*** acormier has joined #openstack-ansible | 00:17 | |
*** hamzy_ has joined #openstack-ansible | 00:17 | |
*** acormier has quit IRC | 00:21 | |
*** dave-mccowan has joined #openstack-ansible | 00:22 | |
*** threestrands has joined #openstack-ansible | 00:27 | |
*** threestrands has quit IRC | 00:27 | |
*** threestrands has joined #openstack-ansible | 00:27 | |
*** acormier has joined #openstack-ansible | 00:48 | |
*** acormier has quit IRC | 00:49 | |
*** acormier has joined #openstack-ansible | 00:49 | |
*** weezS has joined #openstack-ansible | 01:01 | |
*** weezS has quit IRC | 01:07 | |
*** gfa has quit IRC | 01:10 | |
*** gfa has joined #openstack-ansible | 01:11 | |
*** vnogin has joined #openstack-ansible | 01:15 | |
*** vnogin has quit IRC | 01:20 | |
logan- | mnaser: ceph_client https://github.com/openstack/openstack-ansible-ceph_client | 01:20 |
---|---|---|
*** woodard has quit IRC | 01:21 | |
*** woodard has joined #openstack-ansible | 01:21 | |
*** tux_ has joined #openstack-ansible | 01:27 | |
*** threestrands has quit IRC | 01:29 | |
*** Jeffrey4l has joined #openstack-ansible | 01:34 | |
openstackgerrit | zhongshengping proposed openstack/openstack-ansible-os_nova master: Adjust deprecated nova console variables https://review.openstack.org/532836 | 01:44 |
*** Jeffrey4l_ has joined #openstack-ansible | 01:46 | |
*** Jeffrey4l has quit IRC | 01:47 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-rabbitmq_server stable/pike: Zuul: Remove project name https://review.openstack.org/542533 | 01:47 |
*** rama_y has quit IRC | 01:48 | |
*** TxGirlGeek has quit IRC | 01:50 | |
*** tux_ has quit IRC | 01:51 | |
*** tux_ has joined #openstack-ansible | 01:56 | |
*** tux_ has quit IRC | 01:57 | |
mnaser | logan-: that role doesnt seem to include any repos or add them | 02:01 |
mnaser | just install ceph client side things | 02:01 |
logan- | well for example there are repos added here https://github.com/openstack/openstack-ansible-ceph_client/blob/master/tasks/ceph_preinstall_apt.yml | 02:03 |
logan- | which distro are you working on? | 02:04 |
logan- | rhel/centos? | 02:04 |
logan- | and I guess reading into what you said above a little more, the issue is with the gnocci wheel build right? so something that should be installed in the repo container | 02:04 |
logan- | so maybe you'd want to add something to the repo_build role (where the wheel is built) and also os_gnocchi (where the wheel is installed/used) | 02:06 |
logan- | or alternatively ceph_client instead of os_gnocchi if it is a general ceph client need not specific to gnocchi | 02:07 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_aodh stable/newton: Zuul: Remove project name https://review.openstack.org/542541 | 02:08 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_almanach stable/pike: Zuul: Remove project name https://review.openstack.org/542545 | 02:28 |
*** hamzy_ has quit IRC | 02:53 | |
*** tux_ has joined #openstack-ansible | 02:55 | |
*** hamzy_ has joined #openstack-ansible | 02:58 | |
*** jwitko has quit IRC | 03:03 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_swift stable/ocata: Zuul: Remove project name https://review.openstack.org/542558 | 03:08 |
tux_ | Ceilometer eating lots of CPU on my AIO box | 03:10 |
tux_ | show do i keep shutdown service so i can enable when i need? | 03:10 |
tux_ | so do i keep shutdown service so i can enable when i need? | 03:10 |
tux_ | i just ceilometer container for now | 03:21 |
*** tux_ has quit IRC | 03:28 | |
*** gkadam has joined #openstack-ansible | 03:34 | |
*** acormier has quit IRC | 03:43 | |
*** dave-mccowan has quit IRC | 03:48 | |
*** chhagarw has joined #openstack-ansible | 03:57 | |
*** udesale has joined #openstack-ansible | 04:02 | |
*** bhujay has joined #openstack-ansible | 04:03 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-tests stable/newton: Zuul: Remove project name https://review.openstack.org/542581 | 04:08 |
*** chyka has joined #openstack-ansible | 04:41 | |
*** rama_y has joined #openstack-ansible | 04:44 | |
*** chyka has quit IRC | 04:46 | |
*** poopcat has quit IRC | 05:05 | |
*** gkadam_ has joined #openstack-ansible | 05:10 | |
*** vnogin has joined #openstack-ansible | 05:15 | |
*** vnogin has quit IRC | 05:20 | |
*** rama_y has quit IRC | 05:24 | |
*** Deagon has joined #openstack-ansible | 05:26 | |
*** Deagon has quit IRC | 05:28 | |
*** rama_y has joined #openstack-ansible | 05:32 | |
*** ivveh has quit IRC | 05:33 | |
*** dariko has joined #openstack-ansible | 05:41 | |
*** rama_y has quit IRC | 05:47 | |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 05:48 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_ceilometer stable/pike: Zuul: Remove project name https://review.openstack.org/542601 | 05:49 |
*** dariko has quit IRC | 05:58 | |
*** chyka has joined #openstack-ansible | 06:10 | |
*** chyka has quit IRC | 06:14 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-lxc_container_create stable/pike: Zuul: Remove project name https://review.openstack.org/542621 | 06:29 |
jafeha | good morning | 06:30 |
hwoarang | good morning | 06:30 |
*** chyka_ has joined #openstack-ansible | 06:34 | |
openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Remove release note for yum mirror change https://review.openstack.org/542191 | 06:38 |
openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts stable/pike: Zuul: Remove project name https://review.openstack.org/542460 | 06:39 |
openstackgerrit | Merged openstack/openstack-ansible-galera_client master: Remove release note for yum mirror change https://review.openstack.org/542189 | 06:43 |
openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/pike: Remove release note for yum mirror change https://review.openstack.org/542192 | 06:43 |
openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/ocata: Remove release note for yum mirror change https://review.openstack.org/542190 | 06:43 |
*** acormier has joined #openstack-ansible | 06:44 | |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 06:47 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-lxc_hosts stable/newton: Zuul: Remove project name https://review.openstack.org/542634 | 06:49 |
*** acormier has quit IRC | 06:49 | |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Updated from global requirements https://review.openstack.org/535051 | 07:08 |
openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Updated from global requirements https://review.openstack.org/535069 | 07:10 |
openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/pike: Zuul: Remove project name https://review.openstack.org/542457 | 07:13 |
tescra | am i to add the ips of infra nodes for HA proxy in file/etc/openstack_deploy/conf.d/haproxy.yml for production environment? | 07:14 |
tescra | in pike | 07:14 |
tescra | also create respective roles? | 07:15 |
tescra | for HA | 07:15 |
tescra | or it is installed by defalt? | 07:15 |
jafeha | tescra, you'd usually define internal and external loadbalancer vips in the openstack_user_config.yml | 07:15 |
*** sar has joined #openstack-ansible | 07:16 | |
jafeha | haproxy is installed to all your infra hosts by default, so it's HA by default | 07:17 |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 07:17 |
jafeha | you can choose to use your own loadbalancer though | 07:17 |
tescra | thanks @ jafeha | 07:19 |
*** chyka_ has quit IRC | 07:19 | |
*** taseer1 has joined #openstack-ansible | 07:21 | |
openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/ocata: Zuul: Remove project name https://review.openstack.org/542120 | 07:21 |
*** Taseer has quit IRC | 07:22 | |
*** taseer2 has joined #openstack-ansible | 07:22 | |
openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Adjust deprecated nova console variables https://review.openstack.org/532836 | 07:25 |
openstackgerrit | Merged openstack/openstack-ansible-os_glance stable/ocata: tests: test-glance-upgrades.sh: Remove the exit trap https://review.openstack.org/541471 | 07:25 |
*** taseer2 is now known as Taseer | 07:25 | |
*** eumel8 has joined #openstack-ansible | 07:26 | |
*** taseer1 has quit IRC | 07:26 | |
openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541272 | 07:28 |
evrardjp | tescra: on top of that if you want dedicated nodes it's also possible. | 07:30 |
jafeha | good morning evrardjp | 07:31 |
openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Zuul: Remove project name https://review.openstack.org/540308 | 07:35 |
openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Updated from global requirements https://review.openstack.org/535074 | 07:35 |
evrardjp | good morning jafeha | 07:35 |
evrardjp | FYI that's what your colleagues have seen: https://jira.mariadb.org/browse/MDEV-15254 | 07:35 |
*** pcaruana has joined #openstack-ansible | 07:35 | |
evrardjp | that's why rsync was the fix | 07:36 |
jafeha | i have seen you guys had real fun yesterday with the whole galera fun | 07:36 |
jafeha | yeah, that sounds like the stuff we were suffering. | 07:37 |
jafeha | still recovereing on some fronts as it looks | 07:37 |
*** bauruine has joined #openstack-ansible | 07:38 | |
jafeha | evrardjp, do you already have a plan how to deal with this? like version pinning or sth like that? | 07:38 |
evrardjp | jafeha: yes, we have a series of patches to use 10.1.30 for now, setting to repo to use 10.1.30 instead of a 10.1 repo | 07:39 |
evrardjp | all deployers having 10.1.31 should uninstall it, and re-run the playbooks. | 07:39 |
evrardjp | I haven't bumped the galera_client and galera_server role for the stable branches, and I will do it today | 07:39 |
jafeha | mkay | 07:39 |
evrardjp | in the meantime, I am trying to find a long term fix | 07:40 |
tescra | TASK [lxc_hosts : Ensure that the LXC cache has been prepared] ******************************************************* Thursday 08 February 2018 23:39:13 -0800 (0:00:00.538) | 07:40 |
tescra | fails | 07:40 |
evrardjp | but if you moved to use rsync instead of xtrabackup-v2 for sst, everything should be alright for you | 07:40 |
tescra | for pike | 07:40 |
evrardjp | networking issue tescra ? | 07:41 |
jafeha | i need another look at our galera setup, as we're using rsync to sync right now | 07:41 |
evrardjp | yeah rsync works | 07:41 |
tescra | FAILED - RETRYING: Ensure that the LXC cache has been prepared (111 retries left). FAILED - RETRYING: Ensure that the LXC cache has been prepared (111 retries left). FAILED - RETRYING: Ensure that the LXC cache has been prepared (111 retries left). FAILED - RETRYING: Ensure that the LXC cache has been prepared (110 | 07:41 |
evrardjp | if you want to rollback to xtrabackup-v2 that would fail with 10.1.31 | 07:41 |
jafeha | ic. | 07:41 |
evrardjp | tescra: that's normal | 07:41 |
evrardjp | it's an asynchronous event | 07:42 |
evrardjp | it will retry until everything has succeed. | 07:42 |
evrardjp | at the end if the last line is a fail, you have a hard failure | 07:42 |
evrardjp | but right now your task hasn't failed in itself | 07:42 |
jafeha | i'll have another look how broken things are, i can't point my finger to it, but i think our galera cluster has some more issues than just the replication | 07:43 |
evrardjp | oh i see | 07:43 |
evrardjp | well rsync is slow | 07:43 |
evrardjp | so if you have a network connectivity issue, it will freeze and rsync, which can disrupt the write cycles I guess | 07:43 |
evrardjp | but don't trust my word on that, I am definitely a noob | 07:43 |
jafeha | networking "should" be fine | 07:44 |
jafeha | at least no saturation there | 07:44 |
tescra | evrardjp it changed after sometime that's OK? | 07:45 |
tescra | Thanks for explain | 07:46 |
evrardjp | tescra: yeah basically we start these command in the background, with ansible async . Then we test until these commands have completed or a timeout is exceeded. | 07:48 |
jafeha | evrardjp, are there any outstanding cinder issues? getting a HTTP 500 on instance creation | 07:49 |
evrardjp | not that I am aware of | 07:49 |
evrardjp | check your lb ? | 07:49 |
evrardjp | jafeha: you're on ubuntu, right? | 07:50 |
jafeha | yep | 07:51 |
