thetrav | ? | 00:00 |
---|---|---|
thetrav | the security hardening variable? | 00:00 |
cloudnull | assuming your deployment host is not being used for "something" else in the cluster the sec role shouldn't be applied to it. | 00:00 |
thetrav | the deployment host is not a target, it hasn't had the role applied ot it | 00:00 |
cloudnull | ok. | 00:00 |
thetrav | I was referring to it cycling through the tasks in the role for the target hosts | 00:01 |
cloudnull | ah, | 00:01 |
thetrav | deciding to skip on an individual task level | 00:01 |
thetrav | rather than deciding at a higher level | 00:01 |
cloudnull | yea. thats an ansible-ism i wish was different. | 00:01 |
cloudnull | the condition is set here . https://github.com/openstack/openstack-ansible/blob/master/playbooks/security-hardening.yml#L25-L26 | 00:02 |
cloudnull | and regardless of enabled or disabled it iterates through everything in "hosts". | 00:02 |
thetrav | still too many open files | 00:03 |
cloudnull | bah... | 00:03 |
thetrav | different exceptions though | 00:03 |
cloudnull | ooohh :) | 00:03 |
openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible: Introduce a playbook for deploying Gnocchi https://review.openstack.org/324125 | 00:04 |
thetrav | http://cdn.pasteraw.com/fwbwq4fequcafj068cstie4ekz6hoe7 | 00:04 |
*** eil397 has quit IRC | 00:04 | |
cloudnull | quick question, have you tried running the command by hand ? -- openstack-ansible setup-hosts.yml -- to see if you have the same issue without the nested ansible call? | 00:05 |
cloudnull | not that the nested call should do that, I do that all the time, just grasping at straws | 00:06 |
thetrav | oh, yes | 00:06 |
thetrav | sorry to be clear | 00:06 |
thetrav | I use ansible to prepare the deploy host and the target hosts | 00:06 |
thetrav | then I run the actual setup-hosts ansible target by hand | 00:06 |
cloudnull | can you run it with a limit to see if you have the same issue? | 00:07 |
cloudnull | something like -- openstack-ansible setup-hosts.yml --limit hosts[0] | 00:07 |
cloudnull | rather: openstack-ansible setup-hosts.yml --limit 'hosts[0]' | 00:07 |
thetrav | ? | 00:08 |
thetrav | so that only applies to a single host? | 00:09 |
cloudnull | yes, the first host. | 00:09 |
thetrav | I set forks to 1 if that helps | 00:09 |
thetrav | I'll give it a shot | 00:09 |
thetrav | worked | 00:12 |
thetrav | ok, so I could theoretically iterate through one host at a time | 00:12 |
thetrav | however I'm just going to try with security_hardening turned off | 00:13 |
thetrav | cause if that fixes it then I've got a reasonable point to raise a bug from I think | 00:13 |
*** wadeholler has quit IRC | 00:14 | |
*** sdake has quit IRC | 00:15 | |
cloudnull | ++ | 00:15 |
cloudnull | you could try with two hosts openstack-ansible setup-hosts.yml --limit 'hosts[0],hosts[1]' | 00:16 |
cloudnull | and see if it works w/ >1 | 00:16 |
cloudnull | but that's a strange issue for sure. | 00:16 |
cloudnull | I think its worth raising an issue over regardless. | 00:16 |
cloudnull | While I dont want to blame the sec role for the issues, it very well may be the problem | 00:17 |
cloudnull | mhayden: is out of reach this week, he's the one whom wrote most of the sec role, I believe he's at that RH summit, but w/ an issue he may be able to dig into it. | 00:18 |
cloudnull | I have to relocate home, will be back online later. | 00:18 |
cloudnull | but let me know how it goes. | 00:18 |
thetrav | kk cya, thanks a ton for your help | 00:22 |
cloudnull | im interested to know what the root cause is, so if you nail it down please let me know. | 00:22 |
cloudnull | regardless TTFN | 00:22 |
thetrav | yep, so early results are showing I've got approx 800 less files open without the security_hardening file in the mix | 00:27 |
thetrav | I expect this will work out | 00:27 |
*** ManojK has quit IRC | 00:30 | |
*** appprod0 has quit IRC | 00:32 | |
thetrav | ok, so it's not just security | 00:41 |
thetrav | security is the most noticable one, however the open files is steadily creeping up over the whole run | 00:41 |
thetrav | it's at about 800 now at the galera stage | 00:41 |
thetrav | indicative of a leak | 00:41 |
stevelle | I'm wondering if it's something like registered vars are causing the leak, given how bad the security role for you | 00:43 |
thetrav | Could be | 00:43 |
stevelle | that would seem really odd though | 00:43 |
thetrav | I'm once again struck by the fact that most of the handles are to /dev/null | 00:44 |
thetrav | which to me implies arbitrary output stream rather than actual file for useful stuff | 00:44 |
thetrav | I wish I knew how to do proper profiling | 00:46 |
coolj_ | thetrav: how many ansible forks are you running with? | 00:50 |
thetrav | 1 | 00:50 |
thetrav | sorry, my initial attempt was the default, which I believe is 4 (one per host) | 00:50 |
thetrav | I lowered it to 1 to simplify the output | 00:50 |
thetrav | at one point I tried with 10 | 00:51 |
coolj_ | cool, figured it wasn't related to forks but just wanted to verify | 00:51 |
thetrav | made no difference | 00:51 |
thetrav | or more specifically, it still failed for the same reason | 00:51 |
thetrav | obviously there was _some_ difference | 00:51 |
*** thorst_ has joined #openstack-ansible | 00:52 | |
*** ManojK has joined #openstack-ansible | 00:53 | |
*** wadeholler has joined #openstack-ansible | 00:53 | |
coolj_ | what version of ansible? | 00:55 |
thetrav | heh | 00:55 |
thetrav | 1.9.x | 00:55 |
thetrav | the one that the boostrap script installs | 00:56 |
coolj_ | i was thinking maybe this was the fix but i guess you already have it in that case | 00:56 |
coolj_ | https://github.com/ansible/ansible/commit/402a6d0533058a18947b5c5ee2f0f4b35e620175 | 00:56 |
*** thorst_ has quit IRC | 00:57 | |
*** thorst_ has joined #openstack-ansible | 00:58 | |
*** klamath has quit IRC | 01:01 | |
*** klamath has joined #openstack-ansible | 01:02 | |
*** ManojK has quit IRC | 01:03 | |
*** aernhart has quit IRC | 01:06 | |
*** ManojK has joined #openstack-ansible | 01:06 | |
*** thorst_ has quit IRC | 01:06 | |
*** fishcried has joined #openstack-ansible | 01:08 | |
*** fishcried has quit IRC | 01:10 | |
*** ManojK has quit IRC | 01:11 | |
*** fawadkhaliq has joined #openstack-ansible | 01:20 | |
*** fawadkhaliq has quit IRC | 01:24 | |
*** klamath has quit IRC | 01:24 | |
*** thorst_ has joined #openstack-ansible | 01:39 | |
*** thorst_ has quit IRC | 01:40 | |
*** thorst_ has joined #openstack-ansible | 01:41 | |
*** thorst_ has quit IRC | 01:50 | |
*** coolj_ is now known as coolj | 01:52 | |
*** sacharya has joined #openstack-ansible | 01:59 | |
*** fawadkhaliq has joined #openstack-ansible | 02:01 | |
*** fawadkhaliq has quit IRC | 02:05 | |
*** ManojK has joined #openstack-ansible | 02:16 | |
*** thorst_ has joined #openstack-ansible | 02:19 | |
thetrav | coolj_ that fix appears to apply to ansible 2.x, not clear to me whether it's applied to 1.9.x, if it's not then that's certainly the type of issue that would cause this behaviour | 02:21 |
*** thorst_ has quit IRC | 02:21 | |
*** fawadkhaliq has joined #openstack-ansible | 02:22 | |
*** thorst_ has joined #openstack-ansible | 02:22 | |
*** sacharya_ has joined #openstack-ansible | 02:24 | |
*** fawadkhaliq has quit IRC | 02:26 | |
*** sacharya has quit IRC | 02:27 | |
*** thorst_ has quit IRC | 02:30 | |
*** sdake has joined #openstack-ansible | 02:40 | |
*** fawadkhaliq has joined #openstack-ansible | 02:43 | |
*** ManojK has quit IRC | 02:44 | |
*** fawadkhaliq has quit IRC | 02:46 | |
*** sdake has quit IRC | 02:56 | |
*** ManojK has joined #openstack-ansible | 02:58 | |
*** weezS has joined #openstack-ansible | 03:04 | |
*** sdake has joined #openstack-ansible | 03:08 | |
*** jorge_munoz has quit IRC | 03:10 | |
*** jorge_munoz has joined #openstack-ansible | 03:14 | |
*** sdake has quit IRC | 03:27 | |
*** thorst_ has joined #openstack-ansible | 03:28 | |
thetrav | ok, so I had a go using the ansible_2.1 stable branch and came up against the same problem | 03:34 |
