| *** eil397 has quit IRC | 00:04 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-repo_server: Reorganize test playbooks https://review.openstack.org/312302 | 00:06 |
|---|---|---|
| *** fawadkhaliq has quit IRC | 00:09 | |
| *** schwicht has joined #openstack-ansible | 00:14 | |
| *** jamesdenton has joined #openstack-ansible | 00:16 | |
| *** raddaoui has quit IRC | 00:17 | |
| *** pjm6 has quit IRC | 00:17 | |
| *** thorst has joined #openstack-ansible | 00:18 | |
| *** schwicht has quit IRC | 00:18 | |
| *** jamesden_ has joined #openstack-ansible | 00:21 | |
| *** jamesden_ has quit IRC | 00:22 | |
| *** jamesdenton has quit IRC | 00:23 | |
| *** jamesdenton has joined #openstack-ansible | 00:25 | |
| *** jamesden_ has joined #openstack-ansible | 00:26 | |
| *** jamesdenton has quit IRC | 00:28 | |
| *** jamesden_ has quit IRC | 00:28 | |
| *** sacharya has joined #openstack-ansible | 00:34 | |
| *** thorst has quit IRC | 00:36 | |
| *** thorst has joined #openstack-ansible | 00:36 | |
| *** sacharya has quit IRC | 00:39 | |
| *** thorst has quit IRC | 00:41 | |
| *** thorst has joined #openstack-ansible | 00:45 | |
| *** schwicht has joined #openstack-ansible | 00:47 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: Doc: Notice to disable security hardening role during minor upgrades https://review.openstack.org/311202 | 00:57 |
| *** kiranv has quit IRC | 01:08 | |
| *** ozialien10 has quit IRC | 01:12 | |
| *** fishcried has joined #openstack-ansible | 01:12 | |
| *** fishcried has quit IRC | 01:14 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: Doc: Notice to disable security role during major upgrades https://review.openstack.org/311211 | 01:15 |
| *** mummer has quit IRC | 01:15 | |
| *** daneyon has joined #openstack-ansible | 01:19 | |
| *** sdake has joined #openstack-ansible | 01:22 | |
| *** daneyon_ has joined #openstack-ansible | 01:25 | |
| *** daneyon has quit IRC | 01:28 | |
| *** chandanc_ has joined #openstack-ansible | 01:29 | |
| *** daneyon_ has quit IRC | 01:36 | |
| *** joker_ has quit IRC | 01:42 | |
| *** jthorne_ has joined #openstack-ansible | 01:45 | |
| *** jthorne has quit IRC | 01:45 | |
| *** jthorne_ has quit IRC | 01:46 | |
| *** jthorne has joined #openstack-ansible | 01:46 | |
| *** afranc has quit IRC | 01:49 | |
| *** chandanc__ has joined #openstack-ansible | 01:56 | |
| *** chandanc_ has quit IRC | 01:56 | |
| *** chandanc__ has quit IRC | 01:57 | |
| *** thorst has quit IRC | 02:01 | |
| *** thorst has joined #openstack-ansible | 02:01 | |
| *** klamath has quit IRC | 02:01 | |
| *** klamath has joined #openstack-ansible | 02:01 | |
| *** klamath has quit IRC | 02:02 | |
| *** klamath has joined #openstack-ansible | 02:03 | |
| *** sdake_ has joined #openstack-ansible | 02:08 | |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible: Fix dynamic_inventory.py test bug https://review.openstack.org/312808 | 02:08 |
| lbragstad | o/ hey folks - is anyone else getting running the os_keystone role in isolation? Failed to template {{ hostvars[item]['keystone_pubkey'] | b64decode }}: | 02:08 |
| lbragstad | I see a fix has been proposed and backported for os_nova - https://bugs.launchpad.net/openstack-ansible/+bug/1531783 | 02:09 |
| openstack | Launchpad bug 1531783 in openstack-ansible liberty "os_nova not using all configured compute_hosts" [High,In progress] - Assigned to Kevin Carter (kevin-carter) | 02:09 |
| *** thorst has quit IRC | 02:09 | |
| *** sdake has quit IRC | 02:11 | |
| *** schwicht has quit IRC | 02:17 | |
| coolj_ | lbragstad: yeah, i have seen that - i believe it's this bug https://bugs.launchpad.net/openstack-ansible/+bug/1531783 | 02:18 |
| openstack | Launchpad bug 1531783 in openstack-ansible liberty "os_nova not using all configured compute_hosts" [High,In progress] - Assigned to Kevin Carter (kevin-carter) | 02:19 |
| coolj_ | oh that's the one you linked haha | 02:19 |
| lbragstad | coolj_ :) i thought that looked familiar | 02:19 |
| lbragstad | I see the keystone bit in there - but I also see the odyssey4me patched it in os_keystone | 02:19 |
| coolj_ | seems like whatever is causing the *_pubkey facts not to be set is deeper than just using slurp | 02:24 |
| lbragstad | coolj_ actually - i'm wondering if it's the hostnames i'm using in my inventory | 02:25 |
| lbragstad | i do have '-' in there | 02:25 |
| coolj_ | i have a - in my hostnames as well (I'm jcallicoat on lp btw), but it only failed for a few hosts and they all had - in the hostname...i'm thinking something in the actual key content but i don't see any strange characters | 02:26 |
| lbragstad | coolj_ o/ oh hey man! | 02:26 |
| *** sdake_ has quit IRC | 02:27 | |
| coolj_ | hi! :) | 02:27 |
| lbragstad | coolj_ do you pass hostnames in on the command line with `-l <hostname>` | 02:27 |
| lbragstad | when you run ansible-playbook? | 02:27 |
| coolj_ | no, it was all done through dynamic inventory and just letting the plays find their hosts that way | 02:28 |
| lbragstad | gothca | 02:28 |
| lbragstad | huh - no way | 02:29 |
| lbragstad | that was it... I was passing in the host I wanted to run against with `-l <hostname>` and it failed consistently... remove the `-l <hostname>` and run it through inventory and it works fine. | 02:29 |
| coolj_ | huh | 02:31 |
| lbragstad | not sure how common that is... but I can document it in https://bugs.launchpad.net/openstack-ansible/+bug/1531783 | 02:32 |
| openstack | Launchpad bug 1531783 in openstack-ansible liberty "os_nova not using all configured compute_hosts" [High,In progress] - Assigned to Kevin Carter (kevin-carter) | 02:32 |
| *** fawadkhaliq has joined #openstack-ansible | 02:32 | |
| *** sacharya has joined #openstack-ansible | 02:39 | |
| *** kiranv_ has joined #openstack-ansible | 02:54 | |
| *** thorst has joined #openstack-ansible | 03:07 | |
| *** schwicht has joined #openstack-ansible | 03:13 | |
| *** thorst has quit IRC | 03:15 | |
| *** schwicht has quit IRC | 03:18 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: [WIP] DOCS: Configuration section - cleanup https://review.openstack.org/312815 | 03:35 |
| *** fawadkhaliq has quit IRC | 03:41 | |
| *** kiranv_ has quit IRC | 04:13 | |
| *** thorst has joined #openstack-ansible | 04:13 | |
| *** kiranv_ has joined #openstack-ansible | 04:13 | |
| *** asettle has quit IRC | 04:13 | |
| *** kiranv_ is now known as QuantumShadow | 04:13 | |
| *** schwicht has joined #openstack-ansible | 04:14 | |
| *** sdake has joined #openstack-ansible | 04:18 | |
| *** schwicht has quit IRC | 04:19 | |
| *** thorst has quit IRC | 04:19 | |
| *** sdake has quit IRC | 04:47 | |
| *** sdake has joined #openstack-ansible | 04:47 | |
| *** chhavi has joined #openstack-ansible | 04:50 | |
| *** hybridpollo has quit IRC | 04:54 | |
| *** chhavi has quit IRC | 05:06 | |
| *** iceyao has joined #openstack-ansible | 05:09 | |
| *** chhavi has joined #openstack-ansible | 05:09 | |
| *** chandanc has joined #openstack-ansible | 05:10 | |
| *** iceyao_ has joined #openstack-ansible | 05:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Fix generation of LXC hostnames https://review.openstack.org/312752 | 05:14 |
| *** iceyao has quit IRC | 05:16 | |
| *** thorst has joined #openstack-ansible | 05:18 | |
| *** fawadkhaliq has joined #openstack-ansible | 05:18 | |
| *** thorst has quit IRC | 05:25 | |
| *** javeriak has joined #openstack-ansible | 05:28 | |
| *** iceyao_ has quit IRC | 05:34 | |
| *** chhavi has quit IRC | 05:39 | |
| *** chhavi has joined #openstack-ansible | 05:43 | |
| *** iceyao has joined #openstack-ansible | 05:47 | |
| *** bigdogstl has joined #openstack-ansible | 05:51 | |
| *** chhavi has quit IRC | 05:54 | |
| *** chhavi has joined #openstack-ansible | 05:54 | |
| *** javeriak has quit IRC | 05:56 | |
| *** bigdogstl has quit IRC | 05:56 | |
| *** openstackgerrit has quit IRC | 06:03 | |
| *** openstackgerrit has joined #openstack-ansible | 06:03 | |
| *** daneyon has joined #openstack-ansible | 06:08 | |
| *** daneyon_ has joined #openstack-ansible | 06:09 | |
| *** QuantumShadow has quit IRC | 06:10 | |
| *** saneax_AFK is now known as saneax | 06:12 | |
| *** daneyon has quit IRC | 06:13 | |
| *** schwicht has joined #openstack-ansible | 06:15 | |
| *** iceyao_ has joined #openstack-ansible | 06:16 | |
| *** asettle has joined #openstack-ansible | 06:18 | |
| *** schwicht has quit IRC | 06:20 | |
| *** joker_ has joined #openstack-ansible | 06:20 | |
| *** iceyao has quit IRC | 06:20 | |
| *** thorst has joined #openstack-ansible | 06:23 | |
| *** daneyon_ has quit IRC | 06:24 | |
| *** daneyon has joined #openstack-ansible | 06:24 | |
| *** thorst has quit IRC | 06:30 | |
| *** sdake has quit IRC | 06:32 | |
| *** chandanc has quit IRC | 06:35 | |
| *** chandanc has joined #openstack-ansible | 06:36 | |
| *** daneyon has quit IRC | 06:37 | |
| pellaeon | Hi, I'm trying to set up ceilometer for my liberty installation, in the doc it says I need to install mongodb manually, but didn't say where | 06:39 |
| pellaeon | http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-ceilometer.html | 06:39 |
| pellaeon | Where is the recommended place where I install mongodb ? | 06:41 |
| *** adib has quit IRC | 06:44 | |
| *** sj has quit IRC | 07:05 | |
| *** schwicht has joined #openstack-ansible | 07:18 | |
| *** chandanc has quit IRC | 07:19 | |
| *** sacharya has quit IRC | 07:20 | |
| *** schwicht has quit IRC | 07:23 | |
| *** tiagogomes has joined #openstack-ansible | 07:27 | |
| *** thorst has joined #openstack-ansible | 07:27 | |
| mattt | pellaeon: heh, good question | 07:28 |
| mattt | pellaeon: to be honest we don't have a valid mongodb story in openstack-ansible, i think the assumption is that mongodb lives outside of openstack-ansible | 07:30 |
| *** thorst has quit IRC | 07:35 | |
| pellaeon | mattt: I see, I'll setup a separate host, thanks | 07:38 |
| mattt | pellaeon: that's something we will need to discuss as a project, because i think we need a recommended solution for people deploying ceilometer | 07:39 |
| *** asettle has quit IRC | 07:40 | |
| mattt | pellaeon: also, we need to establish what version of mongodb to use -- i did some research into mongo a few months back and everything i read discouraged using distro mongodb packages | 07:41 |
| *** mikelk has joined #openstack-ansible | 07:41 | |
| mattt | pellaeon: in theory you could deploy it into your openstack-ansible environment, but you'd have to create a new environment for it in /etc/openstack_deploy/env.d | 07:42 |
| mattt | pellaeon: either that or install it on a host itself or in an existing container | 07:43 |
