differences. early on we tired to do it all ONE way and that was fraught with peril. | 17:39 |
| cloudnull | i think we're still paying the price for some of those decisions | 17:39 |
| odyssey4me | the cost of maintaining it all is pretty high though, because you're always context switching | 17:39 |
| chandankumar | distro == how they compile the stuff, restart the services and where they dump the file, but distro makes some things stable so that one can use it | 17:39 |
| evrardjp | odyssey4me: I agree with you, it's harder, because it means more pressure on maintainers | 17:40 |
| evrardjp | "I don't know opensuse way" | 17:40 |
| odyssey4me | yeah - we've also got a few things where we symlink things so that the differences between distributions are immaterial, but that's more maintenance | 17:40 |
| evrardjp | yet, I believe cloudnull is right too: There is a price of doing everything the same way: "Oh god selinux broke" | 17:40 |
| odyssey4me | so, I think it really comes down to what you're trying to achieve | 17:41 |
| evrardjp | odyssey4me: to be honest, I kinda hate that, it's less obvious for the operations at first sight | 17:41 |
| evrardjp | odyssey4me: yeah, nothing perfect in here :) | 17:41 |
| odyssey4me | I would think that if we only had one way of deploying - eg: with nspawn containers - and we made a rule that the containers would only be one distro - then life would be quite simple. | 17:42 |
| odyssey4me | But alas, distro vendor licensing makes that near impossible. | 17:42 |
| evrardjp | Well I am thinking of simplifying things for opensuse -- I want to have less moving parts and get opensuse back on track on my own spare time | 17:42 |
| evrardjp | I believe the problems would still happen: people don't know this and maybe don't want to maintain it | 17:43 |
| evrardjp | next to that, I have $dayjob which kinda has the same kind of problem, and I wonder what are the best way forward | 17:43 |
| evrardjp | I don't want to repeat mistakes :) | 17:43 |
| odyssey4me | If there was an easy way to map out the differences and just carry them like some sort of constant, that'd be nice. | 17:45 |
| evrardjp | you mean a lib that projects like ansible/puppet/anything could use? | 17:46 |
| evrardjp | EVIL | 17:46 |
| evrardjp | or you mean a standard? | 17:46 |
| *** SmearedBeard has quit IRC | 17:46 | |
| evrardjp | https://xkcd.com/927/ | 17:46 |
| odyssey4me | lol, yeah | 17:46 |
| odyssey4me | let's not do that | 17:46 |
| *** bgmccollum has quit IRC | 17:47 | |
| chandankumar | jrosser: cloudnull thanks! | 17:48 |
| odyssey4me | chandankumar arxcruz I've hit a bit of a block with regards to stackviz and I wonder if you're available to talk it through a bit? | 17:49 |
| arxcruz | odyssey4me: sure | 17:49 |
| arxcruz | whatsup ? | 17:49 |
| odyssey4me | currently stackviz is installed by default in the tempest role (which I wonder if appropriate, given that stackviz is more of a test tool than a production tool) | 17:50 |
| chandankumar | odyssey4me: /me will help tomorrow. mind is not working, in sleeping/watching movie mode | 17:50 |
| odyssey4me | it's installed from the tarball: https://github.com/openstack/openstack-ansible-os_tempest/blob/master/defaults/main.yml#L334 | 17:50 |
| chandankumar | but arxcruz can help you! He is a super awesome guy! | 17:51 |
| arxcruz | odyssey4me: okay... | 17:51 |
| *** chandankumar is now known as raukadah | 17:51 | |
| odyssey4me | I'm working on changing the way that the python builds are done to get rid of our 'repo build' and exclusively use the venv build role, but I'm hitting the situation where I can build a stackviz wheel but I have no way to then tell pip to use that wheel, because there is no constraint to work with | 17:52 |
| *** cmart has joined #openstack-ansible | 17:52 | |
| odyssey4me | Of course I can change the way it installs to use git+ like we do for everything else, then it would install using the same mechanism and model and life can carry on - but then I wonder how that would affect TripleO's test process. | 17:52 |
| arxcruz | odyssey4me: it won't | 17:53 |
| arxcruz | we are starting to use os_tempest | 17:53 |
| arxcruz | as far as stackviz generate the data, in the right directory, we are collecting it | 17:53 |