*** thorst_ has quit IRC | 03:35 | |
*** ManojK has quit IRC | 03:35 | |
*** zerda2 has joined #openstack-ansible | 03:42 | |
*** persia has quit IRC | 03:49 | |
*** persia has joined #openstack-ansible | 03:56 | |
*** albertcard has quit IRC | 04:01 | |
*** weezS has quit IRC | 04:04 | |
*** thorst_ has joined #openstack-ansible | 04:33 | |
*** thorst_ has quit IRC | 04:41 | |
*** fawadkhaliq has joined #openstack-ansible | 04:46 | |
*** phalmos has joined #openstack-ansible | 04:49 | |
*** fawadkhaliq has quit IRC | 04:50 | |
*** fishcried has joined #openstack-ansible | 05:06 | |
*** phalmos has quit IRC | 05:12 | |
*** saneax_AFK is now known as saneax | 05:13 | |
*** M00nr41n has quit IRC | 05:13 | |
*** fishcried has quit IRC | 05:14 | |
*** eil397 has joined #openstack-ansible | 05:21 | |
*** eil397 has quit IRC | 05:28 | |
*** winggundamth has quit IRC | 05:31 | |
*** fawadkhaliq has joined #openstack-ansible | 05:34 | |
*** fawadkhaliq has quit IRC | 05:36 | |
*** fawadk has joined #openstack-ansible | 05:36 | |
*** winggundamth has joined #openstack-ansible | 05:37 | |
*** thorst_ has joined #openstack-ansible | 05:38 | |
*** admin0 has joined #openstack-ansible | 05:47 | |
*** thorst_ has quit IRC | 05:47 | |
*** admin0 has quit IRC | 05:50 | |
*** markvoelker has quit IRC | 05:58 | |
*** winggundamth has quit IRC | 06:05 | |
*** M00nr41n has joined #openstack-ansible | 06:07 | |
*** winggundamth has joined #openstack-ansible | 06:10 | |
*** javeriak has joined #openstack-ansible | 06:21 | |
*** bootsha has joined #openstack-ansible | 06:21 | |
*** jamielennox is now known as jamielennox|away | 06:44 | |
*** thorst_ has joined #openstack-ansible | 06:44 | |
*** pcaruana has joined #openstack-ansible | 06:46 | |
*** thorst_ has quit IRC | 06:51 | |
*** oneswig has joined #openstack-ansible | 06:57 | |
*** pester has quit IRC | 06:58 | |
*** markvoelker has joined #openstack-ansible | 06:58 | |
*** asettle has joined #openstack-ansible | 07:03 | |
*** fxpester has joined #openstack-ansible | 07:03 | |
*** markvoelker has quit IRC | 07:04 | |
*** admin0 has joined #openstack-ansible | 07:09 | |
*** oneswig has quit IRC | 07:14 | |
*** oneswig has joined #openstack-ansible | 07:20 | |
*** oneswig has quit IRC | 07:21 | |
*** Drago1 has joined #openstack-ansible | 07:32 | |
*** Drago1 has quit IRC | 07:33 | |
*** Drago1 has joined #openstack-ansible | 07:33 | |
openstackgerrit | Merged openstack/openstack-ansible: Flush memcached on first listen IP only https://review.openstack.org/334484 | 07:37 |
*** karimb has joined #openstack-ansible | 07:39 | |
*** thorst_ has joined #openstack-ansible | 07:48 | |
*** karimb has quit IRC | 07:55 | |
*** thorst_ has quit IRC | 07:56 | |
*** asettle has quit IRC | 07:58 | |
*** markvoelker has joined #openstack-ansible | 08:00 | |
evrardjp | good morning everyone | 08:00 |
*** sacharya_ has quit IRC | 08:01 | |
*** oneswig has joined #openstack-ansible | 08:02 | |
*** markvoelker has quit IRC | 08:04 | |
*** bsv has joined #openstack-ansible | 08:11 | |
*** karimb has joined #openstack-ansible | 08:13 | |
*** oneswig has quit IRC | 08:26 | |
*** raddaoui has quit IRC | 08:27 | |
*** jthorne_ has joined #openstack-ansible | 08:29 | |
*** jthorne has quit IRC | 08:29 | |
*** karimb has quit IRC | 08:35 | |
*** bootsha has quit IRC | 08:43 | |
*** asettle has joined #openstack-ansible | 08:45 | |
admin0 | morning all | 08:47 |
bsv | admin0: moin | 08:48 |
*** thetrav has quit IRC | 08:49 | |
*** thorst_ has joined #openstack-ansible | 08:53 | |
*** Drago1 has quit IRC | 08:54 | |
*** markvoelker has joined #openstack-ansible | 09:00 | |
*** thorst_ has quit IRC | 09:01 | |
*** markvoelker has quit IRC | 09:05 | |
*** wadeholler has quit IRC | 09:10 | |
*** bootsha has joined #openstack-ansible | 09:14 | |
*** asettle has quit IRC | 09:21 | |
*** javeriak has quit IRC | 09:27 | |
*** bootsha has quit IRC | 09:38 | |
*** asettle has joined #openstack-ansible | 09:40 | |
*** bootsha has joined #openstack-ansible | 09:49 | |
*** thorst_ has joined #openstack-ansible | 10:00 | |
*** johnmilton has quit IRC | 10:00 | |
*** markvoelker has joined #openstack-ansible | 10:01 | |
*** javeriak has joined #openstack-ansible | 10:02 | |
ioni | hello | 10:02 |
ioni | http://git.openstack.org/cgit/openstack/openstack-ansible-os_cinder/commit/?h=stable/mitaka&id=a4afeb94600e6fb1006a009a512525266a0fd4fd | 10:02 |
ioni | because of this modification, now the storage hosts are not optional anymore | 10:03 |
ioni | fatal: [ctrl1_cinder_api_container-cd77ee0a] => error while evaluating conditional: inventory_hostname in groups['cinder_volume'][0] | 10:03 |
*** markvoelker has quit IRC | 10:06 | |
*** thorst_ has quit IRC | 10:06 | |
*** weshay has quit IRC | 10:27 | |
evrardjp | not sure it should trigger a bug | 10:46 |
evrardjp | the line itself I mean | 10:46 |
evrardjp | when inventory_hostname isn't in groups['cinder_volume'][0] the include will be skipped | 10:46 |
evrardjp | the task could fail if cinder_volume group doesn't exist | 10:47 |
*** fawadk has quit IRC | 10:47 | |
evrardjp | but it should exist and be empty, right | 10:47 |
evrardjp | or non empty | 10:47 |
evrardjp | but still exist | 10:47 |
evrardjp | let me test that out | 10:49 |
*** johnmilton has joined #openstack-ansible | 10:51 | |
*** javeriak has quit IRC | 10:52 | |
evrardjp | ioni: did you change your env.d ? | 10:55 |
*** wadeholler has joined #openstack-ansible | 10:58 | |
*** wadeholler has quit IRC | 11:01 | |
ioni | evrardjp, no | 11:01 |
*** wadeholler has joined #openstack-ansible | 11:01 | |
ioni | i just reverted that commit and it worked | 11:01 |
*** markvoelker has joined #openstack-ansible | 11:02 | |
ioni | evrardjp, in inventory, I do not have such group | 11:02 |
evrardjp | that's the issue | 11:02 |
ioni | I do not have it because I do not have a storage_hosts at this time | 11:03 |
evrardjp | you don't have even an empty cinder_volune group | 11:03 |
ioni | in manually, it said that is optionally | 11:03 |
ioni | *manual | 11:03 |
evrardjp | the groups should be empty | 11:03 |
evrardjp | but exist | 11:03 |
evrardjp | let me try that quickly | 11:03 |
ioni | if you want, i can send you my config | 11:04 |
evrardjp | yup please do | 11:04 |
ioni | i do have instead have the storage-infra_hosts | 11:05 |
evrardjp | I just tried on my machine | 11:06 |
ioni | evrardjp, https://paste.xinu.at/z6j/ | 11:06 |
evrardjp | I have cinder_volume group empty when I don'thave the storage hosts | 11:06 |
evrardjp | you have another issue | 11:06 |
*** weshay has joined #openstack-ansible | 11:06 | |
evrardjp | oh my bahd I'm not on the latest branch | 11:06 |
evrardjp | let me try another time | 11:07 |
*** markvoelker has quit IRC | 11:07 | |
*** smatzek has joined #openstack-ansible | 11:07 | |
ioni | evrardjp, i'm on mitaka | 11:07 |
ioni | i just updated to 13.1.4 | 11:07 |
ioni | it used to work flawless with 13.1.3 | 11:07 |
evrardjp | I retried it worked for me | 11:08 |
evrardjp | oh I know why | 11:08 |
evrardjp | let me retry one last time | 11:08 |
evrardjp | oh no, the commit I thought could cause this hasn't merged | 11:10 |
evrardjp | oh you're on mitaka anyway? then it's 100% sure it works fine | 11:10 |
ioni | mitaka | 11:10 |
ioni | latest commit is the update sha for 13.1.4 | 11:10 |
evrardjp | I'm one commit further | 11:12 |
evrardjp | but it doesn't change a thing | 11:12 |
evrardjp | this works for me | 11:12 |