| mattt | pellaeon: in our gate AIO, we just install it on the host itself | 07:43 |
| *** javeriak has joined #openstack-ansible | 07:43 | |
| *** fawadkhaliq has quit IRC | 08:06 | |
| *** sacharya has joined #openstack-ansible | 08:20 | |
| *** sacharya has quit IRC | 08:26 | |
| *** javeriak has quit IRC | 08:29 | |
| *** thorst has joined #openstack-ansible | 08:32 | |
| *** thorst has quit IRC | 08:40 | |
| *** jorge_munoz has joined #openstack-ansible | 08:49 | |
| *** chhavi has quit IRC | 08:56 | |
| *** electrofelix has joined #openstack-ansible | 08:56 | |
| *** chhavi has joined #openstack-ansible | 09:12 | |
| *** fauxbert has joined #openstack-ansible | 09:16 | |
| fauxbert | Hi, I was wondering if I could get some help with a problem I am encountering with os-neutron-install.yml , on the create network facts it barfs with a JSon serialization error | 09:17 |
| fauxbert | TypeError: KeyError('type',) is not JSON serializable | 09:18 |
| andymccr | fauxbert: hey! so do you have any extra logs you could show us for that? | 09:18 |
| fauxbert | failed: [node05] => {"failed": true, "parsed": false} OpenSSH_6.6.1, OpenSSL 1.0.1f 6 Jan 2014 debug1: Reading configuration data /etc/ssh/ssh_config debug1: /etc/ssh/ssh_config line 22: Applying options for * debug1: auto-mux: Trying existing master debug1: mux_client_request_session: master session id: 2 Traceback (most recent call last): File "<stdin>", line 1895, in <module> File "<stdin>", line 1892, in main File | 09:18 |
| fauxbert | that's about all I get | 09:19 |
| fauxbert | I've ran the provider_networks yaml through a yaml validator | 09:19 |
| andymccr | which task is failing? | 09:19 |
| fauxbert | TASK: [Create the neutron provider networks facts] **************************** | 09:19 |
| *** javeriak has joined #openstack-ansible | 09:20 | |
| fauxbert | as far as I can tell it uses provider_networks, bind_prefix and is_metal to combine into pndata | 09:20 |
| *** sacharya has joined #openstack-ansible | 09:22 | |
| fauxbert | this is on ubuntu 14.04.4 LTS btw | 09:23 |
| *** sdake has joined #openstack-ansible | 09:24 | |
| *** sacharya has quit IRC | 09:27 | |
| *** thorst has joined #openstack-ansible | 09:38 | |
| andymccr | sorry fauxbert looking now, i had a quick meeting to run into. | 09:40 |
| fauxbert | no worries | 09:40 |
| andymccr | ok fauxbert what does your provider_networks look like in your openstack_user_config.yml ? | 09:44 |
| fauxbert | you ok if I use pastebin for this? | 09:44 |
| andymccr | yeh please do | 09:44 |
| *** thorst has quit IRC | 09:45 | |
| fauxbert | http://pastebin.com/scfkLM3U | 09:46 |
| fauxbert | it parses clean through yamllint (the whole openstack_user_config.yml) | 09:47 |
| andymccr | try look in your openstack_inventory.json file and go to the provider_networks bit | 09:57 |
| andymccr | im wondering if its the vxlan range bit, with the comma - the syntax looks fine otherwise | 09:58 |
| andymccr | the inventory file is basically a mixture of the environment settings (in env.d) and the user config files (openstack_user_config.yml and any files in conf.d) | 09:58 |
| andymccr | it takes the 2 sets and dynamically creates the inventory based on the environment definitions. so that inventory file is technically what ansible uses. | 09:59 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible-os_tempest: [WIP] Standardise tempest role https://review.openstack.org/310766 | 10:03 |
| fauxbert | sorry was called away for a few , I'll check now | 10:04 |
| fauxbert | http://pastebin.com/k0jh2YE2 | 10:06 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Use slurp to collect the keystone ssh keys https://review.openstack.org/312882 | 10:06 |
| fauxbert | i've looked at the provider_networks variable by inserting a debug statement before the create network facts task | 10:13 |
| fauxbert | it looks ok to me | 10:13 |
| andymccr | yeah same. | 10:14 |
| andymccr | i dont see anything odd about it | 10:14 |
| evrardjp | will vlan 130 be used by your tenants? | 10:14 |
| evrardjp | try without it, it's just one vlan :p | 10:15 |
| andymccr | i tested that | 10:15 |
| andymccr | it works fine | 10:15 |
| *** iceyao_ has quit IRC | 10:15 | |
| fauxbert | I'll give it a shot | 10:16 |
| andymccr | as an aside - you can remove some of the group_binds for your br-mgmt, since you specify all_containers and hosts the other 4 are redundant because the network will be setup on all hosts and containers already | 10:17 |
| evrardjp | fauxbert in theory "," should work, so you either are hitting a new bug, or you have other stuff wrong | 10:17 |
| fauxbert | I wouldn't bet against having other stuff wrong ;) | 10:17 |
| andymccr | fauxbert: try it, but i did just test with a comma in a vlan list and it works fine | 10:18 |
| fauxbert | vlan130 is kinda critical for us, as it's the external vlan | 10:18 |
| andymccr | so i dont think its that | 10:18 |
| evrardjp | fauxbert if it's external, and you can use the admin to define once and for all I think it should work | 10:18 |
| evrardjp | you define it yourself | 10:18 |
| evrardjp | or at least it used to work | 10:18 |
| evrardjp | you should try :D | 10:19 |
| fauxbert | nah still same issue, even without the 130, | 10:20 |
| fauxbert | I am wondering if it's because I am using the stable/mitaka branch... | 10:20 |
| fauxbert | (though more likely that it's because I screwed up somewhere) | 10:21 |
| evrardjp | could you paste your complete openstack_ file ? (maybe obfuscate the ips if it's sensitive ?) | 10:21 |
| fauxbert | sure | 10:22 |
| fauxbert | give me 2 minutes | 10:22 |
| evrardjp | take your time, don't worry :D | 10:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Removed container_release property from environment files https://review.openstack.org/311802 | 10:23 |
| odyssey4me | mattt can you please review https://review.openstack.org/312793 asap as it's blocking master gating | 10:24 |
| fauxbert | http://pastebin.com/J6t0JVJj | 10:25 |
| evrardjp | fauxbert I think you can add a space in your first item in used_ips | 10:27 |
| evrardjp | I'll check everything just I'll give you advise when I read this | 10:27 |
| fauxbert | okidoki | 10:27 |
| fauxbert | sorry , a space before the - ? | 10:29 |
| *** sdake has quit IRC | 10:29 | |
| fauxbert | it's got 2 spaces right now, I think that's the correct indentation ? | 10:29 |
| evrardjp | yes, for readability | 10:29 |
| odyssey4me | evrardjp some final tweaks in https://review.openstack.org/#/c/311162 needed | 10:29 |
| evrardjp | also br-mgmt doesn't have a type | 10:29 |
| andymccr | thats the issue | 10:29 |
| andymccr | i recreated it its br-mgmt lacking a type | 10:30 |
| evrardjp | fauxbert next steps after that is, if you have the opportunity: scratch everything and start with a new clean base :D | 10:31 |
| andymccr | just add type: "raw" and you should be good to go | 10:31 |
| fauxbert | well it passed that section now :) | 10:32 |
| fauxbert | thank you ! (failed elsewhere but I'd do some checking myself before I bother you guys again) | 10:32 |
| andymccr | id probably still recommend changing up your group_binds for br-mgmt to only include hosts and all_containers | 10:33 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Docs: Mandatory ssh public key https://review.openstack.org/311162 | 10:33 |
| fauxbert | will do | 10:33 |
| andymccr | glad its getting further | 10:33 |
| evrardjp | fauxbert yes you should do that :D | 10:33 |
| *** mrhillsman has joined #openstack-ansible | 10:38 | |
| odyssey4me | mattt https://review.openstack.org/312588 | 10:41 |
| odyssey4me | git-harry https://review.openstack.org/#/q/branch:kilo+project:%255Eopenstack/openstack-ansible.*+status:open | 10:41 |
| *** thorst has joined #openstack-ansible | 10:41 | |
| *** deadnull_ has joined #openstack-ansible | 10:42 | |
| *** asettle has joined #openstack-ansible | 10:46 | |
| *** thorst has quit IRC | 10:50 | |
| *** javeriak has quit IRC | 10:51 | |
| *** asettle has quit IRC | 10:51 | |
| *** bbmbx__ has quit IRC | 10:54 | |
| mattt | odyssey4me: +2, didn't +W only because lack of cherry-pick message ... if a non-issue you can go ahead and +W | 11:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Updates all SHAs for 12.0.12 https://review.openstack.org/309511 | 11:03 |
| *** javeriak has joined #openstack-ansible | 11:04 | |
| *** schwicht has joined #openstack-ansible | 11:11 | |
| *** johnmilton has joined #openstack-ansible | 11:19 | |
| *** LanceHaig has quit IRC | 11:20 | |
| *** LanceHaig has joined #openstack-ansible | 11:20 | |
| odyssey4me | evrardjp mattt https://review.openstack.org/#/c/312692 | 11:26 |
| *** schwicht has quit IRC | 11:34 | |
| *** thorst has joined #openstack-ansible | 11:40 | |
| *** retreved has joined #openstack-ansible | 11:44 | |
| *** woodard has joined #openstack-ansible | 11:45 | |
| *** smatzek has joined #openstack-ansible | 11:45 | |
| *** woodard has quit IRC | 11:45 | |
| *** woodard has joined #openstack-ansible | 11:45 | |
| *** woodard has quit IRC | 11:47 | |
| *** woodard has joined #openstack-ansible | 11:47 | |
| openstackgerrit | Merged openstack/openstack-ansible: Remove XFS filesystem from the daily mlocate cron job https://review.openstack.org/310460 | 11:50 |
| smatzek | evrardjp: Thanks for the review the ceph/uca option. I'm new to ansible and this is my first submission to OSA. Can you clarify what you'd like to see changed, simply change ceph_add_uca_repo to something like ceph_source ? | 11:50 |
| odyssey4me | smatzek I'm kinda thinking through it at the moment. We have a few intersecting initiatives here to take care of. | 11:51 |
| evrardjp | ok I'll wait for odyssey4me 's comment then | 11:51 |
| odyssey4me | One is that we're looking at multi distro implementation - two versions of Ubuntu (Trusty and Xenial), another for CentOS7. | 11:52 |
| odyssey4me | another intersecting need is to simplify how a user consumes these things | 11:53 |
| smatzek | odyssey4me: I had a pull request merged in ceph-ansible yesterday adding UCA support there in a similar fashion, for what it's worth. https://github.com/ceph/ceph-ansible/pull/747 | 11:53 |
| *** weshay has joined #openstack-ansible | 11:53 | |
| odyssey4me | so it seems that perhaps we could implement a dict with a key that switches the source appropriately, then have all the tasks key off that | 11:53 |
| odyssey4me | something like 'ceph_client_package_source' with the keys 'ubuntu_uca', 'ubuntu_ceph', 'ubuntu_ceph_eu' - then under those keys we have the vars for the appropriate repo, keys, packages, etc | 11:54 |