evrardjp | k | 07:51 |
jafeha | loadbalancer seems fine | 07:51 |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541269 | 07:51 |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Zuul: Remove project name https://review.openstack.org/538627 | 07:51 |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Updated from global requirements https://review.openstack.org/535061 | 07:51 |
evrardjp | well it should report the 500 | 07:51 |
evrardjp | shouldn't it ? :p | 07:52 |
*** mbuil has joined #openstack-ansible | 07:52 | |
evrardjp | hello mbuil | 07:52 |
Taseer | evrardjp: can this now be merged ? => https://review.openstack.org/#/c/522491/ | 07:52 |
evrardjp | Taseer: woot! I will have a look | 07:52 |
evrardjp | I see green I will check if it's doing proper tests and all | 07:53 |
jafeha | evrardjp, on a closer look there seems to be something broken | 07:53 |
evrardjp | yup | 07:54 |
Taseer | ok | 07:54 |
*** hamza21 has joined #openstack-ansible | 07:59 | |
openstackgerrit | Merged openstack/openstack-ansible-ops stable/pike: Zuul: Remove project name https://review.openstack.org/541913 | 08:01 |
Taseer | evrardjp: what did you mean by "change functional tests in a non-voting fashion" ? | 08:03 |
jafeha | evrardjp, seems ok now. had to restart one cinder-api container. | 08:04 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 08:04 |
evrardjp | Taseer: done | 08:04 |
evrardjp | Taseer: we are setting false expectations by showing functional tests that are green, while there is nothing that functionally tests it. | 08:05 |
evrardjp | so I am moving these tests to basic functional tests, and we'll able to merge even if it doesn't pass test. | 08:05 |
Taseer | okay | 08:05 |
evrardjp | with the hope this will get fixed as soon as possible in another commit | 08:05 |
evrardjp | this commit has lasted too long already | 08:06 |
mbuil | hello evrardjp | 08:06 |
evrardjp | how are things? | 08:07 |
openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/newton: Zuul: Remove project name https://review.openstack.org/541919 | 08:08 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-memcached_server stable/newton: Zuul: Remove project name https://review.openstack.org/542707 | 08:09 |
*** hw_wutianwei has quit IRC | 08:11 | |
*** hw_wutianwei has joined #openstack-ansible | 08:11 | |
evrardjp | hwoarang: sorry I didn't babysit the https://review.openstack.org/#/c/538933/ patch. will keep under my nose today | 08:11 |
hwoarang | evrardjp: no worries i am also keeping an eye on it | 08:12 |
evrardjp | this mariadb has taken some time, and I had quite a series of meetings yesterday :/ | 08:12 |
openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/ocata: Zuul: Remove project name https://review.openstack.org/541869 | 08:13 |
openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client stable/ocata: Correct task tag https://review.openstack.org/541025 | 08:17 |
*** peri has joined #openstack-ansible | 08:18 | |
tescra | hi for some reason my putty connection got disconnected? | 08:18 |
tescra | and server is at reomte location | 08:19 |
*** sar has quit IRC | 08:20 | |
tescra | can i restart the session and rerun the playbook openstack-ansible setup-hosts.yml or it will cause discrepancy | 08:24 |
tescra | TASK [lxc_container_create : Container network includes] was in progress | 08:27 |
tescra | can i start executing play book from here ? | 08:28 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_ironic stable/newton: Zuul: Remove project name https://review.openstack.org/542713 | 08:29 |
openstackgerrit | Merged openstack/openstack-ansible-memcached_server stable/newton: Zuul: Remove project name https://review.openstack.org/542707 | 08:30 |
tescra | is there any tag? | 08:34 |
*** sar has joined #openstack-ansible | 08:34 | |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Remove pip_install dependency https://review.openstack.org/525546 | 08:36 |
*** olivierbourdon38 has joined #openstack-ansible | 08:42 | |
tescra | can anyone suggest what to do? | 08:43 |
jafeha | tescra, i'd suggest to use tmux, screen or some other terminal multiplexer to avoid those disconnects | 08:46 |
jafeha | or issues with them | 08:46 |
tescra | but what to do now replay the play book? | 08:47 |
jafeha | i'd probably do a complete re-run of the setup-hosts | 08:47 |
tescra | ok but it had executed few plays so the changes made will be overwritten or cause error? | 08:49 |
tescra | thanks jafeha | 08:50 |
*** gameon has joined #openstack-ansible | 08:50 | |
jafeha | you should see a lot of green then, ansible won't execute tasks that do no change | 08:50 |
gameon | Hi all - I have a question which I'm struggling to find any info on. I have a Kilo deployment, and have a provider network which is set as external. This network is NOT visible from within tenant projects. However, using a new Pike deplyoment, the provider network IS visible (and deleteable even!!) from tenant projects. Has anyone got a solution to this? RBAC didn't seem to do anything, as soon as I set the network to extern | 08:52 |
gameon | (I realise this isn't strictly OSA related but I'm curious if anyone else has seen this behavior. This is a 16.0.6 deployment) | 08:53 |
hwoarang | evrardjp: could you quickly get https://review.openstack.org/#/c/542458/ in ? | 08:53 |
hwoarang | i broke stuff ;p | 08:53 |
tescra | thanks again Jafeha | 08:53 |
*** armaan has joined #openstack-ansible | 08:54 | |
evrardjp | hwoarang: oh yeah I remember we talked about it yesterday. | 08:54 |
evrardjp | Sorry I have missed it, I am talking with maria ppl to find out what's wrong | 08:54 |
evrardjp | I think you can help I've narrowed it down to 2 shell scripts. | 08:55 |
evrardjp | :D | 08:55 |
hwoarang | heh | 08:56 |
openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: Zuul: Remove project name https://review.openstack.org/541848 | 08:59 |
openstackgerrit | Merged openstack/openstack-ansible-galera_client master: Revert "RedHat: Ensure that conflicting MariaDB packages are removed" https://review.openstack.org/542458 | 09:04 |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 09:05 |
openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: If debug is enabled, provide verbose output for host discovery https://review.openstack.org/542287 | 09:07 |
openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_monasca-agent stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/542735 | 09:07 |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Replace virtualenv-tools by a script https://review.openstack.org/533657 | 09:10 |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_trove master: Updated from OpenStack Ansible Tests https://review.openstack.org/536441 | 09:15 |
*** pbandark has joined #openstack-ansible | 09:17 | |
*** Jeffrey4l_ has quit IRC | 09:17 | |
*** Jeffrey4l has joined #openstack-ansible | 09:18 | |
Miouge | evrardjp: any tips on how I can add tests to check that the ceph-rgw is working? Also swift and ceph-rgw should probably be mutually exclusive since they use the same HaProxy port? | 09:18 |
evrardjp | Miouge: agreed. | 09:18 |
evrardjp | does the tempest swift testing suite apply? | 09:19 |
evrardjp | else maybe we should change anything, and it's already functionally tested then, we could just change the conditional to say if ceph and swift, only deploy one of those | 09:19 |
evrardjp | which is a scenario thing | 09:19 |
evrardjp | and a group vars thing | 09:20 |
evrardjp | I don't know how you test it rn | 09:20 |
Miouge | Since swift and rgw are mutually exclusive, I wonder if openstack-ansible-deploy-ceph-ubuntu-xenial install swift as well. That test currently passes and install rgw, but if swift is also there it should fail. | 09:23 |
*** acormier has joined #openstack-ansible | 09:24 | |
tescra | hi on replay of the playbook it seems to be stuck at same task where the session disconnected...any idea? | 09:24 |
*** acormier has quit IRC | 09:24 | |
*** acormier has joined #openstack-ansible | 09:25 | |
*** sar has quit IRC | 09:25 | |
tescra | @ jafeha evrardjp | 09:26 |
*** acormier has quit IRC | 09:26 | |
*** acormier has joined #openstack-ansible | 09:27 | |
evrardjp | Miouge: that's a scenario, so we can adapt it. I think we should adapt the scenario to not deploy swift, but use ceph + rgw | 09:31 |
evrardjp | I can change that on your patch if you want | 09:31 |
*** acormier has quit IRC | 09:31 | |
evrardjp | tescra: that's too vague, not sure I can help. Re-Running playbooks should not be a problem | 09:31 |
tescra | ok thanks | 09:32 |
evrardjp | tescra: maybe the output could be helpful? | 09:33 |
tescra | no output the session terminated again at same play | 09:34 |
evrardjp | lack of ramn? | 09:34 |
tescra | in deploy host? | 09:35 |
tescra | or infra ? | 09:35 |
Miouge | evrardjp: if you know how/where to change that, go ahead. I greped the repo, that didnāt help me | 09:36 |
evrardjp | tescra: well that's too vague, but I'd say deploy right now. | 09:36 |
evrardjp | show me a log and I can probably help | 09:36 |
evrardjp | Miouge: no worries, will handle it. | 09:37 |
tescra | ok 1 minute | 09:37 |
evrardjp | tescra: thanks :) | 09:37 |
tescra | syslog:Feb 7 21:13:25 deploy-node systemd[1]: Started Set console keymap. Feb 7 21:13:25 deploy-node systemd-modules-load[430]: Inserted module 'ib_iser' Feb 7 21:13:25 deploy-node systemd[1]: Started Load Kernel Modules. Feb 7 21:13:25 deploy-node systemd[1]: Starting Apply Kernel Variables... Feb 7 21:13:25 deploy-node systemd[1]: Mounting FUSE Control File System... Feb 7 21:13:25 deploy-node systemd[1]: Mounted F | 09:39 |
Taseer | evrardjp: the api verification failed | 09:40 |
odyssey4me | evrardjp I think it might be time to just merge https://review.openstack.org/#/c/522491/, then inspect afterwards and register bugs for things that need to be done? | 09:40 |
Taseer | but the patch can be merged, I guess :) | 09:40 |
tescra | authlog:Feb 9 01:40:28 deploy-node sshd[24735]: pam_unix(sshd:session): session closed for user osadmin Feb 9 01:40:28 deploy-node su[24829]: pam_unix(su:session): session closed for user root Feb 9 01:40:28 deploy-node sudo: pam_unix(sudo:session): session closed for use r root | 09:41 |
evrardjp | Taseer: yes, but you'll have to fix it later. | 09:42 |
openstackgerrit | Merged openstack/openstack-ansible-galera_client stable/newton: Zuul: Remove project name https://review.openstack.org/542471 | 09:42 |
openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client stable/pike: Fix pep8 doc config lint failures https://review.openstack.org/541847 | 09:42 |
odyssey4me | otherwise I can review whats there, and we go through many more review rounds? | 09:42 |
evrardjp | odyssey4me: that's exactly my point: let's stop here, and iterate | 09:43 |
evrardjp | tescra: could you run everything as root? It would make your life simpler. Use ssh keys. | 09:44 |
evrardjp | else you'd have to pass things in the cli, to force the user, the method, etc. | 09:44 |
evrardjp | tescra: but I don't see the full flog there | 09:45 |
odyssey4me | alright, 86 iterations already - how about we go for 100 ;) | 09:45 |
evrardjp | odyssey4me: haha | 09:45 |
tescra | update-alternatives 2018-02-07 22:20:02: run with --install /bin/nc nc /bin/nc.traditional 10 --slave /bin/netcat netcat /bin/nc.traditional --slave /usr/share/man/man1/nc.1.gz nc.1.gz /usr/share/man/man1/nc.traditional.1.gz --slave /usr/share/man/man1/netcat.1.gz netcat.1.gz /usr/share/man/man1/nc.traditional.1.gz | 09:47 |
openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/newton: Zuul: Remove project name https://review.openstack.org/542634 | 09:47 |
openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client stable/pike: Correct task tag https://review.openstack.org/541024 | 09:47 |
tescra | am root users | 09:47 |
tescra | while executing playbooks | 09:47 |
evrardjp | tescra: I'd need a playbook run log | 09:48 |
evrardjp | not a system log | 09:48 |