evrardjp | could you drop you openstack_inventory.json somewhere ? | 11:12 |
ioni | sure | 11:12 |
evrardjp | check if there is no sensitive data in it before :p | 11:13 |
ioni | yep | 11:13 |
ioni | evrardjp, https://paste.xinu.at/QFWE/ | 11:15 |
ioni | i see there is cinder_volume there | 11:16 |
evrardjp | yes so it should work | 11:16 |
evrardjp | let me retry one last time | 11:17 |
ioni | the inventory script doesn't list the cinder_volume group | 11:18 |
ioni | the entire error is https://paste.xinu.at/eIm/ | 11:19 |
evrardjp | that's what I am checking | 11:19 |
evrardjp | it doesn't indeed | 11:20 |
*** karimb has joined #openstack-ansible | 11:23 | |
*** karimb has quit IRC | 11:25 | |
*** fawadkhaliq has joined #openstack-ansible | 11:25 | |
*** karimb has joined #openstack-ansible | 11:25 | |
evrardjp | ioni: ok reproduced it | 11:26 |
ioni | nice | 11:26 |
*** thorst_ has joined #openstack-ansible | 11:26 | |
evrardjp | ioni: which playbook ? | 11:27 |
ioni | setup-openstack | 11:27 |
evrardjp | os-cinder-install ? | 11:27 |
ioni | http://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/app-minorupgrade.html | 11:27 |
ioni | # openstack-ansible setup-openstack.yml | 11:27 |
ioni | i'm not sure, let me check the backlog | 11:27 |
evrardjp | no need I'll find | 11:28 |
evrardjp | it was just out of laziness that I asked | 11:28 |
ioni | TASK: [os_cinder | Ensure cinder api is available] **************************** | 11:29 |
ioni | that was the task | 11:29 |
evrardjp | yes I saw I found the issue | 11:29 |
evrardjp | basically we have multiple ways to solve this | 11:30 |
evrardjp | 1) code 2) docs 3) change our inventory | 11:30 |
evrardjp | I'll go for the code change because it's fast | 11:30 |
evrardjp | but I think the general way would be to NOT use the setup-openstack in your case | 11:30 |
evrardjp | and skip the cinder playbook | 11:31 |
evrardjp | because you don't need it | 11:31 |
evrardjp | because you don't use cinder | 11:31 |
evrardjp | so that's the second approach | 11:31 |
evrardjp | could you create an issue ? | 11:32 |
evrardjp | The issue basically is | 11:32 |
ioni | i wonder if setup-openasck is going to fail on fresh installations also, using the optionally storage_hosts | 11:33 |
evrardjp | wait | 11:33 |
evrardjp | I have another idea | 11:33 |
evrardjp | I'll check what doc you followed | 11:34 |
evrardjp | this way I know what I should do | 11:35 |
ioni | http://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/app-minorupgrade.html | 11:35 |
ioni | i'm on minor updates | 11:35 |
evrardjp | I'll create the issue | 11:35 |
evrardjp | we have a bug triage today, if you want to speak about it | 11:36 |
ioni | alright, if i'm available | 11:38 |
ioni | evrardjp, http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-hostlist.html | 11:38 |
ioni | The default configuration includes an optional storage host. To install without storage hosts, comment out the stanza beginning with the storage_hosts: line. | 11:38 |
*** deverter_ has quit IRC | 11:44 | |
evrardjp | ioni : https://bugs.launchpad.net/openstack-ansible/+bug/1596918 | 11:44 |
openstack | Launchpad bug 1596918 in openstack-ansible "Cinder isn't facultative anymore" [Undecided,New] | 11:44 |
*** deverter has joined #openstack-ansible | 11:45 | |
ioni | thanks | 11:45 |
ioni | you understand the issue better than me :D | 11:46 |
*** sdake has joined #openstack-ansible | 11:58 | |
*** javeriak has joined #openstack-ansible | 12:00 | |
*** sdake_ has joined #openstack-ansible | 12:00 | |
evrardjp | automagically: should we backport this ? https://review.openstack.org/#/c/325380/ | 12:02 |
*** neilus has quit IRC | 12:02 | |
*** sdake has quit IRC | 12:03 | |
*** neilus has joined #openstack-ansible | 12:03 | |
*** fawadkhaliq has quit IRC | 12:04 | |
*** neilus has quit IRC | 12:05 | |
*** zerda2 has quit IRC | 12:05 | |
*** neilus has joined #openstack-ansible | 12:05 | |
*** raddaoui has joined #openstack-ansible | 12:08 | |
bkreitch | on AIO cinder-backup is not running inside a container? | 12:10 |
*** neilus1 has joined #openstack-ansible | 12:16 | |
*** neilus1 has quit IRC | 12:16 | |
*** neilus1 has joined #openstack-ansible | 12:17 | |
*** neilus has quit IRC | 12:18 | |
*** jamielennox|away is now known as jamielennox | 12:21 | |
*** neilus1 has quit IRC | 12:22 | |
*** berendt has joined #openstack-ansible | 12:23 | |
*** karimb has quit IRC | 12:24 | |
*** markvoelker has joined #openstack-ansible | 12:24 | |
*** fawadkhaliq has joined #openstack-ansible | 12:28 | |
*** fawadkhaliq has quit IRC | 12:32 | |
*** javeriak has quit IRC | 12:34 | |
*** woodard has joined #openstack-ansible | 12:35 | |
*** jthorne_ has quit IRC | 12:39 | |
*** jthorne has joined #openstack-ansible | 12:40 | |
*** kstev has joined #openstack-ansible | 12:40 | |
*** neilus has joined #openstack-ansible | 12:42 | |
*** neilus has quit IRC | 12:42 | |
*** bootsha has quit IRC | 12:43 | |
*** neilus has joined #openstack-ansible | 12:45 | |
*** ManojK has joined #openstack-ansible | 12:46 | |
*** klamath has joined #openstack-ansible | 12:54 | |
*** fawadkhaliq has joined #openstack-ansible | 12:59 | |
*** klamath_ has joined #openstack-ansible | 13:02 | |
*** klamath has quit IRC | 13:02 | |
*** klamath_ has quit IRC | 13:02 | |
*** fawadkhaliq has quit IRC | 13:03 | |
*** v1k0d3n has joined #openstack-ansible | 13:03 | |
*** klamath has joined #openstack-ansible | 13:03 | |
*** messy has joined #openstack-ansible | 13:03 | |
*** M00nr41n has quit IRC | 13:05 | |
*** karimb has joined #openstack-ansible | 13:07 | |
*** michaelgugino has joined #openstack-ansible | 13:15 | |
openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-os_swift: Implement Xenial Support https://review.openstack.org/334543 | 13:20 |
*** sdake_ has quit IRC | 13:23 | |
*** sdake has joined #openstack-ansible | 13:24 | |
*** jamielennox has quit IRC | 13:25 | |
*** saneax has quit IRC | 13:25 | |
*** jamesdenton has joined #openstack-ansible | 13:31 | |
*** kylek3h has joined #openstack-ansible | 13:35 | |
*** bsv has quit IRC | 13:40 | |
*** jayc has joined #openstack-ansible | 13:41 | |
*** ManojK has quit IRC | 13:44 | |
*** bryan_att has joined #openstack-ansible | 13:49 | |
*** KLevenstein has joined #openstack-ansible | 13:57 | |
*** ametts has joined #openstack-ansible | 13:57 | |
*** Zucan has joined #openstack-ansible | 13:57 | |
*** KLevenstein has quit IRC | 14:01 | |
*** KLevenstein has joined #openstack-ansible | 14:04 | |
*** cloader89 has joined #openstack-ansible | 14:06 | |
*** Mudpuppy has joined #openstack-ansible | 14:07 | |
*** BjoernT has joined #openstack-ansible | 14:13 | |
automagically | morning all | 14:15 |
automagically | evrardjp: I learned a new word today, āfacultativeā thanks to that bug report | 14:17 |
evrardjp | ? | 14:17 |
evrardjp | it's not a common word ? | 14:18 |
evrardjp | and which bug report you're talking about? | 14:18 |
ioni | the one opened for me | 14:18 |
evrardjp | oh | 14:19 |
evrardjp | ok | 14:19 |
automagically | Not that common in American english | 14:19 |
evrardjp | my bad | 14:19 |
ioni | facultative is used in Romania often as "facultativ" | 14:19 |
evrardjp | optional | 14:19 |
cloudnull | morning | 14:20 |
evrardjp | morning cloudnull | 14:20 |
*** ManojK has joined #openstack-ansible | 14:22 | |
automagically | o/ cloudnull | 14:22 |
cloudnull | hows it? | 14:22 |
cloudnull | anyone willing to give these a shove https://review.openstack.org/#/q/topic:missing-kernel-module :) | 14:23 |
automagically | Going to attempt a backport of the Neutron OVS work on master into mitakaā¦.should be interesting | 14:23 |
cloudnull | nice! | 14:23 |
* automagically looking | 14:23 | |
