| *** lbragsta_ has quit IRC | 00:28 | |
| *** thorst has joined #openstack-ansible | 00:31 | |
| *** thorst has quit IRC | 00:36 | |
| *** gouthamr has joined #openstack-ansible | 01:01 | |
| *** thorst has joined #openstack-ansible | 01:03 | |
| *** thorst has quit IRC | 01:03 | |
| *** acormier_ has joined #openstack-ansible | 01:06 | |
| *** hw_wutianwei has joined #openstack-ansible | 01:07 | |
| *** acormier has quit IRC | 01:09 | |
| *** acormier_ has quit IRC | 01:20 | |
| *** acormier has joined #openstack-ansible | 01:21 | |
| *** acormier has quit IRC | 01:25 | |
| *** hybridpollo has quit IRC | 01:49 | |
| *** hybridpollo has joined #openstack-ansible | 02:06 | |
| *** acormier has joined #openstack-ansible | 02:22 | |
| *** acormier has quit IRC | 02:27 | |
| *** lbragsta_ has joined #openstack-ansible | 02:48 | |
| *** gouthamr has quit IRC | 02:49 | |
| *** thorst has joined #openstack-ansible | 03:04 | |
| *** vishwanathj has quit IRC | 03:08 | |
| *** vishwanathj has joined #openstack-ansible | 03:08 | |
| *** thorst has quit IRC | 03:10 | |
| *** gkadam has joined #openstack-ansible | 03:30 | |
| *** udesale has joined #openstack-ansible | 03:42 | |
| *** japestinho has joined #openstack-ansible | 04:01 | |
| *** lbragsta_ has quit IRC | 04:02 | |
| *** lbragsta_ has joined #openstack-ansible | 04:04 | |
| *** chhavi has joined #openstack-ansible | 04:07 | |
| *** lbragsta_ has quit IRC | 04:35 | |
| *** hybridpollo has quit IRC | 04:41 | |
| *** Dinesh_Bhor has joined #openstack-ansible | 04:56 | |
| *** thorst has joined #openstack-ansible | 05:05 | |
| *** thorst has quit IRC | 05:10 | |
| *** weezS has joined #openstack-ansible | 05:24 | |
| *** weezS has quit IRC | 05:31 | |
| *** Oku_OS-away is now known as Oku_OS | 06:07 | |
| *** udesale__ has joined #openstack-ansible | 06:07 | |
| *** jbadiapa has joined #openstack-ansible | 06:08 | |
| *** udesale has quit IRC | 06:10 | |
| *** japestinho has quit IRC | 06:10 | |
| *** kong has quit IRC | 06:17 | |
| *** thorst has joined #openstack-ansible | 06:27 | |
| *** thorst has quit IRC | 06:31 | |
| *** cshen has joined #openstack-ansible | 06:44 | |
| *** sxc731 has joined #openstack-ansible | 06:52 | |
| *** phalmos has quit IRC | 07:01 | |
| *** udesale has joined #openstack-ansible | 07:02 | |
| *** udesale__ has quit IRC | 07:04 | |
| *** dixiaoli has joined #openstack-ansible | 07:04 | |
| *** yolanda_ has joined #openstack-ansible | 07:05 | |
| *** dixiaoli has left #openstack-ansible | 07:06 | |
| *** yolanda_ has quit IRC | 07:06 | |
| *** 7ITABD5MB has joined #openstack-ansible | 07:06 | |
| *** 07IAALFJ9 has joined #openstack-ansible | 07:06 | |
| *** 07IAALFJ9 has quit IRC | 07:07 | |
| *** mbuil has joined #openstack-ansible | 07:08 | |
| *** pbandark has joined #openstack-ansible | 07:08 | |
| *** 7ITABD5MB has quit IRC | 07:08 | |
| *** yolanda_ has joined #openstack-ansible | 07:08 | |
| *** yolanda_ is now known as yolanda | 07:11 | |
| *** drifterza has joined #openstack-ansible | 07:15 | |
| *** japestinho has joined #openstack-ansible | 07:17 | |
| *** charcol has quit IRC | 07:25 | |
| hwoarang | good morning | 07:33 |
|---|---|---|
| *** sxc731 has quit IRC | 07:46 | |
| *** sxc731 has joined #openstack-ansible | 07:47 | |
| *** sxc731 has left #openstack-ansible | 08:03 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_cinder master: Implement uWSGI for cinder-api https://review.openstack.org/486149 | 08:06 |
| *** markus_z has joined #openstack-ansible | 08:06 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.openstack.org/486156 | 08:07 |
| *** markus_z has quit IRC | 08:16 | |
| yifei | Hi all. I have deployed openstack cluster with OSA. But I cannot find sfc related APIs in neutron cli or openstack cli. Is there anyone can help me about this? | 08:18 |
| *** thorst has joined #openstack-ansible | 08:27 | |
| *** thorst has quit IRC | 08:32 | |
| *** markus_z has joined #openstack-ansible | 08:42 | |
| *** mardim_ has quit IRC | 08:45 | |
| *** mardim has joined #openstack-ansible | 08:45 | |
| *** electrofelix has joined #openstack-ansible | 09:24 | |
| *** stuartgr has joined #openstack-ansible | 09:30 | |
| *** japestinho has quit IRC | 09:47 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.openstack.org/486156 | 09:54 |
| *** udesale__ has joined #openstack-ansible | 10:04 | |
| *** udesale has quit IRC | 10:07 | |
| hwoarang | odyssey4me: how about http://paste.opensuse.org/89053813 ? the problem with the cryptography is in the upstream wheel | 10:08 |
| *** kong has joined #openstack-ansible | 10:16 | |
| *** yifei has quit IRC | 10:16 | |
| *** thorst has joined #openstack-ansible | 10:29 | |
| *** thorst has quit IRC | 10:33 | |
| nyloc | Hi, I tried to ask this in #openstack-dns to get an answer from the designate people, but got no resons in 5 days so I guess their channel is quiet dead support wise. Maybe someone here can help me it looks like neutron is trying to use the wrong protocol (http instead of https) when I add a new floating ip that should create an entry in my DNS-Server. Here is the error from the neutron logs: http://paste.openstack.org/show/616289/ | 10:34 |
| nyloc | In my neutron.conf I have the right url in the section [designate] | 10:35 |
| nyloc | url = https://172.29.248.20:9001/v2 | 10:35 |
| yolanda | hi, any news on the error No passwd entry for user 'broot' ? | 10:37 |
| yolanda | i cannot deploy from master | 10:37 |
| *** pbandark1 has joined #openstack-ansible | 10:38 | |
| *** electrofelix has quit IRC | 10:38 | |
| *** pbandark has quit IRC | 10:39 | |
| *** pbandark1 is now known as pbandark | 10:39 | |
| *** udesale has joined #openstack-ansible | 10:42 | |
| *** electrofelix has joined #openstack-ansible | 10:43 | |
| *** udesale__ has quit IRC | 10:43 | |
| andymccr | nyloc: and if you manually change that to http it works ok? | 10:44 |
| andymccr | yolanda: just off an aio build? | 10:44 |
| nyloc | andymccr: I'm not sure what you mean, what should I manually change to http? | 10:45 |
| andymccr | well you said in the designate section in your neutron.conf its https so if you change that to http does it work? | 10:46 |
| andymccr | change/restart services | 10:46 |
| yolanda | hi andymccr , see http://paste.ubuntu.com/25136182/ | 10:46 |