| odyssey4me | ok, so if I switch it up to use git instead, that would be fine? | 17:53 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Add the ability to set the rabbitmq bind address https://review.openstack.org/644379 | 17:53 |
| arxcruz | odyssey4me: yeah | 17:53 |
| arxcruz | we used tarball, because... i don't remember why | 17:53 |
| arxcruz | ah, yeah, because the git comes with the whole node stuff | 17:54 |
| arxcruz | we just wanted to not waste too much time on that | 17:54 |
| arxcruz | odyssey4me: oh, yeah, so, the tarball have the html already generated | 17:55 |
| arxcruz | the git doesn't | 17:55 |
| arxcruz | if you use git, you need to generate all those node stuff to generate the html | 17:55 |
| arxcruz | that's why we weren't using it | 17:55 |
| odyssey4me | oh dear :/ | 17:55 |
| arxcruz | before, I had generate it, and added on tripleo side | 17:55 |
| arxcruz | odyssey4me: maybe we can ask to have in the tarball the requirements | 17:56 |
| odyssey4me | ok, so y best option here is to just work out a way around this instead | 17:56 |
| odyssey4me | *my | 17:56 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: [DNM] Convert dynamic includes to static imports https://review.openstack.org/644380 | 17:56 |
| odyssey4me | ok, I think I have an idea which is pretty simple and will allow us to just keep using it as-is. | 17:57 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Add the ability to set the rabbitmq bind address https://review.openstack.org/644379 | 17:58 |
| *** gyee has joined #openstack-ansible | 17:58 | |
| arxcruz | odyssey4me: cool :) | 18:06 |
| *** cshen has joined #openstack-ansible | 18:06 | |
| arxcruz | render the html takes like 10 minutes | 18:06 |
| arxcruz | which is too much for us :/ | 18:06 |
| *** luksky has quit IRC | 18:10 | |
| * mnaser thinks take things as distros suggest is the path to least resistance | 18:12 | |
| mnaser | typically, distros ship pretty reliable and well integrated/tested things like mariadb/rabbit/etc | 18:12 |
| mnaser | so install from source should work regardless of distro, in the samy way | 18:13 |
| odyssey4me | doing something like that is easier to maintain from a code standpoint, but stifles innovation somewhat | 18:14 |
| odyssey4me | we've managed to innovate quite a bit and do thing well before any distro has even tried in many cases, but perhaps that's all now tech debt and we'll have to change it | 18:14 |
| mnaser | odyssey4me: well this is only for the infrastructure components and i doubt we'd have much innovation out of rabbit/galera/etc. | 18:16 |
| mnaser | we deploy openstack from source the same way across all environments, and when doing distro installs, we do them the way the distro wants it | 18:17 |
| *** cshen has quit IRC | 18:18 | |
| *** BjoernT has joined #openstack-ansible | 18:32 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-role-python_venv_build master: Add toggle for python wheel build process https://review.openstack.org/644391 | 18:34 |
| NobodyCam | Good Morning OSA folks | 18:34 |
| NobodyCam | off the wall question, does octavia require a flat network? | 18:35 |
| *** BjoernT_ has joined #openstack-ansible | 18:36 | |
| NobodyCam | I attempted to add as vlan and get `Invalid input for operation: Flat provider networks are disabled.` it is attempting to add as a flt | 18:36 |
| *** BjoernT has quit IRC | 18:37 | |
| *** DanyC has quit IRC | 18:39 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest master: Ensure stackviz installs from tarball URL only https://review.openstack.org/637503 | 18:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance master: Update role for new source build process https://review.openstack.org/620340 | 18:43 |
| *** ivve has joined #openstack-ansible | 18:45 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-os_tempest stable/rocky: Set tempest concurrency to int only https://review.openstack.org/643789 | 18:47 |
| *** BjoernT_ has quit IRC | 19:00 | |
| *** dave-mccowan has joined #openstack-ansible | 19:05 | |
| *** chhagarw has quit IRC | 19:06 | |
| *** BjoernT has joined #openstack-ansible | 19:10 | |
| *** BjoernT has quit IRC | 19:11 | |
| *** BjoernT has joined #openstack-ansible | 19:14 | |
| *** bgmccollum has joined #openstack-ansible | 19:18 | |