| smatzek | switching the source via a single flag would be good. To be honest I didn't like the boolean nature of the flag I added. However, users will still need to tweak other variables to target the exact ceph version or apt repo dist they need. | 11:54 |
| odyssey4me | then the tasks simply act on whether some of those things are defined or not | 11:54 |
| *** timl0101 has joined #openstack-ansible | 11:56 | |
| smatzek | ceph-ansible can install from 4? different sources, and they handle with a mix of flags, comments tell the user which set to enable together, not to enable multiple, and they have a sanity check task to ensure the user isn't enabling multiple. | 11:56 |
| odyssey4me | oh, and of course we need to make it simpler to adjust the host used for the mirror to cater for the internal use use-case... that should be changable without having to redefine the whole dict | 11:56 |
| odyssey4me | wow, that sounds complex | 11:57 |
| odyssey4me | the only issue I forsee is a situation where a deployer deploys from one source, then switches to use another source | 11:57 |
| smatzek | yeah, it is with settings on one flag 'upstream' or 'distro' affecting whether or not lower flags are even looked at | 11:57 |
| odyssey4me | to be honest we'd rather be rid of this ceph_client role - the whole ceph implementation needs a revisit but that hasn't really bubbled up as a priority | 11:58 |
| odyssey4me | but I don't think that's going to happen too soon, so we must work with what's there | 11:59 |
| odyssey4me | smatzek if you can work out something that works through switching a single flag, ideally implements as few new vars as possible (maybe even removes some existing vars) then that'd be great | 12:01 |
| odyssey4me | we'll likely follow it on with something like https://github.com/openstack/openstack-ansible-galera_client/commit/0361308dbbf38720f9e723de92d70d8fd0e16958 which then enables multi-OS implementations too | 12:01 |
| smatzek | with CentOS support this role will get much more complex with the need to support RPM installs from yum repos, and then you'll get someone coming along wanting to install from some RHEL specific repo. | 12:02 |
| *** saneax is now known as saneax_AFK | 12:02 | |
| odyssey4me | yeah, we don't have to specifically enable centos support right now - we just want to make it reasonably obvious how it can be done | 12:02 |
| mhayden | morning folks | 12:03 |
| openstackgerrit | Merged openstack/openstack-ansible: Cap paramiko to <2 https://review.openstack.org/312588 | 12:04 |
| openstackgerrit | Merged openstack/openstack-ansible: Cap paramiko to <2 https://review.openstack.org/312587 | 12:05 |
| *** saneax_AFK is now known as saneax | 12:08 | |
| odyssey4me | smatzek my apologies for basically asking you to rewrite your patch - I was kinda torn between approving what you've done, then rewriting it in a follow-on... but then I figured it'd be better to let you know my thinking and help work through to something better. | 12:10 |
| *** psilvad has joined #openstack-ansible | 12:13 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312580 | 12:14 |
| *** iceyao has joined #openstack-ansible | 12:16 | |
| *** psilvad has quit IRC | 12:18 | |
| smatzek | odyssey4me: no problem. I like the galera_client commit you pointed me at. On the variable count concerns. Is it a concern on the number of variables overall or a concern on the number of variables a deployer would need to normally set? For instance, could we do ceph_source: uca|ceph|?? and depending on the source have one or two variables to set like uca_dist: mitaka and ceph_release: hammer ? | 12:20 |
| smatzek | we need a way for deployers to both pick the repo source and pick what version of Ceph to pull from that source. | 12:20 |
| odyssey4me | smatzek for ease of use - only one var should switch things, but then yes a deployer should optionally be able to change the ceph version and the mirror URL source (to cater for internal mirrors) | 12:21 |
| smatzek | odyssey4me: yes, I think we're thinking the same thing. With what I said earlier, we'd have defaults to pulls something like 'hammer' or 'jewel' if you're pulling from ceph.com and a default for 'mitaka', etc if you're pulling from UCA. You'd only need to tweak that var if you wanted something other than the provided default from that repo source. | 12:23 |
| odyssey4me | yep, sounds good to me | 12:23 |
| odyssey4me | master is focused on newton, so I'm guessing there would not be a UCA source for that though - it's come straight out of the Xenial source? | 12:24 |
| odyssey4me | I would guess that for Xenial the package in the distro would be used, but for trusty the package from UCA would be used? | 12:24 |
| odyssey4me | so the multi-os enablement helps there to deal with the differences | 12:24 |
| smatzek | UCA doesn't have a Newton distro up yet. based on their web page newton will only be on Xenial. trusty at UCA will ride out its LTS with mitaka. | 12:25 |
| smatzek | so a wider list of ceph binary sources for debian is really: ceph.com, uca, or just use what's in my distro default repo- don't add one. ceph-ansible has this option as well. | 12:26 |
| *** psilvad has joined #openstack-ansible | 12:26 | |
| smatzek | the issue with "use my distro default" is that for Trusty this is Ceph Giant which was retired by Ceph 1 year ago, and was not an LTS release to begin with. | 12:27 |
| smatzek | which is exactly why I'm trying to add UCA support so I can get a new LTS Ceph release on top of trusty ppc64le, which I can only get from UCA :) | 12:28 |
| odyssey4me | yeah, if we can skip the 'just use my distro' choice then let's do so - although I think for Xenial we'll have to do that... I'll leave that choice to you for now and we can always revise later | 12:28 |
| odyssey4me | it may be a good idea to add some level of docs for this with the patch which describes when you would select each option as you've done in the commit message - I don't think we have docs for this yet because that paradigm was only implemented after this was introduced | 12:30 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_horizon: Enable new LBaaS panel in Horizon https://review.openstack.org/312733 | 12:30 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_horizon: Enable IPv6 support in Horizon https://review.openstack.org/312741 | 12:30 |
| *** javeriak has quit IRC | 12:31 | |
| smatzek | in the release note yml or in a separate doc? | 12:32 |
| mhayden | so IPv6 is enabled in Horizon's default config (in Newton, at least) | 12:34 |
| mhayden | my gut says we should default to "on" in Newton, and leave it off (but configurable) in liberty/mitaka | 12:34 |
| mattt | odyssey4me: the ceph-ansible mechanism for defining what to install /is/ complex | 12:35 |
| odyssey4me | smatzek in /doc/source - I wouldn't worry too much about that right now... if you're commenting the defaults file well and have added a good release note then that may do well enough... with the changes we're doing in the doc strategy it may make better sense to leave the details out of the install guide and rather just make sure the role default comments are clear | 12:35 |
| odyssey4me | mhayden yeah, that sounds fine | 12:36 |
| mhayden | woot | 12:37 |
| *** pjm6 has joined #openstack-ansible | 12:38 | |
| pjm6 | good morning all | 12:38 |
| mattt | smatzek: we wouldn't want xenial to use the distro ceph version right ? | 12:41 |
| mattt | that's jewel, which isn't LTS | 12:42 |
| mattt | oh wait, it is | 12:42 |
| *** jayc_ has joined #openstack-ansible | 12:43 | |
| smatzek | yeah, Jewel is LTS. I haven't spun up a xenial VM to verify what it has by default. | 12:43 |
| smatzek | I think we probably do want a "use my distro version" option, and that option would simply not add a repo in the _preinstall.yml, but still run through the install yml. | 12:43 |
| odyssey4me | mhayden with regards to the lbaas dashboard - what you've done only works for lbaasv2 and completely removes the ability to use the lbaasv1 dashboard - is there a way to somehow allow both? | 12:44 |
| mhayden | odyssey4me: in newton, LBaaSv1 is gone from the neutron codebase | 12:45 |
| mhayden | so i have to work a different fix for mitaka | 12:45 |
| mhayden | in mitaka, i'll need to handle both | 12:45 |
| odyssey4me | mhayden ah, that's fine then - so the backport would have to cater for more options | 12:45 |
| mhayden | unfortunately, yes :) | 12:45 |
| odyssey4me | ok, I'm good with that then | 12:45 |
| mhayden | i'm trying to think about the best way to do that in mitaka, since there's one variable for enabling the lbaas dashboard... | 12:46 |
| mhayden | i like the simplicity of just having "horizon_enable_neutron_lbaas" | 12:46 |
| mhayden | but perhaps i need horizon_enable_neutron_lbaas and horizon_enable_neutron_lbaasv2? | 12:46 |
| deadnull_ | hmm | 12:47 |
| deadnull_ | I am fine either way | 12:47 |
| deadnull_ | cc mhayden odyssey4me | 12:47 |
| mattt | smatzek: yeah that makes sense | 12:47 |
| odyssey4me | yeah, I was kinda thinking that - but if horizon_enable_neutron_lbaas is the only enduring var beyond mitaka then that isn't gret | 12:47 |
| odyssey4me | *great | 12:47 |
| mhayden | so in mitaka, if someone flips horizon_enable_neutron_lbaas to true, how should i detect which one? | 12:47 |
| mattt | smatzek: so we have another issue we need to work around | 12:47 |
| mattt | smatzek: because we've always worked off the assumption that people will be using hammer from ceph.com | 12:48 |
| odyssey4me | I'm sort-of thinking that perhaps the horizon role could determine whether neutron_plugin_base is defined, then can detect the version from there | 12:49 |
| mattt | smatzek: https://github.com/openstack/openstack-ansible-os_nova/blob/master/defaults/main.yml#L399 | 12:49 |
| mhayden | odyssey4me: i'm okay with that | 12:49 |
| mattt | smatzek: we had to build that package manually, since ceph do not bundle their python bits separately | 12:49 |
| deadnull_ | instead of a bool, maybe specify the verison? | 12:49 |
| mhayden | i'd need to be careful since someone might be running it without the neutron role perhaps | 12:49 |
| *** schwicht has joined #openstack-ansible | 12:51 | |
| *** bapalm has joined #openstack-ansible | 12:54 | |
| mhayden | odyssey4me: would it be possible to remove the -2 on this one? https://review.openstack.org/#/c/310605/ | 12:55 |
| smatzek | mattt: Can you help me understand that? You built your own python-cephlibs? What is in there that is not in python-ceph that the ceph_client role installs ? https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/ceph_client/vars/main.yml#L22 | 12:58 |
| mhayden | odyssey4me: thanks for the deuce on the lbaas wheel! | 13:02 |