| nyloc | andymccr: No It can't because designate is not reachable on the public url with http protocol | 10:48 |
| nyloc | andymccr: I'm not sure where the designateclient inside of neutron takes this url from it is not the designate endpoint from the catalog nor the url I provided in the config. | 10:49 |
| andymccr | hmm. mugsie - any thoughts? ^ | 10:49 |
| mugsie | andymccr: give me a few mins and I will dig in | 10:50 |
| andymccr | mugsie: hahah you are the best :D | 10:50 |
| asettle | mugsie: you're a sucker man :p | 10:51 |
| mugsie | asettle: I know, right ? | 10:51 |
| andymccr | yolanda: im looking into that - will try recreate and see where we get | 10:51 |
| asettle | *shakes head* | 10:51 |
| yolanda | andymccr, thx | 10:52 |
| *** smatzek has joined #openstack-ansible | 10:55 | |
| *** udesale has quit IRC | 10:55 | |
| andymccr | yolanda: that worked for me in an AIO - but the error seems to suggest a root perms issue, are you able to sudo as root? the template module is setting the perms on the file to be root - it seems odd it would've got that far though and only fail there. | 10:59 |
| yolanda | andymccr, what is weird, is that it mentions user 'broot', not user 'root' | 11:01 |
| *** drifterza has quit IRC | 11:02 | |
| andymccr | hmm | 11:03 |
| yolanda | also testing with older SHA sums works | 11:03 |
| andymccr | yolanda: that is interesting. SHA's for OSA overall? on master? or specific role SHA's? | 11:04 |
| nyloc | I tried to have a look at the code and found the two lines: | 11:05 |
| nyloc | auth = token_endpoint.Token(CONF.designate.url, context.auth_token) | 11:05 |
| nyloc | client = d_client.Client(session=_SESSION, auth=auth) | 11:05 |
| nyloc | This is where the driver gets his designate client, I have no idea but maybe someone here is seeing a problem there? | 11:05 |
| andymccr | on master we dont pin the role sha's - so changing the openstack-ansible repo sha itself won't adjust that - although if you change a specific role's sha it would - but the lxc_container_create role hasnt changed in quite a while. | 11:06 |
| *** kristian__ has joined #openstack-ansible | 11:06 | |
| andymccr | unless the issue is in the openstack-ansible repo itself | 11:06 |
| nyloc | I was thinking that somehow mixing auth_token with designate.url from CONF may be a problem | 11:06 |
| *** Dinesh_Bhor has quit IRC | 11:07 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_octavia master: Octavia will create the network + upload image upon request itself https://review.openstack.org/482664 | 11:08 |
| *** kristia__ has joined #openstack-ansible | 11:09 | |
| *** kristian__ has quit IRC | 11:09 | |
| yolanda | sha's for osa... i'm testing now with 0648818c64239b534d00db381c4609f28e40bda9 | 11:14 |
| yolanda | andymccr, there was this problem reported on OPNFV mailing list some days ago, and i was hitting that on friday as well | 11:15 |
| andymccr | yolanda: ahh ok - let me know if that sha works we can maybe figure out the issue based on the changes - i know we changed some things to accommodate py3, that means it must be some of the bootstrap scripts though. since it'll always pull the latest role sha's anyway | 11:16 |
| nyloc | andymccr: Now it gets even stranger if I change the url in the neutron.conf to https://this_will_error_out I get an error that has the https protocol: http://paste.openstack.org/show/616292/ | 11:18 |
| andymccr | nyloc: so in your conf its http://this_will_error_out ? | 11:19 |
| nyloc | andymccr: no https | 11:20 |
| nyloc | and the error keeps the https in that case | 11:20 |
| andymccr | hmm and if you set it to http in the neutron.conf it still uses https? | 11:20 |
| nyloc | if I use the real public endpoint for designate https gets replaced with http | 11:20 |
| nyloc | andymccr: No then it is http for the error url and for the real endpoint url | 11:21 |
| mugsie | nyloc: are you sure you didn't have http://<designate-url> in there at some point? | 11:21 |
| mugsie | the token_auth plugin is *really* simple | 11:22 |
| mugsie | it just stores the endpoint, and returns it as needed | 11:22 |
| nyloc | mugsie: I'm pretty sure but you wrote at some time is there a cache somewhere that might produce this? | 11:22 |
| mugsie | if a service that loaded that config var was not restarted it may be in memory | 11:23 |
| mugsie | there is no reason that the URL will go https -> http | 11:23 |
| nyloc | I just put in http://172.29.248.20:9001/v2 just to see if it magically gets changed to https, and I even restartet my whole cloud to clear caches ;) | 11:23 |
| *** acormier_ has joined #openstack-ansible | 11:24 | |
| mugsie | what happens with ^ ? | 11:24 |
| nyloc | mugsie: what do you meanwith ^ with the http it just stays at http no change there ;) | 11:26 |
| mugsie | but the IP got changed? | 11:26 |
| nyloc | yeah if I change the ip to anything invalid the ip changes and the protocol also gets https | 11:27 |
| nyloc | only for the valid ip the ip is right but the protocol is always http | 11:27 |
| nyloc | hmm, I try my internal endpoint that one is http, lets see what happens | 11:28 |
| nyloc | mugsie: If I change the IP to my internal desginate endpoint http://172.29.236.20:9001/v2 the error still states http://172.29.248.20:9001 | 11:29 |
| mugsie | oh .... | 11:30 |
| mugsie | and the url in the catalog has https ? | 11:30 |
| nyloc | yes root@os-control1-utility-container-d92e6b3d:~# openstack endpoint list | grep designate | 11:32 |
| nyloc | | 2d42f7c7ea844e31a7e05b350c853785 | RegionOne | designate | dns | True | admin | http://172.29.236.20:9001 | | 11:32 |
| nyloc | | b62edb6786424e7ba4f151aadac2aea5 | RegionOne | designate | dns | True | internal | http://172.29.236.20:9001 | | 11:32 |
| nyloc | | e427198bf85b460e86838a570464f4ab | RegionOne | designate | dns | True | public | https://172.29.248.20:9001 | | 11:32 |
| nyloc | mugsie: Anywhere else where I could check for a wrong ip/protocol? | 11:33 |
| mugsie | nyloc: when you start neutron in debug mode, can you look at the config table that is printed and see what value is printed there? | 11:34 |
| nyloc | mugsie: Ok, I check how to start in debug mode | 11:34 |
| nyloc | debug = True | 11:35 |
