| *** sdake has quit IRC | 00:01 | |
| *** sdake_ has quit IRC | 00:08 | |
| *** sdake has joined #openstack-ansible | 00:11 | |
| *** openstack has joined #openstack-ansible | 00:26 | |
| *** fawadkhaliq has quit IRC | 00:30 | |
| *** metral is now known as metral_zzz | 00:30 | |
| *** metral_zzz is now known as metral | 00:30 | |
| *** fawadkhaliq has joined #openstack-ansible | 00:33 | |
| *** skamithi has joined #openstack-ansible | 00:54 | |
| *** skamithi has quit IRC | 00:57 | |
| *** skamithi1 has joined #openstack-ansible | 00:58 | |
| *** fawadkhaliq has quit IRC | 00:58 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:02 | |
| *** fawadkhaliq has quit IRC | 01:02 | |
| *** daneyon has quit IRC | 01:04 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:06 | |
| *** fawadkhaliq has quit IRC | 01:06 | |
| *** skamithi1 has quit IRC | 01:07 | |
| *** skamithi has joined #openstack-ansible | 01:08 | |
| *** skamithi has quit IRC | 01:11 | |
| *** skamithi has joined #openstack-ansible | 01:11 | |
| *** skamithi has quit IRC | 01:12 | |
| *** skamithi has joined #openstack-ansible | 01:13 | |
| *** johnmilton has joined #openstack-ansible | 01:17 | |
| *** izaakk has quit IRC | 01:17 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:18 | |
| *** fawadkhaliq has quit IRC | 01:19 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:30 | |
| *** fawadkhaliq has quit IRC | 01:31 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:42 | |
| *** skamithi14 has quit IRC | 01:44 | |
| *** skamithi13 has joined #openstack-ansible | 01:44 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_neutron: Remove iptables checksum rule script https://review.openstack.org/293067 | 01:45 |
|---|---|---|
| *** skamithi14 has joined #openstack-ansible | 01:50 | |
| *** skamithi13 has quit IRC | 01:50 | |
| *** fawadkhaliq has quit IRC | 01:51 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:51 | |
| *** johnmilton has quit IRC | 01:54 | |
| *** mgoddard has quit IRC | 01:54 | |
| *** fawadkhaliq has quit IRC | 02:06 | |
| *** skamithi14 has quit IRC | 02:12 | |
| *** skamithi13 has joined #openstack-ansible | 02:12 | |
| *** thorst_ has quit IRC | 02:18 | |
| *** johnmilton has joined #openstack-ansible | 02:23 | |
| *** fawadkhaliq has joined #openstack-ansible | 02:25 | |
| *** fawadkhaliq has quit IRC | 02:32 | |
| *** johnmilton has quit IRC | 02:42 | |
| *** cemason has joined #openstack-ansible | 03:11 | |
| *** thorst has joined #openstack-ansible | 03:19 | |
| *** thorst has quit IRC | 03:27 | |
| *** fawadkhaliq has joined #openstack-ansible | 04:14 | |
| *** skamithi has quit IRC | 04:18 | |
| *** thorst has joined #openstack-ansible | 04:25 | |
| *** thorst has quit IRC | 04:32 | |
| *** stackhead has joined #openstack-ansible | 04:33 | |
| *** stackhead has quit IRC | 04:40 | |
| *** raddaoui has joined #openstack-ansible | 04:49 | |
| *** asettle has quit IRC | 04:50 | |
| *** sdake has quit IRC | 04:50 | |
| *** jwitko has joined #openstack-ansible | 04:55 | |
| *** asettle has joined #openstack-ansible | 05:11 | |
| *** raddaoui has quit IRC | 05:21 | |
| *** thorst has joined #openstack-ansible | 05:30 | |
| *** admin0 has joined #openstack-ansible | 05:35 | |
| *** thorst has quit IRC | 05:37 | |
| *** admin0 has quit IRC | 05:38 | |
| *** admin0 has joined #openstack-ansible | 05:39 | |
| *** raddaoui has joined #openstack-ansible | 05:42 | |
| *** fawadkhaliq has quit IRC | 05:42 | |
| *** fawadkhaliq has joined #openstack-ansible | 05:43 | |
| *** raddaoui has quit IRC | 05:48 | |
| *** Bofu2U has quit IRC | 05:51 | |
| *** fawadkhaliq has quit IRC | 05:52 | |
| *** fawadkhaliq has joined #openstack-ansible | 05:53 | |
| *** winggundamth has joined #openstack-ansible | 06:08 | |
| *** saneax_AFK is now known as saneax | 06:17 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder: Test that the cinder-api is listening post-convergence https://review.openstack.org/293046 | 06:18 |
| *** phiche has joined #openstack-ansible | 06:29 | |
| *** Bofu2U has joined #openstack-ansible | 06:30 | |
| *** thorst has joined #openstack-ansible | 06:35 | |
| *** Mudpuppy_ has quit IRC | 06:41 | |
| *** thorst has quit IRC | 06:42 | |
| *** asettle has quit IRC | 06:49 | |
| *** admin0 has quit IRC | 06:55 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/290778 | 06:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Functional testing for Glance API and registry https://review.openstack.org/292402 | 06:55 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Add support for Horizon customization module https://review.openstack.org/290954 | 06:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Add support for Horizon customization module https://review.openstack.org/290954 | 07:01 |
| *** phiche has quit IRC | 07:02 | |
| *** fawadkhaliq has quit IRC | 07:03 | |
| *** fawadkhaliq has joined #openstack-ansible | 07:04 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add condition to local IP for overlay net https://review.openstack.org/273793 | 07:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Docs/Reno: Add support for Horizon customization module https://review.openstack.org/275494 | 07:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Account for oslo_config deprecations in heat.conf template https://review.openstack.org/290748 | 07:08 |
| openstackgerrit | Merged openstack/openstack-ansible: Removing openstack_environment.yml. It is no longer useful. https://review.openstack.org/288698 | 07:10 |
| *** phiche has joined #openstack-ansible | 07:15 | |
| openstackgerrit | Merged openstack/openstack-ansible: Docs/Reno: Add support for Horizon customization module https://review.openstack.org/275494 | 07:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_server: Disable slave repo servers while syncing https://review.openstack.org/279452 | 07:19 |
| *** unlaudable has joined #openstack-ansible | 07:21 | |
| *** fawadkhaliq has quit IRC | 07:25 | |
| *** fawadkhaliq has joined #openstack-ansible | 07:26 | |
| *** fawadkhaliq has quit IRC | 07:26 | |
| *** fawadkhaliq has joined #openstack-ansible | 07:27 | |
| *** eric_lopez has quit IRC | 07:28 | |
| *** elo has joined #openstack-ansible | 07:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance: Functional testing for Glance API and registry https://review.openstack.org/292402 | 07:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_server: Trim apt package list https://review.openstack.org/289510 | 07:40 |
| *** thorst has joined #openstack-ansible | 07:40 | |
| *** mariusv has quit IRC | 07:44 | |
| *** lunarlamp has joined #openstack-ansible | 07:48 | |
| *** thorst has quit IRC | 07:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add LBaaS v2 support (OSA) https://review.openstack.org/287243 | 07:55 |
| *** unlaudable has quit IRC | 08:03 | |
| *** Mudpuppy has joined #openstack-ansible | 08:04 | |
| *** Mudpuppy has quit IRC | 08:09 | |
| *** admin0 has joined #openstack-ansible | 08:10 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Fix supplied identity uri https://review.openstack.org/293037 | 08:11 |
| *** admin0 has quit IRC | 08:11 | |
| odyssey4me | o/ hughsaunders | 08:13 |
| *** starmer has joined #openstack-ansible | 08:13 | |
| *** unlaudable has joined #openstack-ansible | 08:15 | |
| *** MCoLo has joined #openstack-ansible | 08:19 | |
| *** permalac has joined #openstack-ansible | 08:20 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/287913 | 08:21 |
| openstackgerrit | git-harry proposed openstack/openstack-ansible: Fix supplied identity uri https://review.openstack.org/293303 | 08:21 |
| *** mikelk has joined #openstack-ansible | 08:21 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/287913 | 08:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/293307 | 08:26 |
| *** cemason1 has joined #openstack-ansible | 08:26 | |
| *** cemason has quit IRC | 08:28 | |
| *** starmer has quit IRC | 08:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/293307 | 08:32 |
| *** admin0 has joined #openstack-ansible | 08:36 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/293307 | 08:38 |
| *** ig0r_ has quit IRC | 08:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: Re-initialise when restarting 1-node cluster https://review.openstack.org/290716 | 08:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add ability to configure Neutron FWaaS https://review.openstack.org/287913 | 08:41 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Add ability to configure Neutron FWaaS https://review.openstack.org/287915 | 08:45 |
| *** thorst has joined #openstack-ansible | 08:46 | |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible-os_nova: WIP: Add Nova config for os_ironic role https://review.openstack.org/293315 | 08:46 |
| *** thorst has quit IRC | 08:52 | |
| *** starmer has joined #openstack-ansible | 08:55 | |
| *** neilus has joined #openstack-ansible | 08:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update LBaaS docs/config to use simple plugin names https://review.openstack.org/293320 | 09:01 |
| openstackgerrit | Merged openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/290778 | 09:02 |
| odyssey4me | hughsaunders https://review.openstack.org/290778 should be backported to liberty/kilo, shouldn't it? | 09:03 |
| * hughsaunders looks | 09:03 | |
| hughsaunders | yep, both | 09:04 |
| hughsaunders | I'll do that | 09:04 |
| odyssey4me | thanks | 09:04 |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/293324 | 09:05 |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/293325 | 09:05 |
| *** mgoddard has joined #openstack-ansible | 09:12 | |
| *** skamithi has joined #openstack-ansible | 09:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add ability to configure Neutron FWaaS https://review.openstack.org/275894 | 09:14 |
| *** elopez has quit IRC | 09:19 | |
| *** skamithi has quit IRC | 09:22 | |
| openstackgerrit | Merged openstack/openstack-ansible: Gate: Use OpenStack-CI apt mirrors https://review.openstack.org/291598 | 09:35 |