| odyssey4me | mhayden done - tweaked a little to be clear about backports and such | 13:02 |
| pjm6 | guys anyone know if I can use horizon test server with SSL? | 13:03 |
| pjm6 | when I do manage*.py --runserver 0.0.0.0:443 (disabling apache2) it says that can't run with SSL | 13:03 |
| smatzek | mattt: is this a case of the package install of the python code not being used, or being overwritten by the pip install of the cephlibs in the venv for services that run in containers? | 13:05 |
| mattt | smatzek: hah, i actually forgot for a second why this was done | 13:10 |
| mattt | smatzek: but yes, it's because the python venv should be self-contained and there is no way to get the python-ceph files installed into the venv | 13:11 |
| *** psilvad has quit IRC | 13:12 | |
| *** jamesdenton has joined #openstack-ansible | 13:12 | |
| *** asettle has joined #openstack-ansible | 13:13 | |
| smatzek | mattt: that's a mess, and it really does prohibit the use of any ceph release other than the one you've built into cephlibs. It secretly pins you to hammer if you're running any ceph client in a container. Nova compute would pick up whatever ceph release you targetted though. | 13:14 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Switch from dict to individual variables https://review.openstack.org/312506 | 13:14 |
| mattt | smatzek: yep | 13:14 |
| mattt | smatzek: not ideal, which is why i wanted to raise it so you're aware | 13:14 |
| mhayden | good catch on the commit message, odyssey4me -- i totally missed that | 13:15 |
| smatzek | mattt: Thanks. This poses a problem for me since we were planning on using Jewel and have a short time frame to get it done. | 13:16 |
| mattt | smatzek: previously we symlinked into the venv (also terrible), but perhaps the lesser of two evils? | 13:17 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312580 | 13:17 |
| smatzek | mattt: that may be the lesser of two evils. Also ugly, but it gives the deployer freedom to choose ceph release and distibution source | 13:18 |
| mattt | smatzek: let me put in a review to get that commit back in | 13:18 |
| mattt | smatzek: and we'll get it reviewed and see what people think | 13:18 |
| smatzek | mattt: awesome, thanks | 13:18 |
| mattt | smatzek: np | 13:18 |
| *** psilvad has joined #openstack-ansible | 13:19 | |
| *** BjoernT has joined #openstack-ansible | 13:20 | |
| *** BjoernT is now known as Bjoern_zZzZzZzZ | 13:20 | |
| *** sacharya has joined #openstack-ansible | 13:23 | |
| *** Ger-chervyak has joined #openstack-ansible | 13:28 | |
| *** timl0101 has quit IRC | 13:28 | |
| *** sacharya has quit IRC | 13:28 | |
| evrardjp | spotz_zzz I answered on your message on this commit https://review.openstack.org/#/c/312421, but I'd be happy to have your insights before doing another patchset | 13:28 |
| *** Ger-chervyak has joined #openstack-ansible | 13:28 | |
| odyssey4me | mattt can you pass by these reviews please https://review.openstack.org/312677 / https://review.openstack.org/312402 / https://review.openstack.org/312387 ? | 13:31 |
| mattt | odyssey4me: k | 13:31 |
| odyssey4me | mhayden does https://review.openstack.org/#/q/I713ac489998e34cf0a916a8acc5ae752e657d163,n,z need to go back into liberty as well ? | 13:32 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Switch from dict to individual variables https://review.openstack.org/312506 | 13:32 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312580 | 13:33 |
| *** tiagogomes has quit IRC | 13:33 | |
| mattt | odyssey4me: on the first, do you actually need to specify origin ? | 13:33 |
| odyssey4me | mattt it depends on the git version I think - origin is a more univeral application and that's consistent with all the other patches on that topic | 13:34 |
| mhayden | odyssey4me: looking | 13:34 |
| mhayden | odyssey4me: actually, liberty is okay | 13:34 |
| mhayden | yeah, just verified -- liberty's lbaasv2 docs are OK | 13:35 |
| *** Bjoern_zZzZzZzZ is now known as BjoernT | 13:35 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add release note for paramiko issue workaround https://review.openstack.org/312953 | 13:35 |
| mhayden | odyssey4me: i fixed a few things mattt found in https://review.openstack.org/#/c/312506/ -- should be good to go now (hopefully) | 13:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add release note for paramiko issue workaround https://review.openstack.org/312954 | 13:36 |
| odyssey4me | mattt also https://review.openstack.org/#/q/status:open+project:openstack/openstack-ansible+topic:bug/1577469 when you're done with the others | 13:37 |
| cloudnull | morning | 13:43 |
| andymccr | morning friend | 13:43 |
| pjm6 | cloudnull, o/ | 13:43 |
| pjm6 | andyhky, o/ | 13:43 |
| pjm6 | * andymccr, o/ | 13:44 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:45 | |
| odyssey4me | mattt git-harry FYI http://docs.openstack.org/releasenotes/openstack-ansible/liberty.html#id6 | 13:45 |
| *** jayc__ has joined #openstack-ansible | 13:46 | |
| *** jthorne_ has joined #openstack-ansible | 13:46 | |
| *** jthorne has quit IRC | 13:46 | |
| *** openstackgerrit has quit IRC | 13:47 | |
| *** openstackgerrit has joined #openstack-ansible | 13:48 | |
| *** jayc_ has quit IRC | 13:49 | |
| cloudnull | hows it going on this fine day ? | 13:53 |
| *** ametts has joined #openstack-ansible | 13:55 | |
| evrardjp | I had ice cream, so I'm in a good mood. Need reviews? | 13:55 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Change pip install task state to 'latest' https://review.openstack.org/312387 | 13:56 |
| *** asettle has quit IRC | 13:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Change pip install task state to 'latest' https://review.openstack.org/312402 | 13:59 |
| *** deadnull_ has quit IRC | 14:01 | |
| *** tiagogomes has joined #openstack-ansible | 14:02 | |
| *** klamath has quit IRC | 14:02 | |
| *** klamath has joined #openstack-ansible | 14:03 | |
| *** klamath has quit IRC | 14:03 | |
| cloudnull | im just catching up on all the things | 14:03 |
| cloudnull | ice cream sounds good though | 14:03 |
| *** klamath has joined #openstack-ansible | 14:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Switch from dict to individual variables https://review.openstack.org/312506 | 14:04 |
| mhayden | i knocked the workflow+1 off this one when i rebased -- if someone could add it back, i'd be much obliged :) https://review.openstack.org/#/c/312580/ | 14:05 |
| odyssey4me | cloudnull can you quickly do a run through on https://review.openstack.org/#/q/status:open+project:openstack/openstack-ansible+topic:bug/1577469 ? | 14:05 |
| cloudnull | odyssey4me: looking now | 14:06 |
| *** michaelgugino has joined #openstack-ansible | 14:08 | |
| *** jthorne_ has quit IRC | 14:09 | |
| cloudnull | anyone want to give https://review.openstack.org/#/c/277199/ another look ? | 14:09 |
| *** jthorne has joined #openstack-ansible | 14:10 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312580 | 14:10 |
| cloudnull | i think with the SSL patch we've deterimed that it all works as intended. so it'd be good to get that in so we have reps on it | 14:12 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Switch from dict to individual variables https://review.openstack.org/312980 | 14:12 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312982 | 14:13 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312982 | 14:13 |
| openstackgerrit | Merged openstack/openstack-ansible: Add release note for paramiko issue workaround https://review.openstack.org/312953 | 14:17 |
| openstackgerrit | Merged openstack/openstack-ansible: Add release note for paramiko issue workaround https://review.openstack.org/312954 | 14:17 |
| michaelgugino | mhayden: Have you been able to prevent services from starting during install during 16.04? invoke-rc.d seems to be summarily ignored. | 14:17 |
| *** jayc__ is now known as jayc | 14:20 | |
| mhayden | oh you've gotta be kidding me -- ubuntu is auto-starting with systemd too? | 14:20 |
| * mhayden vents | 14:20 | |
| * mhayden googles | 14:22 | |
| *** Ger-chervyak has quit IRC | 14:22 | |
| prometheanfire | git-harry: can you make fun of odyssey4me for me | 14:22 |
| prometheanfire | https://github.com/openstack/openstack-ansible/blob/11.2.14/playbooks/inventory/group_vars/all.yml#L17 | 14:22 |
| prometheanfire | that makes me sad :( | 14:22 |
| *** Ger-chervyak has joined #openstack-ansible | 14:22 | |
| mhayden | michaelgugino: i'm asking in #systemd | 14:23 |
| prometheanfire | git-harry: I'm trying to build 11.2.13 now to see if that works | 14:24 |
| mattt | smatzek: so reverting https://github.com/openstack/openstack-ansible-os_nova/commit/1f63049542c5c3b50fbdf31228151af40a86cd0c (with similar reverts in os_glance and os_cinder) will work for you? | 14:24 |
| *** Mudpuppy has joined #openstack-ansible | 14:25 | |
| git-harry | prometheanfire: Using 11.2.13 probably will work but I'd prefer to wait for 11.2.15 | 14:25 |
| odyssey4me | prometheanfire will be tagging 11.2.15 today, or at least requesting the tag today | 14:25 |
| *** jaypipes has joined #openstack-ansible | 14:25 | |
| mhayden | michaelgugino: got linked to https://freedesktop.org/wiki/Software/systemd/Preset/ | 14:26 |
| odyssey4me | and yes, I make mistakes and forget to edit that file from time to time :p | 14:26 |
| michaelgugino | I joined that chan to see the discussion, thanks. I'll dig into that and let you know what I find. | 14:26 |
| odyssey4me | in that particular case I completely forgot to SHA bump :/ | 14:26 |
| smatzek | mattt: yes at first glance it does look like that would fix it. | 14:27 |
| *** spotz_zzz is now known as spotz | 14:27 | |
| *** krotscheck has quit IRC | 14:27 | |
| prometheanfire | odyssey4me: thanks :D | 14:28 |
| cloudnull | odyssey4me: maybe we need to roll up https://github.com/openstack/openstack-ansible/blob/master/scripts/update-revision.sh into the source banch update script | 14:28 |
| prometheanfire | odyssey4me: I know, I' | 14:28 |
| prometheanfire | odyssey4me: I know, I'm just not in the uk to poke at you for it myself :P | 14:29 |
| cloudnull | that script updates the revisions and while it does almost nothing in master, mitaka, it did more in liberty and kilo | 14:29 |
| evrardjp | mhayden thanks for the Preset link | 14:29 |
| mhayden | de nada | 14:30 |
| mhayden | hopefully it works | 14:30 |
| cloudnull | sorry kilo and juno it did more | 14:30 |
| odyssey4me | cloudnull I've been thinking if adding a function that automagically figures out the next version, but I haven't had the time. In this particular case I didn't bump SHA's at all... I left the task to do it until the morning, then promptly forgot | 14:30 |