| nyloc | that was easy | 11:35 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-tests master: defaults: main: Always build cryptography from source https://review.openstack.org/486580 | 11:35 |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-tests master: test-vars: Always build cryptography from source https://review.openstack.org/486580 | 11:36 |
| *** acormier_ has quit IRC | 11:39 | |
| *** acormier has joined #openstack-ansible | 11:39 | |
| nyloc | mugsie: neutron doesn't print the settings in the [designate] section on server start | 11:40 |
| mugsie | of course -_- | 11:40 |
| nyloc | mugsie: So it is supposed not to pint them no error there? | 11:41 |
| *** acormier has quit IRC | 11:41 | |
| mugsie | could it be using a different service ? | 11:41 |
| mugsie | I dont really remeber neutrons arch | 11:41 |
| *** acormier has joined #openstack-ansible | 11:42 | |
| nyloc | mugsie: maybe the neutron api service | 11:42 |
| nyloc | I will have a look | 11:42 |
| mugsie | nyloc: oh - that sounds familiar | 11:43 |
| nyloc | mugsie: Oh I confused it with nova neutron doesn't seem to have other services except the agents. | 11:44 |
| mugsie | is the neutron api not hosted in apache? | 11:45 |
| nyloc | mugsie: hmm, I don't know I will check if I have a container that may have it. | 11:46 |
| *** acormier has quit IRC | 11:46 | |
| *** drifterza has joined #openstack-ansible | 11:47 | |
| nyloc | mugsie: Na it looks like the neutron-server is the right place: https://docs.openstack.org/neutron/latest/install/common/get-started-networking.html | 11:48 |
| mugsie | what other neutron services are running? | 11:49 |
| nyloc | I think just the default agentls ml2, dhcp and so on | 11:51 |
| nyloc | This is a complete startup log from neutron-sever in debug mode: http://paste.openstack.org/show/616294/ | 11:52 |
| *** thorst has joined #openstack-ansible | 11:54 | |
| *** thorst_ has joined #openstack-ansible | 11:56 | |
| *** thorst_ has quit IRC | 11:56 | |
| *** thorst_ has joined #openstack-ansible | 11:57 | |
| *** drifterza has quit IRC | 11:57 | |
| *** thorst has quit IRC | 11:58 | |
| *** goldenfri has quit IRC | 12:00 | |
| nyloc | mugsie: I just noticed that is working half way, with the internalURL set in the neutron.conf The A records are created in my DNS just the reverse records are not updated. | 12:02 |
| mugsie | Oh .... | 12:02 |
| mugsie | do a openstack zone list --all-projects | 12:03 |
| mugsie | with an admin accoutn | 12:03 |
| nyloc | 03316303-5b0b-4833-8e56-1f54cc7b05b3 80cbd392d96e4607a7dddd865ac60a0e os.fuzzy.de. PRIMARY 1500897538 ACTIVE NONE | 12:05 |
| nyloc | mugsie: Am I right that there should be a reverse zone somewhere? | 12:06 |
| mugsie | it should be created by neutron afaik | 12:06 |
| nyloc | yeah I looked in the driver.py for designate and they try to create it if it does not exists but here comes in the wrong url again.... argh back to the start | 12:07 |
| nyloc | So:https://github.com/openstack/neutron/blob/master/neutron/services/externaldns/drivers/designate/driver.py#L49 | 12:07 |
| nyloc | is working but https://github.com/openstack/neutron/blob/master/neutron/services/externaldns/drivers/designate/driver.py#L60 is not | 12:08 |
| mugsie | yeah - looks like it | 12:08 |
| nyloc | mugsie: Maybe it's just be but where does the admin client get his url set? | 12:09 |
| nyloc | the normal client does that when creating the Token | 12:09 |
| nyloc | maybe I'm missing CONF.designate.auth_type | 12:11 |
| nyloc | and the lower part is broken but nobody noticed it because they all have this setting? | 12:12 |
| mugsie | maybe - this is all new code for me | 12:13 |
| mugsie | it seems to have been added in ocata | 12:13 |
| mugsie | #openstack-neutron might have a little more detail on it | 12:13 |
| mugsie | mlavelle wrote this integration - but I cannot see him online | 12:14 |
| *** woodard has joined #openstack-ansible | 12:14 | |
| nyloc | mugsie: Thanks a lot, I try to throw in a auth_type = password and see what happens :P | 12:14 |
| mugsie | :) | 12:14 |
| *** woodard has quit IRC | 12:19 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.openstack.org/486156 | 12:19 |
| nyloc | mugsie: I got to a new error but that doesn't help much :P EmptyCatalog: The service catalog is empty. | 12:24 |
| *** askb has quit IRC | 12:24 | |
| mugsie | OK ... that sounds like the admin user in the config file is not configure right? | 12:24 |
| nyloc | I tried using auth_type = password and username admin and password myadminpassword | 12:25 |
| nyloc | but I guess that is not how the auth_type thing works :P | 12:25 |
| mugsie | try auth_type = False ? | 12:26 |
| mugsie | then set the params here: https://github.com/openstack/neutron/blob/master/neutron/services/externaldns/drivers/designate/driver.py#L55-L59 | 12:27 |
| nyloc | mugsie: yeah thats what I had before which gave me the wrong url | 12:28 |
| mugsie | with auth_type false? | 12:28 |
| nyloc | yeah with auth_type not set but I was just missing the project_name setting, now auth_type = password works but it also uses the wrong endpoint | 12:35 |
| nyloc | mugsie: You know if it is possible to overwrite the endpoint with the keyston load from config plugin? | 12:38 |
| *** gkadam has quit IRC | 12:41 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-tests master: test-vars: Always build cryptography from source https://review.openstack.org/486580 | 12:41 |
| *** yifei has joined #openstack-ansible | 12:47 | |
| *** vnogin has joined #openstack-ansible | 12:48 | |
| nyloc | mugsie: I give up for now I tried everything I could think of for now it is allow_reverse_dns_lookup = False | 12:52 |
| mhayden | happy monday | 12:52 |
| nyloc | mugsie: Thanks a lot for your help | 12:52 |
| andymccr | mornin' mhayden | 12:58 |
| *** kylek3h has joined #openstack-ansible | 13:03 | |
| *** esberglu has joined #openstack-ansible | 13:07 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: [TEST] Test neutron-server uwsgi patch /w Nova https://review.openstack.org/486604 | 13:08 |
| *** yifei has quit IRC | 13:08 | |
| *** lucasxu has joined #openstack-ansible | 13:11 | |
| *** jamesdenton has joined #openstack-ansible | 13:18 | |
| *** jamesdenton has quit IRC | 13:25 | |
| *** jamesdenton has joined #openstack-ansible | 13:25 | |
| *** goldenfr_ has joined #openstack-ansible | 13:26 | |
| *** kristia__ has quit IRC | 13:40 | |
| *** Dinesh_Bhor has joined #openstack-ansible | 13:40 | |