*** shardy has joined #openstack-ansible | 09:49 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_nova stable/newton: Zuul: Remove project name https://review.openstack.org/542747 | 09:50 |
Taseer | evrardjp: I will fix it. I just need some help. | 09:52 |
Taseer | from a person in congress | 09:53 |
openstackgerrit | Merged openstack/openstack-ansible-os_almanach stable/pike: Zuul: Remove project name https://review.openstack.org/542545 | 09:53 |
tescra | evrardjp path for playbook run log? can't find in /var/log of depoly node | 09:55 |
openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/pike: Use the good configuration directive for memcache servers https://review.openstack.org/540931 | 09:57 |
evrardjp | Taseer: yeah, but we can do it in a later commit, let's just start by inclusion in osa | 09:59 |
evrardjp | tescra: just re-run it and capture the output? | 10:00 |
evrardjp | else it's in your ansible log folder location maybe /var/log/somethingansible? | 10:00 |
openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create stable/pike: Zuul: Remove project name https://review.openstack.org/542621 | 10:01 |
openstackgerrit | Merged openstack/openstack-ansible-os_cinder stable/pike: Zuul: Remove project name https://review.openstack.org/538544 | 10:05 |
*** nsingh has joined #openstack-ansible | 10:06 | |
*** niraj_singh has quit IRC | 10:08 | |
Taseer | evrardjp: so I guess, it just needs your approval with the long awaited +2 :) | 10:08 |
openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/newton: Zuul: Remove project name https://review.openstack.org/542541 | 10:09 |
openstackgerrit | Merged openstack/openstack-ansible-tests stable/newton: Zuul: Remove project name https://review.openstack.org/542581 | 10:09 |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-galera_client master: Set proper yum repo priority on CentOS 7 https://review.openstack.org/542762 | 10:15 |
tescra | no file for ansible /var/log$ ls -a . auth.log dmesg installer ntpstats .. bootstrap.log dpkg.log kern.log syslog alternatives.log btmp faillog lastlog unattended-upgrades apt dist-upgrade fsck lxd wtmp | 10:16 |
tescra | also my session terminates before i can see play recap | 10:17 |
openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server stable/pike: Zuul: Remove project name https://review.openstack.org/542533 | 10:17 |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_nova master: tests: test-nova-upgrades.sh: Drop Ansible verbosity https://review.openstack.org/537379 | 10:20 |
Miouge | tescra: are you runnining that in a screen? | 10:20 |
*** Smeared_Beard has quit IRC | 10:22 | |
openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/ocata: Zuul: Remove project name https://review.openstack.org/542558 | 10:22 |
*** vnogin has joined #openstack-ansible | 10:25 | |
odyssey4me | Taseer ok, it looks like a good start there in the os_congress role - a few things to clean up and update... the non-voting functional test is not a huge concern from my standpoint right now - but I'd like to try and get it at least to a good quality place and up to a minimum standard | 10:25 |
odyssey4me | evrardjp Taseer given that the os_congress role already has some stuff in it which is broken, we need to try and get something better in soon to replace the current broken, IMO... so Taseer if you could apply my suggestions then we should be good to go and can work on making it functional afterwards in subsequent patches | 10:26 |
tescra | shall i execute setupeverything instead of setup-hosts.yml | 10:33 |
tescra | same thing repeated 3thrice.. | 10:34 |
tescra | no logs for ansible no play recap | 10:35 |
openstackgerrit | Merged openstack/openstack-ansible-os_monasca-agent stable/pike: Updated from OpenStack Ansible Tests https://review.openstack.org/542735 | 10:40 |
*** SmearedBeard has joined #openstack-ansible | 10:41 | |
Taseer | odyssey4me: ok, I will address your comments | 10:46 |
*** gkadam_ has quit IRC | 10:46 | |
*** armaan has quit IRC | 10:46 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_keystone stable/pike: Zuul: Remove project name https://review.openstack.org/542777 | 10:50 |
tescra | ok: [infra3_cinder_api_container-a9cf1875 -> 10.40.100.23] => (item={'value': {'netmask': '255.255.252.0', 'interface': 'eth2', 'bridge': 'br-storage', 'address': '10.40.244.144', 'type': 'veth'}, 'key': 'storage_address'}) | 10:51 |
tescra | no progress hereafter | 10:51 |
tescra | TASK [lxc_container_create : Run container veth wiring script] | 10:52 |
evrardjp | tescra: maybe you have networking issue | 10:53 |
*** vnogin has quit IRC | 10:59 | |
*** vnogin has joined #openstack-ansible | 11:01 | |
*** hw_wutianwei has quit IRC | 11:02 | |
odyssey4me | thanks Taseer - apologies for not getting to review that sooner | 11:02 |
odyssey4me | evrardjp your thoughts please? https://review.openstack.org/#/c/542288/ passed again on a recheck, but it's based on https://review.openstack.org/#/c/541463/ which isn't passing... shall we just consolidate and see how that goes? | 11:06 |
odyssey4me | hwoarang did you see my comment in https://review.openstack.org/537381 and the others like it? | 11:08 |
evrardjp | odyssey4me: afk for a few minutes, will review. | 11:10 |
openstackgerrit | Manuel Buil proposed openstack/openstack-ansible master: l3_agent should only appear in computes when dvr https://review.openstack.org/541310 | 11:11 |
openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Revert "Switch the LXD compute driver test to non-voting" https://review.openstack.org/541372 | 11:16 |
openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Updated from OpenStack Ansible Tests https://review.openstack.org/536441 | 11:16 |
openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/newton: Zuul: Remove project name https://review.openstack.org/542713 | 11:19 |
*** olivierbourdon38 has quit IRC | 11:22 | |
hwoarang | odyssey4me: not yet got busy with soem opnfv bits | 11:23 |
odyssey4me | hwoarang no rush - just letting you know 'cos my vote's hidden behind a +2 :) | 11:24 |
*** vnogin has quit IRC | 11:25 | |
tescra | ok: [infra3_cinder_api_container-a9cf1875 -> 10.40.100.23] => (item={'value': {'type': 'veth', 'address': '10.40.244.91', 'netmask': '255.255.252.0', 'interface': 'eth2', 'bridge': 'br-storage'}, 'key': 'storage_address'}) | 11:27 |
evrardjp | odyssey4me: imo they don't depend on each other. | 11:27 |
evrardjp | they do for your use case | 11:28 |
odyssey4me | yes, they don't - but that didn't pass until I put them on top of each other | 11:28 |
evrardjp | tescra: please paste the whole log into a pastebin. | 11:28 |
odyssey4me | I'll rebase to verify | 11:28 |
evrardjp | I think it's not linked, but like you said, a more insidious cause: a race condition | 11:29 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova stable/ocata: If debug is enabled, provide verbose output for host discovery https://review.openstack.org/542288 | 11:29 |
*** udesale has quit IRC | 11:35 | |
*** armaan has joined #openstack-ansible | 11:39 | |
armaan | evrardjp: Hello! | 11:39 |
*** stuartgr has joined #openstack-ansible | 11:43 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Document known issue with galera 10.1.31 https://review.openstack.org/542790 | 11:47 |
*** vnogin has joined #openstack-ansible | 11:47 | |
evrardjp | armaan: hey | 11:49 |
armaan | evrardjp: how are you? | 11:50 |
evrardjp | good and you ? Fixing the mariadb issues accross all branches, thanks to their terrible packaging! | 11:50 |
evrardjp | long story short, I'd wait a little before upgrading :( | 11:50 |
evrardjp | one after another... | 11:51 |
armaan | evrardjp: struggling with ceph performance issues, damn latency :/ | 11:51 |
evrardjp | oh yeah that's definitely linked. | 11:51 |
evrardjp | odyssey4me: what do you think of https://review.openstack.org/#/c/542790/ ? | 11:51 |
evrardjp | also, me terribad with rst. | 11:52 |
tacco | hi there | 11:52 |
evrardjp | hello tacco | 11:52 |
armaan | evrardjp: ok then, i will wait :), thanks for letting me know... | 11:52 |
evrardjp | when I pinged you, you got a bell? | 11:52 |
evrardjp | that's a terrible joke. | 11:53 |
armaan | :P | 11:53 |
evrardjp | armaan: yeah, pay attention to your upgrades too | 11:53 |
evrardjp | if you're doing package updates on your system --- meh! you lose! | 11:53 |
tacco | yeah this mariadb xtrabackup stuff is always a bit unstable thats why i would prefer to use rsync repl. method even if this is a bit less performant | 11:53 |
odyssey4me | evrardjp mind if I tweak it? | 11:53 |
evrardjp | go ahead. | 11:53 |
evrardjp | tacco: yeah, and you get read locks | 11:54 |
armaan | evrardjp: ohh, i will take extra care! | 11:54 |
tacco | yes | 11:54 |
*** sar has joined #openstack-ansible | 11:54 | |
evrardjp | which is no go for many ppl. And on top of it, it's not recommended by mariadb ppl :) | 11:54 |
evrardjp | but yes, at least rsync works :) | 11:54 |
tacco | yes but if you are in big trouble, you better have some read locks instead of a dead cluster | 11:55 |
tacco | :D | 11:55 |
evrardjp | fair | 11:55 |
evrardjp | I just think they shouldn't have released a broken thing. | 11:55 |
evrardjp | :p | 11:55 |
tacco | true | 11:55 |
evrardjp | odyssey4me: building the docs on my local machine to see what it looks like | 11:56 |
evrardjp | hope they're gonna be pretty. | 11:56 |
evrardjp | Maybe I should get a crash course to writing rst instead of md | 11:56 |
evrardjp | md is so convenient | 11:56 |
evrardjp | tacco: oh btw | 11:58 |
evrardjp | WELCOME! | 11:59 |
evrardjp | (you have to hear the brasilian party music in the background) | 11:59 |
tacco | thanks a lot | 11:59 |
evrardjp | odyssey4me: FYI tacco is a colleague of jafeha :) | 11:59 |
evrardjp | what a good team! | 12:00 |
tacco | hope that we can contribute some good stuff | 12:00 |
evrardjp | that's already started, jafeha helped people quite a lot in this chan, plus we have the scenario for ovs on the way | 12:01 |
tacco | great | 12:03 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Document known issue with galera 10.1.31 https://review.openstack.org/542790 | 12:04 |
odyssey4me | evrardjp fixed up | 12:04 |
*** sar has quit IRC | 12:04 | |
odyssey4me | evrardjp in future, https://docs.openstack.org/reno/latest/user/usage.html#note-file-syntax is helpful :) | 12:05 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible stable/pike: Document known issue with galera 10.1.31 https://review.openstack.org/542790 | 12:05 |
odyssey4me | killed the whitespace | 12:05 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Integrate OVS AIO Scenario https://review.openstack.org/534318 | 12:07 |
evrardjp | odyssey4me: bookmarked! | 12:08 |
hwoarang | odyssey4me: ok got your comments now i am going to update the open requests | 12:08 |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_swift master: tests: test-swift-upgrades.sh: Drop Ansible verbosity https://review.openstack.org/537381 | 12:09 |
evrardjp | but ara doesn't show the file loading order easily! | 12:10 |
*** sar has joined #openstack-ansible | 12:11 | |
evrardjp | :) | 12:11 |
evrardjp | odyssey4me: thanks for the work done in the last days on metal. | 12:11 |
openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: tasks: Ensure Open vSwitch is started for all providers that need it https://review.openstack.org/538933 | 12:17 |
*** rodolof has joined #openstack-ansible | 12:19 | |
*** acormier has joined #openstack-ansible | 12:26 | |
*** acormier has quit IRC | 12:27 | |
*** acormier has joined #openstack-ansible | 12:27 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 12:29 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 12:29 |
*** acormier has quit IRC | 12:31 | |
odyssey4me | evrardjp perhaps that's best given as feedback to dmsimard ;) | 12:34 |
odyssey4me | (the file load ordering) | 12:34 |
odyssey4me | evrardjp and sure - happy to have helped, and very glad we have that example demonstrated now... | 12:35 |
evrardjp | yup, will do it :) | 12:35 |
evrardjp | I think we definitely improved coverage this cycle, which allow us great refactors in the future. | 12:36 |