| mhayden | i'm spinning up a 16.04 to tinkey | 14:30 |
| mhayden | tinker, even | 14:30 |
| odyssey4me | heh, and again we use automagically's name in vain | 14:30 |
| evrardjp | shouldn't ansible handle this correctly mhayden ? | 14:31 |
| mhayden | ansible doesn't know the difference, IIRC | 14:31 |
| mhayden | ubuntu/debian are the only distros i know of that start daemons immediately before you can even configure them :/ | 14:31 |
| evrardjp | yeah, that sucks | 14:32 |
| *** krotscheck has joined #openstack-ansible | 14:33 | |
| cloudnull | michaelgugino: does writing a /usr/sbin/policy-rc.d file work to block service auto startups? | 14:34 |
| michaelgugino | mhayden: evrardjp Ansible doesn't know what it's doing | 14:34 |
| michaelgugino | cloudnull: no | 14:34 |
| cloudnull | :'( | 14:34 |
| evrardjp | michaelgugino it's maybe time to improve the service module then | 14:34 |
| mhayden | i'll have to lengthen my rant blog post | 14:34 |
| michaelgugino | My concern though is there might be a script in the mariadb-galera-server package that is calling to /usr/bin/mysqld directly, as opposed to using systemctl et al | 14:34 |
| evrardjp | link mhayden ? | 14:34 |
| mhayden | evrardjp: https://major.io/2014/06/26/install-debian-packages-without-starting-daemons/ | 14:35 |
| cloudnull | I had thought deb-systemd-invoke would do the right thing. | 14:35 |
| cloudnull | but then again this is stemd we're talking about | 14:35 |
| * cloudnull nudges mhayden and rackertom | 14:36 | |
| * mhayden puts his arm around cloudnull | 14:36 | |
| michaelgugino | deb-systemd-invoke is installed, seems to do nothing | 14:36 |
| cloudnull | but then again this is ubuntu we're talking about... | 14:37 |
| cloudnull | :D | 14:37 |
| * mhayden hands cloudnull some crayons | 14:37 | |
| cloudnull | i love crayons | 14:37 |
| mhayden | okay it's not that bad | 14:37 |
| mhayden | i'll adjust my comment | 14:37 |
| * mhayden hands cloudnull some colored pencils | 14:38 | |
| michaelgugino | I started digging through the deb build scripts that mariadb has on github. It's a mess and I can't make sense of it because I've never built a .deb (I'm a RHEL guy) | 14:38 |
| *** mageshgv has joined #openstack-ansible | 14:38 | |
| cloudnull | colored pencils are not as cool as crayons | 14:38 |
| * odyssey4me hands cloudnull some colored permanent markers | 14:38 | |
| odyssey4me | :D | 14:38 |
| cloudnull | are they the smelly kind? | 14:38 |
| * cloudnull walks off | 14:39 | |
| * mhayden sniffs | 14:39 | |
| michaelgugino | well, presets failed for mariadb-galera-server >:O | 14:40 |
| cloudnull | https://www.youtube.com/watch?v=0N-zS92_dUw | 14:40 |
| mhayden | i think it's something in dpkg land that is overriding systemd | 14:40 |
| michaelgugino | They may actually work for packages that are packaged by someone other than mariadb | 14:40 |
| evrardjp | michaelgugino that's sad | 14:40 |
| mhayden | yeah, i just installed nginx and it started immediately :/ | 14:42 |
| michaelgugino | bah. This is the problem with systemd. It probably works, but nobody knows how to call their programs correctly | 14:43 |
| mhayden | michaelgugino: actually, this is ubuntu using systemd improperly | 14:43 |
| mhayden | this doesn't happen on fedora, centos, rhel, arch... | 14:43 |
| michaelgugino | right, the problem is with people not using it correctly, that's what I said ;) | 14:44 |
| cloudnull | i just did this on suse, seems to work fine too. | 14:44 |
| cloudnull | so its a ubuntu problem | 14:44 |
| cloudnull | likely a debian problem | 14:45 |
| odyssey4me | cores - need some reviews on https://review.openstack.org/312365 and https://review.openstack.org/312366 please | 14:45 |
| michaelgugino | could be. | 14:45 |
| mhayden | michaelgugino: WOW | 14:45 |
| mhayden | i masked nginx, installed it, and it's running | 14:45 |
| mhayden | so dpkg basically tells systemd to get lost | 14:45 |
| mhayden | and does what it wants | 14:45 |
| michaelgugino | I tried disable * | 14:45 |
| deadnull | I don't like when distributions try to do things for the admin... tsk tsk | 14:46 |
| michaelgugino | The problem I'm running into is that the config files for galera cluster cause the initial boot to hang, because the server is trying to join a cluster that doesn't exist and cannot connect. | 14:47 |
| mhayden | so dpkg is forcefully removing the marked unit file and starting the daemon anyway | 14:47 |
| * mhayden is impressed | 14:47 | |
| palendae | mhayden: impressive | 14:48 |
| *** chhavi has quit IRC | 14:48 | |
| mhayden | sheesh | 14:48 |
| michaelgugino | But, the processes also don't seem to exit correct :/, and I've got a half functioning mysql db running in the background | 14:48 |
| mhayden | also, an apt-get remove doesn't stop the service | 14:48 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-security: Add dependencies for paramiko 2.0 https://review.openstack.org/313007 | 14:50 |
| *** sdake has joined #openstack-ansible | 14:50 | |
| evrardjp | mattt mhayden https://review.openstack.org/#/c/313007/ | 14:51 |
| cloudnull | anyone ask in #ubuntu yet? | 14:52 |
| mhayden | evrardjp: got it! | 14:52 |
| evrardjp | thanks | 14:52 |
| cloudnull | ill float the question out there to see if anyone knows the internal black magic | 14:53 |
| * mhayden takes a deep breath and joins #ubuntu | 14:54 | |
| *** gravi has joined #openstack-ansible | 14:54 | |
| *** iceyao has quit IRC | 14:54 | |
| jamesdenton | Supreme Leader Ubuntu knows what is best for you, mhayden | 14:54 |
| mhayden | hah | 14:54 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Updates all SHAs for 12.0.12 https://review.openstack.org/309511 | 14:54 |
| palendae | Should have asked Shuttleworth while you had him in the elevator | 14:55 |
| Mudpuppy | O.o | 14:55 |
| *** jorge_munoz_ has joined #openstack-ansible | 14:56 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Check for two IP addresses assigned to same host https://review.openstack.org/308052 | 14:58 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Test and refactor argument parsing https://review.openstack.org/312665 | 14:58 |
| michaelgugino | mystery solved mhayden | 14:59 |
| *** jorge_munoz has quit IRC | 14:59 | |
| *** jorge_munoz_ is now known as jorge_munoz | 14:59 | |
| michaelgugino | No #!/bin/bash at the top of invoke-rc.d | 14:59 |
| *** Brew has joined #openstack-ansible | 15:00 | |
| cloudnull | mhayden: i think the answer is no in the ubuntu channel, as in they have no idea. | 15:00 |
| *** jthorne has quit IRC | 15:00 | |
| *** jthorne has joined #openstack-ansible | 15:01 | |
| *** gravi has quit IRC | 15:02 | |
| mhayden | cloudnull: can we just switch to centos as the default | 15:02 |
| odyssey4me | michaelgugino if we use the package in the ubuntu repo (instead of downloading from mariadb), does it improve the situation? | 15:02 |
| cloudnull | yes, if by centos you mean anything but im +2 | 15:02 |
| *** phalmos has joined #openstack-ansible | 15:03 | |
| mhayden | cloudnull: lolwut | 15:03 |
| mhayden | michaelgugino: ORLY? | 15:04 |
| mhayden | seems dirty | 15:04 |
| cloudnull | odyssey4me: no the package doesn't make a difference. it seems to be changes in the way debian wraps systemd | 15:04 |
| cloudnull | michaelgugino: so removing https://github.com/openstack/openstack-ansible-galera_server/blob/master/files/policy-rc.d#L1 makes it happy? | 15:05 |
| cloudnull | or does it need to be "bash" specific ? | 15:05 |
| palendae | That's policy-rc.d | 15:06 |
| cloudnull | ah, yea. my mistake | 15:06 |
| *** BjoernT has quit IRC | 15:09 | |
| *** daneyon has joined #openstack-ansible | 15:09 | |
| michaelgugino | The fix is echo exit 101 > /usr/sbin/policy-rc.d | 15:11 |
| michaelgugino | odyssey4me: still no galera compiled in in 10.0 in ubuntu's repos. | 15:11 |
| odyssey4me | michaelgugino yeah, but there are add-on packages right? | 15:11 |
| cloudnull | michaelgugino: so we should be able to remove https://github.com/openstack/openstack-ansible-galera_server/blob/master/files/policy-rc.d#L1-L2 and make mariadb/galera happy | 15:12 |
| mhayden | looks like i found the logic in init-system-helpers | 15:13 |
| cloudnull | we'd need to do the same in rabbit_server too https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/files/policy-rc.d#L1-L2 | 15:13 |
| mhayden | michaelgugino: https://anonscm.debian.org/git/collab-maint/init-system-helpers.git/tree/autoscripts/postinst-systemd-start | 15:14 |
| mhayden | huge WTF on line 3 | 15:14 |
| cloudnull | mhayden: your ubuntu is making me sad | 15:15 |
| mhayden | but it looks like it's still checking for the policy file -> https://anonscm.debian.org/git/collab-maint/init-system-helpers.git/tree/script/deb-systemd-invoke | 15:15 |
| mhayden | check line 60 | 15:15 |
| michaelgugino | right, deb-systemd-invoke is supposed to respect the file | 15:16 |
| michaelgugino | odyssey4me: I believe mariadb-galera-server is source patched, so I don't think there's simply an addon we can use from the repos to get the cluster to work | 15:16 |
| mhayden | hmm, i got the policy-rc.d to work | 15:16 |
| michaelgugino | me too. I had to make it that one liner from the linked page | 15:17 |
| michaelgugino | https://jpetazzo.github.io/2013/10/06/policy-rc-d-do-not-start-services-automatically/ | 15:19 |
| mhayden | michaelgugino: https://gist.github.com/major/88c708684e29aaecba63dca83e2b23f0 | 15:19 |
| *** schwicht has quit IRC | 15:20 | |
| mhayden | that seems to work for me | 15:20 |
| *** weezS has joined #openstack-ansible | 15:20 | |
| michaelgugino | yeah, me too | 15:20 |
| mhayden | this still feels so silly for something that should be a default | 15:20 |
| *** schwicht has joined #openstack-ansible | 15:20 | |
| michaelgugino | I completely agree. | 15:21 |
| palendae | Yeah | 15:21 |
| michaelgugino | wasted a whole afternoon on this yesterday. | 15:21 |
| palendae | Hasn't Debian/Ubuntu been a terrible offender of just starting services on install since inception though? | 15:21 |
| michaelgugino | yes, they have | 15:21 |
| cloudnull | so then we need to modify https://github.com/openstack/openstack-ansible-galera_server/blob/master/files/policy-rc.d#L1-L2 and https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/files/policy-rc.d#L1-L2 to use bash instead of sh | 15:22 |
| palendae | They're just making it harder to fix with systemd I guess | 15:22 |
| *** basilAB has quit IRC | 15:23 | |
| michaelgugino | cloudnull: I've already got a huge patch in the works, I'll work that in | 15:23 |
| *** basilAB has joined #openstack-ansible | 15:23 | |