| *** gouthamr has joined #openstack-ansible | 13:40 | |
| *** kristian__ has joined #openstack-ansible | 13:41 | |
| *** yifei has joined #openstack-ansible | 13:43 | |
| dmsimard | I've bumped on https://github.com/pypa/setuptools/issues/1086 ( http://logs.openstack.org/61/486361/3/check/gate-ara-integration-openstack-ansible-func-ubuntu-xenial/fbe81c8/logs/ara/result/0fcfd9b9-7092-4750-99fa-16a9bde97845/ ) on and off during the weekend, anyone know why ? | 13:47 |
| dmsimard | Sometimes things passed, sometimes they didn't ... | 13:47 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.openstack.org/486156 | 13:48 |
| andymccr | dmsimard: looks like https://github.com/pypa/setuptools/issues/1086 | 13:50 |
| dmsimard | andymccr: right, that's my first link :p | 13:51 |
| andymccr | i know we had tha tissue last week but i was under the impression it had gone away because we had avoided 36.2.0 now. after the images update, although based on that log perhaps that isnt the case :( | 13:51 |
| cloudnull | mornings | 13:51 |
| dmsimard | andymccr: but was wondering about the intermittent nature of the issue | 13:51 |
| *** chyka has joined #openstack-ansible | 13:51 | |
| andymccr | dmsimard: ahh yeah I'm slow apparently :P | 13:51 |
| dmsimard | np /me still waking up too | 13:51 |
| andymccr | dmsimard: i believe it will depend on the version of the package on the instance. | 13:52 |
| andymccr | so if setuptools 36.2.0 is already there we have an issue. i thought the infra images were update to avoid that though. | 13:52 |
| dmsimard | andymccr: ah, maybe the new images hadn't yet been pushed out or something | 13:52 |
| dmsimard | btw ara on py3 fully working, needed an actual fix on your end https://review.openstack.org/#/c/485938/7/scripts/gate-check-commit.sh | 13:53 |
| dmsimard | so I'll take out the pip install hack and we should be able to merge the fix once ara 0.14 is out | 13:53 |
| andymccr | dmsimard: ahh nice! | 13:54 |
| dmsimard | (if you do it before, ara will be enabled and <0.14 doesn't work with py3 so there'll be a bunch of errors all over the place) | 13:54 |
| andymccr | dmsimard: thanks a bunch for the fix | 13:54 |
| dmsimard | hoping to ship 0.14 this week | 13:54 |
| *** chyka has quit IRC | 13:57 | |
| *** firebat has joined #openstack-ansible | 14:02 | |
| firebat | Hey guys what’s the best way to manage a custom deploy repo of openstack… just fork from the main? | 14:03 |
| *** cpuga has joined #openstack-ansible | 14:05 | |
| firebat | *openstack-ansible that is | 14:05 |
| andymccr | firebat: what are you trying to customize? | 14:06 |
| *** drifterza has joined #openstack-ansible | 14:07 | |
| *** sxc731 has joined #openstack-ansible | 14:08 | |
| firebat | I’m trying to add in FWaaS, LBaaS, and domain based seperation customizations for my companies deploy. Additionally I’m creating a set of bootstrapping tools that I’m sure the higher ups will want in the same repo even though I’d like to keep it separate. | 14:08 |
| andymccr | firebat: ideally the aim should be to have any customization in tree - so you can follow tree, you should be able to specify specifics for your company/deployment as vars etc - and bootstrapping sure that makes sense outside of tree | 14:09 |
| dmsimard | Hey, thought I'd share something I found openshift-ansible was doing that I find very clever | 14:09 |
| andymccr | forking is usually tricky, because things change and get fixed/improved etc | 14:09 |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: add deploy node into mn-aio-xenial https://review.openstack.org/485441 | 14:09 |
| firebat | andymccr: Makes sense… what should I do about /etc/openstack-deploy? | 14:10 |
| dmsimard | I was patching our deployment code for our openshift instance and ended up with this error: http://paste.openstack.org/raw/616306/ | 14:10 |
| dmsimard | It turns out they have a variety of health checks configured that they can call at any time | 14:10 |
| dmsimard | https://github.com/openshift/openshift-ansible/blob/7b0acaff56fe08e4d302a86ed4b00db6739f16a9/playbooks/common/openshift-cluster/config.yml#L11 | 14:10 |
| firebat | I’m trying to make it a simple process to bootstrap a deploy host, pull down specific configs, and then letting it deploy against its own set of hosts | 14:10 |
| dmsimard | https://github.com/openshift/openshift-ansible/blob/1da90af63656f127b21720248b5c8c25ebc728ed/roles/openshift_health_checker/action_plugins/openshift_health_check.py | 14:10 |
| dmsimard | https://github.com/openshift/openshift-ansible/blob/240b923c95e04e9467c607c502c857dd86d14179/roles/openshift_health_checker/openshift_checks/memory_availability.py | 14:10 |
| andymccr | firebat: so you mean bootstrapping the vars files in /etc/openstack_deploy to meet your deployment? | 14:11 |
| firebat | andymccr: Yep… using probably ansible vault or this custom tool we built | 14:11 |
| andymccr | firebat: that makes sense - id say anything up to (and including) the configuration of the vars and setup for the deployment makes total sense to have a repo for | 14:11 |
| firebat | andymccr: Also it might be a future goal of ours to pull down machine info from MaaS for those config files… | 14:12 |
| firebat | andymccr: Ok cool then yeah that’s what I’ll do for our vars and stuff and I’ll leave tree alone | 14:12 |
| andymccr | firebat: we setup some stuff for an AIO but its just an AIO for testing, its not really meant to be used as a genuine approach, although we do have bootstrap-ansible.sh script for configuring all the packages you will need (without configuring /etc/openstack_deploy) so it would be worth using that | 14:12 |
| *** cpuga has quit IRC | 14:13 | |
| firebat | andymccr: Do you think I should tag to match your tags so people don’t mess up pulling down old configs? Yeah I’ve been relying some on bootstrap-ansible.sh for my test deploys… so your thinking I can just pull from it for my tool for certain things? | 14:13 |
| firebat | Or should my tool call bootstrap-ansible.sh | 14:14 |
| firebat | *you're | 14:14 |
| andymccr | firebat: i'd call it directly - mostly because we update it to match the intree playbooks/etc. so for example we may use a newer version of ansible or require some new packages for some other reason, so we'll update that in tree | 14:14 |