*** hamza21 has quit IRC | 12:41 | |
*** serverascode has quit IRC | 12:48 | |
*** serverascode has joined #openstack-ansible | 12:49 | |
*** rodolof has quit IRC | 12:54 | |
*** armaan has quit IRC | 12:56 | |
evrardjp | odyssey4me: has super fancy rst skillz | 12:57 |
openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/newton: Zuul: Remove project name https://review.openstack.org/542747 | 12:59 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/ocata: Document known issue with galera 10.1.31 https://review.openstack.org/542800 | 13:01 |
evrardjp | odyssey4me: ^ ocata one | 13:01 |
evrardjp | doing the newtonz now | 13:01 |
evrardjp | oh there is no newtonz. | 13:02 |
evrardjp | good! | 13:02 |
*** acormier has joined #openstack-ansible | 13:05 | |
openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/pike: Zuul: Remove project name https://review.openstack.org/542777 | 13:06 |
*** acormier has quit IRC | 13:06 | |
*** acormier has joined #openstack-ansible | 13:06 | |
openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible master: Make Opendaylight as the BGP speaker using Quagga https://review.openstack.org/523907 | 13:07 |
openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 13:09 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_trove stable/pike: Zuul: Remove project name https://review.openstack.org/542811 | 13:11 |
*** fdegir has quit IRC | 13:14 | |
*** fdegir has joined #openstack-ansible | 13:15 | |
*** kklimonda has quit IRC | 13:15 | |
*** kklimonda has joined #openstack-ansible | 13:16 | |
*** woodard has quit IRC | 13:18 | |
*** woodard has joined #openstack-ansible | 13:18 | |
mgariepy | morning everyone | 13:18 |
mhayden | morning | 13:18 |
evrardjp | morning mgariepy mhayden | 13:19 |
evrardjp | mnaser: for gnocchi that needs ceph, it's probable that ceph bindings aren't installed into the venv | 13:19 |
mnaser | evrardjp: so how should I add it? Any other role to follow? | 13:19 |
evrardjp | for example for cinder we do this to install basic ceph stuff on the container: https://github.com/openstack/openstack-ansible-os_cinder/blob/f807cdccfbc148fc335a2fd89d13ec8ea85aa6ba/tasks/main.yml#L79-L89 | 13:20 |
evrardjp | then I am not sure where we have an example for linking the ceph c bindings | 13:20 |
evrardjp | mhayden: you played with that recently, right? | 13:21 |
evrardjp | mnaser: then, if you need cradox, just add cradox under the list of packages to install with pip, if that's a pip package | 13:21 |
evrardjp | mnaser: oh the linking is done in the role already: https://github.com/openstack/openstack-ansible-ceph_client/blob/master/tasks/ceph_install_python_libs.yml#L26-L33 | 13:22 |
mnaser | evrardjp: right but in developer mode install on the gate wouldnāt that not matter? | 13:22 |
evrardjp | it wouldn't matter for being built, but you still have to stay to install the package if it's not in pip package's setup.py/requirements. | 13:23 |
evrardjp | I am not aware of how gnocchi does it | 13:23 |
evrardjp | let me quickly check | 13:23 |
evrardjp | mhayden: do you know why we aren't using RDO packages for mariadb? | 13:24 |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: fix misspelling of 'configuration' and 'list' https://review.openstack.org/534195 | 13:25 |
mhayden | evrardjp: i was using the mariadb repos for centos 7 so that it would match the other distros | 13:26 |
mhayden | RDO's mariadb is a little older | 13:26 |
mhayden | i was worried about configuration conflicts and such | 13:26 |
mhayden | did we ever get to the bottom of what's up with 10.1.31? or is it just wait and see | 13:27 |
evrardjp | I stopped this morning. I was into deep, but I had to stop for other things. | 13:28 |
evrardjp | I can hand that over to you, or we can do a debugging session together. | 13:28 |
openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts master: Ensure openstack_hosts sets nodepool variables https://review.openstack.org/542352 | 13:29 |
evrardjp | mhayden: I did like you did, asking #maria. didn't really help | 13:29 |
odyssey4me | evrardjp looks like it might be a race - the same patch, two different attempts: http://logs.openstack.org/88/542288/1/check/openstack-ansible-upgrade-ubuntu-xenial/7a05b52/job-output.txt.gz#_2018-02-08_17_15_22_049899 and http://logs.openstack.org/88/542288/2/check/openstack-ansible-upgrade-ubuntu-xenial/6ddf7fe/job-output.txt.gz#_2018-02-08_19_57_37_170897 | 13:29 |
evrardjp | I don't like this. | 13:30 |
evrardjp | introducing a sleep 20 is the solution! | 13:30 |
evrardjp | oh but it's in tempest | 13:31 |
odyssey4me | evrardjp we already have a form of a sleep | 13:31 |
odyssey4me | we wait until the compute node registers | 13:31 |
odyssey4me | so clearly, for an upgrade, this is not enough | 13:31 |
odyssey4me | I'll keep looking into it. | 13:31 |
odyssey4me | and no, this is not in tempest - tempest fails because there are no compute nodes in cell1 so it can't schedule the instance build and fails | 13:32 |
evrardjp | oh I might have an idea | 13:32 |
evrardjp | is it from O to P? | 13:33 |
evrardjp | or from N to O ? | 13:33 |
odyssey4me | N to O | 13:33 |
odyssey4me | this is the 'sleep': https://github.com/openstack/openstack-ansible-os_nova/blob/stable/ocata/handlers/main.yml#L67-L76 | 13:34 |
evrardjp | odyssey4me: could you display, in your tests, the result of a task shell: "{{ _db_nova_bin }}/nova-manage cell_v2 list_cells ? | 13:37 |
*** dariko has joined #openstack-ansible | 13:37 | |
odyssey4me | could do - but the output there shows the two cells already, cell0 and cell1 | 13:39 |
odyssey4me | see http://logs.openstack.org/88/542288/2/check/openstack-ansible-upgrade-ubuntu-xenial/6ddf7fe/job-output.txt.gz#_2018-02-08_19_57_37_170711 and http://logs.openstack.org/88/542288/1/check/openstack-ansible-upgrade-ubuntu-xenial/7a05b52/job-output.txt.gz#_2018-02-08_17_15_22_049405 | 13:40 |
*** kstev has joined #openstack-ansible | 13:40 | |
*** mubix has quit IRC | 13:41 | |
evrardjp | odyssey4me: yeah I wasn't on the right line, it took a while to load | 13:42 |
*** mubix has joined #openstack-ansible | 13:42 | |
*** wspthr has quit IRC | 13:42 | |
*** kstev has quit IRC | 13:44 | |
*** wspthr has joined #openstack-ansible | 13:45 | |
*** kstev has joined #openstack-ansible | 13:46 | |
*** dave-mccowan has joined #openstack-ansible | 13:46 | |
*** dave-mcc_ has joined #openstack-ansible | 13:54 | |
*** dave-mccowan has quit IRC | 13:55 | |
*** hamza21 has joined #openstack-ansible | 13:58 | |
*** acormier has quit IRC | 14:01 | |
*** acormier has joined #openstack-ansible | 14:02 | |
*** johnsom has quit IRC | 14:03 | |
*** johnsom has joined #openstack-ansible | 14:04 | |
*** acormier_ has joined #openstack-ansible | 14:05 | |
*** acormier has quit IRC | 14:06 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-memcached_server stable/pike: Zuul: Remove project name https://review.openstack.org/542842 | 14:11 |
openstackgerrit | Merged openstack/openstack-ansible-os_swift master: tests: test-swift-upgrades.sh: Drop Ansible verbosity https://review.openstack.org/537381 | 14:12 |
*** vnogin has quit IRC | 14:12 | |
*** wspthr has quit IRC | 14:17 | |
*** dave-mcc_ has quit IRC | 14:17 | |
*** vnogin has joined #openstack-ansible | 14:18 | |
openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 14:19 |
* mnaser is in the office again | 14:19 | |
*** adcpdk has joined #openstack-ansible | 14:19 | |
mnaser | okay so, os_gnocchi installs librados2-devel | 14:20 |
*** wspthr has joined #openstack-ansible | 14:20 | |
mnaser | and centos ships hammer release if we don't install storage sig | 14:20 |
*** acormier_ has quit IRC | 14:21 | |
*** acormier has joined #openstack-ansible | 14:22 | |
adcpdk | idlemind thank you so much for such a nice troubleshooting yesterday. I finally got my networking setup and working as a flat network! Such a nice feeling when there are people out there to help and go out with you and fight with monsters you were scared of))) It was my first experience of using IRC channels in my life and I won't forget that. I will try to do my best to be around to help the newbies like me ;-) | 14:23 |
mnaser | so my small question is, what role can i add to dependencies to make sure it adds the ceph modern repos | 14:25 |
openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Updated from OpenStack Ansible Tests https://review.openstack.org/518909 | 14:25 |
*** dave-mccowan has joined #openstack-ansible | 14:29 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_nova stable/ocata: Zuul: Remove project name https://review.openstack.org/542851 | 14:31 |
tescra | hi my userconfig file is ok asper yamllint but openstack-ansible --syntax-check shows no plays | 14:39 |
*** dave-mccowan has quit IRC | 14:40 | |
openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Fix logic to check for insecure Keystone https://review.openstack.org/539305 | 14:43 |
*** berendt has quit IRC | 14:46 | |
*** berendt has joined #openstack-ansible | 14:46 | |
mnaser | https://github.com/openstack/openstack-ansible/blob/master/playbooks/ceph-install.yml#L35-L52 | 14:51 |
mnaser | does this happen in another role? | 14:51 |
*** hogepodge has quit IRC | 14:51 | |
mnaser | or should i add that to the ceph gnocchi role? | 14:51 |
*** hogepodge has joined #openstack-ansible | 14:52 | |
*** acormier has quit IRC | 14:57 | |
*** adrianreza has quit IRC | 14:57 | |
*** adrianreza has joined #openstack-ansible | 14:57 | |
*** sar has quit IRC | 14:58 | |
*** dave-mccowan has joined #openstack-ansible | 15:03 | |
*** vadim_ has joined #openstack-ansible | 15:04 | |
*** rodolof has joined #openstack-ansible | 15:04 | |
*** acormier has joined #openstack-ansible | 15:04 | |
*** acormier has quit IRC | 15:06 | |
*** acormier has joined #openstack-ansible | 15:06 | |
*** bhujay has quit IRC | 15:07 | |
logan- | mnaser: https://github.com/openstack/openstack-ansible-ceph_client/blob/master/tasks/ceph_preinstall_yum.yml probably needs to be updated | 15:10 |
ashak | I tried running a bunch of the OSA playbooks earlier after I made a change to some variable that I hoped would allow me to be able to upload image files via horizon directly to glance... but during that run, something tried to restart mariadb on one of my two galera containers... Since then i've failed to be able to recover that galera container. I followed everything here | 15:10 |
ashak | https://docs.openstack.org/openstack-ansible/newton/developer-docs/ops-galera-recovery.html (because I couldn't find a similar one for ocata or pike). But whatever happens mariadb is never able to start on the container it failed on. | 15:10 |
ashak | Even when that entire container is rebuilt | 15:10 |
*** rama_y has joined #openstack-ansible | 15:10 | |
vadim_ | hi, I think https://github.com/openstack/openstack-ansible-galera_server/commit/f2bfbd38513ac8d61ba4e02a4d5ef6cbbca259cc commit is missing from ocata branch | 15:11 |
cloudnull | ashak: there was recently an issue with the mariadb (galera) package which broke galera clustering | 15:12 |
cloudnull | recent prs went in to fix that | 15:12 |
cloudnull | but you'll have to pull in the pr using the branch instead of using a tag for now | 15:13 |
cloudnull | that's the pr to fix it | 15:13 |
cloudnull | https://github.com/openstack/openstack-ansible-galera_server/commit/8a42facf6b98f77ef3cd04ca619e3c5173ab39e9 | 15:13 |
cloudnull | for the server | 15:13 |
cloudnull | https://github.com/openstack/openstack-ansible-galera_client/commit/3524b8352f0728b2955a002540a4fc4646159132 | 15:13 |
cloudnull | for the client | 15:13 |
ashak | cloudnull: arse. Ok, thanks | 15:14 |
*** olivierbourdon38 has joined #openstack-ansible | 15:16 | |
mnaser | logan-: i guess i can include_role: from_tasks: there | 15:16 |
logan- | right | 15:17 |
logan- | thats how glance/cinder/nova work right now | 15:17 |
logan- | https://github.com/openstack/openstack-ansible-os_nova/blob/9e44a4dd6dd8cdef984d61afc8b68cffc45472ea/tasks/main.yml#L112-L123 | 15:17 |
logan- | and https://github.com/openstack/openstack-ansible-os_cinder/blob/master/tasks/main.yml#L79-L89 | 15:17 |
logan- | for example | 15:17 |