| cloudnull | cool | 15:23 |
| *** sacharya has joined #openstack-ansible | 15:24 | |
| *** mummer has joined #openstack-ansible | 15:24 | |
| *** sacharya has quit IRC | 15:29 | |
| mhayden | michaelgugino: it actually looks like systemd masks are workig | 15:29 |
| mhayden | i did a ln -s /dev/null /etc/systemd/system/apache2.service | 15:29 |
| mhayden | then installed apache2 | 15:29 |
| mhayden | and it didn't start | 15:29 |
| michaelgugino | very nice. | 15:30 |
| mhayden | that could be easier | 15:30 |
| michaelgugino | you should blog these approaches ;) | 15:30 |
| mhayden | i guess i should | 15:30 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server: Reorganize test playbooks https://review.openstack.org/312302 | 15:31 |
| pjm6 | guys I have a doubt | 15:32 |
| michaelgugino | I'll try the service masking for mariadb-galera. That is still dropping an init file, so I'm not sure if systemd masking will work. | 15:32 |
| pjm6 | when we use launch an instance | 15:32 |
| pjm6 | we have a private IP and then we can use a floating IP | 15:32 |
| *** karimb has joined #openstack-ansible | 15:33 | |
| pjm6 | my question is why when we add an bridge interface in the ETH0 | 15:33 |
| pjm6 | we lost connectivity of info? | 15:33 |
| pjm6 | its because the interface eth0 now will be the br0 | 15:33 |
| pjm6 | and this breaks the neutron configuration? | 15:33 |
| pjm6 | the point is to make a openvpn server that will allow users to connect to the private network | 15:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Add dependencies for paramiko 2.0 https://review.openstack.org/313007 | 15:34 |
| odyssey4me | mhayden evrardjp ^ I added the missing bits to make it go | 15:34 |
| michaelgugino | let's get some quick reviews on this one: https://review.openstack.org/#/c/312808/ | 15:35 |
| michaelgugino | fixes a bug with linters test | 15:35 |
| odyssey4me | michaelgugino yeah, saw that - was wondering about it because palendae has something of a similar line | 15:37 |
| odyssey4me | michaelgugino I think it's https://review.openstack.org/308052 | 15:37 |
| palendae | Sounds like there's still confusion | 15:38 |
| michaelgugino | odyssey4me: that's already merged I thought | 15:38 |
| palendae | No, multiple host names to one IP is merged | 15:38 |
| palendae | Multiple IPs to one host name is awaiting merge | 15:38 |
| michaelgugino | https://github.com/openstack/openstack-ansible/blob/master/playbooks/inventory/dynamic_inventory.py#L52 | 15:39 |
| palendae | michaelgugino: Multiple Hosts, One IP | 15:39 |
| palendae | Not one host, multiple IPs | 15:39 |
| palendae | 2 different problems, but similar | 15:39 |
| *** galstrom_zzz is now known as galstrom | 15:39 | |
| michaelgugino | I think two ip's assigned to the same host is applicable in some cases | 15:39 |
| odyssey4me | not in our inventory | 15:40 |
| michaelgugino | true | 15:40 |
| odyssey4me | well, to be clear - not two ip's to one host in the openstack_user_config or conf.d files | 15:40 |
| palendae | In any case, those are the two scenarios; your patch is for the multiple hosts with the same IP. It's valid, but the commit message states it's for duplicate IPs | 15:40 |
| michaelgugino | I'm following you guys, let me check the commit msg | 15:41 |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible: Fix dynamic_inventory.py test bug https://review.openstack.org/312808 | 15:43 |
| *** Zucan has joined #openstack-ansible | 15:43 | |
| michaelgugino | palendae: You should implement the same fix in your test | 15:43 |
| palendae | Not sure I'm following; the dictionary insert order shouldn't matter if you're looking up by the specific key | 15:43 |
| michaelgugino | inventory is a dictionary, it's not guaranteed which IP or hostname will show up first. | 15:43 |
| palendae | Right... | 15:44 |
| palendae | Tests are using ordered dicts though | 15:44 |
| palendae | Using that .sort method doesn't necessarily guarantee accurate order, either | 15:44 |
| palendae | It'd do alphanumeric sort | 15:44 |
| palendae | Which may not reflect what actually got passed in to the function | 15:45 |
| michaelgugino | we only need to it sort for the test | 15:45 |
| michaelgugino | https://review.openstack.org/#/c/311802/ | 15:45 |
| michaelgugino | you'll see the linter check failed there | 15:45 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Enable IPv6 support in Horizon https://review.openstack.org/312741 | 15:45 |
| michaelgugino | AssertionError: 'ip address:172.29.236.100 has already been assigned to host:hap, cannot assign same ip to host:aio1' != 'ip address:172.29.236.100 has already been assigned to host:aio1, cannot assign same ip to host:hap' | 15:45 |
| *** jayc_ has joined #openstack-ansible | 15:45 | |
| palendae | Ok; if it's ordering for the test, it shouldn't be in the exception itself | 15:45 |
| michaelgugino | I think putting it in the exception is the most obvious place | 15:46 |
| *** mikelk has quit IRC | 15:46 | |
| odyssey4me | git-harry FYI Kilo tag request is done: https://review.openstack.org/313028 | 15:48 |
| *** neilus has quit IRC | 15:48 | |
| *** jayc has quit IRC | 15:49 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Enable ha router support in Horizon https://review.openstack.org/313031 | 15:49 |
| palendae | Oh | 15:52 |
| *** eil397 has joined #openstack-ansible | 15:52 | |
| palendae | That needs this part of the change https://review.openstack.org/#/c/308052/14/tox.ini | 15:52 |
| palendae | That makes the dictionaries reliable | 15:52 |
| *** fauxbert has quit IRC | 15:54 | |
| *** saneax is now known as saneax_AFK | 15:59 | |
| palendae | michaelgugino: Have a link to the actual Jenkins log for the failure? Gerrit's not showing it now | 15:59 |
| palendae | I'd like to see what PYTHONHASHSEED it got | 16:00 |
| *** jmccrory_ has joined #openstack-ansible | 16:00 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Enable next gneration instance launch in Horizon https://review.openstack.org/313037 | 16:01 |
| michaelgugino | palendae: http://logs.openstack.org/02/311802/2/gate/gate-openstack-ansible-linters/2751e60/console.html | 16:02 |
| michaelgugino | linters runtests: PYTHONHASHSEED='681289476' | 16:03 |
| palendae | Thanks | 16:03 |
| michaelgugino | I was reading somewhere recently that the hashseed can be the same, but if the underlying system isn't, you'll get unpredictable outcomes | 16:03 |
| palendae | Right | 16:03 |
| palendae | Can be | 16:03 |
| odyssey4me | meeting in #openstack-meeting-4 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, | 16:04 |
| odyssey4me | ametts, v1k0d3n, severion, bgmccollum, darrenc, JRobinson__, asettle | 16:04 |
| palendae | But changing that sort won't be reliable either | 16:04 |
| palendae | If the systems weren't read in alphanumerical order, that sort will be backwards | 16:04 |
| palendae | e.g. zhost was processed first but ahost came second, that sort will say ahost was the first one | 16:04 |
| michaelgugino | palendae: sort will be reliable. That's why I created a list and sorted it | 16:04 |
| michaelgugino | that's what we want, for it to say that ahost was the same one | 16:04 |
| michaelgugino | *first one | 16:04 |
| palendae | ahost came second | 16:05 |
| palendae | In the processing | 16:05 |
| michaelgugino | there is no second in a dictionary | 16:05 |
| palendae | No, but there is in string | 16:05 |
| michaelgugino | as long as the user has both hostnames, they should be able to sort it out | 16:05 |
| palendae | With stated arguments | 16:05 |
| *** timl0101 has joined #openstack-ansible | 16:05 | |
| *** tiagogomes has quit IRC | 16:06 | |
| michaelgugino | it's the first host only in the context that it was randomly selected from the dictionary as the first host | 16:06 |
| palendae | Sure | 16:06 |
| michaelgugino | perhaps the error string should be updated as well to be more clear | 16:07 |
| palendae | Yeah, by doing that .sort you're making the string inaccurate | 16:07 |
| palendae | It conveys that it parsed ahost first in that scenario, when it parsed zhost | 16:08 |
| palendae | You're right in that it probably doesn't matter a ton | 16:08 |
| palendae | Also, yeah, the python hash seed can be set the same and the system will behave differently; locally my tests pass with the failing seed | 16:09 |
| palendae | I'd be for correcting the error message to simply say both hosts were assigned the same IP, drop any statement of order in the text | 16:09 |
| michaelgugino | yeah, these kinds of bugs suck ;) | 16:09 |
| palendae | They do | 16:10 |
| palendae | I'd like to replace everything internally with ordered structures | 16:10 |
| palendae | But not wholesale quite yet | 16:10 |
| *** karimb has quit IRC | 16:10 | |
| palendae | Even just an OrderedDict | 16:10 |
| michaelgugino | no need for ordered, we want this dynamic_inventory.py script to evolve into something else, no sense in refactoring something that will hopefully go away | 16:11 |
| palendae | It'll evolve, yes, but we'll always have some kind of ordered tests | 16:11 |
| mhayden | michaelgugino: oops, i woke up jkeating :) https://twitter.com/iamjkeating/status/728255588849745920 | 16:11 |
| palendae | And all code should go away :) | 16:11 |
| Nepoc | Greetings, could someone check their openstack-ansible liberty install if they are able to select the zh-cn language for their user and actually have it update the entire UI? Currently it only changes dates and one column name for me.. However it works fine in dev-stack | 16:11 |
| palendae | we refactor anyway | 16:11 |
| *** Ger-chervyak has quit IRC | 16:12 | |
| michaelgugino | lol. I'm not familiar with that person mhayden | 16:12 |
| michaelgugino | Rub some docker on them sounds like a plan though. | 16:12 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Automatically enable neutron ha router capabilities https://review.openstack.org/313042 | 16:13 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Enable next gneration instance launch in Horizon https://review.openstack.org/313037 | 16:17 |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_gnocchi: Update SHAs and config templates https://review.openstack.org/312774 | 16:21 |
| mhayden | michaelgugino: he's a former rackspace guy who works for "Bluebox, an IBM company" :) | 16:23 |
| mhayden | he's super talented with ansible | 16:23 |
| michaelgugino | ah, cool. | 16:23 |
| michaelgugino | I avoid twitter. Too much fluff | 16:24 |
| mhayden | yeah, i don't tweet much | 16:24 |
| * mhayden winks | 16:24 | |
| palendae | mhayden: Ask him when Bluebox switched to Kolla | 16:26 |
| mhayden | wait, they did? | 16:26 |