| firebat | Exactly what I was thinking ok cool feels good to not be a complete idiot | 14:15 |
| andymccr | firebat: haha - hopefully it's not too bad, bt i think we've all been there :P | 14:15 |
| andymccr | dmsimard: that is pretty cool - i like the use of action plugins. | 14:16 |
| *** woodard has joined #openstack-ansible | 14:18 | |
| *** weezS has joined #openstack-ansible | 14:18 | |
| Tahvok | Should the horizon http -> https redirect be configured? It's not in my case. I'm using f5 script if that matters | 14:18 |
| *** acormier has joined #openstack-ansible | 14:22 | |
| cloudnull | Tahvok: are you using https://github.com/rcbops/rpc-openstack/blob/master/scripts/f5-config.py | 14:24 |
| Tahvok | cloudnull: yes | 14:24 |
| cloudnull | it doesn't look like the redirect is being created | 14:25 |
| Tahvok | I have checked the output, and there's nothing regarding horizon redirect | 14:25 |
| cloudnull | if you're wanting 80 to redirect to 443 | 14:25 |
| Tahvok | For some reason horizon does not answer on port 80 | 14:25 |
| Tahvok | So I guessed the redirect is mandatory | 14:25 |
| Tahvok | Because it's working fine on https | 14:26 |
| cloudnull | yes you'll need to create an 80 > 443 redirect on the F5 | 14:26 |
| cloudnull | looks like something we could add to that script | 14:26 |
| cloudnull | if you're so inclined I'm sure the RPC team would be happy to have that change :D | 14:26 |
| firebat | Hey I’m getting a cinder error, and can’t seem to figure out what’s wrong with the config… does anyone recognize this warning? (there are weird errors below): https://pastebin.com/PU2evDsW | 14:27 |
| Tahvok | Yeah, I though it was part of the script already but I didn't see it in the output, that's why I asked, maybe I did something wrong | 14:27 |
| cloudnull | nope nothing you did wrong | 14:28 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Use sudo with rsync for logs https://review.openstack.org/486633 | 14:34 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_nova master: [TEST] Test neutron-server uwsgi patch /w Nova https://review.openstack.org/486604 | 14:34 |
| andymccr | firebat: could it be that the backend isnt present/created - e.g. if its an lvm backend does the logical volume exist? | 14:36 |
| andymccr | firebat: in an AIO we create one as a loopback device etc, but in production you would need to create your own storage LV (since we dont know what you would be using drive wise or size wise etc). | 14:36 |
| firebat | Yeah that’s what has me confused I setup the lvm already | 14:36 |
| andymccr | ahh ok, check the names are consistent? | 14:37 |
| firebat | Yeah both cinder-volumes | 14:37 |
| firebat | Maybe I have the loopback in there and that’s the issue as it’s looping back when it doesn’t need to? | 14:37 |
| *** drifterza has quit IRC | 14:38 | |
| *** kristian__ has quit IRC | 14:39 | |
| *** drifterza has joined #openstack-ansible | 14:39 | |
| *** kristian__ has joined #openstack-ansible | 14:39 | |
| *** marst has quit IRC | 14:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server stable/ocata: Escape periods in DontCache regex https://review.openstack.org/486139 | 14:42 |
| openstackgerrit | David Moreau Simard proposed openstack/openstack-ansible master: Dynamically retrieve the location of ARA to work on both py2 & py3 https://review.openstack.org/485938 | 14:42 |
| dmsimard | logan-: ^ :) | 14:43 |
| *** kristian__ has quit IRC | 14:43 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-os_swift master: Flag service for restart if the venv changes https://review.openstack.org/485225 | 14:45 |
| *** marst has joined #openstack-ansible | 14:47 | |
| *** chyka has joined #openstack-ansible | 14:50 | |
| *** yifei has quit IRC | 14:51 | |
| *** jamesdenton has quit IRC | 14:51 | |
| *** jamesdenton has joined #openstack-ansible | 14:52 | |
| *** jamesdenton has quit IRC | 14:59 | |
| *** vnogin has quit IRC | 15:00 | |
| *** jwitko has joined #openstack-ansible | 15:00 | |
| *** kristian__ has joined #openstack-ansible | 15:01 | |
| *** kristian__ has quit IRC | 15:02 | |
| *** kristian__ has joined #openstack-ansible | 15:02 | |
| *** kristian__ has quit IRC | 15:02 | |
| *** kristian__ has joined #openstack-ansible | 15:02 | |
| *** jamesdenton has joined #openstack-ansible | 15:02 | |
| *** lucasxu has quit IRC | 15:03 | |
| *** markus_z has quit IRC | 15:07 | |
| *** kristian__ has quit IRC | 15:07 | |
| *** kristian__ has joined #openstack-ansible | 15:07 | |
| *** vnogin has joined #openstack-ansible | 15:07 | |
| *** weezS has quit IRC | 15:08 | |
| *** vnogin has quit IRC | 15:12 | |
| mhayden | dmsimard: kudos to you for tackling py2/3 issues | 15:17 |
| mhayden | them is hard | 15:17 |
| evrardjp | hello everyone | 15:17 |
| dmsimard | mhayden: it's especially not fun | 15:18 |
| mhayden | OHAI2U EVRARDJP | 15:18 |
| dmsimard | mhayden: like, I'm not a professional programmer or anything. Banging your head on encoding things is just not fun. You're not developing features or anything. Meh. | 15:18 |
| *** Oku_OS is now known as Oku_OS-away | 15:18 | |
| dmsimard | Other than encoding, utf8, unicode and all that, it was surprisingly not that hard to get py2 and py3 working simultaneously | 15:19 |
| *** vnogin has joined #openstack-ansible | 15:20 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_keystone master: defaults: main: Always build cryptography from source https://review.openstack.org/486656 | 15:23 |
| *** vnogin has quit IRC | 15:25 | |
| mbuil | hey, could somebody have a look at the patch: https://review.openstack.org/#/c/486006/? I need it to progress on the tacker installer :) | 15:25 |
| evrardjp | hwoarang: for cryptography do you think we should do this as a group var instead ? | 15:26 |
| evrardjp | group var all | 15:26 |
| evrardjp | or another mechanism to do it automatically | 15:26 |
| hwoarang | i have no preference to be honest | 15:27 |
| *** kristia__ has joined #openstack-ansible | 15:27 | |
| *** kristian__ has quit IRC | 15:28 | |
| evrardjp | It's just out of laziness. | 15:28 |
| hwoarang | are you talking about the os_keystone patchset? | 15:29 |
| hwoarang | or the one in openstack-ansible-tests | 15:29 |
| mhayden | finally got grafana graphs going again for the OSA reviews -> http://osa-grafana.mhtx.net | 15:34 |