| *** thorst has joined #openstack-ansible | 09:50 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** openstackgerrit_ is now known as openstackgerrit | 09:53 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** openstackgerrit_ has joined #openstack-ansible | 09:54 | |
| *** openstackgerrit_ is now known as openstackgerrit | 09:54 | |
| *** openstackgerrit_ has joined #openstack-ansible | 09:55 | |
| *** openstackgerrit has quit IRC | 09:55 | |
| *** openstackgerrit has joined #openstack-ansible | 09:56 | |
| *** thorst has quit IRC | 09:57 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_keystone: [WIP] Dummy dynamic inventory for tests https://review.openstack.org/292911 | 09:58 |
| *** karimb has joined #openstack-ansible | 10:03 | |
| *** skamithi has joined #openstack-ansible | 10:07 | |
| *** skamithi has quit IRC | 10:07 | |
| mattt | no idea where ansible keeps facts when running tests, but this is really annoying me :) | 10:11 |
| mattt | anyone have any idea? | 10:11 |
| *** electrofelix has joined #openstack-ansible | 10:15 | |
| mattt | odyssey4me: finally got https://review.openstack.org/#/c/292911/ to gate, so this is a pattern we can use rather than having the test playbook manually add things to the inventory | 10:15 |
| jiteka | mattt: hey | 10:15 |
| jiteka | mattt: are you using fact cahing ? | 10:15 |
| jiteka | caching ? | 10:15 |
| mattt | jiteka: apparently :) | 10:15 |
| mattt | jiteka: we do for sure, but when i run a dummy test playbook i have no idea where those facts are going | 10:16 |
| mattt | (there is no /etc/ansible and they're not in ~/.ansible) | 10:16 |
| jiteka | mattt: I found an interesting page about that http://jpmens.net/2015/01/29/caching-facts-in-ansible/ | 10:16 |
| *** starmer has quit IRC | 10:16 | |
| jiteka | mattt: by default it's in memory | 10:17 |
| mattt | jiteka: what is tripping me up is i run something, change inventory etc. and re-run the test playbook but it's still seeing old values | 10:17 |
| admin0 | inside there /etc/openstack-deploy/ | 10:18 |
| mattt | admin0: i'm running the ansible tests as a non-root user, and /etc/openstack-deploy doesn't exist | 10:18 |
| admin0 | oh | 10:18 |
| mattt | yeah it's super weird | 10:18 |
| jiteka | mattt: so you mean even when you change inventory (so you gather fact on other hosts) you still see previous hosts facts values ?! | 10:19 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_keystone: [WIP] Dummy dynamic inventory for tests https://review.openstack.org/292911 | 10:20 |
| mattt | jiteka: i think so :( | 10:20 |
| mattt | i've just seen a bunch of weirdness running these functional tests and i suspect it has something to do w/ facts | 10:20 |
| git-harry | mattt: I think inventory precedence is fairly low | 10:20 |
| git-harry | Are you sure it's not set somewhere else | 10:21 |
| jiteka | mattt: you can try this option as well --flush-cache | 10:21 |
| mattt | git-harry: i'm not sure, but like i changed inventory, tore down containers, re-ran the tests and it still fails trying to use an old IP | 10:21 |
| jiteka | mattt: and choose as fact caching mechanism json | 10:21 |
| mattt | i could be herping the derp, but i ran into similar issues fiddling w/ these tests which i attributed to facts | 10:21 |
| jiteka | mattt: if you want to dig more into it | 10:22 |
| mattt | cool, thanks jiteka | 10:22 |
| jiteka | :) | 10:22 |
| jiteka | mattt: let me know if it made the trick, I'm curious | 10:22 |
| jiteka | :) | 10:22 |
| mattt | will do! | 10:34 |
| Bofu2U | morn | 10:35 |
| odyssey4me | mattt great, will take a look shortly :) | 10:45 |
| odyssey4me | jiteka unfortunately --flush-cache is broken in Ansible 1.9x :) | 10:46 |
| jiteka | awwwh | 10:46 |
| jiteka | damned | 10:46 |
| jiteka | :D | 10:47 |
| odyssey4me | mattt some of the test playbooks are set not to gather facts, also you can force the gathering of facts by implementing the setup module in the play :) | 10:47 |
| jiteka | odyssey4me: gather_facts: False ? | 10:48 |
| jiteka | odyssey4me: gather_facts: True use setup module right ? | 10:48 |
| jiteka | odyssey4me: or you mean using setup module in another way ? | 10:49 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder: Remove the messaging tasks from the role https://review.openstack.org/288103 | 10:50 |
| odyssey4me | gather_facts: True will do it at the start of the play, but you can initiate it (I believe) at any time using a task inside a playbook | 10:50 |
| jiteka | mattt: by the way are you only dealing with remote host fact problem or local fact as well ? | 10:52 |
| jiteka | mattt: by default you should be able to take a look here /etc/ansible/facts.d for local facts | 10:52 |
| jiteka | odyssey4me: ok, I never tried to use setup module as a task as I use fact from gather_fact and set_fact for other purposes | 10:53 |
| *** johnmilton has joined #openstack-ansible | 10:55 | |
| *** thorst has joined #openstack-ansible | 10:55 | |
| *** jiteka has quit IRC | 11:02 | |
| *** thorst has quit IRC | 11:02 | |
| *** neilus has quit IRC | 11:03 | |
| *** jiteka has joined #openstack-ansible | 11:04 | |
| *** neilus has joined #openstack-ansible | 11:04 | |
| odyssey4me | #success OpenStack-Ansible now has a Designate role ready for testing: http://git.openstack.org/cgit/openstack/openstack-ansible-os_designate | 11:05 |
| openstackstatus | odyssey4me: Added success to Success page | 11:05 |
| admin0 | \o/ | 11:05 |
| mattt | woot! | 11:07 |
| odyssey4me | mattt hughsaunders if you have a moment, I'd like to get this chain through the door today: https://review.openstack.org/287243 / https://review.openstack.org/293307 / https://review.openstack.org/293320 | 11:10 |
| odyssey4me | git-harry / hughsaunders it seems confirmed that we're hitting the 30 min job limit on the roles, I guess I'll have to see if we're in a position to optimise the use of infra resources there before we ask for an increase in timeout | 11:11 |
| mattt | odyssey4me: sure, will have a look | 11:11 |
| mattt | reading automagically's core contributor doc | 11:15 |
| odyssey4me | mattt great :) | 11:15 |
| mattt | odyssey4me: what is meant by '+A changes' ? | 11:15 |
| mattt | i'm assuming that's workflowing a review ? | 11:15 |
| odyssey4me | mattt +a = +w | 11:15 |
| mattt | ok cool | 11:16 |
| mattt | just checking :P | 11:16 |
| odyssey4me | yeah, I'm not sure why the +A is used in OpenStack in general | 11:16 |
| mattt | +approved i guess ? | 11:16 |
| odyssey4me | yeah, perhaps - maybe we should make that clearer for those who don't have the tribal knowledge | 11:16 |
| *** openstackgerrit has quit IRC | 11:18 | |
| *** openstackgerrit has joined #openstack-ansible | 11:18 | |
| mattt | also wonder if we need to start switching up our meetings a bit | 11:20 |
| mattt | ie. one week at a certain time, the next at another | 11:20 |
| mattt | to give those in australia, etc. a chance to join | 11:20 |
| mattt | odyssey4me: i'd recommend getting some of those involved w/ lbaas to review these patches, the current revisions have no reviews | 11:27 |
| mattt | i'm coming in pretty late into the game not knowing anything about lbaas | 11:27 |
| git-harry | odyssey4me: yeah, I had a quick look through the logs on a galera patch but couldn't see any obvious cause for the timeout, I think it's just too much to do in too short a time. | 11:28 |
| *** javeriak has joined #openstack-ansible | 11:30 | |
| *** karimb has quit IRC | 11:30 | |
| *** robby_ has joined #openstack-ansible | 11:32 | |
| *** admin0 has quit IRC | 11:34 | |
| *** weshay has joined #openstack-ansible | 11:34 | |
| *** thorst has joined #openstack-ansible | 11:35 | |
| *** admin0 has joined #openstack-ansible | 11:37 | |
| *** skamithi has joined #openstack-ansible | 11:37 | |
| *** skamithi13 has quit IRC | 11:39 | |
| *** skamithi has quit IRC | 11:39 | |
| *** skamithi13 has joined #openstack-ansible | 11:39 | |
| *** skamithi14 has joined #openstack-ansible | 11:46 | |
| *** skamithi13 has quit IRC | 11:49 | |
| *** admin0 has quit IRC | 11:51 | |
| odyssey4me | mattt the one involved is the author :p note that the patch in the os_neutron role has already merged... that's simply documentation | 11:56 |
| mattt | odyssey4me: yeah since realised :) | 11:56 |
| odyssey4me | and activation of the horizon bits, which also merged some time ago | 11:56 |
| mattt | looking at the other 2 now | 11:56 |
| *** admin0 has joined #openstack-ansible | 11:58 | |
| openstackgerrit | Merged openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/293325 | 11:59 |
| *** robby_ has quit IRC | 12:02 | |
| odyssey4me | #success OpenStack-Ansible now has a Zaqar role ready for testing: http://git.openstack.org/cgit/openstack/openstack-ansible-os_zaqar | 12:05 |
| openstackstatus | odyssey4me: Added success to Success page | 12:05 |
| *** Bofu2U has quit IRC | 12:05 | |
| *** Mudpuppy has joined #openstack-ansible | 12:06 | |
| *** Mudpuppy has quit IRC | 12:11 | |
| *** winggundamth has quit IRC | 12:11 | |
| openstackgerrit | Merged openstack/openstack-ansible: Fix command not found in scripts lib https://review.openstack.org/293324 | 12:17 |
| odyssey4me | mattt good comment in https://review.openstack.org/#/c/293307/3/tasks/neutron_db_setup.yml - I'm just not sure if it is meant to run offline - the original patch ran it online :/ | 12:20 |
| mattt | odyssey4me: i'm building an instance now to test this all, so more feedback in a few | 12:21 |
| odyssey4me | ok cool, you're better equipped to know what to look for there | 12:21 |
| *** woodard has joined #openstack-ansible | 12:24 | |
| mattt | odyssey4me: i could have sworn i tested this with neutron-fwaas and no migrations had to be explicitly run for that | 12:24 |
| *** woodard has quit IRC | 12:24 | |
| *** woodard has joined #openstack-ansible | 12:25 | |
| *** pcaruana has quit IRC | 12:28 | |
| odyssey4me | mattt it might be that your online migration method works already, and the offline method isn't required, in which case your condition edit is correct | 12:30 |