logan- | it looks like gnocchi does the ceph_client include at the playbook level https://github.com/openstack/openstack-ansible/blob/8224fbcede862078d2a3914863f06339ca6c85cb/playbooks/os-gnocchi-install.yml#L50-L57 | 15:18 |
mnaser | logan-: i guess that is the problem, it needs to happen in the role or otherwise the functional testing is failing | 15:19 |
mnaser | ill get it moved into the role and out of openstack-ansible | 15:20 |
*** chhagarw has quit IRC | 15:20 | |
logan- | also interesting that no venv path is being sent into ceph_client for linking | 15:20 |
*** chhagarw has joined #openstack-ansible | 15:20 | |
mnaser | logan-: the thing is i think ceph_client won't do anything if no ceph_mons are defined | 15:20 |
logan- | yeah | 15:21 |
vadim_ | yes, socket path needs to be fixed even for .30 mariadb | 15:21 |
mnaser | so gnocchi absolutely needs cradox even if you don't use ceph | 15:21 |
mnaser | :< | 15:21 |
logan- | i tihnk there might be a workaround for that if you feed it a ceph.conf dump directly? | 15:21 |
logan- | i can't remember | 15:21 |
logan- | but it normally copies ceph.conf from one of the mons | 15:21 |
mnaser | logan-: right but in the gate functional testing we're installing gnocchi with file storage | 15:21 |
mnaser | just to test if the role is working ok | 15:22 |
logan- | ah | 15:22 |
mnaser | im thinking include_role with the tasks from here https://github.com/openstack/openstack-ansible-ceph_client/blob/master/tasks/main.yml#L27-L33 | 15:22 |
mnaser | that will make sure whatever dependencies are in | 15:22 |
logan- | hmmm yeah if gnocchi needs some package regardless of whether ceph is enabled then this doesnt sound like a ceph_client change/issue | 15:22 |
mnaser | because the os_gnocchi role installs librados itself so i guess it should make sure that it has fresh repos too | 15:23 |
mnaser | if you think of it independently | 15:23 |
mgariepy | vadim_, you are probably better deploying pike on centos. | 15:23 |
vadim_ | why? | 15:24 |
mgariepy | ocata was the first release that ""supported"" centos for most of the openstack-service. why do you want to deploy ocata ? | 15:24 |
vadim_ | it's deployed already. and I plan to upgrade to pike soon. | 15:25 |
mgariepy | ho | 15:25 |
mgariepy | out of curiosity how is it going ? | 15:26 |
*** chhagarw has quit IRC | 15:26 | |
vadim_ | 15.1.10 was ok for a while. 15.1.14 broke galera. 15.1.15 did not fix it. mariadb 31 broke galera again;) overall ok | 15:28 |
*** dariko has quit IRC | 15:28 | |
*** dariko has joined #openstack-ansible | 15:29 | |
mnaser | ok | 15:30 |
mnaser | found it. | 15:30 |
mnaser | https://github.com/openstack/openstack-ansible-os_gnocchi/blob/master/defaults/main.yml#L166-L172 | 15:31 |
mnaser | the ",ceph" there should be dependent if the storage driver is ceph | 15:31 |
mnaser | i think i'll refactor it so that it only installs the extra which it needs (ex: ceph or swift or file depending on driver) | 15:31 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible stable/pike: Zuul: Remove project name https://review.openstack.org/542888 | 15:32 |
tescra | hi now syntax check says dynamic inventory Failed to parse /opt/openstack-ansible/inventory/dynamic_inventory.py with script plugin: Inventory script (/opt/openstack-ansible/inventory/dynamic_inventory.py) had an execution error: Traceback (most recent call last): File "/opt/openstack-ansible/inventory/dynamic_inventory.py", line 80, in <module> output = generate.main(**all_args) File "/opt/openstack-ansible/osa_tool | 15:35 |
tescra | no host found | 15:36 |
openstackgerrit | Marc GariƩpy (mgariepy) proposed openstack/openstack-ansible-galera_server stable/ocata: Fix Galera socket for RedHat https://review.openstack.org/542894 | 15:44 |
mgariepy | vadim_, can you test this patch ^^ | 15:45 |
vadim_ | sure | 15:47 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Install only the required storage driver extras https://review.openstack.org/542898 | 15:49 |
mgariepy | thanks | 15:49 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Switch to Keystone extra https://review.openstack.org/542900 | 15:50 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Update the Tempest plugin to point towards the new plugin https://review.openstack.org/541270 | 15:50 |
ashak | cloudnull: /me grumbles, seems I cannot cherry-pick those commits into my current checkout of tag 16.0.7 for each of those ansible roles due to conflicts. Checking out master for each role contains them but when I run the OSA playbooks it causes other problems. | 15:54 |
ashak | Should I be doing it some other way? Or are these commits likely to make it into a new tag soon enough that I can just let OSA bootstrapping take care of it for me? | 15:56 |
cloudnull | you could cherry pick that change using something like `git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD` | 15:59 |
cloudnull | from here https://review.openstack.org/#/c/542790/ | 15:59 |
cloudnull | or just make those two changes in the role requirements file | 15:59 |
ashak | oh I see | 15:59 |
ashak | I assumed you'd meant patch the two roles directly | 16:00 |
cloudnull | you could do either | 16:00 |
cloudnull | I'd just pull the updated roles down wholesale | 16:00 |
ashak | I did, they don't work :) | 16:00 |
cloudnull | the updated roles didn't work ? | 16:00 |
ashak | fatal: [infra1_galera_container-02a31604]: FAILED! => {"changed": false, "failed": true, "msg": "missing required arguments: name"} | 16:00 |
ashak | in TASK [galera_server : Reload systemd service files] | 16:01 |
*** pcaruana has quit IRC | 16:01 | |
ashak | But I checkout out master for both roles | 16:01 |
ashak | checked out | 16:01 |
*** dariko has quit IRC | 16:01 | |
cloudnull | after a fetch ? | 16:01 |
*** dariko1 has joined #openstack-ansible | 16:01 | |
ashak | Yes | 16:01 |
cloudnull | you may be too far ahead using the HEAD of master. | 16:01 |
cloudnull | for pike | 16:02 |
odyssey4me | git checkout master; git pull; ./scripts/bootstrap-ansible.sh | 16:02 |
odyssey4me | if you want to use pike, then just grab https://review.openstack.org/542790 - or wait for it to merge | 16:02 |
ashak | when might it merge? | 16:02 |
odyssey4me | in the next 90 mins or so if all goes well | 16:03 |
odyssey4me | you can also just use your current pike, make the changes to the ansible-role-requirements.yml file in that review, then bootstrap-ansible and go | 16:03 |
ashak | Ok, thanks | 16:05 |
tescra | root@deploy-node:/opt/openstack-ansible# git branch -a remotes/origin/stable/pike * stable/pike | 16:05 |
tescra | dynamic inventory Failed to parse | 16:06 |
tescra | none of playbook is executing | 16:07 |
tescra | shall i run boothstrap again | 16:07 |
prometheanfire | evrardjp: when requirements branches/unfreezes you will start getting rocky requirements updates to your master(queens) branch, either don't merge them or get a +1 from a requirements core (please) | 16:08 |
odyssey4me | every time you change the git ref, ansible should be bootstrapped again - otherwise you're not using the role designed to work with the repo | 16:08 |
tescra | odyssey4me is that ofr me? | 16:08 |
odyssey4me | tescra yes | 16:08 |
*** tux_ has joined #openstack-ansible | 16:09 | |
odyssey4me | tescra flipping from master to pike is not supported though | 16:09 |
evrardjp[m] | prometheanfire: yup i know. | 16:09 |
evrardjp[m] | the usual | 16:09 |
prometheanfire | evrardjp[m]: cool, just was looking for an ack before branching | 16:10 |
tescra | odyssey4me where shall i place the git head | 16:10 |
tescra | i started with pike | 16:10 |
evrardjp[m] | it's today ? | 16:10 |
*** jamesdenton has joined #openstack-ansible | 16:10 | |
evrardjp[m] | wow | 16:10 |
odyssey4me | tescra what are you trying to deploy? | 16:10 |
tescra | production env | 16:10 |
tescra | with pike | 16:10 |
tescra | on bare metal | 16:10 |
odyssey4me | if you're trying to deploy pike, then use stable/pike and add that patch on top of it | 16:11 |
tescra | how? | 16:11 |
odyssey4me | git checkout stable/pike; git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD | 16:11 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_designate stable/pike: Zuul: Remove project name https://review.openstack.org/542909 | 16:12 |
tescra | thanks | 16:12 |
*** armaan has joined #openstack-ansible | 16:13 | |
odyssey4me | tescra - perhaps better is: git fetch --all; git checkout origin/stable/pike; git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD | 16:13 |
tescra | it syas Already on 'stable/pike' remote: Counting objects: 18, done. remote: Compressing objects: 100% (10/10), done. remote: Total 11 (delta 7), reused 2 (delta 1) Unpacking objects: 100% (11/11), done. From https://git.openstack.org/openstack/openstack-ansible * branch refs/changes/90/542790/3 -> FETCH_HEAD error: could not apply 0e0819b... Document known issue with galera 10.1.31 hint: after resolving the co | 16:13 |
*** olivierbourdon38 has quit IRC | 16:13 | |
tescra | Unpacking objects: 100% (82/82), done. From https://git.openstack.org/openstack/openstack-ansible 6a906e7..8224fbc master -> origin/master eb357e2..30301c2 stable/ocata -> origin/stable/ocata 00d84d9..01368fc stable/pike -> origin/stable/pike ansible-role-requirements.yml: needs merge error: you need to resolve your current index first From https://git.openstack.org/openstack/openstack-ansible * branch | 16:15 |
tescra | ansible-role-requirements.yml: needs merge error: you need to resolve your current index first From https://git.openstack.org/openstack/openstack-ansible * branch refs/changes/90/542790/3 -> FETCH_HEAD error: cherry-pick is not possible because you have unmerged files. hint: Fix them up in the work tree, and then use 'git add/rm <file>' hint: as appropriate to mark resolution and make a commit. fatal: cherry-pi | 16:16 |
odyssey4me | tescra remove your customisations | 16:19 |
odyssey4me | git reset --hard origin/stable/pike | 16:19 |
*** vnogin has quit IRC | 16:20 | |
odyssey4me | git fetch --all; git reset --hard origin/stable/pike; git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD | 16:21 |
tescra | odyssey4me op was HEAD is now at 01368fc Use native packages to deploy keepalived | 16:21 |
odyssey4me | now cherry-pick on top | 16:21 |
tescra | shall i go for cherrypick cmd? | 16:21 |
tescra | now? | 16:22 |
odyssey4me | but that's old too | 16:22 |
tescra | k | 16:22 |
tescra | ?? | 16:22 |
tescra | old? | 16:22 |
odyssey4me | ah no, 01368fc is the head | 16:23 |
*** weezS has joined #openstack-ansible | 16:23 | |
odyssey4me | so cherry-pick on top of that | 16:23 |
*** vnogin has joined #openstack-ansible | 16:23 | |
ashak | odyssey4me: I did that but now Get Modern PIP is failing with trying to download http://172.29.236.3:8181/os-releases/16.0.8/ubuntu-16.04-x86_64/get-pip.py from my repo containers. Am I supposed to run something else after the bootstrap script to update the repo containers? | 16:23 |
odyssey4me | ashak yes, you'll need to now run setup-hosts.yml, setup-infrastructure.yml; setup-openstack.yml... you're effectively doing a minor upgrade now | 16:24 |
tescra | HEAD is now at 01368fc Use native packages to deploy keepalived root@deploy-node:/opt/openstack-ansible# git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD From https://git.openstack.org/openstack/openstack-ansible * branch refs/changes/90/542790/3 -> FETCH_HEAD *** Please tell me who you are. Run git config --global user.email "you@exampl | 16:24 |
tescra | is that ok? | 16:25 |
ashak | odyssey4me: oh of course | 16:25 |
odyssey4me | https://docs.openstack.org/openstack-ansible/pike/user/minor-upgrade.html | 16:25 |
tescra | git config --global user.name "Your Name" to set your account's default identity. Omit --global to set the identity only in this repository. fatal: unable to auto-detect email address (got 'root@deploy-node.(none)') | 16:25 |
ashak | Damn, that's going to take ages | 16:25 |
odyssey4me | tescra no, your git is not configured so it hasn't done the pick | 16:25 |
odyssey4me | you need to configure git | 16:25 |
tescra | by? | 16:26 |