| palendae | ;) | 16:26 |
| palendae | If he wants to docker everything! | 16:26 |
| stevelle | yeah, he says 'container' but seems to mean 'docker' | 16:27 |
| palendae | Definitely docker mentality if nothing else | 16:27 |
| stevelle | cuz the daemon still starts immediately if you install it in an lxc container | 16:27 |
| stevelle | :D | 16:27 |
| palendae | All process ids must live in their own docker container! | 16:28 |
| palendae | process id will always be 2! | 16:28 |
| *** sdake_ has joined #openstack-ansible | 16:28 | |
| *** daneyon has quit IRC | 16:30 | |
| *** sdake has quit IRC | 16:31 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Mention of the supported locales in the documentation https://review.openstack.org/311058 | 16:38 |
| *** sacharya has joined #openstack-ansible | 16:39 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Check for two IP addresses assigned to same host https://review.openstack.org/308052 | 16:40 |
| palendae | michaelgugino: ^ uses the sort for the IP address error now | 16:40 |
| *** david-lyle has joined #openstack-ansible | 16:42 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Enable next gneration instance launch in Horizon https://review.openstack.org/313037 | 16:42 |
| Nepoc | Hello, not sure if anyone read my message but I've been talking to the openstack-horizon people and they seem to think that openstack-ansible is not running the compilemessage when building horizon. | 16:43 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Automatically enable neutron ha router capabilities https://review.openstack.org/313042 | 16:44 |
| cloudnull | hi Nepoc | 16:44 |
| michaelgugino | palendae: exactly what I was thinking re the log message itself | 16:44 |
| Nepoc | Hi cloudnull | 16:45 |
| cloudnull | Nepoc: sorry im missing context, what do you mean "compilemessage"? | 16:45 |
| cloudnull | sorry to make you repeat | 16:45 |
| Nepoc | No problem | 16:45 |
| Nepoc | I am currently unable to change languages in Horizon. For example if I select simplified chinese it will only change dates and a single column "e-mail" on the User page. | 16:46 |
| Nepoc | Users list I should say | 16:47 |
| palendae | I think that's part of django's translation layer https://docs.djangoproject.com/en/1.9/topics/i18n/translation/ | 16:47 |
| Nepoc | Yes | 16:47 |
| palendae | https://docs.djangoproject.com/en/1.9/ref/django-admin/#compilemessages | 16:47 |
| palendae | Which probably relates to evrardjp's patch at https://review.openstack.org/311058 | 16:47 |
| Nepoc | Ah :( | 16:48 |
| palendae | But that's a good point; any work to support future locales needs to, for example, run compilemessages | 16:48 |
| Nepoc | Okay now to figure out how to make this work | 16:48 |
| evrardjp | palendae I'm glad I became an oracle that can predict the future issues. | 16:49 |
| Nepoc | I have to have Chinese enabled in Horizon by the end of the week | 16:49 |
| evrardjp | too bad I'm not. | 16:49 |
| palendae | evrardjp: :p | 16:49 |
| palendae | Nepoc: Just Horizon? | 16:49 |
| Nepoc | That would be enough | 16:49 |
| evrardjp | Nepoc that's not a short deadline right? | 16:49 |
| evrardjp | including console? | 16:49 |
| Nepoc | No console doesn't have to be | 16:49 |
| palendae | As a high level idea, could set a variable to take a locale identifier, if specified then run the compilemessages command | 16:50 |
| evrardjp | Nepoc is your browser correctly set to chinese? | 16:50 |
| evrardjp | just to make sure | 16:50 |
| spotz | ok I tooks a few might have bitten off more then I can chew though:) | 16:50 |
| Nepoc | I have a computer from china I can test with | 16:51 |
| *** daneyon has joined #openstack-ansible | 16:54 | |
| Nepoc | palendae: When/Where should I run compilemessages? | 16:55 |
| palendae | Nepoc: That I'm not sure of. I'd assume the logic would go somewhere in these files though https://github.com/openstack/openstack-ansible-os_horizon/tree/master/tasks | 16:55 |
| Nepoc | I have tried django-admin compilemessages after it has already been installed but it complains that Settings already configured | 16:56 |
| Nepoc | specifically I did: export DJANGO_SETTINGS_MODULE=openstack_dashboard.settings ; cd /openstack/venvs/horizon-12.0.5 ; ./bin/django-admin compilemessages | 16:59 |
| Nepoc | brb | 16:59 |
| *** daneyon has quit IRC | 17:00 | |
| *** sdake_ has quit IRC | 17:00 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Use master git branches when testing https://review.openstack.org/313059 | 17:00 |
| *** eil397 has left #openstack-ansible | 17:01 | |
| *** fignew has quit IRC | 17:02 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Isolate Ansible from the deployment host https://review.openstack.org/304840 | 17:05 |
| *** javeriak has joined #openstack-ansible | 17:05 | |
| *** daneyon has joined #openstack-ansible | 17:05 | |
| *** daneyon_ has joined #openstack-ansible | 17:09 | |
| *** weezS has quit IRC | 17:10 | |
| evrardjp | I'm off, see you another day! | 17:13 |
| *** daneyon has quit IRC | 17:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Multi-distro pattern for os_heat role https://review.openstack.org/311749 | 17:14 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-repo_server: Updated role for multi-distro support https://review.openstack.org/313067 | 17:14 |
| odyssey4me | Nepoc if you can work out how to extend the support to different locales and get patches in to make that happen, then we can revert https://review.openstack.org/311058 once it merges... for now that's there to make people aware that we don't test it in any way and we haven't tried it either. | 17:15 |
| cloudnull | michaelgugino: I updated https://review.openstack.org/#/c/311058/ to remove the bind mount options. | 17:16 |
| cloudnull | odyssey4me: responded to your comments too | 17:16 |
| cloudnull | sorry https://review.openstack.org/#/c/304840/ | 17:16 |
| Nepoc | odyssey4me: Thank you for the info... hopefully I can make some magic happen. | 17:19 |
| michaelgugino | cool. I'll take a look at it more thoroughly. | 17:19 |
| openstackgerrit | Merged openstack/openstack-ansible: Updates all SHAs for 12.0.12 https://review.openstack.org/309511 | 17:20 |
| odyssey4me | Nepoc I hope so. It was through many of the interactions with yourself and others that we realised that it wasn't simple, so we opted to add the doc entry to rather set expectations. | 17:20 |
| cloudnull | who wants to push the button on https://review.openstack.org/#/c/277199/ ;) | 17:22 |
| *** kiranv has joined #openstack-ansible | 17:22 | |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible: Fix dynamic_inventory.py test bug https://review.openstack.org/312808 | 17:23 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-repo_server: Updated role for multi-distro support https://review.openstack.org/313067 | 17:23 |
| *** thorst has quit IRC | 17:23 | |
| *** saneax_AFK is now known as saneax | 17:24 | |
| *** thorst has joined #openstack-ansible | 17:24 | |
| *** gluytium has quit IRC | 17:24 | |
| michaelgugino | palendae: I updated that error message just for you ;) | 17:25 |
| cloudnull | Nepoc: so if you run the following ``pushd /openstack/venvs/horizon-master/lib/python2.7/site-packages/horizon/; /openstack/venvs/horizon-master/bin/horizon-manage.py compilemessages; popd`` it should work ? | 17:25 |
| *** jang has joined #openstack-ansible | 17:25 | |
| cloudnull | assuming you have the locals set? | 17:25 |
| cloudnull | also assuming you're on master | 17:26 |
| cloudnull | # /openstack/venvs/horizon-${RELEASE_VERSION}/bin/horizon-manage.py compilemessages | 17:26 |
| *** thorst has quit IRC | 17:28 | |
| michaelgugino | mhayden: ln -s /dev/null /etc/systemd/system/mysql.service does not work for mariadb-galera-server, but the modified policy-rc.d method does. This will only be an issue for debian/ubuntu based hosts anyway, RPM distros don't start things automatically (imagine that) | 17:30 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Use master git branches when testing https://review.openstack.org/313059 | 17:32 |
| Nepoc | cloudnull: I'll try it out, I'm on liberty | 17:35 |
| *** Ger-chervyak has joined #openstack-ansible | 17:37 | |
| Nepoc | cloudnull: Well there is now 3 more things with translations | 17:39 |
| cloudnull | Nepoc: i also found https://github.com/openstack/horizon/blob/master/openstack_dashboard/settings.py#L198-L220 | 17:40 |
| Nepoc | It certainly created the .mo file | 17:40 |
| cloudnull | you might need to set https://github.com/openstack/horizon/blob/master/openstack_dashboard/settings.py#L216 | 17:40 |
| *** Ger-chervyak has quit IRC | 17:40 | |
| Nepoc | mmm ok | 17:41 |
| Nepoc | I'll try that | 17:41 |
| *** javeriak_ has joined #openstack-ansible | 17:44 | |
| *** javeriak has quit IRC | 17:46 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-repo_server: Updated role for multi-distro support https://review.openstack.org/313067 | 17:47 |
| *** fignew has joined #openstack-ansible | 17:47 | |
| *** jayc_ has quit IRC | 17:49 | |
| Nepoc | There we go... I guess the translation is very limited... I swear it was 100% translated in the past | 17:50 |
| Nepoc | The Compute Overview page has the most translation done to it | 17:51 |
| *** itlinux has joined #openstack-ansible | 17:52 | |
| cloudnull | Nepoc: it seems that the compile command needs to be run in "/openstack/venvs/horizon-${RELEASE_VERSION}/lib/python2.7/site-packages/openstack_dashboard" and "/openstack/venvs/horizon-${RELEASE_VERSION}/lib/python2.7/site-packages/horizon" | 17:58 |
| cloudnull | while related they are different :\ | 17:58 |
| Nepoc | okay I'll do that too, thank you | 17:58 |
| *** chandanc has joined #openstack-ansible | 17:59 | |
| *** electrofelix has quit IRC | 17:59 | |
| *** pjm6 has quit IRC | 18:00 | |
| *** jayc_ has joined #openstack-ansible | 18:03 | |
| *** thorst_ has joined #openstack-ansible | 18:04 | |
| *** weezS has joined #openstack-ansible | 18:04 | |
| Nepoc | There we go! | 18:05 |
| Nepoc | After running that and an apache restart we are in business | 18:05 |
| Nepoc | cloudnull: Thank you for the help | 18:06 |
| Nepoc | Now just need to do that on install | 18:07 |
| cloudnull | sweet! | 18:07 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Updates all SHAs for 11.2.16 https://review.openstack.org/313091 | 18:08 |
| odyssey4me | Nepoc I expect to see a patch in for that soon. :) | 18:09 |
| Nepoc | odyssey4me: I will see what I can do, but I promise nothing! :) | 18:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Updates all SHAs for 12.0.13 https://review.openstack.org/313094 | 18:13 |