| odyssey4me | mattt that work was broken out of the fwaas changes which follow on from that | 12:31 |
| *** fawadkhaliq has quit IRC | 12:31 | |
| odyssey4me | I broke it out to try and ensure that the feature and the re-config for LBaaS were seperated into focused tests. | 12:31 |
| odyssey4me | *focused patches | 12:32 |
| odyssey4me | pabelanger did you manage to complete your tooling to convert an ansible-role-requirements.yml file into something that zuul-cloner can parse? | 12:35 |
| mattt | odyssey4me: yeah it shouldn't be necessary, but i'll verify ... also what happens when neutron_plugin_type != 'ml2' ? | 12:37 |
| mattt | presumably you can use plumgrid with neutron-lbass right ? | 12:37 |
| pabelanger | odyssey4me: sorry, not yet. I can focus on that this week | 12:38 |
| odyssey4me | mattt that caters for plumgrid - if you use a non-ml2 plugin then whatever you have running must handle its own things | 12:38 |
| odyssey4me | our only use-case at the moment for that is plumgrid | 12:38 |
| odyssey4me | pabelanger great, I'll hang ten and see how you fare while I work on other stuff | 12:39 |
| javeriak | matt an implementation for the lbass agent is being created, but at the atm no we dont enable it with plumgrid | 12:39 |
| mattt | javeriak: ok cool :) | 12:40 |
| *** ramishra has joined #openstack-ansible | 12:42 | |
| *** ramishra has quit IRC | 12:43 | |
| mattt | odyssey4me: would it not make sense for https://review.openstack.org/#/c/287243/9/doc/source/install-guide/configure-lbaas.rst to use the same names you're changing to in https://review.openstack.org/#/c/293307/3/defaults/main.yml ? | 12:49 |
| mattt | i've -W until you've commented | 12:51 |
| *** karimb has joined #openstack-ansible | 12:52 | |
| *** jlvillal has quit IRC | 12:52 | |
| *** jlvillal has joined #openstack-ansible | 12:52 | |
| *** inc0 has joined #openstack-ansible | 12:54 | |
| *** Rodrigo_BR has joined #openstack-ansible | 12:54 | |
| odyssey4me | mattt see the follow on patch: https://review.openstack.org/#/c/293320/1 | 12:55 |
| odyssey4me | until the patch merges into the neutron role, the current method should be in the docs | 12:55 |
| odyssey4me | hopefully we can fast follow with the merge of the patches in both repositories that change to the short naming | 12:55 |
| odyssey4me | but I don't want to hold back the LBaaS doc inclusion for the short names stuff | 12:56 |
| mattt | yep makes sense | 12:56 |
| bsv | guys, you know if there is any automatic way to clean up orphaned resources? | 12:58 |
| *** inc0_ has joined #openstack-ansible | 12:58 | |
| *** inc0 has quit IRC | 12:58 | |
| *** javeriak has quit IRC | 12:59 | |
| admin0 | rm -rf all and redo :D | 13:02 |
| admin0 | nah | 13:02 |
| Rodrigo_BR | When I deployed using ceph backend the radosgw is deployed instead of swift ? or I need do this manually ? | 13:02 |
| *** neilus has quit IRC | 13:02 | |
| *** neilus1 has joined #openstack-ansible | 13:02 | |
| *** neilus1 has quit IRC | 13:03 | |
| bsv | admin0: ha, mr. funny guy ;) | 13:03 |
| admin0 | they are orphaned resources .. let them grow .. in time, they will move out :D | 13:03 |
| *** neilus has joined #openstack-ansible | 13:04 | |
| admin0 | define the orphaned resources | 13:04 |
| admin0 | disks , ips, instances | 13:04 |
| admin0 | why were they orphaned in the first place | 13:04 |
| bsv | created project, user, networks, routers, instances... then killed the project "project delete 3b7161aa58684464b21d8973834c7e97" :D | 13:06 |
| bsv | Want to see how the system handles careless os-admins ;) | 13:06 |
| bsv | it does not ! :) | 13:06 |
| *** scarlisle has joined #openstack-ansible | 13:07 | |
| odyssey4me | bsv haha, welcome to OpenStack :p | 13:09 |
| odyssey4me | some stuff behaves well, some does not - it's a loosely coupled system | 13:09 |
| odyssey4me | a lot of the stuff left behind is intentional though, to allow you to recover | 13:10 |
| bsv | ok | 13:10 |
| odyssey4me | there are shadow tables where stuff is archived, and as an Operator you're expected to truncate those from time to time according to your policies | 13:10 |
| odyssey4me | bsv you may be interested in https://launchpad.net/osops - both to glean information, and to contribute | 13:11 |
| bsv | Is it possible to map the orphaned resources to an existing project then? | 13:11 |
| odyssey4me | you may also wish to sign up to the openstack-operators mailing list | 13:11 |
| bsv | odyssey4me: ahh yes, most certainly | 13:12 |
| odyssey4me | bsv well, the orphaned stuff in the DB is easy enough - each service has its own DB | 13:12 |
| bsv | >.< | 13:12 |
| odyssey4me | the clean up of stuff like network namespaces, etc does happen - sometimes it needs some special configuration to enable | 13:12 |
| bsv | So, write a tool that enums all the available services, connect to each service and do a recursive search of the orphaned objects and any object linking to them (rinse and repeat), then end up with a object-id map used for deleting all the objects found? | 13:14 |
| *** kro_sinus_hell is now known as krot_sinus_purga | 13:15 | |
| odyssey4me | bsv well, it may be best to actually engage with the openstack-operators list and to ask how other operators deal with it... it's probably best to also give examples of 'orphaned resources' that you're seeing | 13:15 |
| *** krot_sinus_purga is now known as kro_sniffles | 13:15 | |
| *** cemason1 has quit IRC | 13:15 | |
| bsv | sure | 13:15 |
| openstackgerrit | Merged openstack/openstack-ansible: Add LBaaS v2 support (OSA) https://review.openstack.org/287243 | 13:19 |
| *** ig0r_ has joined #openstack-ansible | 13:24 | |
| *** fawadkhaliq has joined #openstack-ansible | 13:29 | |
| *** automagically_ has joined #openstack-ansible | 13:29 | |
| *** automagically_ has left #openstack-ansible | 13:31 | |
| *** automagically has left #openstack-ansible | 13:32 | |
| *** automagically has joined #openstack-ansible | 13:32 | |
| *** sdake has joined #openstack-ansible | 13:33 | |
| *** ametts has joined #openstack-ansible | 13:33 | |
| *** tiagogomes has quit IRC | 13:34 | |
| *** metachr0n has quit IRC | 13:34 | |
| automagically | morning all | 13:35 |
| *** iceyao has joined #openstack-ansible | 13:35 | |
| odyssey4me | o/ automagically | 13:35 |
| * odyssey4me grabs some lunch | 13:36 | |
| mattt | morning automagically | 13:39 |
| *** admin0_ has joined #openstack-ansible | 13:40 | |
| *** admin0 has quit IRC | 13:40 | |
| *** admin0_ is now known as admin0 | 13:40 | |
| gregfaust | morning :) | 13:40 |
| odyssey4me | o/ gregfaust | 13:40 |
| *** tiagogomes has joined #openstack-ansible | 13:41 | |
| *** admin0 has quit IRC | 13:41 | |
| automagically | Hey gregfaust, glad to see you here | 13:44 |
| mattt | odyssey4me: so we shouldn't have to explicitly run neutron-lbaas migrations, because neutron-lbaas is in the venv irrespective of whether or not you're using it it will run migrations for it | 13:44 |
| odyssey4me | mattt ah, so the expand/contract methods will work for those? | 13:45 |
| mattt | odyssey4me: appears so | 13:45 |
| odyssey4me | or should we add the condition you suggested? | 13:45 |
| mattt | odyssey4me: i don't think you need any addition db sync, so i'd just remove that | 13:45 |
| odyssey4me | cool - please comment that in the review, and in the fwaas review as well | 13:46 |
| mattt | odyssey4me: incase you know, is neutron-lbaas the python package that gets used irrespective of v1 or v2 ? | 13:46 |
| odyssey4me | mattt yeah, afaik it's the same package | 13:46 |
| mattt | ok cool | 13:46 |
| odyssey4me | just a different class | 13:46 |
| *** galstrom_zzz is now known as galstrom | 13:48 | |
| gregfaust | I'm interested in mirroring all of the OSA external dependencies. I'm wondering if there is an OSA dependency list to reference or if I need to look at each of the IRRs. | 13:48 |
| *** neilus has quit IRC | 13:48 | |
| *** neilus1 has joined #openstack-ansible | 13:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Fix inconsistent gate and lint failures for os-swift https://review.openstack.org/292363 | 13:48 |
| odyssey4me | gregfaust if you build an AIO then you'll get the standard dependent wheels - it'll be on the repo server under the tagged folder - an example is http://rpc-repo.rackspace.com/os-releases/12.0.5/requirements_absolute_requirements.txt | 13:49 |
| odyssey4me | gregfaust unless you're looking for all the apt deps? | 13:49 |
| automagically | ALLLLL the deps | 13:50 |
| odyssey4me | gregfaust note that openstack-ansible allows you to add your own python wheels to the repo, so not all OSA builds are equal :) | 13:50 |
| *** admin0 has joined #openstack-ansible | 13:52 | |
| *** neilus1 has quit IRC | 13:53 | |
| gregfaust | odyssey4me, thanks! | 13:54 |
| *** mgoddard_ has joined #openstack-ansible | 14:01 | |
| *** daneyon has joined #openstack-ansible | 14:03 | |
| *** jthorne has joined #openstack-ansible | 14:03 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar: Implement base configuration for OpenStack-CI https://review.openstack.org/293449 | 14:04 |
| *** mgoddard has quit IRC | 14:05 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:06 | |
| *** spotz_zzz has quit IRC | 14:06 | |
| *** mgagne has quit IRC | 14:07 | |
| *** hughsaunders has quit IRC | 14:07 | |
| *** toan has quit IRC | 14:07 | |
| *** mgagne has joined #openstack-ansible | 14:07 | |
| *** alextricity25 has quit IRC | 14:07 | |
| *** hughsaunders has joined #openstack-ansible | 14:07 | |
| *** toan has joined #openstack-ansible | 14:08 | |
| *** alextricity25 has joined #openstack-ansible | 14:08 | |
| *** busterswt has joined #openstack-ansible | 14:08 | |
| *** spotz_zzz has joined #openstack-ansible | 14:09 | |
| *** KLevenstein has joined #openstack-ansible | 14:12 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar: Implement base configuration for OpenStack-CI https://review.openstack.org/293449 | 14:13 |
| alextricity25 | cloudnull: suh dude | 14:13 |
| alextricity25 | cloudnull: What flavor do you use for the multi-node-aio? | 14:13 |