odyssey4me | tescra I'm sure that google can help you. | 16:26 |
tescra | providing email? | 16:26 |
tescra | ok | 16:26 |
odyssey4me | the stderr is telling you what to do already | 16:27 |
tescra | only for deploy host or all servers? | 16:27 |
odyssey4me | just deploy host, and only because you're cherry-picking a change that hasn't merged | 16:27 |
ashak | You guys must love us untrained people and our inabilities to work some of this stuff out/find the correct documents :) | 16:28 |
odyssey4me | that's where we all started ;) | 16:28 |
ashak | I'm still struggling to work out where you got refs/changes/90/542790/3 to use with the git fetch command you suggested | 16:29 |
odyssey4me | tescra ashak the head of the branch is the proposed next release, we test it for two weeks before releasing it... so be aware that you're trying to use the bleeding edge ;) | 16:30 |
ashak | I constantly feel like i'm using the bleeding edge with this stuff ;) | 16:30 |
odyssey4me | ashak if you open https://review.openstack.org/542790 and select 'download' on the top right | 16:30 |
odyssey4me | next to 'cherry-pick' is the text showing how to cherry-pick it, and next to that is a 'copy to clipboard icon' | 16:31 |
ashak | odyssey4me: How would I have worked that out from the two commits you linked originally? | 16:31 |
dcdamien | Hi! I'm doing first fresh deploy after a break and noticed that os_cinder tries to install open-iscsi package on api *container* | 16:31 |
mnaser | :O | 16:31 |
mnaser | os_aodh successfully installed os_gnocchi on centos.. is this finally it.. | 16:32 |
odyssey4me | ashak well, you'd have to watch the review queue in this case... but if you look at https://github.com/openstack/openstack-ansible/blob/stable/pike/ansible-role-requirements.yml#L16 you'll see the 'version' value is a git SHA | 16:32 |
ashak | Jeez.... i'm never going to remember all this :) | 16:32 |
odyssey4me | if you then look at https://github.com/openstack/openstack-ansible-galera_client/commits/stable/pike (the repo being referred to), you'll find that sha there | 16:32 |
odyssey4me | that's how a role is pinned for a release | 16:33 |
logan- | anyone have an upstream bug link handy for the mariadb 10.1.31 galera issue | 16:33 |
odyssey4me | if you ever want to use a different SHA, then you change https://github.com/openstack/openstack-ansible/blob/stable/pike/ansible-role-requirements.yml#L16 to match the SHA you want to use | 16:33 |
odyssey4me | logan- see https://review.openstack.org/#/c/542790/3/releasenotes/notes/mariadb-10.1.31-package-issue-78816cff4e9f2de9.yaml | 16:33 |
ashak | Of course. I'd not thought about it from the point of view of OSA itself. Just the two roles with the problem. | 16:34 |
logan- | ahhh ok thanks | 16:34 |
logan- | missed the reno.. was looking in the commit msg :) | 16:34 |
odyssey4me | ashak yep - the openstack-ansible repo is an integration of all the roles with pinning of the SHA's so that a tagged release always uses the same thing - the sha's are updated at least every two weeks for every release | 16:34 |
dcdamien | apt fails on postinst: http://paste.openstack.org/show/666944/ and as I know it's because of open-iscsi can't work properly in the container | 16:35 |
ashak | odyssey4me: ok, nice one. Thanks for all the info. | 16:35 |
odyssey4me | dcdamien hmm, I think I've seen that before - there's a bug I think | 16:36 |
*** sar has joined #openstack-ansible | 16:36 | |
dcdamien | I'm curious how gates passes runs on CI :P | 16:36 |
odyssey4me | https://bugs.launchpad.net/openstack-ansible/+bug/1704161 | 16:36 |
openstack | Launchpad bug 1704161 in openstack-ansible "open-iscsi post-installation script fails in cinder-api-container" [High,Fix committed] | 16:36 |
odyssey4me | ah yes, we pushed up a change in gate and it now tests that continually... so yes, it's curious why it works in gate and not in your env | 16:37 |
* ashak yawns, is there a way to speed up running all the things involved in a minor upgrade? | 16:37 | |
odyssey4me | ashak you can change the serial values if you want, but then your api will go down as each service updates | 16:38 |
evrardjp | I am le back! | 16:38 |
odyssey4me | but it is a bit slow, and we're hoping to improve that over the next few cycles | 16:38 |
ashak | the serial values? | 16:38 |
tux_ | Any idea what has been included in Minor upgrade? https://docs.openstack.org/openstack-ansible/pike/user/minor-upgrade.html | 16:39 |
mnaser | "functional: commands succeeded" | 16:40 |
mnaser | !!! | 16:40 |
openstack | mnaser: Error: "!!" is not a valid command. | 16:40 |
ashak | oh, ansible's serial stuff to only allow the changes to happen on a certain percentage of hosts at once? | 16:40 |
*** sar has quit IRC | 16:40 | |
evrardjp | ashak: yes. | 16:40 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Zuul: Remove project name https://review.openstack.org/541172 | 16:42 |
odyssey4me | ashak each service has a _serial var which you can use, for example: openstack-ansible setup-openstack.yml -e 'cinder_serial=100%' -e 'glance_serial=100%' -e 'keystone_serial=100%' -e 'nova_serial=100%' | 16:43 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Updated from global requirements https://review.openstack.org/535049 | 16:43 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Replace virtualenv-tools by a script https://review.openstack.org/533645 | 16:43 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_aodh master: Standardize services list and package installation https://review.openstack.org/492580 | 16:44 |
odyssey4me | dcdamien I wonder if it is perhaps working in the gate because that package is installed on the host in the image we use? | 16:44 |
openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 16:45 |
tescra | evrardjp I am at stable pike | 16:45 |
ashak | odyssey4me: is there a place that lists them all? ;) | 16:45 |
tescra | but had root@deploy-node:/opt/openstack-ansible# git fetch https://git.openstack.org/openstack/openstack-ansible refs/changes/90/542790/3 && git cherry-pick FETCH_HEAD From https://git.openstack.org/openstack/openstack-ansible * branch refs/changes/90/542790/3 -> FETCH_HEAD On branch stable/pike You are currently cherry-picking commit 0e0819b. | 16:46 |
tescra | is it stable? | 16:46 |
tescra | for prod | 16:47 |
odyssey4me | ashak yes, grep for serial in the integrated repo and explore the code that comes out of it :) | 16:47 |
ashak | Hah | 16:47 |
mnaser | can i get a second core to have an eye on this? https://review.openstack.org/#/c/541270/ finally got it working, so i can figure out stable branches | 16:47 |
mnaser | (and the dependencies, tho most have merged) | 16:48 |
odyssey4me | tescra there are over 100 live production deployments using OSA, from juno all the way through to queens (unreleased) - so yes, it's stable for production... as long as you understand it's an open source project which requires feedback and help from its users | 16:48 |
dcdamien | odyssey4me: Probably. Can it be tested another way to make it similar to real deployment? | 16:48 |
tescra | thanks odyssesy | 16:48 |
odyssey4me | dcdamien well, if you can confirm if it works with the package installed on the host then we have the issue triaged and can figure out a fix - our tests only use what we're given, so 'production-like tests' require our users to feedback :) | 16:49 |
dcdamien | odyssey4me: ok, I'll test it | 16:50 |
odyssey4me | thanks dcdamien - it'd be nice to put that one to bed | 16:50 |
tescra | but will the previous values of backup_openstack_inventory.tar be overwritten ? How? | 16:51 |
tescra | my actuall conern was that the playbooks were unable to connect to dynamic inverntory | 16:53 |
*** armaan has quit IRC | 16:54 | |
tescra | all of a sudden none of the host were reachable | 16:54 |
mnaser | thanks logan- | 16:55 |
mnaser | and finally the dependency on this - https://review.openstack.org/#/c/542898/ | 16:55 |
mnaser | just to kick it into the gate :> | 16:55 |
mnaser | oh okay i noticed you did a check experimental | 16:55 |
mnaser | my bad :< | 16:55 |
odyssey4me | tescra not sure what you're talking about, but if the dynamic inventory fails then it's usually a configuration error | 16:56 |
logan- | mnaser: yeah hopefully check experimental does an integrated build like I think it does | 16:57 |
mnaser | logan-: it did according to zuul.o.o | 16:57 |
logan- | great | 16:57 |
odyssey4me | IIRC there's a --debug flag, you can run the inventory script on its own, this will do a check of your config: /opt/ansible-runtime/bin/python /opt/openstack-ansible/playbooks/inventory/dynamic_inventory.py --check | 16:57 |
odyssey4me | --debug will do debug information | 16:57 |
logan- | just want to make sure the py_pkgs lookup doesnt fumble the package name due to the templating. https://github.com/openstack/openstack-ansible/blob/8224fbcede862078d2a3914863f06339ca6c85cb/playbooks/repo-build.yml#L56-L62 | 16:58 |
*** armaan has joined #openstack-ansible | 16:58 | |
*** armaan has quit IRC | 16:59 | |
*** armaan has joined #openstack-ansible | 17:00 | |
tescra | no logs for ansible no play recap | 17:00 |
tescra | oddessy4me | 17:04 |
tescra | File "/opt/openstack-ansible/playbooks/inventory/dynamic_inventory.py", line 80, in <module> output = generate.main(**all_args) File "/opt/openstack-ansible/osa_toolkit/generate.py", line 1141, in main user_defined_config File "/opt/openstack-ansible/osa_toolkit/generate.py", line 747, in container_skel_load address_prefix=p_net.get('address_prefix') File "/opt/openstack-ansible/osa_toolkit/generate.py", | 17:06 |
tescra | on checking dynamic nventory | 17:06 |
tescra | it saya infr4 but am having 3 infra only | 17:07 |
odyssey4me | logan- mnaser yep, that'll be a problem | 17:07 |
logan- | odyssey4me: yea im thinking the same thing re: group_vars/all making this go | 17:07 |
odyssey4me | tescra check your config - and also, perhaps you previously had other entries... you can use inventory-manage.py to remove hosts from your inventory, or hack them out manually... or if this is a new environment and nothing's deployed you can delete it and start fresh | 17:08 |
ashak | odyssey4me: Meh, even with the updates, restart mysql is failing on the galera container on my first infra node. I guess it's somehow got itself into a broken state that these changes can't fix | 17:08 |
*** rama_y has quit IRC | 17:08 | |
odyssey4me | logan- mnaser it probably makes sense to set that var in group_vars/all anyway, because group_vars understand the inventory, and can therefore provide the list of the right things to add to it | 17:09 |
tescra | odyssey4me thanks | 17:09 |
ashak | Guess i'd best learn a whole bunch more about galera | 17:09 |
odyssey4me | ashak if you had already installed, you'd have to remove first and reinstall - or just wipe the ocntainers if this is a new build | 17:09 |
odyssey4me | otherwise, yes - get to know more about galera :) | 17:09 |
*** sar has joined #openstack-ansible | 17:09 | |
ashak | it's reasonably new and mostly in testing. But i'm trying to treat it like a production system because it always helps to now what fun you might face when it's actually in use :) | 17:10 |
mnaser | odyssey4me: logan- im a bit confused now so what should i be doing ideally for this? | 17:10 |
*** gameon has quit IRC | 17:13 | |
openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Document known issue with galera 10.1.31 https://review.openstack.org/542800 | 17:13 |
*** vnogin has quit IRC | 17:15 | |
*** vnogin has joined #openstack-ansible | 17:16 | |
*** weezS has quit IRC | 17:17 | |
*** weezS has joined #openstack-ansible | 17:19 | |
openstackgerrit | Maxime Guyot proposed openstack/openstack-ansible master: Ceph RadosGW integration https://review.openstack.org/517856 | 17:20 |
*** peri has quit IRC | 17:23 | |
*** vnogin has quit IRC | 17:25 | |
dmsimard | evrardjp, mnaser: btw I know you already know that but feel free to reach out if you'd like help on RDO things | 17:31 |
mnaser | thanks dmsimard :D | 17:31 |