| *** DanyC has joined #openstack-ansible | 19:53 | |
| *** DanyC_ has joined #openstack-ansible | 19:58 | |
| *** luksky has joined #openstack-ansible | 20:01 | |
| *** DanyC has quit IRC | 20:02 | |
| *** irclogbot_0 has joined #openstack-ansible | 20:12 | |
| *** cshen has joined #openstack-ansible | 20:15 | |
| *** BjoernT has quit IRC | 20:16 | |
| *** fresta_ has quit IRC | 20:17 | |
| *** cshen has quit IRC | 20:19 | |
| *** irclogbot_0 has quit IRC | 20:25 | |
| *** irclogbot_0 has joined #openstack-ansible | 20:27 | |
| *** hamzaachi has joined #openstack-ansible | 20:39 | |
| *** dave-mccowan has quit IRC | 20:39 | |
| *** BjoernT has joined #openstack-ansible | 20:40 | |
| *** cmart has quit IRC | 20:42 | |
| *** dave-mccowan has joined #openstack-ansible | 20:44 | |
| guilhermesp | odyssey4me nice, already unblocked some of my PRs https://review.openstack.org/#/c/643573/5 | 20:49 |
| guilhermesp | and to unblock some others, once the above one is merged, I appreciate any review https://review.openstack.org/#/c/643789/ | 20:50 |
| *** pcaruana has quit IRC | 20:54 | |
| *** ansmith has quit IRC | 20:57 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible stable/rocky: Pass ceph_mons as a list of inventory hosts https://review.openstack.org/644265 | 20:58 |
| *** irclogbot_0 has quit IRC | 21:05 | |
| *** irclogbot_0 has joined #openstack-ansible | 21:06 | |
| *** priteau has quit IRC | 21:12 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-rabbitmq_server master: Add the ability to set the rabbitmq bind address https://review.openstack.org/644379 | 21:25 |
| *** irclogbot_0 has quit IRC | 21:26 | |
| *** irclogbot_0 has joined #openstack-ansible | 21:28 | |
| *** hamzaachi has quit IRC | 21:28 | |
| *** BjoernT has quit IRC | 21:30 | |
| *** cmart has joined #openstack-ansible | 21:49 | |
| *** cmart has quit IRC | 21:54 | |
| *** nurdie has quit IRC | 22:00 | |
| *** DanyC_ has quit IRC | 22:08 | |
| *** cshen has joined #openstack-ansible | 22:15 | |
| *** marst has quit IRC | 22:17 | |
| *** cshen has quit IRC | 22:19 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add OpenStack Zun support to OSA https://review.openstack.org/571400 | 22:25 |
| *** cmart has joined #openstack-ansible | 22:30 | |
| *** fresta has joined #openstack-ansible | 23:05 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Always set pip vars for tempest install https://review.openstack.org/643573 | 23:06 |
| *** ivve has quit IRC | 23:09 | |
| NobodyCam | is there a trick to getting Osa to completely recreate a container? I made a mistake with the network config in openstack_user_config.yml I corrected that destroyed the container removed the it from the inventory, However when I rerun the playbooks the container is not getting a ip address | 23:23 |
| cloudnull | NobodyCam are any of the containers getting an IP address when you stop/restart them ? | 23:34 |
| cloudnull | I'd check to see if lxc-dnsmasq is running / restart the process/ | 23:34 |
| NobodyCam | cloudnull: !!! Thank you ! | 23:34 |
| cloudnull | i've seen that process get wedged on a host | 23:34 |
| cloudnull | resulting in a container not getting an address on eth0 and that blocks just about everything else | 23:35 |
| NobodyCam | I think I had a couple issues, main one being I was one space off in the user_config yaml | 23:41 |
| cloudnull | ah that sucks | 23:41 |
| NobodyCam | :p | 23:42 |
| NobodyCam | took me a couple tries to catch that one | 23:42 |
| cloudnull | there have been aspirations in the past to build a general config validator, its come close but nothing concrete quite yet | 23:42 |
| cloudnull | something that would validate and catch little things like that could be super useful for folks | 23:42 |
| * cloudnull IMHO | 23:42 | |
| nicolasbock | +1 I agree | 23:43 |
| cloudnull | o/ nicolasbock | 23:43 |
| nicolasbock | \o :) | 23:43 |
| NobodyCam | :) we be super helpful if --check-syntact just printed out what was parsed.. Ie, network configs, type, range, ip from, etc... | 23:44 |
| cloudnull | one off errors are pretty common when building the config. | 23:44 |
| cloudnull | yea That would be a good thing to have | 23:44 |
| *** tosky has quit IRC | 23:48 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