| odyssey4me | alextricity25 the bare metal flavor :) | 14:14 |
| *** neilus has joined #openstack-ansible | 14:14 | |
| admin0 | :D | 14:14 |
| *** daneyon has quit IRC | 14:15 | |
| alextricity25 | I'm using general purpose v2 large, and getting this: http://cdn.pasteraw.com/3nsvls4yq4uyz7l0q6jz25lhq3ivl62 | 14:15 |
| alextricity25 | This is on RAX OnMetal | 14:15 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_neutron: Fix typo in tasks/neutron_lbaas.yml https://review.openstack.org/293458 | 14:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/293307 | 14:17 |
| odyssey4me | mattt ^ remove the db migration | 14:17 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_neutron: Fix typo in tasks/neutron_lbaas.yml https://review.openstack.org/293458 | 14:17 |
| *** Mudpuppy has joined #openstack-ansible | 14:17 | |
| *** Mudpuppy has quit IRC | 14:19 | |
| *** Mudpuppy has joined #openstack-ansible | 14:19 | |
| *** galstrom is now known as galstrom_zzz | 14:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add ability to configure Neutron FWaaS https://review.openstack.org/287913 | 14:24 |
| *** raddaoui has joined #openstack-ansible | 14:24 | |
| *** sdake has quit IRC | 14:27 | |
| *** galstrom_zzz is now known as galstrom | 14:27 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Add ability to configure Neutron FWaaS https://review.openstack.org/287915 | 14:27 |
| mattt | odyssey4me: wicked, done | 14:29 |
| *** fawadkhaliq has quit IRC | 14:30 | |
| odyssey4me | hmm, it seems that if you have the same change-id across repositories, zuul doesn't know what to do with it | 14:30 |
| odyssey4me | I'll have to resubmit these patch with a fresh change id | 14:30 |
| *** fawadkhaliq has joined #openstack-ansible | 14:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Add ability to configure Neutron FWaaS https://review.openstack.org/293467 | 14:32 |
| *** tiagogomes has quit IRC | 14:35 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: Role should fail immediately when its preconditions are not met https://review.openstack.org/291873 | 14:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Add ability to configure Neutron FWaaS https://review.openstack.org/293469 | 14:36 |
| *** iceyao has quit IRC | 14:36 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add ability to configure Neutron FWaaS https://review.openstack.org/275894 | 14:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Add ability to configure Neutron FWaaS https://review.openstack.org/293469 | 14:39 |
| cloudnull | alextricity25: I'm using onmetal-io | 14:40 |
| *** saneax is now known as saneax_AFK | 14:40 | |
| alextricity25 | cloudnull: iov2? The one with only 240GB of system disk? Do you attack a block device to it afterwards? | 14:41 |
| cloudnull | alextricity25: v1/2 will work | 14:41 |
| cloudnull | they come with 1.3T as unpartitioned . | 14:41 |
| *** tiagogomes has joined #openstack-ansible | 14:44 | |
| *** daneyon has joined #openstack-ansible | 14:46 | |
| *** mgoddard_ has quit IRC | 14:47 | |
| *** mgoddard has joined #openstack-ansible | 14:47 | |
| alextricity25 | cloudnull: oh I see it now | 14:49 |
| alextricity25 | thanks :) | 14:49 |
| *** spotz_zzz is now known as spotz | 14:49 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron: Fix typo in tasks/neutron_lbaas.yml https://review.openstack.org/293458 | 14:50 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_keystone: [WIP] Fix how constraints are added to pip_install_options https://review.openstack.org/293476 | 14:51 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_keystone: [WIP] Fix how constraints are added to pip_install_options https://review.openstack.org/293476 | 14:52 |
| *** cloudtrainme has joined #openstack-ansible | 14:55 | |
| *** galstrom is now known as galstrom_zzz | 14:55 | |
| cloudnull | alextricity25: anytime . | 14:57 |
| cloudnull | if you have improvements you want to make the process, let me know and PRs are always welcome | 14:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar: Implement base configuration for OpenStack-CI https://review.openstack.org/293449 | 14:58 |
| alextricity25 | For sure :) | 15:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Use upper-constraints file in developer mode https://review.openstack.org/292301 | 15:02 |
| *** galstrom_zzz is now known as galstrom | 15:02 | |
| *** javeriak has joined #openstack-ansible | 15:06 | |
| *** inc0___ has joined #openstack-ansible | 15:08 | |
| *** inc0___ has quit IRC | 15:09 | |
| *** inc0_ has quit IRC | 15:12 | |
| *** kencjohnston_ has joined #openstack-ansible | 15:14 | |
| *** galstrom is now known as galstrom_zzz | 15:22 | |
| *** woodard has quit IRC | 15:22 | |
| *** sdake has joined #openstack-ansible | 15:25 | |
| *** weezS has joined #openstack-ansible | 15:26 | |
| *** karimb has quit IRC | 15:27 | |
| *** unlaudable has quit IRC | 15:30 | |
| *** fawadkhaliq has quit IRC | 15:33 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_keystone: Fix how constraints are added to pip_install_options https://review.openstack.org/293476 | 15:33 |
| *** fawadkhaliq has joined #openstack-ansible | 15:34 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_zaqar: Implement base configuration for OpenStack-CI https://review.openstack.org/293449 | 15:38 |
| *** fawadkhaliq has quit IRC | 15:38 | |
| meteorfox | andymccr: hi | 15:38 |
| *** jroll has quit IRC | 15:41 | |
| *** jroll has joined #openstack-ansible | 15:41 | |
| *** jroll has quit IRC | 15:41 | |
| *** jroll has joined #openstack-ansible | 15:41 | |
| mattt | automagically: can you do the needful on https://review.openstack.org/#/c/292976/ ? | 15:42 |
| mattt | (please and thank you :)) | 15:42 |
| *** tiagogomes has quit IRC | 15:43 | |
| odyssey4me | jmccrory automagically cloudnull d34dh0r53 stevelle mattt hughsaunders andymccr can I get some help smashing this one through? it's the base set of things for anyone to work on the zaqar role if they want to: https://review.openstack.org/293449 | 15:43 |
| * cloudnull looking | 15:43 | |
| hughsaunders | also looking | 15:44 |
| palendae | cloudnull: l2pop is listed in the ml2_conf.ini right? | 15:45 |
| hughsaunders | heh approved by cloudnull | 15:45 |
| cloudnull | yes. | 15:45 |
| cloudnull | palendae: ^ | 15:45 |
| palendae | Ok, it's not in my liberty upgrade stuff | 15:45 |
| *** tiagogomes has joined #openstack-ansible | 15:46 | |
| cloudnull | https://github.com/openstack/openstack-ansible/blob/liberty/playbooks/roles/os_neutron/defaults/main.yml#L260 | 15:46 |
| openstackgerrit | Merged openstack/openstack-ansible-os_zaqar: Implement base configuration for OpenStack-CI https://review.openstack.org/293449 | 15:46 |
| cloudnull | ^ in kilo thats hard coded to true , now its an option | 15:46 |
| cloudnull | idk if you need it | 15:46 |
| palendae | k | 15:46 |
| cloudnull | but something worth looking at | 15:46 |
| palendae | https://bugs.launchpad.net/neutron/+bug/1509312 is the behavior I'm seeing, and like it says, manually creating the entries works | 15:47 |
| openstack | Launchpad bug 1509312 in neutron "unable to use tenant network after kilo to liberty update due to port security extension" [Undecided,Expired] | 15:47 |
| andymccr | hey meteorfox, how's it going? | 15:47 |
| meteorfox | andymccr: doing well, and you? | 15:48 |
| andymccr | good good! | 15:48 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: pin version of ansible-lint we use for testing https://review.openstack.org/292976 | 15:50 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_cinder: OpenStack services should reach Glance via the internal LB VIP https://review.openstack.org/290856 | 15:51 |
| meteorfox | andymccr: cool, I have at least what I hope is a quick question. I want to delete all the data from /srv/nodes/ but because the services are running it tries to replicate them immediately, and I can't clean it up. I know I have to stop the services, but is there an init script, or just a script to stop, and then after I deleted everything start the Swift | 15:52 |
| meteorfox | services with OSA? | 15:52 |
| Nepoc | good morning, quick question. I am trying to override the repo variables with something like requirements_git_repo: "http://{{ repo_server_ip }}/G/openstack/aodh" however it ends up trying to connect to 'http://{#' repo_server_ip '#}/G/openstack/aodh' what kind of variable magic is going on here? | 15:52 |
| andymccr | meteorfox: there's not an automated play within osa to do that, but you could just stop rsync on all of the servers. | 15:53 |
| andymccr | the swift services themselves have init scripts | 15:53 |
| meteorfox | I see | 15:54 |
| andymccr | so if you look in /etc/init/ there should be a script per service | 15:54 |
| Nepoc | Also the what it's trying to connect to is actually that string, it's not substituting the repo_server_ip for the ip | 15:54 |
| logan- | Nepoc.. yeah the python module py_pkgs doesn't use jinja templating afaik. I ran into that a while back I think | 15:55 |
| cloudnull | Nepoc: I assume youre also defining "repo_server_ip" as an option ? | 15:55 |
| meteorfox | andymccr: right I saw that, but it look somewhat cumbersome to do stop each service independently, I was looking for swift-init all stop or something like that | 15:55 |
| meteorfox | but I can try stopping rsync | 15:55 |
| automagically | Nepoc: You can work around this by defining your own “local_packages” variable in a user_*.yml config | 15:56 |
| meteorfox | andymccr: btw, I found rsync max connections to be too low. Currently at 2, if you have a workload that touches many objects/containers it creates a backlog really quickly. I'm experimenting with 8 connections each, but I want to try different parameters | 15:57 |
| automagically | Nepoc: An example of what I mean: https://github.com/openstack/openstack-ansible-repo_build/blob/master/tests/test.yml#L29 | 15:57 |
| *** admin0 has quit IRC | 15:57 | |
| Nepoc | automagically: Excellent, I like the look of that | 15:58 |
| Nepoc | automagically: thank you, trying that out in 10 minutes | 15:58 |