cloudnull | is there anything that folks need / want me to go beat on ? | 14:25 |
* cloudnull still catching up from being on holiday. | 14:25 | |
*** jmckind has joined #openstack-ansible | 14:25 | |
automagically | cloudnull: I was trying to decipher a bunch of commentary that you left in dynamic_inventory in _add_additional_networks function | 14:26 |
automagically | Looks like a fair amount of code in there to transition between an older configuration datamodel and the new one | 14:26 |
automagically | Any chance you could help remove the stuff that you donāt think is needed anymore? | 14:26 |
automagically | Not a pressing need, but a request for if/when you have the time | 14:27 |
cloudnull | Yea I can go and look at that again . | 14:27 |
*** M00nr41n has joined #openstack-ansible | 14:27 | |
cloudnull | I know most of that crap is a pit of dispare | 14:27 |
automagically | I like to think of it as an overgrown garden that needs lots of weeding ;) | 14:28 |
cloudnull | recursive functions that recurese into other functions while manipulating data in memory. | 14:28 |
*** spotz_zzz is now known as spotz | 14:28 | |
cloudnull | hahahaha. lots of weeding... | 14:29 |
*** KLevenstein has quit IRC | 14:32 | |
openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Added the ip_vs kernel module to all openstack hosts https://review.openstack.org/334698 | 14:32 |
asettle | automagically: now the fun begins, 3 timezones, one meeting... | 14:33 |
automagically | Yepā¦should be an interesting challenge | 14:33 |
asettle | automagically: where in the US are you? I know where everyone else is, so I can come up with a plan-ish :) | 14:33 |
automagically | Philadelphia thus EST | 14:34 |
asettle | Of courseeeeee | 14:34 |
asettle | Make that 4 timezones :P | 14:34 |
*** KLevenstein has joined #openstack-ansible | 14:35 | |
*** jamielennox has joined #openstack-ansible | 14:36 | |
mgariepy | evrardjp, the var issue i was experiencing with kilo - ansible 1.9.3 | 14:36 |
mgariepy | is doing the same with mitake 1.9.4 ;D | 14:37 |
mgariepy | cinder_ceph_client is taken from inventory/group_vars/hosts.yml for the ceph_client role. | 14:38 |
evrardjp | that's weird | 14:38 |
evrardjp | could you dump your user_variables and your inventory.json somewhere mgariepy? | 14:41 |
evrardjp | wait a sec I'll give you something to run | 14:41 |
spotz | mhayden: Gonna get the scoop at 11 on midcycles at castle if you want to join for a walk and chat | 14:41 |
*** saneax_AFK has joined #openstack-ansible | 14:42 | |
*** saneax_AFK is now known as saneax | 14:42 | |
*** ManojK has quit IRC | 14:44 | |
*** ManojK has joined #openstack-ansible | 14:44 | |
mgariepy | the var in the user_variables.yaml is correctlyu configuring cinder.conf but then the ceph_client gather the key from ceph-mons and it takes the variable in inventory/group_vars/hosts.yml | 14:46 |
evrardjp | I think in the past I had the ceph_mons in my user variables | 14:47 |
evrardjp | but I'd be happy to have the content of the variables somewhere | 14:48 |
evrardjp | and the openstack_inventory.json | 14:48 |
*** krotscheck_dcm is now known as krotscheck | 14:48 | |
evrardjp | (ofc if you could leave out the sensitive details...) | 14:48 |
evrardjp | dear cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, spotz, vdo, jmccrory, alextricity25, jasondotstar, KLevenstein, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccoll | 14:49 |
evrardjp | um, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec : | 14:49 |
asettle | um yourself evrardjp :P | 14:49 |
spotz | heheh | 14:49 |
evrardjp | the osa bug triage will start in 1h10'. Please have a look at the bug list before starting: https://etherpad.openstack.org/p/osa-bugtriage | 14:49 |
palendae | asettle, I think that line got broken on bgmccollum's name ;) | 14:49 |
evrardjp | thank you | 14:49 |
admin0 | alwaysn when i am driving :( | 14:49 |
admin0 | bad timing evrardjp :D | 14:49 |
asettle | hahaha palendae damn | 14:49 |
mgariepy | http://paste.openstack.org/show/523783 | 14:50 |
mgariepy | http://paste.openstack.org/show/523784 | 14:50 |
mgariepy | evrardjp, ^^ | 14:50 |
evrardjp | thanks mgariepy | 14:50 |
asettle | Ohhhh automagically I didn't know about the hol. Cheers. Wed sounds swell. | 14:51 |
admin0 | evrardjp: i will skip the bugs, continue the work on getting manilla in :D | 14:53 |
*** psilvad has joined #openstack-ansible | 14:53 | |
admin0 | manila sorry .. | 14:53 |
evrardjp | thanks admin0 for the work on manila ! | 14:53 |
admin0 | starting out .. | 14:54 |
admin0 | will commit code as soon as i have something workable | 14:54 |
*** weezS has joined #openstack-ansible | 14:58 | |
cloudnull | admin0: did you ever get the SSL termination bits going ? | 14:59 |
admin0 | in bits and pieces | 14:59 |
* cloudnull just popped into my head | 14:59 | |
admin0 | but still waiting for 1 full ansible way of makign it work | 14:59 |
admin0 | everything else in ansible when i run the playbooks makes me happy and say āgreatā .. except for SSL | 14:59 |
admin0 | i can beta-test your patch :) if its already in neuton cycle | 15:01 |
bkreitch | admin0: let me know if i can help with manila, meanwhile i'm trying to copy it from cinder role | 15:03 |
admin0 | bkreitch: if you have started already, can you make a checkout in review and we can all colloboarate from there | 15:04 |
admin0 | cloudnull, automagically: on the current stable branch, this is still the way to do it right ? https://gist.github.com/odyssey4me/bb79e9f2ae349413dae3 | 15:04 |
automagically | admin0: That looks correct to me | 15:05 |
admin0 | and in neuton will be what cloudnull worked on .. full SSL on all public endpoints ? | 15:06 |
alextricity25 | you know what would be cool....using LXD as a hypervisor | 15:07 |
alextricity25 | Is OSA 16.04 ready? | 15:07 |
alextricity25 | I was looking through the etherpad and it seems like most of the tasks are completed | 15:08 |
*** M00nr41n has quit IRC | 15:08 | |
bkreitch | admin0: i guess there should be openstack-ansible-os_manila project - who has the permission to create it? | 15:09 |
*** sacharya has joined #openstack-ansible | 15:09 | |
admin0 | just give it a name for now | 15:09 |
admin0 | when its ready to have that title, someone will step forward :) | 15:10 |
admin0 | right now it can be bkreitch_awesome_manila_branch as well :D | 15:10 |
*** M00nr41n has joined #openstack-ansible | 15:10 | |
evrardjp | alextricity25: 16.04 is following its course | 15:11 |
admin0 | alextricity25: let 16.04 prove itself first :D ? | 15:11 |
*** sacharya_ has joined #openstack-ansible | 15:12 | |
* admin0 needs to convert some systemd haters | 15:12 | |
palendae | alextricity25, As a hypervisor for nova? I think michaelgugino is working on that | 15:12 |
evrardjp | someone is already working on LXD for nova indeed | 15:12 |
evrardjp | can't remember who | 15:12 |
*** M00nr41n has quit IRC | 15:14 | |
*** sacharya has quit IRC | 15:14 | |
michaelgugino | yes, I'm hoping to have a patchset done this week for lxd on 14.04 and 16.04 | 15:15 |
*** javeriak has joined #openstack-ansible | 15:17 | |
*** M00nr41n has joined #openstack-ansible | 15:20 | |
*** fawadkhaliq has joined #openstack-ansible | 15:20 | |
alextricity25 | Awesome! | 15:20 |
*** Drago has joined #openstack-ansible | 15:23 | |
evrardjp | automagically: could you give a review on this one? https://review.openstack.org/#/c/334586/ | 15:23 |
evrardjp | or any other core | 15:23 |
*** Drago has quit IRC | 15:23 | |
*** Drago has joined #openstack-ansible | 15:23 | |
evrardjp | cloudnull: already did his job there :D | 15:24 |
*** M00nr41n has quit IRC | 15:24 | |
*** fawadkhaliq has quit IRC | 15:25 | |
bkreitch | admin0: branch under openstack-ansible? i'm putting it in https://github.com/bkreitch/openstack-ansible-os_manila for meanwhile | 15:28 |