| evrardjp | the one in keystone | 15:34 |
| evrardjp | mhayden: cool, what does the metrics gathering? | 15:34 |
| mhayden | python ;) | 15:34 |
| evrardjp | let me rephrase, you have a long running thing that queries gerrit api and publishes data to influx? | 15:38 |
| evrardjp | what would be really cool, would be to have that with LP bugs too... | 15:40 |
| evrardjp | great dashboards | 15:41 |
| *** perniciouscaffei has quit IRC | 15:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Enable some off-by-default checks https://review.openstack.org/486065 | 15:44 |
| *** kong has quit IRC | 15:45 | |
| *** lostRhino has joined #openstack-ansible | 15:46 | |
| mhayden | evrardjp: good idea wrt to LP | 15:47 |
| mhayden | i'll take a look | 15:47 |
| mhayden | the LP API was... entertaining... last i looked | 15:48 |
| *** kristia__ has quit IRC | 15:48 | |
| *** kristian__ has joined #openstack-ansible | 15:49 | |
| *** weezS has joined #openstack-ansible | 15:52 | |
| mhayden | do the gate jobs run as 'jenkins' user primarily? | 15:53 |
| mhayden | trying to solve this log file ownership mess | 15:53 |
| *** kristian__ has quit IRC | 15:54 | |
| *** jamesden_ has joined #openstack-ansible | 15:55 | |
| *** kristian__ has joined #openstack-ansible | 15:56 | |
| *** jamesdenton has quit IRC | 15:57 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Use sudo with rsync for logs https://review.openstack.org/486633 | 15:59 |
| *** lostRhino has left #openstack-ansible | 15:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: test-vars: Always build cryptography from source https://review.openstack.org/486580 | 16:01 |
| *** mbuil has quit IRC | 16:05 | |
| *** cpuga has joined #openstack-ansible | 16:11 | |
| *** pbandark1 has joined #openstack-ansible | 16:12 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible master: Fix SHAs for Pike milestone-3 https://review.openstack.org/486682 | 16:13 |
| openstackgerrit | Amy Marrich (spotz) proposed openstack/openstack-ansible master: [DOCS] Update of SSL Config to include HAProxy https://review.openstack.org/485371 | 16:13 |
| *** pbandark has quit IRC | 16:14 | |
| *** pbandark1 is now known as pbandark | 16:14 | |
| *** cpuga has quit IRC | 16:15 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible master: Fix SHAs for Pike milestone-3 https://review.openstack.org/486682 | 16:15 |
| *** drifterza has quit IRC | 16:18 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible master: Octavia scenario https://review.openstack.org/483520 | 16:21 |
| *** jamesden_ has quit IRC | 16:21 | |
| *** jamesdenton has joined #openstack-ansible | 16:23 | |
| * mhayden tips his hat to evrardjp | 16:24 | |
| evrardjp | what did I do? :D | 16:27 |
| evrardjp | oh I don't know what user it runs as. | 16:27 |
| evrardjp | maybe a devstack user knows :p | 16:27 |
| evrardjp | or any other project user :p | 16:28 |
| evrardjp | yeah bug stats are good to improve our coverage | 16:28 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_tacker master: Updated from OpenStack Ansible Tests https://review.openstack.org/486689 | 16:30 |
| *** weezS_ has joined #openstack-ansible | 16:32 | |
| *** weezS has quit IRC | 16:33 | |
| *** weezS_ is now known as weezS | 16:33 | |
| *** ricardoas has quit IRC | 16:36 | |
| *** craigs has joined #openstack-ansible | 16:37 | |
| mhayden | evrardjp: http://osa-grafana.mhtx.net | 16:41 |
| mhayden | scroll to the bottom | 16:41 |
| craigs | @cloudnull - this was the previous fix on erlang version with rabbitmq. - https://review.openstack.org/#/c/473573/ | 16:46 |
| craigs | Tested it did not work... | 16:46 |
| craigs | new bug opened - https://bugs.launchpad.net/openstack-ansible/+bug/1706122 | 16:46 |
| openstack | Launchpad bug 1706122 in openstack-ansible "openstack-ansible rabbitmq installs latest release of erlang which is not supported" [Undecided,New] | 16:46 |
| mhayden | evrardjp: nvm, broke it out into another dashboard http://osa-grafana.mhtx.net/dashboard/db/openstack-ansible-launchpad-stats?refresh=10s&orgId=1 | 16:49 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift master: [WIP] Implement uWSGI for Swift services https://review.openstack.org/486694 | 16:51 |
| cloudnull | craigs: looking | 16:54 |
| *** firebat has quit IRC | 16:54 | |
| cloudnull | craigs: https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/meta/main.yml#L44-L49 | 16:57 |
| cloudnull | it looks like the apt pins are there. | 16:57 |
| cloudnull | are those not working ? | 16:57 |
| cloudnull | its in ocata too https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/stable/ocata/meta/main.yml#L39-L44 | 16:57 |
| cloudnull | but not newton | 16:57 |
| craigs | nope - they were not pinned to a good version. Looks like team rabbit - Our point of view on 20 is still this: if you have existing installations to upgrade, wait for RabbitMQ 3.6.11 | 16:58 |
| craigs | ack - but we needed it for newton, were the problem exists. | 16:59 |
| cloudnull | in your rabbitmq role do you have the additional pins set in the meta ? | 16:59 |
| craigs | we have pinned to fix the problem. | 16:59 |
| craigs | but understandably the fix would be best choice for upgrades to go smoother. | 17:00 |
| cloudnull | my patch lacked these bits | 17:01 |
| cloudnull | https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/meta/main.yml#L44-L49 | 17:01 |
| craigs | pins do become a hassle when they start pilling up. | 17:01 |
| cloudnull | ++ | 17:01 |
| cloudnull | but if you have them in the role it should work ? | 17:01 |
| cloudnull | and that has been backported to ocat a | 17:02 |
| cloudnull | looking at the role I dont thing we backported the erlang solutions repo to newton | 17:02 |
| *** sxc731 has quit IRC | 17:05 | |
| *** weezS has quit IRC | 17:07 | |
| craigs | it would be nice to have it in newton | 17:09 |
| cloudnull | It would, however I'm not sure we can get that specific change into newton atm. it's a bit past the time in which we can change the characteristics of the role. | 17:11 |
| cloudnull | evrardjp: odyssey4me: andymccr: logan-: jmccrory: spotz: hwoarang: ^ thoughts ? | 17:11 |
| cloudnull | mhayden: ^ | 17:11 |
| *** electrofelix has quit IRC | 17:27 | |
| spotz | cloudnull: reading one sec | 17:36 |