| *** woodard has joined #openstack-ansible | 15:58 | |
| *** woodard has quit IRC | 15:59 | |
| andymccr | meteorfox: ok, i see that option isnt exposed atm, we should adjust that. | 15:59 |
| automagically | Nepoc: Also, you may want to use something like this: https://gist.github.com/trumant/b0093f13ee8c57d73988 to get a good baseline value to start with | 15:59 |
| *** woodard has joined #openstack-ansible | 15:59 | |
| Nepoc | automagically: don't suppose someone already has a dump of the local_packages I can take? :) | 16:00 |
| automagically | Nepoc: I’ve got one, but I can’t vouch for how up to date it is. If you run that playbook on your deployment host under /opt/openstack-ansible/playbooks, you’ll get the value you need | 16:01 |
| Nepoc | automagically: sounds good trying it now | 16:02 |
| *** skamithi14 has quit IRC | 16:03 | |
| *** skamithi13 has joined #openstack-ansible | 16:03 | |
| *** jiteka has quit IRC | 16:03 | |
| Nepoc | automagically: well that was easy enough... now to use this dump as fertilizer for a user_offline.yml file | 16:04 |
| *** raddaoui has quit IRC | 16:04 | |
| cloudnull | alextricity25: https://github.com/evrardjp/ansible-haproxy | 16:04 |
| automagically | Nepoc: Glad it helped you out | 16:04 |
| *** fawadkhaliq has joined #openstack-ansible | 16:04 | |
| alextricity25 | cloudnull: thanks! | 16:05 |
| *** raddaoui has joined #openstack-ansible | 16:10 | |
| *** cloudtrainme has quit IRC | 16:12 | |
| *** elopez has joined #openstack-ansible | 16:20 | |
| *** cloudtrainme has joined #openstack-ansible | 16:21 | |
| meteorfox | andymccr: I stopped rsync on both regions and I still see objects/ directories popping up on all storage hosts | 16:22 |
| andymccr | meteorfox: if you're putting objects then you would. the easiest way is probably to write a play that will just stop all the services. | 16:23 |
| mattt | if anyone has any thoughts on https://review.openstack.org/#/c/293476/ that'd be great, as this will need to be done to all IRRs that we've implemented these constraints on | 16:25 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Let rsync max_connections be set by variables https://review.openstack.org/293543 | 16:25 |
| odyssey4me | mattt looking | 16:27 |
| *** raddaoui has quit IRC | 16:27 | |
| *** mgoddard has quit IRC | 16:28 | |
| *** mgoddard has joined #openstack-ansible | 16:28 | |
| odyssey4me | mattt I'm not sure I understand the logic | 16:28 |
| odyssey4me | if pip_install_options is set, it should be used and we should add to it - that was the intent, right? | 16:29 |
| odyssey4me | if I understand you correctly, that's not working? | 16:29 |
| *** fawadkhaliq has quit IRC | 16:32 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:33 | |
| *** weezS has quit IRC | 16:33 | |
| *** fawadkhaliq has quit IRC | 16:33 | |
| odyssey4me | andymccr can config_template not allow it to be set? | 16:35 |
| andymccr | odyssey4me: its hard coded in the template | 16:35 |
| odyssey4me | andymccr doesn't matter - if you use the config_template override you can add and overwrite any setting in the file | 16:35 |
| odyssey4me | there is no need for a var | 16:36 |
| odyssey4me | which template is this? | 16:36 |
| odyssey4me | heh, I see you have a patch in - lemme look quickly | 16:36 |
| odyssey4me | andymccr hmm, rsync doesn't appear to have a [default] section, so I'm not sure if this will work entirely - but it may be worth the try | 16:39 |
| andymccr | you can use config_templates for that, but its a required setting - it seems more sensible to have a variable, otherwise we'd be in a situation where we remove all vars and just use config_template for everything? | 16:40 |
| odyssey4me | if https://github.com/openstack/openstack-ansible-os_swift/blob/master/tasks/swift_storage_hosts_setup.yml#L35-L42 is altered to use config_template to put the file down and there's a var which defines the override, then deployers can add any setting and override any setting | 16:40 |
| odyssey4me | well, yes - we should only really have vars for things that we need to override specifically to connect services together | 16:41 |
| *** severion has joined #openstack-ansible | 16:41 | |
| odyssey4me | the more vars we have, the more a deployer has to learn about OSa instead of learning about the service in question | 16:41 |
| odyssey4me | in this case I agree that it's a bit arb and what you're doing may make sense, but I do suggest that config_template is perhaps used for that conf file in case someone wants to set some other arb setting | 16:42 |
| mattt | odyssey4me: so it kinda happens now, things are sort of working, except i found an edge case yesterday | 16:43 |
| *** eil397 has joined #openstack-ansible | 16:43 | |
| mattt | odyssey4me: because we set a fact for pip_options, if you have two different services installed on the same host, it's going to use the same pip_install_options fact irrespective of whether or not one service has developer mode set to true and the other false | 16:44 |
| mattt | odyssey4me: so if my logic here is correct this should allow you to have different pip_install_options values depending on how the service is being installed | 16:45 |
| odyssey4me | mattt ah, I see | 16:45 |
| odyssey4me | this is because everything is executed in one long play list | 16:45 |
| *** cloudtrainme has quit IRC | 16:46 | |
| odyssey4me | so the fact remains in memory through the entire run | 16:46 |
| mattt | odyssey4me: not necessarily, i think it's more that we for example install keystone, and glance into a single container, and the fact is saved across plays | 16:46 |
| odyssey4me | yes, for the same host | 16:46 |
| mattt | yeah, it's kinda messy, i think i implemented that constraints stuff in a bit of a hurry :( | 16:47 |
| odyssey4me | mattt I'm wondering, instead of churning all those lines | 16:49 |
| odyssey4me | urgh, I just realised that it doesn't make much difference | 16:50 |
| *** fawadkhaliq has joined #openstack-ansible | 16:50 | |
| *** fawadkhaliq has quit IRC | 16:51 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:51 | |
| odyssey4me | I was thinking perhaps set a new var called 'pip_developer_mode_options', then add it after pip_install_options in each extra_args. | 16:52 |
| odyssey4me | but then you have to default it anyway, so your way may actually be neater | 16:52 |
| openstackgerrit | Merged openstack/openstack-ansible: Make the purpose of the example user_variables file clearer https://review.openstack.org/293076 | 16:54 |
| openstackgerrit | Merged openstack/openstack-ansible: Docs/Reno: Make default_schedule_zone configurable https://review.openstack.org/290161 | 16:54 |
| *** admin0 has joined #openstack-ansible | 16:55 | |
| mattt | odyssey4me: cool, if we get someone else to agree then i'll hack the rest out tomorrow morning | 16:57 |
| Nepoc | automagically: That worked for replacing the variables... now I just need to figure out how to match git.openstack.org's setup so I stop getting "Failed to checkout <commit hash>" | 16:59 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 17:02 | |
| odyssey4me | we need another review on https://review.openstack.org/291933 if anyone has the time to take a look | 17:02 |
| odyssey4me | mattt https://review.openstack.org/293469 is another small piece of the FWaaS puzzle (for horizon) | 17:03 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Adding Vagrantfile for local developer testing https://review.openstack.org/288934 | 17:04 |
| Nepoc | Anyone know who I should talk to that can tell me how to replicate the setup on git.openstack.org? | 17:06 |
| odyssey4me | nepoc you can ask in #openstack-infra | 17:07 |
| Nepoc | Thank you | 17:07 |
| odyssey4me | sure | 17:07 |
| jmccrory | mattt: would storing the extra constraints in a default var, and then appending them to pip_install_options in other tasks based on dev_mode | bool work? | 17:09 |
| *** jguy_ has joined #openstack-ansible | 17:09 | |
| jguy_ | Looking for an openvswitch module for OSAD…google keeps picking up a github project called openstack-ansible. But the code does not look like it is part of the official project. Has anyone setup OVS? | 17:09 |
| odyssey4me | jguy_ that project builds icehouse only and was a thertiary education project | 17:10 |
| odyssey4me | jguy_ there is work ongoing to include OVS, but no - OSA doesn't yet do it - we're looking for contributors | 17:10 |
| odyssey4me | jguy_ is there something you need OVS specifically for? | 17:11 |
| jguy_ | I see under “extending”, there is an “Adding galaxy roles”. Is there a ansible galaxy with additional modules, like OVS available? | 17:11 |
| odyssey4me | jguy_ OVS is simply a package to install... to instrument it with OpenStack would probably require some extra prep on host bridges and such. The rest is probably configurable with no code change in OSA. | 17:12 |
| jguy_ | Odyssey4me: That L2GW thing I am trying to get working. I have it loaded, but it interacts with an OVSDB on hardware switches…thinking possibly OVS is needed on the neutron agent. :( | 17:12 |
| *** mikelk has quit IRC | 17:12 | |
| odyssey4me | jguy_ I'm not really clued up with that stuff. Perhaps Apsu could confirm it - he knows his way around networking a bit. :p | 17:13 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Install packages necessary to build Python wheels https://review.openstack.org/291933 | 17:13 |
| jguy_ | Not really sure why the control plane connection is not working…I see the packets arrive, but the TCP sessions is not connecting to anything. | 17:14 |
| jguy_ | So was not sure if the ovs was needed for the ovsdb-client function to work. | 17:14 |
| odyssey4me | jmccrory (cc mattt) I was thinking in a similar way - but it seems to be much of a muchness. | 17:14 |
| jguy_ | The debugging sux on it… Perhaps I need to walk the code. | 17:15 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Enable functional testing https://review.openstack.org/289454 | 17:17 |
| *** cloudtrainme has joined #openstack-ansible | 17:26 | |