*** M00nr41n has joined #openstack-ansible | 15:29 | |
*** TxGirlGeek has joined #openstack-ansible | 15:29 | |
admin0 | see ya all .. going home | 15:31 |
admin0 | bkreitch: someone else will be abel to guide you now :) | 15:31 |
admin0 | i will check back in 2 hours | 15:32 |
bkreitch | ok:) | 15:32 |
*** javeriak_ has joined #openstack-ansible | 15:33 | |
*** javeriak has quit IRC | 15:34 | |
*** admin0 has quit IRC | 15:36 | |
openstackgerrit | Merged openstack/openstack-ansible-security: Docs: Fix rendering of :orphan: https://review.openstack.org/334586 | 15:39 |
*** eil397 has joined #openstack-ansible | 15:39 | |
evrardjp | my bad there is still this one too on the same topic: https://review.openstack.org/#/c/334587 | 15:41 |
evrardjp | andymccr: automagically ^ | 15:42 |
*** TxGirlGeek has quit IRC | 15:43 | |
evrardjp | thanks! | 15:45 |
openstackgerrit | Merged openstack/openstack-ansible-security: Docs: Fix rendering of :orphan: https://review.openstack.org/334587 | 15:47 |
evrardjp | nobody commented here? https://etherpad.openstack.org/p/osa-bugtriage | 15:53 |
*** v1k0d3n has quit IRC | 15:53 | |
*** TxGirlGeek has joined #openstack-ansible | 15:54 | |
*** sdake_ has joined #openstack-ansible | 15:54 | |
eil397 | ^ very nice doc. I've not seen it before | 15:54 |
evrardjp | eil397: you mean bug triage thing? | 15:55 |
evrardjp | or the security ? | 15:55 |
eil397 | ye.s bug triage | 15:55 |
*** adrian_otto has joined #openstack-ansible | 15:55 | |
evrardjp | it's something new I'm trying to start | 15:56 |
evrardjp | ppl can just drop their comments there as preparation for the bug triage meeting | 15:56 |
evrardjp | this way we can actually talk bugs, instead of just discovering them at the moment of the meeting | 15:56 |
eil397 | yes. agree. it looks helpful for me | 15:56 |
*** TxGirlGeek has quit IRC | 15:56 | |
*** TxGirlGeek has joined #openstack-ansible | 15:57 | |
*** sdake has quit IRC | 15:57 | |
evrardjp | talking about bug triage, it's time for it ! | 16:00 |
evrardjp | cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, spotz, vdo, jmccrory, alextricity25, jasondotstar, KLevenstein, admin0, michaelgugino, ametts, v1k0d3n, severion | 16:01 |
evrardjp | bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec | 16:01 |
rromans | . | 16:02 |
d34dh0r53 | o/ | 16:02 |
palendae | o/ | 16:02 |
eil397 | o/ | 16:02 |
asettle | \o/ fun | 16:02 |
michaelgugino | here | 16:02 |
automagically | o/ | 16:02 |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: glance_api_servers must contain a valid url with protocol https://review.openstack.org/335111 | 16:02 |
evrardjp | for the bug triage, we're gonna go on this page, like last week: https://etherpad.openstack.org/p/osa-bugtriage | 16:02 |
*** deverter_ has joined #openstack-ansible | 16:03 | |
evrardjp | and I suggest we go through all of them in the order listed | 16:03 |
*** deverter_ has quit IRC | 16:03 | |
evrardjp | so first one is: | 16:03 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1583788 | 16:03 |
openstack | Launchpad bug 1583788 in openstack-ansible "Security role should use pam_faillock for V-38501 on CentOS" [Wishlist,New] - Assigned to Major Hayden (rackerhacker) | 16:03 |
*** deverter_ has joined #openstack-ansible | 16:03 | |
evrardjp | I suggest we move this to Confirmed, because it's quite straightforward | 16:04 |
evrardjp | and everything is already filled in | 16:04 |
evrardjp | everyone is alright ? | 16:04 |
*** TxGirlGeek has quit IRC | 16:04 | |
*** TxGirlGeek has joined #openstack-ansible | 16:05 | |
eil397 | agree | 16:05 |
evrardjp | we have plenty of topics today, so I'll go quickly through every topic | 16:06 |
evrardjp | don't miss yours! | 16:06 |
evrardjp | next is: | 16:06 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1594467 | 16:06 |
openstack | Launchpad bug 1594467 in openstack-ansible "galera_client fallback_keyserver never used" [Medium,New] | 16:06 |
*** deverter has quit IRC | 16:06 | |
evrardjp | I'd like to classify this as High importance, it can prevents stuff from being properly installed | 16:07 |
spotz | +1 | 16:07 |
michaelgugino | I think fallback behavior is a medium priority, but it's a simple fix either way | 16:07 |
evrardjp | True | 16:07 |
evrardjp | someone wants to take it? | 16:08 |
eil397 | I can take it | 16:08 |
evrardjp | cool thanks! | 16:08 |
d34dh0r53 | was thinking that one was low-hanging-fruit | 16:08 |
d34dh0r53 | forgot to tag it | 16:08 |
evrardjp | it's fine d34dh0r53, it allowed ppl to read the bug :D | 16:09 |
d34dh0r53 | :) | 16:09 |
evrardjp | I did it :p | 16:09 |
d34dh0r53 | cool | 16:09 |
evrardjp | ok, next would be https://bugs.launchpad.net/openstack-ansible/+bug/1590166 | 16:09 |
openstack | Launchpad bug 1590166 in openstack-ansible "Rejoining/recovery of galera cluster fails" [High,New] | 16:09 |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_nova: Nova role should not default vars outside the nova namespace https://review.openstack.org/335113 | 16:09 |
michaelgugino | not sure what we can do about that one. We're still using 10.0 mariadb | 16:12 |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_nova: DOC - Adopting the common role documentation pattern https://review.openstack.org/334084 | 16:12 |
evrardjp | I couldn't test it myself | 16:13 |
evrardjp | because we have a full bug triage, I'll leave it that way, until next bug triage then. | 16:15 |
evrardjp | let's skip the next one because automagically already worked on it | 16:16 |
automagically | works for me ;) | 16:16 |
evrardjp | so we're at https://bugs.launchpad.net/openstack-ansible/+bug/1595669 | 16:16 |
openstack | Launchpad bug 1595669 in openstack-ansible "Separate documentation for STIGs that aren't in Ansible" [Undecided,New] | 16:16 |
evrardjp | I didn't understand this one | 16:18 |
jmccrory | needs more info | 16:18 |
evrardjp | ok | 16:19 |
evrardjp | I'll leave it that way then | 16:19 |
evrardjp | maybe wishlist, maybe not... I don't know. | 16:19 |
automagically | Its wishlist. Major has been tracking a lot of his work stream that way | 16:19 |
evrardjp | automagically: yes, but usually major is at the origin of the bug | 16:20 |
automagically | Ah right, just noticed that he isnāt the one who opened it | 16:20 |
evrardjp | it's not the case here :D | 16:20 |
evrardjp | yup | 16:20 |
evrardjp | I'll leave it that way, and we'll see | 16:20 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1595631 | 16:20 |
openstack | Launchpad bug 1595631 in openstack-ansible "missing lxc_net_mtu configuration in documentation" [Undecided,New] | 16:20 |
evrardjp | Confirmed and Wishlist for me | 16:21 |
automagically | ++ | 16:21 |
evrardjp | someone wants to take it? | 16:22 |
evrardjp | it's a doc change | 16:22 |
Mudpuppy | I'll take it | 16:23 |
evrardjp | cool thanks! | 16:23 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1595290 | 16:24 |
openstack | Launchpad bug 1595290 in openstack-ansible "file limits not configured for swift nodes" [Undecided,New] | 16:24 |
evrardjp | andymccr: you're there? | 16:25 |
evrardjp | I was planning to say "whishlist", but because I didn't got the chance to confirm it, I'd be happy to know your opinion | 16:26 |
evrardjp | for the status | 16:26 |
evrardjp | Incomplete/Confirmed/Won't fix ? | 16:26 |
*** asettle has quit IRC | 16:27 | |
michaelgugino | wish list | 16:28 |
evrardjp | Confrimed michaelgugino? | 16:28 |
andymccr | we use the the config template work for swift, so you should be able to set those if you want/need? | 16:29 |
michaelgugino | I haven't confirmed it | 16:29 |
evrardjp | andymccr: I don't know that's why I am asking | 16:29 |