| spotz | cloudnull: Ok here's my thoughts on newton and I said the same thing to the RDO guys. COA has just been upgraded to Newton from Liberty and without talking to Anne I would thing it's going to remain there for atleast a year. If something isn't right we should fix it as lots of folks will be playing with the AIO version at the least | 17:37 |
| *** thorst_ has quit IRC | 17:38 | |
| kysse | with proper patches newton feels stable enought | 17:39 |
| kysse | havent tried ocata but I think it also needs 6 months of testing and patching before we can use it on production :) | 17:39 |
| kysse | (speakin for myself) | 17:40 |
| craigs | :) | 17:40 |
| kysse | also there is that thing with ceilometer and billing,, no ceilo for ocata :) | 17:40 |
| kysse | atleast with mongo backend | 17:41 |
| *** thorst has joined #openstack-ansible | 17:44 | |
| *** thorst has quit IRC | 17:49 | |
| *** kristian__ has quit IRC | 17:54 | |
| *** kristian__ has joined #openstack-ansible | 17:55 | |
| *** stuartgr has quit IRC | 17:57 | |
| *** kristian__ has quit IRC | 17:59 | |
| *** gkadam has joined #openstack-ansible | 18:00 | |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible master: Octavia scenario https://review.openstack.org/483520 | 18:01 |
| *** poopcat has joined #openstack-ansible | 18:02 | |
| andymccr | cloudnull: what is the impact of not backporting that? If it is impactful we can add a re-no and version bump on release | 18:03 |
| *** guiltybeta has joined #openstack-ansible | 18:06 | |
| *** albertcard1 has joined #openstack-ansible | 18:06 | |
| *** poopcat has quit IRC | 18:07 | |
| guiltybeta | I have a need to setup cpu/disk allocation ratios different per compute host. I added them via host_vars but I also have global overrides I need for nova in user_config. If I leave those in user_config the compute hosts don't use the host_vars. Is there another way I should be doing this? | 18:07 |
| *** gtrxcb has quit IRC | 18:08 | |
| *** chhavi has quit IRC | 18:10 | |
| andymccr | guiltybeta: the issue with overrides is they take precedence, so anything in your user_variables will override everything else - you could specify it in host_vars and group_vars and remove it from user_variables. (host_vars would take precedence over group_vars) | 18:11 |
| *** thorst has joined #openstack-ansible | 18:12 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tacker master: ansible-linters fixes https://review.openstack.org/486006 | 18:12 |
| *** cpuga has joined #openstack-ansible | 18:13 | |
| guiltybeta | andymccr: Thank for the clarification I will utilize group_vars see if I can work it out that way. | 18:15 |
| andymccr | guiltybeta: no problem! it's a result of ansible's var precedence - host_vars would work if you only wanted to change the default on an individual host or few hosts. (role default is basically the lowest precedence) | 18:16 |
| craigs | cloudnull rabbitmq is statying at 1.9x until osa gets to 3.6.11+ or 3.7 | 18:16 |
| craigs | cloudnull: ^ | 18:16 |
| *** cpuga has quit IRC | 18:17 | |
| cloudnull | in newton we're using the system package for erlang | 18:17 |
| cloudnull | https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/stable/newton/vars/debian.yml#L22-L25 | 18:17 |
| cloudnull | andymccr: so the impact is that while it works its not using the recommended erlang for the release of rabbitmq | 18:18 |
| cloudnull | which has a performance impact | 18:18 |
| cloudnull | ^ craigs oversimplified, but is that right ? | 18:18 |
| craigs | should be used - Pin: version 1:19.3-1 | 18:18 |
| craigs | https://www.rabbitmq.com/install-debian.html | 18:18 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible-os_swift master: Flag service for restart if the venv changes https://review.openstack.org/485225 | 18:19 |
| cloudnull | craigs: that pin wont work in trusty by default. | 18:19 |
| cloudnull | we're not using the erlang repo | 18:19 |
| cloudnull | we're just using what ubuntu gave us , which is erlang 18.x | 18:19 |
| cloudnull | however we could backport the erlang repos and pins to newton | 18:20 |
| cloudnull | which i'd be happy to do, if folks think its the right move ? | 18:20 |
| craigs | It seems like the right move, and should not cause any other issues. | 18:21 |
| *** pbandark has quit IRC | 18:23 | |
| *** asura has joined #openstack-ansible | 18:25 | |
| spotz | cloudnull: just need to make sure it doesn't cause an issue somewhere else but I don't think it would | 18:26 |
| *** gtrxcb has joined #openstack-ansible | 18:31 | |
| *** gkadam has quit IRC | 18:38 | |
| *** sxc731 has joined #openstack-ansible | 18:41 | |
| *** albertcard1 is now known as poopcat | 18:43 | |
| cloudnull | ++ | 18:43 |
| *** ricardoas has joined #openstack-ansible | 18:46 | |
| *** mbuil has joined #openstack-ansible | 18:48 | |
| *** jamesden_ has joined #openstack-ansible | 18:49 | |
| *** jamesdenton has quit IRC | 18:49 | |
| *** weezS has joined #openstack-ansible | 19:00 | |
| *** thorst is now known as thorst_afk | 19:15 | |
| mhayden | so i'm trying to get yum.log to get copied to the test results in https://review.openstack.org/486633 | 19:37 |
| mhayden | but the file itself is a 404 | 19:37 |
| mhayden | http://logs.openstack.org/33/486633/2/check/gate-openstack-ansible-tests-ansible-func-centos-7/d9bc193/logs/host/ | 19:38 |
| mhayden | am i missing a step? | 19:38 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: Use sudo with rsync for logs https://review.openstack.org/486633 | 19:39 |
| *** gtrxcb has quit IRC | 19:41 | |
| sxc731 | Hi team, quick q: is it a good idea to put the /openstack directory directory that exists on the infra nodes on an LVM LV and if so is there some magic setting to achieve this? | 19:57 |
| *** kysse has quit IRC | 19:59 | |
| sxc731 | In particular I'm concerned with /openstack/xxxx_galera_container-yyyy which is mount-bound as /var/lib/mysql in the Galera container and currently lands on the root fs of my controller nodes... | 20:00 |
| sxc731 | ...it's entirely possible that I've done smth wrong; not an expert by any means ;-) | 20:01 |
| mhayden | sxc731: you should be able to put that in a logical volume if you want -- you could pre-create it ahead of time | 20:04 |
| mhayden | and pre-mount it | 20:04 |
| mhayden | but if you make a VG called 'lxc', the openstack-ansible-lxc_container_create role will put the containers in there (in separate LVs) | 20:05 |