*** armaan has quit IRC | 17:35 | |
*** armaan has joined #openstack-ansible | 17:36 | |
*** armaan has quit IRC | 17:37 | |
*** armaan has joined #openstack-ansible | 17:38 | |
*** rama_y has joined #openstack-ansible | 17:41 | |
*** rama_y has quit IRC | 17:42 | |
*** rama_y has joined #openstack-ansible | 17:42 | |
eumel8 | evrardjp: still around? | 17:42 |
openstackgerrit | Merged openstack/openstack-ansible stable/pike: Document known issue with galera 10.1.31 https://review.openstack.org/542790 | 17:45 |
openstackgerrit | Logan V proposed openstack/openstack-ansible master: Add gnocchi_storage_driver var to all group https://review.openstack.org/542957 | 17:48 |
*** mbuil has quit IRC | 17:49 | |
openstackgerrit | Logan V proposed openstack/openstack-ansible master: Add gnocchi_storage_driver var to all group https://review.openstack.org/542957 | 17:50 |
*** shardy has quit IRC | 17:51 | |
evrardjp | eumel8: I am sorry I have to go for today. I am not forgetting the simplification of the job, in fact, something just merged to improve it. | 17:54 |
evrardjp | dmsimard: :D | 17:55 |
evrardjp | see you everyone online next week! | 17:55 |
eumel8 | evrardjp: alright, was only just wondering that the job now runs successful: http://zuul.openstack.org/builds.html?job_name=openstack-ansible-deploy-translations-ubuntu-xenial | 17:56 |
*** dave-mccowan has quit IRC | 17:57 | |
*** dave-mccowan has joined #openstack-ansible | 17:58 | |
*** weezS has quit IRC | 18:01 | |
*** vnogin has joined #openstack-ansible | 18:05 | |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Install only the required storage driver extras https://review.openstack.org/542898 | 18:06 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Switch to Keystone extra https://review.openstack.org/542900 | 18:06 |
mnaser | rebased with depends-on logan- ^ | 18:07 |
logan- | cool, hopefully that does the trick | 18:07 |
*** vnogin has quit IRC | 18:09 | |
cloudnull | o/ logan- | 18:09 |
cloudnull | hows it ? | 18:09 |
odyssey4me | logan- mnaser made a suggesiton in https://review.openstack.org/#/c/542898/2 | 18:10 |
logan- | hey cloudnull. always a good day when its friday :) | 18:12 |
logan- | how about you | 18:12 |
cloudnull | same. | 18:12 |
cloudnull | friday fun day :) | 18:12 |
mnaser | hey thats a good call odyssey4me | 18:13 |
mnaser | voila | 18:14 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Install only the required storage driver extras https://review.openstack.org/542898 | 18:14 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_gnocchi master: Switch to Keystone extra https://review.openstack.org/542900 | 18:14 |
mnaser | https://github.com/openstack/openstack-ansible-os_keystone/blob/master/tasks/keystone_pre_install.yml#L16-L23 -- could someone explain whats the point of delegate_to + with_items as ansible_play_hosts ? | 18:15 |
*** olivierbourdon38 has joined #openstack-ansible | 18:15 | |
mnaser | sounds like delegate_to, with_items and the when can be dropped to accomplish the same exact thing? | 18:15 |
logan- | it is there to allow serialization of the playbook | 18:16 |
logan- | https://github.com/openstack/openstack-ansible-os_keystone/commit/86c42e0697049757641656dac60cb9a2e00159f5 | 18:16 |
mnaser | ah ok | 18:16 |
mnaser | feels so dirty :< | 18:16 |
logan- | yeah | 18:16 |
*** SerenaFeng has joined #openstack-ansible | 18:20 | |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 18:25 |
*** acormier has quit IRC | 18:28 | |
*** acormier has joined #openstack-ansible | 18:29 | |
*** acormier_ has joined #openstack-ansible | 18:32 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_aodh stable/pike: Zuul: Remove project name https://review.openstack.org/542971 | 18:32 |
*** acormier has quit IRC | 18:33 | |
*** acormier_ has quit IRC | 18:36 | |
mnaser | also whats the reasoning behind the log dir link/dir checks | 18:38 |
openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Avoid explicit requirement for groups to exist https://review.openstack.org/537774 | 18:51 |
*** chyka has joined #openstack-ansible | 18:52 | |
*** stuartgr has left #openstack-ansible | 18:53 | |
*** chyka has quit IRC | 18:56 | |
openstackgerrit | Merged openstack/openstack-ansible-tests stable/pike: test-set-nodepool-vars: Add nodepool mirror discovery for openSUSE https://review.openstack.org/542456 | 18:57 |
odyssey4me | logan- mnaser unfortunately the integrated build won't actually verify the whole stack build there - you'd need a build that deploys gnocci (or forces all wheel to build) | 18:59 |
logan- | ah true | 19:00 |
odyssey4me | if you have a gap to test an an AIO logan- please go ahead, otherwise I'll fire something up tomorrow | 19:00 |
odyssey4me | time for me to !computer | 19:00 |
odyssey4me | mnaser the logic for those relates to when the role is built on a host without containers - there's some weird symlink things we do which puts the folder in /openstack/log instead of /var/log, then we symlink... not sure why - I think it was to put all logs in the same place | 19:01 |
mnaser | odyssey4me: i see, it's a bit odd, hopefully we can clean that up at some point because it adds to the things that tie up roles to OSA | 19:02 |
odyssey4me | I would really like us to completely re-think logging/syslog next cycle, but unfortunately I don't have the spare cycles for it. | 19:02 |
mnaser | ++ | 19:02 |
odyssey4me | We have systemd-journal which could do far better, and we can be rid of a ton of legacy cruft if we figure out a goal we're all keen on. | 19:02 |
openstackgerrit | Merged openstack/openstack-ansible-os_designate stable/pike: Zuul: Remove project name https://review.openstack.org/542909 | 19:03 |
ashak | odyssey4me: removing the container with the broken instance seems to have fixed it as you suggested | 19:03 |
mnaser | odyssey4me: and im pretty sure with nspawn, systemd-journal logs can be accessed from the host.. i think | 19:03 |
odyssey4me | mnaser perhaps on interest, I did this patch yesterday which I think is a little cleaner: https://github.com/openstack/openstack-ansible-os_tempest/commit/4b9f1010fe884bfc17f9782b15343187700d27bf | 19:03 |
mnaser | cloudnull: might know more :) | 19:03 |
odyssey4me | mnaser yes they can, which is much more awesome | 19:03 |
mnaser | i'm hacking around this weekend around adding an 'keystone_installation_type' option, which you can guess does what | 19:04 |
odyssey4me | but for our lxc users it'd be nice to work out something equivalent - perhaps have all containers on a host journal to the host - then the host ships to the log host | 19:04 |
mnaser | "how do i make this change without breaking the world and making both source and binary installations work" | 19:05 |
mnaser | so far i'm trying to refactor the stages which need modification such as pre install and install to seperate files which are included based on install method, with common things remaining in the main file | 19:06 |
mnaser | to minimize impact | 19:06 |
logan- | im going to try the repo build on this pike aio i have running thru the gate odyssey4me, so no worries | 19:06 |
odyssey4me | mnaser you might want to take some notes from pabelanger's work in https://github.com/openstack/ansible-role-nodepool which caters for packaged and source deploys, and there's some scaffolding to allow the init stuff to be deployed (machine containers and hosts) or not (eg: app containers like docker)... there are some pretty neat patterns in there | 19:07 |
mnaser | odyssey4me: nice ill check that out right now | 19:07 |
odyssey4me | mnaser also see https://github.com/openstack/openstack-ansible-specs/blob/master/specs/queens/deployment-stages.rst which I'd be very happy to partner up on, that work needs contributors as it's a big job on its own and I'll be trying to do it along with https://github.com/openstack/openstack-ansible-specs/blob/master/specs/queens/python-build-install-simplification.rst which dovetails nicely into it | 19:09 |
odyssey4me | the staged deployments proposal lends itself very nicely to also having different installation methods | 19:09 |
odyssey4me | awesome, thank logan- | 19:10 |
logan- | is there a ptg etherpad up? | 19:10 |
odyssey4me | alright, time for me to head out for the night - thanks all! | 19:10 |
odyssey4me | logan- not as far as I've seen, but it about that time that evrardjp informs us :) | 19:10 |
logan- | have a good weekend odyssey4me | 19:11 |
*** lbragstad has quit IRC | 19:18 | |
*** SerenaFeng has quit IRC | 19:19 | |
ashak | should I be able to upload images direct from my browser via the web interface with a fairly standard default setup? | 19:22 |
*** vnogin has joined #openstack-ansible | 19:24 | |
ashak | Hrm, the actual request the browser makes isn't to horizon at all, it's to port 9292. | 19:25 |
ashak | oh directly to glance | 19:25 |
idlemind | ugh finally, dispatcher.d setup to activate my veth via up) and pre-down) for br-vlan | 19:26 |
idlemind | #networkmanager | 19:26 |
ashak | my issues looks like a CORS problem: Response to preflight request doesn't pass access control check: No 'Access-Control-Allow-Origin' header is present on the requested resource. Origin 'https://172.29.236.4' is therefore not allowed access. | 19:29 |
idlemind | ashak i wrote a bug for that | 19:29 |
idlemind | let me find it | 19:29 |
idlemind | https://bugs.launchpad.net/openstack-ansible/+bug/1747684 | 19:30 |
openstack | Launchpad bug 1747684 in openstack-ansible "Default Values Do Not Allow Image Uploads to Glance from Horizon" [Undecided,New] | 19:30 |
ashak | ooh, ta | 19:30 |
idlemind | you can modify your user_variables to get past 'er | 19:31 |
idlemind | but if you could post on the bug that it impacted you too maybe that'll help the pro's take a look at it | 19:31 |
ashak | I'll have a read and will do that if I can confirm it's the same problem (which is sounds like it is) | 19:32 |
idlemind | do we build the ifcfg-lxbr0 or is that part of someother process on redhat/centos? | 19:35 |
*** lbragstad has joined #openstack-ansible | 19:37 | |
openstackgerrit | Marc GariƩpy (mgariepy) proposed openstack/openstack-ansible master: [DOC] rectify galera recovery for systemd https://review.openstack.org/542994 | 19:37 |
*** idlemind has quit IRC | 19:37 | |
*** idlemind has joined #openstack-ansible | 19:38 | |
openstackgerrit | Merged openstack/openstack-ansible-galera_server stable/ocata: Fix Galera socket for RedHat https://review.openstack.org/542894 | 19:40 |
vadim_ | mgariepy, socket patch works, quotes in template can be omitted, though. | 19:42 |
*** kstev has quit IRC | 19:42 | |
*** vnogin has quit IRC | 19:43 | |
*** kstev has joined #openstack-ansible | 19:46 | |
mgariepy | vadim_, cool. | 19:51 |
mgariepy | well it's merged now :) | 19:51 |
idlemind | k, i think i'm going to stop letting kickstart attempting to do any network configuration on my hosts beyond putting a single ip on the server; it's nearly useless past that | 19:53 |
mgariepy | idlemind, try with preseed instead of kickstart :P\ | 19:53 |
mgariepy | then you can do the same with disk partitioning haha | 19:54 |
idlemind | mgariepy thx actually disk partitioning works like a champ beyond naming things a little against what my ocd desires | 19:55 |
idlemind | i'll have to look into the preseed stuff | 19:55 |
idlemind | i already have a post-install ansible script that does other things like adds cockpit, disables selinux (at least until we get 'er sorted in openstack) amongst other things | 19:56 |
mgariepy | ok | 19:57 |
idlemind | i'll look at the preseed for partitioning to see if it'll give me more control over naming for sure | 19:57 |
mgariepy | preseed + partitionning is not fun | 19:58 |
ashak | +1 to that :) | 19:58 |
mgariepy | idlemind, http://paste.openstack.org/show/666960/ | 19:59 |
mgariepy | all that for gpt + efi + raid1 ;p | 20:00 |
mgariepy | haha | 20:00 |
idlemind | and nothing about names, so must use the same self-generated ones you get w/kickstart but i can hit the docs and see what i can dig up | 20:01 |
idlemind | http://paste.openstack.org/show/666961/ | 20:02 |
idlemind | works reliably for a bios boot w/a goofy amount of different size volumes (usb boot volume + 2 x 2tb, 2 x 4tb) | 20:02 |