andymccr | yeah im not too clear what the bug is specifically referring to. | 16:29 |
evrardjp | it changes a lot from Invalid to confirmed :D | 16:30 |
eil397 | evrardjp: I can take to check. | 16:30 |
evrardjp | ok so incomplete | 16:30 |
andymccr | if its the swift options - they are already exposed via config templates | 16:30 |
andymccr | so nothing to do. | 16:30 |
evrardjp | ok | 16:30 |
andymccr | just my opinion :) | 16:30 |
michaelgugino | maybe docs need updating? | 16:30 |
andymccr | michaelgugino: as far as im aware, the config template docs are reasonably solid, but open to adding anything that is missing or if somebody has seen gaps that need filling. | 16:31 |
andymccr | i think lets clarify the bug and what is specifically being requested then we can make decisions on that | 16:31 |
automagically | The role docs could benefit from some additions on how config_template is used within each role | 16:31 |
andymccr | happy for that to go in. or at least a link to each conf file that config template can be used with and a link to how to use it. | 16:32 |
evrardjp | I agree with both of you | 16:32 |
evrardjp | the bug has been updated | 16:32 |
evrardjp | next one is | 16:33 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1595143 | 16:33 |
openstack | Launchpad bug 1595143 in openstack-ansible "During the upgrade M->N mysql doesn't start properly" [Undecided,New] | 16:33 |
evrardjp | I didn't have resources to test this one | 16:34 |
evrardjp | MySQL restart is part of the upgrade process, so I don't think the comment is really relevant here | 16:35 |
evrardjp | IIRC | 16:35 |
evrardjp | has anyone an opinion on this one? | 16:35 |
michaelgugino | I think I agree with the commenter. Why would mysql be restarted from M->N? We're not upgrading any galera versions | 16:36 |
evrardjp | michaelgugino: isn't the -e 'galera_upgrade=true' part of the standard upgrade procedure ? | 16:37 |
evrardjp | on the setup infra ? | 16:37 |
evrardjp | it is | 16:37 |
evrardjp | http://docs.openstack.org/developer/openstack-ansible/upgrade-guide/manual-upgrade.html#upgrade-infrastructure | 16:37 |
michaelgugino | I think that it should probably be removed. We've been on 10.0 for some time now. | 16:37 |
michaelgugino | Mitaka definitely uses 10.0 | 16:38 |
jmccrory | does galera_upgrade=true notify a restart? | 16:38 |
michaelgugino | so, there's not a galera upgrade for M->N | 16:38 |
*** appprod0 has joined #openstack-ansible | 16:38 | |
evrardjp | jmccrory: checking | 16:39 |
michaelgugino | galera_upgrade modifies a large chunk of the logic in the galera_server role | 16:40 |
*** neilus has quit IRC | 16:40 | |
michaelgugino | I believe we may have removed that from the role testing this cycle as well, I know we did for Xenial at very least | 16:40 |
kysse | ( Ķ”Ā° ĶŹ Ķ”Ā°) ubuntu pls | 16:40 |
michaelgugino | https://github.com/openstack/openstack-ansible-galera_server/commit/7bec84f67aa22a2f546bfbfa8be2d81d48ab4184 | 16:41 |
evrardjp | The triage of this one will not be as easy as the others then | 16:41 |
michaelgugino | so, I think the docs should be updated for newton to remove the galera_upgrade bits | 16:41 |
evrardjp | ok | 16:41 |
michaelgugino | that is not need for this release | 16:42 |
evrardjp | I'll mention that in the bug then | 16:42 |
jmccrory | only restarts are on maria package install or config changes | 16:42 |
*** cloader89 has quit IRC | 16:42 | |
evrardjp | and classify as confirmed -- it's a doc bug | 16:42 |
jmccrory | and there's a config change between m>n | 16:42 |
evrardjp | oh | 16:42 |
michaelgugino | config change should trigger the handler without the galera_upgrade | 16:43 |
*** ManojK has quit IRC | 16:43 | |
*** permalac has joined #openstack-ansible | 16:43 | |
*** bryan_att has quit IRC | 16:43 | |
jmccrory | galera_upgrade is a red herring | 16:43 |
evrardjp | let's forget about galera_upgrade | 16:43 |
evrardjp | galera_upgrade_post.yml is the file holding the task | 16:43 |
evrardjp | the first task starts the mysql | 16:44 |
*** asettle has joined #openstack-ansible | 16:44 | |
evrardjp | we have two issues here: a mysql restart issue after a standard upgrade -- and a doc bug | 16:44 |
evrardjp | or am I wrong? | 16:44 |
michaelgugino | +1 | 16:44 |
michaelgugino | the galera_server role is not particularly idempotent | 16:45 |
*** ManojK has joined #openstack-ansible | 16:45 | |
*** permalac_ has quit IRC | 16:45 | |
evrardjp | ok let's go to the next one: | 16:45 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1594584 | 16:45 |
openstack | Launchpad bug 1594584 in openstack-ansible "Kilo nova extra migrations may not have successfully completed" [Undecided,New] | 16:45 |
*** admin0 has joined #openstack-ansible | 16:46 | |
*** prometheanfire has quit IRC | 16:47 | |
jmccrory | ah forgot to add more details here, i'll do that today | 16:47 |
evrardjp | ok | 16:48 |
evrardjp | I'll bump then | 16:48 |
evrardjp | next is: | 16:49 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1593247 | 16:49 |
openstack | Launchpad bug 1593247 in openstack-ansible "IPv6 breaks HA router because of keepalived version" [Undecided,New] | 16:49 |
*** admin0 has quit IRC | 16:49 | |
evrardjp | did someone test this? | 16:49 |
evrardjp | I think it's legit | 16:49 |
evrardjp | I just couldn't test it | 16:49 |
evrardjp | I remember in the past using keepalived ppa for this | 16:49 |
*** pcaruana has quit IRC | 16:49 | |
evrardjp | for ipv6 keepalived support I mean | 16:50 |
*** karimb has quit IRC | 16:50 | |
evrardjp | So if someone agrees, I'd say confirmed medium | 16:51 |
evrardjp | or triaged/medium | 16:51 |
michaelgugino | idk. Is there a stable source of a package we can use in it's place? | 16:51 |
michaelgugino | ie, UCA or backports? Is it broken in Xenial? CentOS? | 16:51 |
eil397 | am I right. that is ipv6 network is provider network. routing should not be affected by keepaliaved | 16:51 |
michaelgugino | is IPv6 a supported use case of OSA at this time? | 16:51 |
*** Drago has quit IRC | 16:52 | |
eil397 | s/is/if/g | 16:52 |
evrardjp | michaelgugino: why not? | 16:52 |
evrardjp | eil397: it's only for neutron ha routers | 16:53 |
evrardjp | it makes sense to me | 16:53 |
*** TxGirlGeek has quit IRC | 16:53 | |
evrardjp | michaelgugino: there is a good ppa | 16:54 |
*** TxGirlGeek has joined #openstack-ansible | 16:54 | |
eil397 | evrardjp: sorry. missed that it is abou l3 router | 16:54 |
evrardjp | didn't check in UCA, but that's implementation details | 16:54 |
evrardjp | I mean if we agree that there is work to be done there, let's tag this as confirmed and (whishlist or higher) | 16:55 |
evrardjp | I'll take it and not refresh its state | 16:56 |
evrardjp | Last one for today | 16:57 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1579919 | 16:57 |
openstack | Launchpad bug 1579919 in openstack-ansible "msgpack pip issue after Juno to Kilo upgrade" [Undecided,New] - Assigned to Andrew Meserole (andrew-meserole) | 16:57 |
evrardjp | jmccrory: what's your opinion? | 16:57 |
evrardjp | confirmed medium? | 16:57 |
evrardjp | you want to discuss a little more about it ? | 16:57 |
jmccrory | sounds right, i think repo servers are getting destroyed/rebuilt during that upgrade now | 16:58 |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_cinder: Simplify cinder.conf logic around cinder_glance_api_servers var https://review.openstack.org/335130 | 16:58 |
evrardjp | not all of them jmccrory | 16:58 |
jmccrory | https://review.openstack.org/#/c/328485/ | 16:59 |
evrardjp | was it kilo to liberty or juno to kilo ? | 16:59 |
*** prometheanfire has joined #openstack-ansible | 16:59 | |
evrardjp | should this be backported to kilo while kilo is still not EOL? | 16:59 |