| mhayden | you can configure that VG name with lxc_container_vg_name | 20:05 |
| mhayden | https://docs.openstack.org/openstack-ansible-lxc_container_create/latest/ | 20:05 |
| sxc731 | mhayden: many thanks. I have the 'lxc' VG and the root fs of all the containers does land on there so that's working well | 20:06 |
| sxc731 | But my galera container has this: | 20:07 |
| sxc731 | lxc.mount.entry = /openstack/toctrl3_galera_container-ca804b8d var/lib/mysql none bind,create=dir 0 0 | 20:07 |
| sxc731 | which causes the mariaDB store to land in /openstack (which is the root fs). Have I done smth wrong? | 20:07 |
| mhayden | hmm, that's a good question -- i normally deploy on the filesystem | 20:08 |
| mhayden | w/o LVM | 20:08 |
| mhayden | you haven't done anything wrong | 20:08 |
| sxc731 | Or to rephrase the question: is there a mechanism to configure where the mariaDB store lands (using OSA)? | 20:08 |
| *** mamitchl has joined #openstack-ansible | 20:09 | |
| sxc731 | Reason I noticed is my infra nodes' root fs is tiny and... mariaDB fails to come up ;-) | 20:09 |
| *** kysse has joined #openstack-ansible | 20:11 | |
| mhayden | i understand what you mean -- i've just not deployed it that way yet ;) | 20:12 |
| *** weezS has quit IRC | 20:12 | |
| *** cpuga has joined #openstack-ansible | 20:14 | |
| sxc731 | OK cool, thank you again kindly... I guess I'll have to hack smth pre OSA-deployment as you suggest | 20:18 |
| *** cpuga has quit IRC | 20:19 | |
| hwoarang | could i get a quick +2 on that one https://review.openstack.org/#/c/485738/ please to unblock neutron on opensuse? :) TIA | 20:21 |
| sxc731 | mhayden: should this (customizable mariaDB store location) be raised as a feature request or do you feel the pre-deployment approach is right? | 20:22 |
| mhayden | hwoarang: I CAN'T PUSH IT ANY HARDER, CAPTAIN! | 20:22 |
| hwoarang | i know i know sir | 20:22 |
| mhayden | sxc731: i believe you can control the bind mount via a variablr | 20:23 |
| * mhayden is looking ofr it | 20:23 | |
| mhayden | wow can't type today | 20:23 |
| sxc731 | mhayden: yeah I thought so given that the lxc.mount.entry is special for the galera container but my 'grep -r' couldn't figure it out... | 20:24 |
| sxc731 | ... smth to do with "lxc_container_bind_mounts | 20:25 |
| openstackgerrit | German Eichberger proposed openstack/openstack-ansible master: Octavia scenario https://review.openstack.org/483520 | 20:32 |
| *** guiltybeta has quit IRC | 20:37 | |
| *** kristian__ has joined #openstack-ansible | 20:40 | |
| dmsimard | mhayden: so, there should be a "icanhazip" ansible module | 20:42 |
| dmsimard | just sayin | 20:42 |
| *** kristian__ has quit IRC | 20:44 | |
| mhayden | haha what would you expect it to do? | 20:49 |
| mhayden | just allow you to store the external IP as a var? | 20:50 |
| *** kristian__ has joined #openstack-ansible | 20:50 | |
| cloudnull | "{{ lookup('pipe', 'curl https://icanhazip.com') }}" | 20:52 |
| mhayden | hah | 20:52 |
| SamYaple | cloudnull: funfact: i know the guy that runs that site | 20:57 |
| cloudnull | lol | 20:58 |
| *** woodard_ has joined #openstack-ansible | 21:00 | |
| *** woodard has quit IRC | 21:04 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-tests master: [WIP] Use sudo with rsync for logs https://review.openstack.org/486633 | 21:04 |
| *** woodard_ has quit IRC | 21:05 | |
| *** sxc731_ has joined #openstack-ansible | 21:09 | |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible-plugins stable/newton: Do not convert integers to strings https://review.openstack.org/486754 | 21:11 |
| prometheanfire | mhayden: odyssey4me cloudnull not sure if you were paying attention to https://bugs.launchpad.net/openstack-ansible/+bug/1705469 but I think that's a good fix. | 21:12 |
| openstack | Launchpad bug 1705469 in openstack-ansible "newton config-template change breaks previously working configuration" [Undecided,New] | 21:12 |
| prometheanfire | python doesn't have a good way of inspecting unicode for typeness without try-except | 21:12 |
| *** mbuil has quit IRC | 21:12 | |
| cloudnull | lgtm | 21:14 |
| *** sxc731 has left #openstack-ansible | 21:15 | |
| prometheanfire | neato | 21:15 |
| *** sxc731 has joined #openstack-ansible | 21:15 | |
| *** sxc731 has quit IRC | 21:18 | |
| *** smatzek has quit IRC | 21:20 | |
| *** perniciouscaffei has joined #openstack-ansible | 21:24 | |
| *** sxc731_ has quit IRC | 21:28 | |
| *** sxc731_ has joined #openstack-ansible | 21:30 | |
| *** sxc731_ has quit IRC | 21:33 | |
| *** esberglu has quit IRC | 21:46 | |
| *** chyka has quit IRC | 21:52 | |
| *** chyka has joined #openstack-ansible | 21:52 | |
| *** marst has quit IRC | 21:52 | |
| *** kong has joined #openstack-ansible | 21:55 | |
| openstackgerrit | Merged openstack/openstack-ansible-plugins stable/newton: Do not convert integers to strings https://review.openstack.org/486754 | 22:00 |
| *** woodard has joined #openstack-ansible | 22:02 | |
| *** woodard has quit IRC | 22:03 | |
| *** woodard has joined #openstack-ansible | 22:04 | |
| *** esberglu has joined #openstack-ansible | 22:05 | |
| *** esberglu has quit IRC | 22:10 | |
| *** kylek3h has quit IRC | 22:13 | |
| *** cpuga has joined #openstack-ansible | 22:16 | |
| *** cpuga has quit IRC | 22:21 | |
| *** sliver has joined #openstack-ansible | 22:26 | |
| *** gouthamr has quit IRC | 22:27 | |
| *** jwitko_ has joined #openstack-ansible | 22:27 | |
| *** jwitko__ has joined #openstack-ansible | 22:27 | |
| *** acormier has quit IRC | 22:31 | |
| *** acormier has joined #openstack-ansible | 22:32 | |
| *** acormier has quit IRC | 22:36 | |
| *** gouthamr has joined #openstack-ansible | 22:41 | |
| *** gtrxcb has joined #openstack-ansible | 22:57 | |
| *** craigs has quit IRC | 23:11 | |
| *** Lirt has quit IRC | 23:15 | |
| *** Lirt has joined #openstack-ansible | 23:15 | |
| *** dxiri has joined #openstack-ansible | 23:22 | |
| *** chyka has quit IRC | 23:26 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican master: Perform an atomic policy file change https://review.openstack.org/461081 | 23:29 |
| *** acormier has joined #openstack-ansible | 23:33 | |
| *** acormier has quit IRC | 23:37 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!