logan- | kickstart is so much nicer to provision with | 20:03 |
logan- | disk wise anyway | 20:03 |
idlemind | i'll take a look at preseed too | 20:03 |
mgariepy | zfs is so much better dans mdraid + lvm also. | 20:03 |
idlemind | i'd also like to figure out a way other than a dedicated pxe boot server to send custom ks's to my different hosts | 20:04 |
jrosser | i wish zfs root worked properly | 20:04 |
idlemind | longer term i'd like to go ceph bluestore directly so my perf will get limited by that i'd suppose | 20:04 |
idlemind | for now just ghetto lvm disk | 20:04 |
jrosser | then all this nonsense would go away | 20:04 |
mgariepy | debian-installer got me sad this morning, wait for someone to press a key on the kb before it continues | 20:05 |
mgariepy | on all the server i had to install.. | 20:05 |
mgariepy | zfs root would be awesome | 20:06 |
jrosser | we have it working partially | 20:06 |
jrosser | all the real activity like /var on a zfs mirror | 20:07 |
jrosser | and the bare minimum needed to boot remains not on zfs | 20:07 |
mgariepy | one of my colleague do netboot and the / on zfs. | 20:07 |
mgariepy | but it's somewhat hackish ;) haha | 20:07 |
jrosser | getting that all to come up working our a pxeboot was challenging | 20:08 |
jrosser | *out of | 20:08 |
mgariepy | for my ctrl plane i do: / raid1 and then my container on zfs. | 20:08 |
mgariepy | better than nothing i guess. | 20:08 |
idlemind | guess i never understood the love affair with zfs, is it all perf driven? | 20:10 |
jrosser | i guess if anything it's the opposite of that | 20:11 |
jrosser | features + data integrity + not being RAID | 20:11 |
jrosser | and of course when combined with lxc, copy-on-write | 20:12 |
idlemind | or combined w/cinder you'd get copy on write like ceph provides which would be a huge boon to image creation time but i don't think zfs has a driver does it? | 20:13 |
idlemind | does zfs require ecc memory? | 20:13 |
mgariepy | idlemind, everything needs ecc don't you think ? | 20:13 |
mgariepy | it's always better. | 20:14 |
idlemind | not saying it isn't but for my lab i'd prefer it's not required | 20:14 |
idlemind | and almost no commodity gear you find will have it | 20:14 |
openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/pike: Zuul: Remove project name https://review.openstack.org/542971 | 20:14 |
*** acormier has joined #openstack-ansible | 20:21 | |
*** kstev has quit IRC | 20:25 | |
*** olivierbourdon38 has quit IRC | 20:27 | |
*** yolanda has quit IRC | 20:28 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_gnocchi stable/newton: Zuul: Remove project name https://review.openstack.org/543015 | 20:33 |
*** hamza21 has quit IRC | 20:33 | |
*** dave-mccowan has quit IRC | 20:39 | |
*** acormier has quit IRC | 20:45 | |
*** acormier has joined #openstack-ansible | 20:47 | |
*** acormier has quit IRC | 20:47 | |
*** vnogin has joined #openstack-ansible | 20:48 | |
openstackgerrit | Merged openstack/openstack-ansible stable/pike: Zuul: Remove project name https://review.openstack.org/542888 | 20:48 |
*** acormier has joined #openstack-ansible | 20:48 | |
*** acormier has quit IRC | 20:50 | |
*** acormier has joined #openstack-ansible | 20:50 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_ceilometer stable/ocata: Zuul: Remove project name https://review.openstack.org/543022 | 20:53 |
adcpdk | idlemind thank you so much for such a nice troubleshooting yesterday. I finally got my networking setup and working as a flat network! Such a nice feeling when there are people out there to help and go out with you and fight with monsters you were scared of))) It was my first experience of using IRC channels in my life and I won't forget that. I will try to do my best to be around to help the newbies like me ;-) | 21:01 |
*** adcpdk has quit IRC | 21:06 | |
cloudnull | ^ that's awesome! | 21:12 |
cloudnull | community for the win! | 21:12 |
cloudnull | https://github.com/ansible/ansible/pull/35913 -- logan- have a look at that | 21:13 |
cloudnull | should greatly speed up variable render time | 21:13 |
cloudnull | however! hopefully it doesn't break everything | 21:14 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-tests master: DNM: Remove database contract for Keystone https://review.openstack.org/543026 | 21:15 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_keystone master: DNM: Deploy Keystone from binary packages https://review.openstack.org/543027 | 21:16 |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_keystone master: DNM: Deploy Keystone from binary packages https://review.openstack.org/543027 | 21:16 |
idlemind | adcpdk thx i'm going to blush | 21:20 |
idlemind | i too have found this community to be probably the best in the openstack ecosystem in terms of collaboration and one of the better open source projects as a whole | 21:20 |
mnaser | dmsimard: looks like osa uses openstack-pike repos instead of master, should we use master or master-head? | 21:20 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible stable/ocata: Remove jinja from failed_when https://review.openstack.org/541392 | 21:21 |
*** adcpdk has joined #openstack-ansible | 21:21 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible master: Use package_state=present for CentOS 7 testing https://review.openstack.org/541017 | 21:21 |
dmsimard | mnaser: where ? | 21:21 |
mnaser | dmsimard: https://github.com/openstack/openstack-ansible-openstack_hosts/blob/master/vars/redhat-7.yml#L101-L113 | 21:21 |
*** vnogin has quit IRC | 21:21 | |
dmsimard | mnaser: those are stable repos, not trunk | 21:22 |
*** chyka has joined #openstack-ansible | 21:22 | |
dmsimard | mnaser: purpose is not the same since OSA only uses RDO for deps | 21:22 |
mnaser | dmsimard: correct, but i see 2 master repos in rdo, wondering which one is better | 21:22 |
mnaser | dmsimard: aware, i'm going to be using master packages for my binary deployment fun (see reviews above) | 21:22 |
dmsimard | mnaser: For trunk you want to use https://trunk.rdoproject.org/centos7-master/delorean-deps.repo | 21:23 |
*** vnogin has joined #openstack-ansible | 21:23 | |
*** armaan has quit IRC | 21:23 | |
dmsimard | mnaser: for actual openstack packages, you want to use https://trunk.rdoproject.org/centos7-master/current-passed-ci/delorean.repo | 21:23 |
*** armaan has joined #openstack-ansible | 21:24 | |
mnaser | dmsimard: ok perfect! | 21:24 |
dmsimard | mnaser: however note that current-passed-ci is a symlink that can change so you typically want to resolve the hashed repository and set up the hashed repository (i.e, https://trunk.rdoproject.org/centos7/17/a5/17a57d99b10f5872520e1a395214f80c9cfd1a65_c30c070f ) | 21:24 |
mnaser | dmsimard: for dev purposes though, could we use unhashed so it always tests on latest (at least now in the early iterations) | 21:25 |
dmsimard | mnaser: unhashed means your job could start and have half the installed packages installed from a version and the other half on another version if it changed midway | 21:25 |
dmsimard | not sure if you see what I mean | 21:25 |
mnaser | dmsimard: oh i see what you mean | 21:25 |
mnaser | that makes sense now | 21:25 |
dmsimard | you have two ways of retrieving the hash | 21:26 |
dmsimard | you can parse that delorean.repo file | 21:26 |
dmsimard | and you also have a yaml file: https://trunk.rdoproject.org/centos7-master/current-passed-ci/commit.yaml (just realized we need to set up a text mimetype for that, use curl :)) | 21:26 |
mnaser | i think ill get_url it for now | 21:27 |
dmsimard | you can probably do a get_url and then use from_yaml or something | 21:27 |
*** adcpdk has quit IRC | 21:27 | |
dmsimard | the formula for the trunk repository hash is ... I foret | 21:27 |
dmsimard | forget | 21:27 |
mnaser | wasnt it <openstack_commit_id>_<first 8 of delorean current commit> or something | 21:28 |
mnaser | :p | 21:28 |
dmsimard | $commithash_$distrohash[:-8] ? | 21:28 |
dmsimard | yeah | 21:28 |
dmsimard | I always forget which one comes first and how many characters it is .. | 21:28 |
evrardjp | logan-: odyssey4me: the PTG's etherpad was discussed at last community meeting | 21:29 |
evrardjp | https://etherpad.openstack.org/p/osa-rocky-ptg | 21:29 |
evrardjp | it's also on the list of etherpads | 21:30 |
evrardjp | I took the liberty to add us to a photo slot | 21:30 |
idlemind | ugh, need to find me some reading on ansible'ing some cloud admin tasks (create projects / users) | 21:30 |
evrardjp | idlemind: check in our ops repo | 21:30 |
evrardjp | :) | 21:30 |
idlemind | hopefully in an idempotent way so can run, run and run again as needed. | 21:30 |
idlemind | cool cool will do | 21:30 |
spotz | hey logan- is etcd a requirement for Pike in general or just recommended. I saw you did our play:) | 21:32 |
*** armaan has quit IRC | 21:33 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-haproxy_server stable/newton: Zuul: Remove project name https://review.openstack.org/543028 | 21:33 |
*** adcpdk has joined #openstack-ansible | 21:41 | |
*** chyka has quit IRC | 21:43 | |
*** chyka has joined #openstack-ansible | 21:43 | |
*** sar has quit IRC | 21:44 | |
*** vnogin has quit IRC | 21:45 | |
*** tux_ has quit IRC | 21:45 | |
spotz | evrardjp: You sticking around or heading out? | 21:45 |
*** chyka has quit IRC | 21:48 | |
openstackgerrit | Mohammed Naser proposed openstack/openstack-ansible-os_keystone master: DNM: Deploy Keystone from binary packages https://review.openstack.org/543027 | 21:50 |
*** adcpdk has quit IRC | 21:59 | |
*** adcpdk has joined #openstack-ansible | 21:59 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Add nspawn container driver https://review.openstack.org/477017 | 22:03 |
*** chigang_ has quit IRC | 22:06 | |
*** chigang_ has joined #openstack-ansible | 22:06 | |
*** sar has joined #openstack-ansible | 22:08 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_swift stable/pike: Zuul: Remove project name https://review.openstack.org/543037 | 22:13 |
*** ildikov has quit IRC | 22:13 | |
*** ildikov has joined #openstack-ansible | 22:14 | |
idlemind | hey, does the rsyslog container have a web-ui for viewing the logs (could one be added to it by chance)? | 22:15 |
idlemind | ssh + attach + less = not as fun | 22:16 |
*** boxrick has quit IRC | 22:19 | |
*** boxrick has joined #openstack-ansible | 22:20 | |
*** pbandark has quit IRC | 22:23 | |
evrardjp | spotz: sorry, I was out, and I will stop working today. Let's postpone to Mondays :) | 22:27 |
spotz | evrardjp: we'll get time eventually! | 22:27 |
evrardjp | :) | 22:28 |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-os_nova stable/pike: Zuul: Remove project name https://review.openstack.org/543041 | 22:33 |
*** dave-mccowan has joined #openstack-ansible | 22:40 | |
*** acormier has quit IRC | 22:41 | |
*** armaan has joined #openstack-ansible | 22:43 | |
*** trident has joined #openstack-ansible | 22:52 | |
*** acormier has joined #openstack-ansible | 22:52 | |
*** dariko1 has quit IRC | 22:55 | |
*** acormier has quit IRC | 23:03 | |
*** AndyWojo has quit IRC | 23:04 | |
*** AndyWojo has joined #openstack-ansible | 23:04 | |
*** armaan has quit IRC | 23:06 | |
*** rodolof has quit IRC | 23:11 | |
*** geekinutah has quit IRC | 23:16 | |
*** geekinutah has joined #openstack-ansible | 23:16 | |
*** vnogin has joined #openstack-ansible | 23:31 | |
*** chyka has joined #openstack-ansible | 23:34 | |
openstackgerrit | James E. Blair proposed openstack/openstack-ansible-lxc_hosts stable/ocata: Zuul: Remove project name https://review.openstack.org/543052 | 23:34 |
*** esberglu has quit IRC | 23:37 | |
*** chyka has quit IRC | 23:37 | |
*** chyka has joined #openstack-ansible | 23:38 | |
*** chyka has quit IRC | 23:42 | |
*** eumel8 has quit IRC | 23:45 | |
*** vnogin has quit IRC | 23:46 | |
*** lbragstad has quit IRC | 23:49 | |
openstackgerrit | James E. Blair proposed openstack/ansible-hardening stable/pike: Zuul: Remove project name https://review.openstack.org/543053 | 23:54 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!