jmccrory | oh right, that's kilo | 17:00 |
evrardjp | yeah, it's another world | 17:00 |
evrardjp | juno! | 17:00 |
evrardjp | to kilo! | 17:00 |
evrardjp | anyway | 17:01 |
evrardjp | opinion? | 17:01 |
evrardjp | Won't fix? | 17:01 |
evrardjp | Confirmed? | 17:01 |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Define galera_address in the all group_vars https://review.openstack.org/335131 | 17:01 |
evrardjp | I like this ^ automagically | 17:02 |
jmccrory | evrardjp: this might need to go back https://review.openstack.org/#/c/302419/ | 17:02 |
automagically | more to come in that vein evrardjp | 17:02 |
evrardjp | automagically: I'm thinking of writing an os-generic-install.yml for a while, didn't got the chance yet | 17:02 |
evrardjp | it's beer 4 weeks that I say that | 17:03 |
evrardjp | jmccrory: what's your opinion then? | 17:03 |
evrardjp | too hard to fix ? :D | 17:03 |
jmccrory | confirmed, i'll do some more testing today | 17:03 |
evrardjp | I mean it's juno kilo we are talking about | 17:03 |
evrardjp | It's better to not leave the wounded behind, I agree | 17:03 |
jmccrory | think it's a pip lock file | 17:03 |
evrardjp | oh it's not fully a repo issue | 17:04 |
evrardjp | ok interesting | 17:04 |
jmccrory | yeah, think upgrades are needed to to get people out of unsupported versions | 17:04 |
jmccrory | fixes to upgrades* | 17:04 |
evrardjp | I'll leave it the bug as is, and give feedback to the bug about possible solutions | 17:04 |
palendae | I'd tend to agree with that | 17:04 |
*** asettle has quit IRC | 17:04 | |
palendae | Exit strategies from EOL versions are important | 17:04 |
evrardjp | until the time we have a proper fix | 17:04 |
evrardjp | I think we all agree on this | 17:05 |
evrardjp | that's why I want to document as much as we can with the resources we have | 17:05 |
evrardjp | ok, sorry for exceeding my time for the bug triage | 17:05 |
evrardjp | you can have a normal activity, or continue reading the list if you want | 17:05 |
evrardjp | :D | 17:06 |
*** jamesdenton has quit IRC | 17:07 | |
*** fawadkhaliq has joined #openstack-ansible | 17:07 | |
*** admin0 has joined #openstack-ansible | 17:07 | |
*** weezS has quit IRC | 17:08 | |
evrardjp | thanks everyone for attending | 17:08 |
*** Zucan has quit IRC | 17:09 | |
*** elo has joined #openstack-ansible | 17:18 | |
evrardjp | could someone give me his opinion about this? | 17:20 |
evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1596918 | 17:20 |
openstack | Launchpad bug 1596918 in openstack-ansible "Cinder isn't facultative anymore" [Undecided,New] | 17:20 |
evrardjp | the second approach is quite simple | 17:20 |
evrardjp | I personally prefer the first one | 17:20 |
*** cloader89 has joined #openstack-ansible | 17:31 | |
*** adrian_otto has quit IRC | 17:38 | |
*** adrian_otto has joined #openstack-ansible | 17:40 | |
*** fawadkhaliq has quit IRC | 17:40 | |
*** adrian_otto has quit IRC | 17:43 | |
*** michaelgugino has quit IRC | 17:43 | |
*** adrian_otto has joined #openstack-ansible | 17:44 | |
*** adrian_otto has quit IRC | 17:47 | |
*** adrian_otto has joined #openstack-ansible | 17:51 | |
*** jamesdenton has joined #openstack-ansible | 17:52 | |
openstackgerrit | Marc GariƩpy proposed openstack/openstack-ansible: Fix ceph client keyrings override. https://review.openstack.org/335149 | 17:57 |
*** jorge_munoz_ has joined #openstack-ansible | 18:00 | |
*** jorge_munoz has quit IRC | 18:03 | |
*** jorge_munoz_ is now known as jorge_munoz | 18:03 | |
*** electrofelix is now known as electrofelix|afk | 18:03 | |
*** javeriak_ has quit IRC | 18:05 | |
*** albertcard has joined #openstack-ansible | 18:05 | |
*** TxGirlGeek has quit IRC | 18:06 | |
evrardjp | automagically: why do we remove this bind_prefix feature ? | 18:10 |
*** klamath has quit IRC | 18:13 | |
*** TxGirlGeek has joined #openstack-ansible | 18:13 | |
*** weezS has joined #openstack-ansible | 18:17 | |
*** pcaruana has joined #openstack-ansible | 18:25 | |
*** Drago has joined #openstack-ansible | 18:25 | |
*** Drago has quit IRC | 18:28 | |
*** Drago has joined #openstack-ansible | 18:28 | |
*** Drago has quit IRC | 18:28 | |
*** Drago has joined #openstack-ansible | 18:28 | |
*** adrian_otto1 has joined #openstack-ansible | 18:32 | |
*** adrian_otto has quit IRC | 18:35 | |
*** adrian_otto1 has quit IRC | 18:36 | |
*** adrian_otto has joined #openstack-ansible | 18:36 | |
*** metral is now known as metral_zzz | 18:37 | |
*** admin0 has quit IRC | 18:39 | |
*** ManojK has quit IRC | 18:44 | |
*** ManojK has joined #openstack-ansible | 18:45 | |
*** GMAzrael has quit IRC | 18:45 | |
*** pacoblox has joined #openstack-ansible | 18:52 | |
*** michaelgugino has joined #openstack-ansible | 19:00 | |
*** bootsha has joined #openstack-ansible | 19:01 | |
*** sdake has joined #openstack-ansible | 19:02 | |
*** sdake_ has quit IRC | 19:03 | |
*** appprod0 has quit IRC | 19:07 | |
openstackgerrit | Merged openstack/openstack-ansible-specs: Add spec for only supporting venv installs https://review.openstack.org/334649 | 19:07 |
openstackgerrit | Merged openstack/openstack-ansible: [docs] Address tox errors https://review.openstack.org/334298 | 19:07 |
*** asettle has joined #openstack-ansible | 19:11 | |
*** karimb has joined #openstack-ansible | 19:13 | |
*** admin0 has joined #openstack-ansible | 19:16 | |
*** johnmilton has quit IRC | 19:16 | |
*** sdake_ has joined #openstack-ansible | 19:16 | |
*** admin0 has quit IRC | 19:17 | |
*** pacoblox has quit IRC | 19:18 | |
*** sdake has quit IRC | 19:20 | |
*** karimb has quit IRC | 19:22 | |
*** appprod0 has joined #openstack-ansible | 19:24 | |
*** mummer has joined #openstack-ansible | 19:25 | |
*** wadeholl_ has joined #openstack-ansible | 19:25 | |
*** wadehol__ has joined #openstack-ansible | 19:26 | |
*** wadeholler has quit IRC | 19:29 | |
*** wadeholl_ has quit IRC | 19:30 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: conditionally include the scsi_dh kernel module https://review.openstack.org/335202 | 19:33 |
*** metral_zzz is now known as metral | 19:36 | |
*** bootsha has quit IRC | 19:37 | |
*** jimmy has joined #openstack-ansible | 19:37 | |
*** mkrish004 has joined #openstack-ansible | 19:37 | |
*** raddaoui has quit IRC | 19:37 | |
mkrish004 | Hi Guys, I am trying out OSA 12.0.9 liberty, doest it default support IPv6 tenant network ? | 19:38 |
*** ametts has quit IRC | 19:40 | |
*** pacoblox has joined #openstack-ansible | 19:42 | |
*** jamesdenton has quit IRC | 19:43 | |
*** pacoblox has quit IRC | 19:43 | |
openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Define galera_address in the all group_vars https://review.openstack.org/335131 | 19:44 |
*** v1k0d3n has joined #openstack-ansible | 19:47 | |
cloudnull | mkrish004: you can deploy w/ ipv6 | 19:48 |
cloudnull | I believe mhayden has his env up with ipv6 as a default option | 19:48 |
*** jamesdenton has joined #openstack-ansible | 19:48 | |
cloudnull | I've personally not done testing with ipv6 and liberty | 19:48 |
cloudnull | but mitaka+ should support it out the gate last I recal. | 19:49 |
mkrish004 | @cloudull, how to enable in the configuration in order to bring up the env with ipv6...please | 19:52 |
*** ametts has joined #openstack-ansible | 19:56 | |
*** wadehol__ has quit IRC | 19:59 | |
*** wadeholler has joined #openstack-ansible | 19:59 | |
mkrish004 | @mhayden, can you please help me to configure the OSA for IPv6 ? | 20:01 |
*** jorge_munoz has quit IRC | 20:04 | |
*** wadeholler has quit IRC | 20:04 | |
*** jorge_munoz has joined #openstack-ansible | 20:05 | |
*** v1k0d3n has quit IRC | 20:06 | |
*** Drago has quit IRC | 20:16 | |
*** bootsha has joined #openstack-ansible | 20:17 | |
*** asettle has quit IRC | 20:20 | |
*** pacoblox has joined #openstack-ansible | 20:23 | |
cloudnull | sorry mkrish004, mhayden may be the best one to discuss the process. | 20:23 |