| *** KLevenstein has quit IRC | 17:33 | |
| *** phiche has quit IRC | 17:33 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Removing openstack_environment.yml testing symlink https://review.openstack.org/293606 | 17:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: [WIP] Tests: Make containers use the host's apt/pip config https://review.openstack.org/293611 | 17:40 |
| *** weezS has joined #openstack-ansible | 17:41 | |
| *** raddaoui has joined #openstack-ansible | 17:42 | |
| *** elopez has quit IRC | 17:42 | |
| *** fawadkhaliq has quit IRC | 17:49 | |
| *** fawadkhaliq has joined #openstack-ansible | 17:50 | |
| *** mgoddard_ has joined #openstack-ansible | 17:59 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: [WIP] Tests: Make containers use the host's apt/pip config https://review.openstack.org/293611 | 17:59 |
| *** mgoddard has quit IRC | 18:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: [WIP] Tests: Make containers use the host's apt/pip config https://review.openstack.org/293611 | 18:05 |
| Apsu | jguy_: ovsdb-client talks to ovsdb, which manages the userspace database OVS uses. | 18:06 |
| Apsu | jguy_: Not super useful to be running ovsdb without using the OVS system and switches | 18:07 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 18:07 | |
| *** skamithi14 has joined #openstack-ansible | 18:13 | |
| *** skamithi13 has quit IRC | 18:15 | |
| jguy_ | @apsu: I know that, but I was not sure if I have to impliment the neutron networking with openvswitch, or if the neutron-server could interact with the ovsdb-client to populate the l2gateway tables for sending to the hardware switch. | 18:17 |
| odyssey4me | automagically cloudnull jmccrory stevelle how're things going over there? | 18:18 |
| automagically | i.e. in the big ole Drumpfed up USA? | 18:18 |
| stevelle | huh wha? :) | 18:19 |
| cloudnull | WAT!? | 18:19 |
| stevelle | ^ I've got an andymccr image macro for that | 18:19 |
| Apsu | jguy_: Eh.... I suspect you'll need to use OVS with neutron through and through, because you have to pick your ml2 plugin and it applies to everything, afaik | 18:20 |
| jguy_ | That is what I was afraid of. :( | 18:21 |
| Apsu | Well it's not like OVS doesn't work. It is the default configuration. | 18:22 |
| jguy_ | In openstack-ansible? | 18:23 |
| *** elopez has joined #openstack-ansible | 18:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: [WIP] Tests: Make containers use the host's apt/pip config https://review.openstack.org/293611 | 18:26 |
| Apsu | jguy_: No, in openstack proper | 18:27 |
| *** Mudpuppy_ has joined #openstack-ansible | 18:28 | |
| odyssey4me | it was for a few cycles, but that is changing | 18:28 |
| jguy_ | Yeah…However in the Liberty install guide, OVS is gone. | 18:28 |
| odyssey4me | it is the preferred option for the network engineers, but not for the systems engineers | 18:28 |
| *** Mudpuppy_ has quit IRC | 18:28 | |
| *** karimb has joined #openstack-ansible | 18:28 | |
| jguy_ | Being a network engineer, I find OVS a real pain to work with. Not going to miss it when it is gone. | 18:30 |
| *** Mudpuppy_ has joined #openstack-ansible | 18:30 | |
| *** Mudpuppy has quit IRC | 18:31 | |
| *** Mudpuppy_ has quit IRC | 18:31 | |
| *** Mudpuppy has joined #openstack-ansible | 18:32 | |
| Apsu | I think most network engineers are unfamiliar with how advanced stock Linux networking is | 18:33 |
| Apsu | Most of them don't even realize net-tools and bridge-utils are ancient and broken | 18:33 |
| jguy_ | That is all changing. Look at Cumulus, Pica8, etc | 18:34 |
| Apsu | Sure | 18:34 |
| Apsu | I've got a couple of each. Well, traded the Picas for a Cumulus. But yeah | 18:34 |
| Apsu | Broadcom has their ODMs too | 18:34 |
| *** raddaoui has quit IRC | 18:35 | |
| jguy_ | this test I am working on with the l2gw is connecting to a cumulus ovsdb-server. They don’t run openvswitch anywhere that I can see… That is why I was thinking the actual openvswitch is not needed... | 18:35 |
| Apsu | I haven't tried it. Have at it | 18:36 |
| *** tiagogomes has quit IRC | 18:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Tests: Make containers use the host's apt/pip config https://review.openstack.org/293611 | 18:48 |
| *** electrofelix has quit IRC | 18:49 | |
| *** fawadkhaliq has quit IRC | 18:49 | |
| *** jorge_munoz has joined #openstack-ansible | 18:50 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Remove super-verbosity from ansible-playbook execution https://review.openstack.org/293647 | 18:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Tests: Remove super-verbosity from ansible-playbook execution https://review.openstack.org/293647 | 18:51 |
| lbragstad | o/ hey folks - I'm curious if it's possible to expose some of the oslo.log stuff in the os_keystone role. I'm not seeing anything in the variable defaults for those but I could be looking in the wrong place https://github.com/os-cloud/openstack-ansible-os_keystone/blob/master/defaults/main.yml | 18:52 |
| *** mgoddard_ has quit IRC | 18:54 | |
| odyssey4me | lbragstad you may wish to start making use of https://github.com/openstack/openstack-ansible-os_keystone :) and tracking it | 18:54 |
| *** mgoddard has joined #openstack-ansible | 18:54 | |
| odyssey4me | also, anything not covered in the template conf file can be added using this method: http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-openstack.html | 18:55 |
| lbragstad | odyssey4me ah - you're a smart man.. i assume that's the role the gets pulled in when i do something like https://github.com/rackerlabs/capstone/blob/master/deploy/deploy.yml#L6 ? | 18:55 |
| lbragstad | odyssey4me nope - looks like I'm out of date! https://github.com/rackerlabs/capstone/blob/master/deploy/ansible-role-requirements.yml#L29 | 18:56 |
| odyssey4me | lbragstad yep, we imported the role a week or two ago | 18:56 |
| odyssey4me | we've been working on improving the functional tests since then | 18:56 |
| lbragstad | odyssey4me awesome - that's good news | 18:57 |
| lbragstad | odyssey4me we're looking to deploy keystone with some specific logging configurations, do you know if there is a way for us to deploy oslo.log specific configurations in keystone through the os_keystone role? | 18:57 |
| lbragstad | or is that something that will come later? | 18:57 |
| odyssey4me | so if the entries you want to add need to be in keystone.conf, then you'll see in https://github.com/openstack/openstack-ansible-os_keystone/blob/master/tasks/keystone_post_install.yml#L16-L29 that keystone.conf is laid down with the var called 'keystone_keystone_conf_overrides' for overrides | 18:57 |
| odyssey4me | does that apply to what you're looking to do? | 18:58 |
| odyssey4me | otherwise you'll have to let me know a bit more about how you would do this without ansible :) | 18:58 |
| lbragstad | odyssey4me looks like we already use that https://github.com/rackerlabs/capstone/blob/master/deploy/deploy.yml#L50 | 18:58 |
| odyssey4me | yep, so if you want to add another section then add it at the same level as 'auth' | 18:59 |
| odyssey4me | what're the entries you want as a result? | 18:59 |
| lbragstad | odyssey4me we wanted to override https://github.com/openstack/keystone/blob/master/etc/keystone.conf.sample#L133 | 19:00 |
| lbragstad | so that we could specify out own logging config file | 19:00 |
| lbragstad | our* | 19:00 |
| lbragstad | jorge_munoz was that the only thing we changed in our keystone.conf? | 19:00 |
| odyssey4me | so that's under default, right? | 19:00 |
| lbragstad | odyssey4me yes - it appears so | 19:01 |
| odyssey4me | so above 'auth', add 'DEFAULT' at the same level of indentation, then with two spaces add the option | 19:02 |
| jorge_munoz | jorge_munoz: Yes, that were we would like add our logging file with log rotation. | 19:02 |
| odyssey4me | lemme pastebin it | 19:02 |
| odyssey4me | lbragstad jorge_munoz like this: http://pastebin.com/WyKm2e4s | 19:03 |
| jorge_munoz | odyssey4me: awesome, thanks! | 19:03 |
| lbragstad | odyssey4me you're a lifesaver | 19:04 |
| odyssey4me | it takes some getting used to, but once you figure it out it's really easy to use | 19:04 |
| lbragstad | odyssey4me ++ | 19:04 |
| lbragstad | thanks for the help | 19:04 |
| odyssey4me | anything set in the template can be completely overwritten using that method | 19:04 |
| odyssey4me | and you are able to use jinja2 syntax in your var definition too, so your values don't have to be static | 19:05 |
| *** raddaoui has joined #openstack-ansible | 19:08 | |
| lbragstad | odyssey4me perfect | 19:10 |
| odyssey4me | lbragstad jorge_munoz whoops, I missed a ':' after DEFAULT... it's should be 'DEFAULT:' | 19:15 |
| *** Mudpuppy has quit IRC | 19:16 | |
| lbragstad | odyssey4me so http://cdn.pasteraw.com/20rslg3g5gcqu4w6760wm6hchqe8tis | 19:16 |
| odyssey4me | lbragstad that's it! | 19:16 |
| *** Mudpuppy has joined #openstack-ansible | 19:16 | |
| lbragstad | odyssey4me I owe you yet another beer in austin | 19:16 |
| odyssey4me | foo.log is the perfect log file name, of course :) | 19:16 |
| lbragstad | absolutely | 19:17 |
| odyssey4me | haha, I shall be there on the Sunday - we should beer ahead of the summit | 19:17 |
| lbragstad | odyssey4me agreed | 19:17 |
| *** Mudpuppy has quit IRC | 19:19 | |
| *** Mudpuppy has joined #openstack-ansible | 19:20 | |
| *** Mudpuppy_ has joined #openstack-ansible | 19:20 | |
| *** Mudpuppy has quit IRC | 19:25 | |
| *** karimb has quit IRC | 19:26 | |
| *** kro_sniffles is now known as kro_mnckn_doc | 19:27 | |
| *** raddaoui has quit IRC | 19:28 | |
| *** Mudpuppy has joined #openstack-ansible | 19:31 | |
| odyssey4me | automagically I didn't quite mean that. I meant to just check in and see how things are going. :) It seems that things have settled a bit more. The tests are a bit rough and there's a ton of duplication, but we can improve in the Newton cycle. | 19:33 |
| automagically | odyssey4me: Fair enough | 19:33 |
| *** Mudpuppy_ has quit IRC | 19:34 | |