cloudnull | maybe prometheanfire can help ? | 20:23 |
*** metral is now known as metral_zzz | 20:25 | |
*** pacoblox has quit IRC | 20:25 | |
*** metral_zzz is now known as metral | 20:28 | |
*** metral is now known as metral_zzz | 20:30 | |
*** jorge_munoz_ has joined #openstack-ansible | 20:32 | |
*** jorge_munoz has quit IRC | 20:32 | |
*** jorge_munoz_ is now known as jorge_munoz | 20:32 | |
*** pacoblox has joined #openstack-ansible | 20:32 | |
*** pacoblox has quit IRC | 20:32 | |
*** v1k0d3n has joined #openstack-ansible | 20:33 | |
*** KLevenstein has quit IRC | 20:35 | |
*** KLevenstein has joined #openstack-ansible | 20:36 | |
michaelgugino | cloudnull: if you could look at https://review.openstack.org/#/c/334543/ I would appreciate it. Builds are failing at random points for 16.04 | 20:37 |
*** v1k0d3n has quit IRC | 20:37 | |
michaelgugino | I had it passing the check, added a release note, and now it's failing consistently. | 20:37 |
* cloudnull looking now. | 20:38 | |
*** asettle has joined #openstack-ansible | 20:42 | |
mhayden | mkrish004: i'm at a conf this week so i am a bit slow to reply | 20:43 |
*** ManojK has quit IRC | 20:43 | |
mhayden | mkrish004: you can add an IPv6 subnet to a network in neutron fairly easily -- adding it in horizon in liberty *should* work | 20:44 |
mhayden | or you can use neutronclient (cli) | 20:44 |
mhayden | but there's nothing special to do in OSA for it, that's just neutron networking | 20:44 |
*** ManojK has joined #openstack-ansible | 20:45 | |
*** aernhart has joined #openstack-ansible | 20:46 | |
*** ametts has quit IRC | 20:49 | |
*** pacoblox has joined #openstack-ansible | 20:49 | |
*** bryan_att has joined #openstack-ansible | 20:50 | |
*** smatzek has quit IRC | 20:52 | |
*** aernhart has quit IRC | 20:52 | |
chris__hultin | Can anyone lend a hand setting up an external network with Mitaka? | 20:55 |
mkrish004 | @mhayden, thanks for reply on your busy sched, so for the neutron and dhcp configuration ...? do we have any links to follow | 20:56 |
*** oneswig has joined #openstack-ansible | 20:56 | |
*** KLevenstein has quit IRC | 20:58 | |
*** pacoblox has quit IRC | 20:59 | |
prometheanfire | chris__hultin: why? | 21:00 |
prometheanfire | cloudnull: why? | 21:00 |
prometheanfire | chris__hultin: sorry, so many names starting with c | 21:00 |
chris__hultin | prometheanfire: Ah, oK. | 21:00 |
*** KLevenstein has joined #openstack-ansible | 21:00 | |
*** sdake has joined #openstack-ansible | 21:01 | |
*** weshay has quit IRC | 21:02 | |
mhayden | mkrish004: i believe there is an example in the neutron docs | 21:02 |
mhayden | you should be able to do: neutron help subnet-create | 21:02 |
mhayden | there are docs in the neutron/networking guide that explain the different options you have for handing out addresses | 21:03 |
mhayden | in my prod env, i let my router/firewall (outside of openstack) hand out addresses via SLAAC | 21:04 |
*** smatzek has joined #openstack-ansible | 21:04 | |
mhayden | so neutron can predict what the SLAAC address will be | 21:04 |
*** sdake_ has quit IRC | 21:04 | |
mhayden | there are options to run a DHCPv6 or SLAAC (radvd) server with neutron, too | 21:04 |
*** pacoblox has joined #openstack-ansible | 21:07 | |
*** woodard has quit IRC | 21:12 | |
*** bootsha has quit IRC | 21:13 | |
*** Mudpuppy has quit IRC | 21:13 | |
*** jamesdenton has quit IRC | 21:14 | |
openstackgerrit | Anton Khaldin proposed openstack/openstack-ansible-galera_client: Add failed_when to fix bug with fallback source for key. https://review.openstack.org/335233 | 21:15 |
*** deverter_ has quit IRC | 21:16 | |
*** pacoblox has quit IRC | 21:21 | |
*** weezS has quit IRC | 21:23 | |
*** v1k0d3n has joined #openstack-ansible | 21:27 | |
*** julian1_ has joined #openstack-ansible | 21:28 | |
*** KLevenstein has quit IRC | 21:31 | |
*** pacoblox has joined #openstack-ansible | 21:31 | |
*** pacoblox has quit IRC | 21:31 | |
*** Drago has joined #openstack-ansible | 21:32 | |
*** Drago has quit IRC | 21:32 | |
*** smatzek has quit IRC | 21:32 | |
*** Drago has joined #openstack-ansible | 21:33 | |
*** KLevenstein has joined #openstack-ansible | 21:33 | |
*** kstev has quit IRC | 21:34 | |
*** v1k0d3n has quit IRC | 21:34 | |
*** pacoblox has joined #openstack-ansible | 21:34 | |
*** thorst_ has quit IRC | 21:37 | |
*** julian1_ has quit IRC | 21:37 | |
*** julian1_ has joined #openstack-ansible | 21:37 | |
*** thorst_ has joined #openstack-ansible | 21:37 | |
*** sdake has quit IRC | 21:39 | |
*** thorst__ has joined #openstack-ansible | 21:40 | |
*** asettle has quit IRC | 21:40 | |
mkrish004 | @mhayden, Does the neutron agent containers need any IPv6 address? | 21:42 |
*** thorst_ has quit IRC | 21:42 | |
mkrish004 | if i am running DHCPv6 inside container ? | 21:42 |
*** thorst__ has quit IRC | 21:44 | |
*** pacoblox has quit IRC | 21:45 | |
*** ManojK has quit IRC | 21:46 | |
*** adrian_otto has quit IRC | 21:53 | |
*** pacoblox has joined #openstack-ansible | 21:54 | |
*** pacoblox has quit IRC | 21:54 | |
*** kylek3h has quit IRC | 21:54 | |
*** messy has quit IRC | 21:55 | |
*** jmckind has quit IRC | 21:57 | |
alextricity25 | Are bugs for specific roles tracked in the openstack-ansible launchpad or do they have their own respective launchpad page? | 21:59 |
*** pacoblox has joined #openstack-ansible | 22:00 | |
*** thorst_ has joined #openstack-ansible | 22:00 | |
*** ManojK has joined #openstack-ansible | 22:01 | |
*** oneswig has quit IRC | 22:04 | |
*** cloader89 has quit IRC | 22:04 | |
*** johnmilton has joined #openstack-ansible | 22:04 | |
*** ManojK has quit IRC | 22:16 | |
*** woodard has joined #openstack-ansible | 22:17 | |
*** woodard has quit IRC | 22:17 | |
*** woodard has joined #openstack-ansible | 22:18 | |
*** TxGirlGeek has quit IRC | 22:28 | |
*** KLevenstein has quit IRC | 22:29 | |
*** jayc has quit IRC | 22:29 | |
*** pacoblox has quit IRC | 22:31 | |
*** spotz is now known as spotz_zzz | 22:32 | |
*** Drago has quit IRC | 22:37 | |
*** BjoernT has quit IRC | 22:40 | |
mrda | Morning | 22:41 |
*** jorge_munoz has quit IRC | 22:41 | |
eil397 | mrda: morning : - ) | 22:47 |
*** mkrish004 has quit IRC | 22:47 | |
eil397 | alextricity25: as I know, openstack-ansible launchpad is for all ^openstack/openstack-ansible.* projects. | 22:49 |
*** TM1 has joined #openstack-ansible | 22:50 | |
alextricity25 | eil397: thanks | 22:51 |
*** weezS has joined #openstack-ansible | 22:59 | |
*** ManojK has joined #openstack-ansible | 23:01 | |
*** weezS has quit IRC | 23:06 | |
*** thorst_ has quit IRC | 23:06 | |
*** thorst has joined #openstack-ansible | 23:07 | |
mrda | o/ | 23:10 |
*** BjoernT has joined #openstack-ansible | 23:12 | |
*** thorst has quit IRC | 23:15 | |
*** BjoernT has quit IRC | 23:17 | |
*** adrian_otto has joined #openstack-ansible | 23:17 | |
*** kylek3h has joined #openstack-ansible | 23:22 | |
*** kylek3h has quit IRC | 23:23 | |
*** psilvad has quit IRC | 23:30 | |
*** psilvad has joined #openstack-ansible | 23:30 | |
*** adrian_otto1 has joined #openstack-ansible | 23:30 | |
*** psilvad has quit IRC | 23:31 | |
*** adrian_otto has quit IRC | 23:33 | |
*** thetrav has joined #openstack-ansible | 23:37 | |
thetrav | can anyone look at https://bugs.launchpad.net/openstack-ansible/+bug/1596744 and tell me if I've raised it correctly? Whether I need to do anything else? | 23:38 |
openstack | Launchpad bug 1596744 in openstack-ansible "too many open files" [Undecided,New] | 23:38 |
* cloudnull looking | 23:41 | |
*** eil397 has quit IRC | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!