| automagically | Agreed that we still have a ways to go with the tests, but significant progress has been made | 19:34 |
| odyssey4me | yup, we need to start working on bumping SHA's to at at least mitaka-3 for now | 19:35 |
| odyssey4me | rc1 is due in the next two days | 19:35 |
| odyssey4me | so if anyone's in a position to try it out in specific roles, that's the next step | 19:36 |
| odyssey4me | we should probably start with keystone | 19:36 |
| admin0 | what step did I miss to get this errorr: fatal: [c11_glance_container-0f51e04c] => error while evaluating conditional: (not is_metal | bool) and (glance_default_store == "file") and (glance_nfs_client is not defined) | 19:39 |
| automagically | odyssey4me: Are you thinking bump them first in the IRR tests, then if that looks good, move to the main repo? | 19:40 |
| odyssey4me | admin0 did you set a glance_default_store at all in user_variables? | 19:40 |
| odyssey4me | automagically yes, exactly that | 19:40 |
| admin0 | sorry .. missed that part | 19:40 |
| odyssey4me | admin0 the default is file, which isn't great but it has a place | 19:41 |
| odyssey4me | if you didn't set it then that's exposed a bug | 19:41 |
| *** cloudtrainme has quit IRC | 19:41 | |
| admin0 | i have not set it up | 19:41 |
| admin0 | glance default store = file = goes to storage_host right ? | 19:42 |
| odyssey4me | admin0 nope, it's store on the file system of the glance_registry server, I think - not sure | 19:42 |
| odyssey4me | generally one would hope that you're storing it on an NFS shared storage system | 19:43 |
| odyssey4me | otherwise you're in for a world of pain | 19:43 |
| odyssey4me | the storage_host is there for the cinder lvm-backed storage, which one hopes you don't use for anything except ephemeral storage | 19:44 |
| odyssey4me | it's a single point of failure to use lvm-backed storage | 19:44 |
| odyssey4me | better to use NFS, iSCSI or Ceph | 19:44 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server: Disable slave repo servers while syncing https://review.openstack.org/279452 | 19:44 |
| *** raddaoui has joined #openstack-ansible | 19:45 | |
| odyssey4me | hughsaunders can you add a release note to the openstack-ansible repo for that merge ^ | 19:45 |
| admin0 | odyssey4me: i am writing for this page: https://www.openstackfaq.com/openstack-ansible-liberty-setup/ .. | 19:45 |
| hughsaunders | odyssey4me: k | 19:45 |
| odyssey4me | thanks hughsaunders | 19:45 |
| admin0 | do you recommend that i also include ceph in there and have one direct document to answer most questions here | 19:45 |
| admin0 | my goal is to write one document as example that might solve 50%+ of repeated questions here | 19:46 |
| * admin0 goes on to create a 3 node ceph setup in the mean time .. if ceph is the correct way to go, it will be the way to go | 19:50 | |
| *** galstrom_zzz is now known as galstrom | 19:54 | |
| odyssey4me | admin0 generally speaking ceph provides a good general purpose storage platform at least - and storage for nova, glance and cinder which is shared | 19:54 |
| odyssey4me | so live migrations work nicely | 19:54 |
| odyssey4me | instances are built from snapshots, so they're built much quicker | 19:54 |
| *** ig0r_ has quit IRC | 19:55 | |
| admin0 | ok | 19:55 |
| admin0 | i will make a ceph cluster ready .. may need some help to integrate them | 19:55 |
| odyssey4me | and upgrades are much easier because you can live migrate instances from one host to another - the instance pauses briefly for the memory flush to disk and restore from disk | 19:55 |
| odyssey4me | afk for a bit | 19:56 |
| admin0 | ok | 19:56 |
| SamYaple | odyssey4me: memory isnt actualyl flushed to disk with livemigrate at all | 19:59 |
| SamYaple | its sent live until the amount left can be transfered in under 500ms (by default) | 19:59 |
| SamYaple | admin0: in Mitaka (it merged yay!) well finally be able to do CoW snapshots of instances too | 20:00 |
| admin0 | :D | 20:00 |
| *** cloudtrainme has joined #openstack-ansible | 20:02 | |
| *** Mudpuppy has quit IRC | 20:04 | |
| *** Mudpuppy has joined #openstack-ansible | 20:04 | |
| *** jguy_ has quit IRC | 20:05 | |
| Nepoc | Okay so repo_clone_git.yml is driving me crazy with my git mirrors. I have everything working except for novnc. Yet the git commands run manually work fine. I get "Failed to checkout 670dbddb54264fd0082d0aca1b3acb0f1814b1d2" when ansible attempts to clone and checkout to that commit. | 20:08 |
| admin0 | is there a quick-and-dirty ceph howto somewhere | 20:08 |
| admin0 | ready for ansible to use | 20:08 |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible: Release note for Repo Sync Patch https://review.openstack.org/293691 | 20:09 |
| *** cloudtrainme has quit IRC | 20:14 | |
| *** cloudtrainme has joined #openstack-ansible | 20:16 | |
| odyssey4me | admin0 mattt is the one to ask for that info - he knows it all well | 20:17 |
| admin0 | mattt: ping ! | 20:18 |
| odyssey4me | Nepoc odd :/ | 20:18 |
| odyssey4me | SamYaple oh wow, I didn't know it was network transferred - hectic! | 20:18 |
| odyssey4me | afk again | 20:19 |
| Nepoc | odyssey4me: such is my life today | 20:20 |
| * admin0 gives Nepoc a pizza | 20:21 | |
| SamYaple | odyssey4me: its even cooler moving forward too there will be the option for forward migration which means its a hard cutover the second you say migrate, and it _pulls_ the ram. this saves lots of time and bandwidth as well as lowers the downtime for an instance, even heavily used instance | 20:21 |
| SamYaple | downside is there is if something breaks the instance goes into stopped state | 20:21 |
| Nepoc | admin0: I was getting hungry... | 20:21 |
| admin0 | that was to put a smile on your life today :D | 20:22 |
| Nepoc | admin0: haha thank you | 20:22 |
| * admin0 gives Nepoc fish and chips :D | 20:22 | |
| Nepoc | admin0: now I'm just drooling | 20:22 |
| * admin0 wants to make Nepoc more hungry .. evil laugh | 20:22 | |
| * admin0 makes a wok dish with chickens for Nepoc :D | 20:22 | |
| *** phiche has joined #openstack-ansible | 20:23 | |
| Nepoc | well I got past that stupid issue... I told it to suck it up and use the master branch. | 20:23 |
| *** skamithi14 has quit IRC | 20:23 | |
| *** skamithi13 has joined #openstack-ansible | 20:23 | |
| Nepoc | Now to a worse problem... "Build package constraints file" "TypeError: argument of type 'StrictUndefined' is not iterable" | 20:24 |
| *** skamithi13 has joined #openstack-ansible | 20:24 | |
| *** Mudpuppy has quit IRC | 20:25 | |
| *** Mudpuppy has joined #openstack-ansible | 20:26 | |
| automagically | mhayden: Seeing a bunch of failing tempest tests in gate check now around “The requested availability zone is not available” Do you think it may be related to: https://github.com/openstack/openstack-ansible-os_nova/commit/1908a3d95d846c848026363511d577eef72e7754 which merged earlier today? | 20:27 |
| automagically | Example gate check log: http://logs.openstack.org/06/293606/1/check/gate-openstack-ansible-dsvm-commit/603c28e/console.html#_2016-03-16_20_19_13_275 | 20:27 |
| *** Mudpuppy has quit IRC | 20:28 | |
| *** Mudpuppy_ has joined #openstack-ansible | 20:28 | |
| *** scarlisle has quit IRC | 20:32 | |
| *** admin0 has quit IRC | 20:32 | |
| *** johnmilton has quit IRC | 20:32 | |
| *** admin0 has joined #openstack-ansible | 20:32 | |
| odyssey4me | SamYaple that would qualify as amazeballs :) | 20:32 |
| odyssey4me | Nepoc that's weird. It seems that the ansible module that clones the repo does a clone, then checkout | 20:33 |
| odyssey4me | so I'm not sure why you're hitting that issue | 20:33 |
| odyssey4me | automagically maybe a bad default | 20:34 |
| odyssey4me | ? | 20:34 |
| Nepoc | Yeah ... oh well hopefully master is fine | 20:34 |
| *** Mudpuppy_ has quit IRC | 20:34 | |
| Nepoc | The clone/checkout works manually though | 20:34 |
| *** Mudpuppy has joined #openstack-ansible | 20:34 | |
| odyssey4me | automagically I was hesitant on that patch, because a config_override can happily provide that - it's just less friendly | 20:34 |
| odyssey4me | Nepoc an ansible bug? we seem to find a lot of them | 20:35 |
| odyssey4me | Newton will bring Ansible 2.x - I can't wait! | 20:35 |
| *** Mudpuppy has quit IRC | 20:36 | |
| odyssey4me | Nepoc very odd :/ | 20:36 |
| *** Mudpuppy has joined #openstack-ansible | 20:37 | |
| mattt | admin0: howdy | 20:41 |
| admin0 | hello :: looking for quick and dirty ceph howto to use for ansible | 20:41 |
| *** asettle has joined #openstack-ansible | 20:42 | |
| mattt | admin0: for deploying ceph, for deploying ceph with openstack-ansible? | 20:42 |
| Nepoc | odyssey4me: doesn't seem like an ansible bug, I'll have to write a one liner to test it out. | 20:42 |
| admin0 | anything, everything that you got :D | 20:42 |
| admin0 | if the later, its good as well | 20:42 |
| admin0 | direct via ansible | 20:43 |
| automagically | Later all | 20:43 |
| odyssey4me | mattt he needs to deploy the ceph cluster first, then instrument OSA | 20:43 |
| odyssey4me | o/ automagically | 20:43 |
| mattt | admin0: to deploy ceph itself i have only used https://github.com/ceph/ceph-ansible, this is what we use for rpc-openstack | 20:43 |
| mattt | night automagically | 20:43 |
| mattt | admin0: i'd suggest you have a look at rpc-openstack to see how we deploy it and integrate it with openstack-ansible (ie. leverage existing inventory etc.) | 20:45 |
| mattt | admin0: https://github.com/rcbops/rpc-openstack/blob/master/rpcd/etc/openstack_deploy/env.d/ceph.yml for example, and https://github.com/rcbops/rpc-openstack/blob/master/rpcd/etc/openstack_deploy/conf.d/ceph.yml.aio | 20:45 |
| odyssey4me | admin0 yeah, we support the use of those roles to deploy ansible... and simply instrument the access to a ceph cluster | 20:45 |
| admin0 | thanks mattt and odyssey4me .. i will play with the rpc-openstack | 20:46 |
| mattt | admin0: https://github.com/rcbops/rpc-openstack/blob/master/rpcd/etc/openstack_deploy/user_extras_variables.yml#L98-L126 may be of interest too | 20:46 |
| *** retreved has joined #openstack-ansible | 20:46 | |
| odyssey4me | mattt there must be an actual playbook somewhere? | 20:47 |
| jmccrory | Nepoc: if you still have log output from your repo_build run, were any of these tasks skipped? https://github.com/openstack/openstack-ansible-repo_build/blob/master/tasks/repo_set_facts.yml | 20:47 |
| mattt | odyssey4me: https://github.com/rcbops/rpc-openstack/blob/master/rpcd/playbooks/ceph-all.yml | 20:47 |
| admin0 | i can diasable journal also right ? .. if i am planning for say 24-36 osd per server, the journal becomes the single hot point of failure | 20:47 |
| mattt | it's pretty straightforward tho (basically just calls the ceph-ansible roles) | 20:47 |
| admin0 | and ssds are well, not good for contious writes | 20:47 |
| mattt | admin0: you don't want to disable journal, no | 20:48 |
| odyssey4me | admin0 FYI, and useful for your howto... a lot of environments may want to use an existing iSCSI storage system (IBM, NetApp, EMC, etc)... we provide for multiple back-ends and many of the benfits of using Ceph are available for such systems | 20:48 |
| Nepoc | jmccrory: not skipped but I found my problem | 20:49 |
| mattt | admin0: heading out for the night, i'll be online tomorrow and can help w/ any ceph-related questions that you have | 20:49 |
| odyssey4me | admin0 also, SSD's are often used in Ceph when you want high IO | 20:49 |
| jmccrory | ah good, what was it? | 20:49 |
| mattt | i don't know too much from an operations perspective tho, but can help w/ the ansible bits :) | 20:49 |
| admin0 | thanks mattt | 20:49 |
| odyssey4me | admin0 while the SSD's may burn out more often, Ceph takes care of moving the data to another spot where it's welcome :) | 20:49 |
| Nepoc | jmccrory: I was missing the "original" part of the remote_package_parts | 20:49 |
| *** retreved has quit IRC | 20:51 | |
| admin0 | one thought i had was to remove the journal altogehter, but configure each disk as raid0 .. and use the (super awesome hardware) raid controller’s cache feature to make up for that | 20:51 |
| odyssey4me | admin0 naughty, bad juju - dig more into Ceph and you'll see that this is not how you Cloud | 20:52 |
| *** yaya has joined #openstack-ansible | 20:52 | |
| odyssey4me | throw away all your storage knowledge and start again :p | 20:52 |
| admin0 | if say a cluster is being planned for say 10000-20000 vms with nova, glance and cinder all using ceph, i easily see the journal as a single point of failure | 20:53 |
| odyssey4me | admin0 I tried all sorts of stuff like that, and it gave me nothing. Ceph operates best with SSD's for journaling and pretty much raw access to the hardware. | 20:53 |
| admin0 | whats the ssd burn out rate ? | 20:53 |
| mattt | admin0: it is a single point of failure as far as the host is concerned, but you should have x more copies of the data on other hosts | 20:53 |
| odyssey4me | but, of course, you can do anythiing you like :) | 20:53 |
| admin0 | odyssey4me: was just thinking of ideas | 20:54 |
| odyssey4me | Ceph operates best with more hosts. You get a higher read rate and a better spread of data protection. | 20:54 |
| odyssey4me | admin0 of course - try everything, because you must... this is how we learn :) | 20:55 |
| admin0 | 24 servers with 36 drives each , 40gb of connectivity with each other | 20:55 |
| admin0 | but i know that if 1 drive dies, there is a sudden spike of replication traffic .. | 20:55 |
| odyssey4me | admin0 yes, backfill traffic is something to monitor and manage | 20:56 |
| odyssey4me | you can control the threads for backfilling as I recall | 20:56 |
| admin0 | i will not talk more about ceph .. else matt is not leaving :D | 20:57 |
| admin0 | leave mattt :D good night | 20:57 |
| spotz | night mattt | 20:57 |
| mattt | night admin0, spotz :) | 20:58 |
| *** mattoliverau has quit IRC | 20:59 | |
| *** matt6434 has joined #openstack-ansible | 20:59 | |
| *** matt6434 is now known as mattoliverau | 21:00 | |
| *** thorst has quit IRC | 21:01 | |
| *** thorst has joined #openstack-ansible | 21:03 | |
| *** thorst has quit IRC | 21:07 | |
| *** KLevenstein has joined #openstack-ansible | 21:09 | |
| *** raddaoui has quit IRC | 21:11 | |
| *** Mudpuppy has quit IRC | 21:12 | |
| *** doublek has joined #openstack-ansible | 21:18 | |
| *** kro_mnckn_doc has quit IRC | 21:19 | |
| *** woodard has quit IRC | 21:24 | |
| *** Rodrigo_BR has quit IRC | 21:30 | |
| *** darrenc is now known as darren_afk | 21:31 | |
| *** lkoranda has quit IRC | 21:32 | |
| *** scarlisle has joined #openstack-ansible | 21:33 | |
| *** yaya has quit IRC | 21:35 | |
| *** johnmilton has joined #openstack-ansible | 21:37 | |
| *** lkoranda has joined #openstack-ansible | 21:38 | |
| *** ametts has quit IRC | 21:39 | |
| *** yaya has joined #openstack-ansible | 21:40 | |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible-ironic: Change default standalone ironic driver https://review.openstack.org/293725 | 21:41 |
| *** johnmilton has quit IRC | 21:42 | |
| *** yaya has quit IRC | 21:44 | |
| *** darren_afk is now known as darrenc | 21:46 | |
| flwang | odyssey4me: hi, do you know who is working on ansible octavia? | 21:47 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 21:48 | |
| *** permalac has quit IRC | 21:50 | |
| *** automagically_ has joined #openstack-ansible | 21:50 | |
| *** raddaoui has joined #openstack-ansible | 21:51 | |
| *** johnmilton has joined #openstack-ansible | 21:54 | |
| stevelle | flwang: mhayden has been leading that work | 21:56 |
| openstackgerrit | Amy Marrich (spotz) proposed openstack/openstack-ansible: Added configuration information for using Ceph with Cinder https://review.openstack.org/290873 | 21:57 |
| odyssey4me | flwang I see that stevelle has directed you correctly | 21:57 |
| Nepoc | Making this offline installer is a nightmare | 21:58 |
| Nepoc | mainly due to lack of tools for creating pypi mirror based on the global-requirements yet following the requirements_constraints | 21:59 |
| flwang | thank you guys, seems it's still an idea, i can't see any code yet | 22:00 |
| admin0 | odyssey4me: https://github.com/rcbops/rpc-openstack/blob/master/rpcd/etc/openstack_deploy/env.d/ceph.yml . this needs to go into the global /etc/ directory right ? | 22:00 |
| admin0 | so inside same env.d as other config file for openstack | 22:00 |
| odyssey4me | Nepoc OpenStack-infra have both pypi and apt mirrors. Have you looked into their tooling? | 22:01 |
| admin0 | never mind :D | 22:01 |
| *** raddaoui has quit IRC | 22:05 | |
| Nepoc | odyssey4me: I did but unfortunately the tool for pypi doesn't seem to handle taking in both files. I've rewritten part of it so I might as well "enhance" it some more to deal with the constraints file as well. | 22:07 |
| Nepoc | odyssey4me: I had to update it previously since it didn't work with anything newer than pip 1.0.5 | 22:08 |
| *** odyssey4me has quit IRC | 22:09 | |
| *** odyssey4me has joined #openstack-ansible | 22:09 | |
| odyssey4me | Nepoc oh? weird | 22:11 |
| *** javeriak has quit IRC | 22:14 | |
| *** javeriak has joined #openstack-ansible | 22:15 | |
| *** KLevenstein has quit IRC | 22:19 | |
| Nepoc | odyssey4me: I just talked to the openstack-infra guys, they are using bandersnatch and doing a full mirror (500GB), where as the mirror I have so far is only 339MB and I have about 70% of what I need. | 22:20 |
| Nepoc | odyssey4me: I guess I'll be maintaining pypi-mirror from now on :) | 22:20 |
| *** daneyon has quit IRC | 22:25 | |
| *** galstrom is now known as galstrom_zzz | 22:29 | |
| *** busterswt has quit IRC | 22:34 | |
| *** raddaoui has joined #openstack-ansible | 22:36 | |
| *** ephoenix has joined #openstack-ansible | 22:38 | |
| ephoenix | hey everyone | 22:39 |
| *** severion has quit IRC | 22:39 | |
| *** severion has joined #openstack-ansible | 22:40 | |
| *** jthorne has quit IRC | 22:41 | |
| *** raddaoui has quit IRC | 22:42 | |
| *** marst has joined #openstack-ansible | 22:43 | |
| admin0 | bye all :D | 22:43 |
| *** severion has quit IRC | 22:44 | |
| *** scarlisle has quit IRC | 22:44 | |
| *** severion has joined #openstack-ansible | 22:44 | |
| *** admin0 has quit IRC | 22:44 | |
| *** manjeets has left #openstack-ansible | 22:48 | |
| *** automagically_ has quit IRC | 22:50 | |
| *** asettle has quit IRC | 22:54 | |
| *** phiche has quit IRC | 22:55 | |
| *** marst has quit IRC | 22:57 | |
| *** retreved has joined #openstack-ansible | 22:58 | |
| *** karlamrhein has joined #openstack-ansible | 23:02 | |
| *** automagically_ has joined #openstack-ansible | 23:02 | |
| *** weezS has quit IRC | 23:03 | |
| *** johnmilton has quit IRC | 23:03 | |
| *** automagically_ has left #openstack-ansible | 23:08 | |
| *** ephoenix has quit IRC | 23:10 | |
| *** spotz is now known as spotz_zzz | 23:10 | |
| *** krotscheck has joined #openstack-ansible | 23:11 | |
| *** karimb has joined #openstack-ansible | 23:13 | |
| *** johnmilton has joined #openstack-ansible | 23:16 | |
| *** bryan_att has joined #openstack-ansible | 23:18 | |
| *** fawadkhaliq has joined #openstack-ansible | 23:19 | |
| *** javeriak has quit IRC | 23:24 | |
| *** karlamrhein has quit IRC | 23:25 | |
| *** skamithi13 has quit IRC | 23:30 | |
| *** skamithi13 has joined #openstack-ansible | 23:31 | |
| *** skamithi13 has quit IRC | 23:32 | |
| *** skamithi13 has joined #openstack-ansible | 23:32 | |
| *** cloudtrainme has quit IRC | 23:35 | |
| *** asettle has joined #openstack-ansible | 23:37 | |
| cloudnull | alextricity25: you around still ? | 23:40 |
| *** asettle has quit IRC | 23:42 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron: Updated neutron plugin_base setup and config(s) https://review.openstack.org/293307 | 23:45 |
| *** eil397 has quit IRC | 23:49 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Let rsync max_connections be set by variables https://review.openstack.org/293543 | 23:53 |
| *** karimb has quit IRC | 23:58 | |
| *** skamithi13 has quit IRC | 23:59 | |
| *** skamithi14 has joined #openstack-ansible | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!