| Nepoc | Another question today for something that I'm hopefully not going to be working on for a couple weeks. Can I get openstack-ansible to install "some" lxc compute nodes? | 18:13 |
| odyssey4me | Nepoc maybe - michaelgugino and I were talking about doing a test implementation of the nova-lxd driver and implementing that as an optional hypervisor | 18:14 |
| michaelgugino | we're going to be working on that very soon. | 18:15 |
| Nepoc | I would need to have say 10 KVM comptue and 2 LXC compute... would that address that scenario? | 18:15 |
| michaelgugino | I'll let my associate know to put it on the top of his list for next week. | 18:15 |
| michaelgugino | Nepoc: yes, we plan on doing a mixed environment. | 18:15 |
| Nepoc | Great :) | 18:16 |
| Nepoc | I can be your Guinea pig | 18:16 |
| odyssey4me | it'll be easy enough to implement the hypervisor, but to have it work in the mixed environment will possibly take a little openstack experience as it'll involve working with scheduler filters, host aggregates and such I would guess | 18:16 |
| odyssey4me | but getting the bits and the right config down is a good starting point | 18:17 |
| Nepoc | The potential downside is that I'm running 12.0.5 (not sure what the latest is at this point) | 18:17 |
| michaelgugino | odyssey4me: I think it should be transparent. You just tag images with the hypervisor type, and that should be it. It will schedule the correct hypervisors. | 18:17 |
| michaelgugino | of course, this is all theory on our end, at this point. We'll have something POC soon, hopefully. | 18:18 |
| odyssey4me | Nepoc we'll aim for inclusion in Newton only. Backporting that to Liberty will be something you'll have to take the risk on and do yourself. There are no guarantees on that working. | 18:18 |
| michaelgugino | What I was told was that mixed environments is supported in Mitaka. I'm looking for BP to mitaka ;) | 18:19 |
| odyssey4me | backport to Mitaka can be considered depending on the amount of code churn it takes and how comfortable everyone is with the implementation | 18:19 |
| *** jmccrory_ has quit IRC | 18:19 | |
| michaelgugino | but, that's not up to me. I think it should be marked experimental, of course. | 18:19 |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-galera_server: [WIP] Implement Xenial Support https://review.openstack.org/312602 | 18:20 |
| *** TxGirlGeek has joined #openstack-ansible | 18:21 | |
| Nepoc | Fun times... So would you say I'm "crazy" if I were to try and get mixed environments to run in liberty? | 18:21 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-repo_server: Updated role for multi-distro support https://review.openstack.org/313067 | 18:23 |
| *** retreved_ has joined #openstack-ansible | 18:25 | |
| cloudnull | Nepoc: no | 18:27 |
| cloudnull | you can set the vars currently in the openstack_user_config there may be some dragons there but (in theory) it should work . | 18:28 |
| odyssey4me | Nepoc not crazy - it's just that the project won't be actively working on supporting it natively considering that Liberty is moving into the 'security-only' fix state of the life cycle | 18:28 |
| *** retreved has quit IRC | 18:28 | |
| odyssey4me | Nepoc we'd rather work on helping you get to Mitaka through an upgrade process :) | 18:28 |
| *** retreved_ has quit IRC | 18:31 | |
| Nepoc | Yeah it's not that simple, we have product that is integrated with openstack. Unfortunately our development is a lot slower paced than openstack. We usually skip a version of openstack with our releases. I've also customized the openstack-ansible installer to install our product on it as well. | 18:31 |
| *** eil397 has joined #openstack-ansible | 18:31 | |
| *** retreved has joined #openstack-ansible | 18:31 | |
| odyssey4me | Nepoc well, hopefully you can upstream a lot of the customisations to make that easier to do | 18:31 |
| Nepoc | Unforunately 90% can't be sent back up... useless to everyone else. | 18:32 |
| *** TxGirlGeek has quit IRC | 18:32 | |
| Nepoc | My next step is to try and seperate it out as much as possible. I've leart a lot so I can remove some of the customizations | 18:33 |
| odyssey4me | jmccrory automagically cloudnull d34dh0r53 stevelle mattt hughsaunders andymccr a couple of reviews need some attention https://review.openstack.org/#/q/topic:py_from_git_begone+status:open | 18:34 |
| Nepoc | I think the biggest changes were to ha-proxy, I had to completely redo the templates and everything that fed into them. I needed custom certificates for each component. | 18:34 |
| Nepoc | That might be useful to some | 18:35 |
| odyssey4me | another review that needs some votes https://review.openstack.org/313007 | 18:35 |
| d34dh0r53 | anyone else getting tons of 502s on gerrit? | 18:37 |
| odyssey4me | d34dh0r53 yep | 18:37 |
| d34dh0r53 | cool, not just me then | 18:37 |
| cloudnull | Nepoc: ++ that'd be useful | 18:39 |
| cloudnull | IMO | 18:39 |
| Nepoc | I just need to make the majority of my changes easily configurable from openstack_deploy, then it will be useful :) | 18:40 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Change pip install task state to 'latest' https://review.openstack.org/312381 | 18:43 |
| cloudnull | Nepoc: typie typie | 18:43 |
| cloudnull | :p | 18:43 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_barbican: Add .swp files to .gitignore https://review.openstack.org/312542 | 18:46 |
| Nepoc | cloudnull: haha | 18:47 |
| Nepoc | cloudnull: I've created mysefl a userstory to do all of this... makes my life easier too in the long run. then I'm not a bottleneck for upgrading to later releases. | 18:48 |
| *** sdake has joined #openstack-ansible | 18:48 | |
| cloudnull | awesome! | 18:48 |
| Nepoc | I'm very interested how an openstack-ansible upgrade will work on my systems. | 18:50 |
| Nepoc | and... lunchtime | 18:50 |
| * cloudnull afk for food too | 18:51 | |
| *** TxGirlGeek has joined #openstack-ansible | 18:52 | |
| *** sdake_ has joined #openstack-ansible | 18:55 | |
| *** sdake has quit IRC | 18:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Remove py_from_git role https://review.openstack.org/312366 | 18:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Remove py_from_git role https://review.openstack.org/312365 | 18:58 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Forward rsyslog to containers on the log_hosts group only https://review.openstack.org/313117 | 19:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically increment the patch ver with sources-branch-updater https://review.openstack.org/313118 | 19:05 |
| *** TxGirlGeek has quit IRC | 19:05 | |
| odyssey4me | prometheanfire cloudnull ^ magic now happens | 19:05 |
| prometheanfire | hi | 19:05 |
| *** Ger-chervyak has joined #openstack-ansible | 19:06 | |
| *** javeriak_ has quit IRC | 19:09 | |
| *** hybridpollo has joined #openstack-ansible | 19:09 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Test and refactor argument parsing https://review.openstack.org/312665 | 19:09 |
| *** smatzek has quit IRC | 19:15 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Add dependencies for paramiko 2.0 https://review.openstack.org/313007 | 19:15 |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Forward rsyslog to containers on the log_hosts group only https://review.openstack.org/313117 | 19:16 |
| -openstackstatus- NOTICE: Gerrit is restarting to address performance issues related to a suspected memory leak | 19:20 | |
| *** Ger-chervyak has quit IRC | 19:20 | |
| *** Ger-chervyak has joined #openstack-ansible | 19:21 | |
| *** kstev has joined #openstack-ansible | 19:24 | |
| openstackgerrit | Tim Laszlo proposed openstack/openstack-ansible: Use task state instead of output to create haproxy log directory https://review.openstack.org/313126 | 19:27 |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-os_neutron: [WIP] Implementing neutron_openvswitch_agent https://review.openstack.org/298765 | 19:31 |
| *** TxGirlGeek has joined #openstack-ansible | 19:39 | |
| *** smatzek has joined #openstack-ansible | 20:01 | |
| *** TxGirlGeek has quit IRC | 20:04 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:04 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that the sources-branch-updater updates the Ironic role https://review.openstack.org/313139 | 20:08 |
| *** Ger-chervyak has quit IRC | 20:09 | |
| *** TxGirlGeek has quit IRC | 20:09 | |
| odyssey4me | night all - time for me to get home | 20:09 |
| cloudnull | have a good one | 20:11 |
| *** Ger-chervyak has joined #openstack-ansible | 20:12 | |
| spotz | night odyssey4me | 20:12 |
| *** TxGirlGeek has joined #openstack-ansible | 20:13 | |
| palendae | Wow, not even hom? | 20:13 |
| palendae | Have a good one!~ | 20:13 |
| *** admin0 has joined #openstack-ansible | 20:16 | |
| *** chandanc has quit IRC | 20:19 | |
| *** smatzek has quit IRC | 20:20 | |
| *** Zucan has quit IRC | 20:22 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Doc: Configuring the network refactor https://review.openstack.org/312421 | 20:22 |
| *** mummer has quit IRC | 20:26 | |
| *** mummer has joined #openstack-ansible | 20:26 | |
| spotz | Let me guess I'm the only one getting 503 in gerrit? | 20:27 |
| cloudnull | nope | 20:28 |
| spotz | woot it's not just me!!!! | 20:28 |
| -openstackstatus- NOTICE: Gerrit is restarting to revert incorrect changes to test result displays | 20:28 | |
| palendae | lol | 20:28 |
| *** michaelgugino has quit IRC | 20:29 | |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Skip rsyslog configuration for juno-container-cleanup.sh https://review.openstack.org/313117 | 20:33 |
| *** sdake has joined #openstack-ansible | 20:35 | |
| *** sdake_ has quit IRC | 20:36 | |
| *** weezS has quit IRC | 20:43 | |
| openstackgerrit | Bjoern Teipel proposed openstack/openstack-ansible: Skip rsyslog configuration for juno-container-cleanup.sh https://review.openstack.org/313117 | 20:45 |
| *** TxGirlGeek has quit IRC | 20:45 | |
| *** thorst_ has quit IRC | 20:46 | |
| *** daneyon_ has quit IRC | 20:47 | |
| cloudnull | who wants to push the button on https://review.openstack.org/#/c/277199/ :) | 20:48 |
| *** schwicht has quit IRC | 20:48 | |
| cloudnull | andymccr automagically jmccrory mattt d34dh0r53 ? | 20:49 |
| cloudnull | :D | 20:49 |
| jmccrory | cloudnull: reading through it | 20:50 |
| cloudnull | jmccrory: thnx :) | 20:50 |
| *** thorst_ has joined #openstack-ansible | 20:53 | |
| *** chandanc has joined #openstack-ansible | 20:55 | |
| mrda | Morning | 20:55 |
| spotz | hey mrda | 20:56 |
| mrda | o/ | 20:56 |
| admin0 | cloudnull: he | 20:57 |
| admin0 | hey | 20:57 |
| *** thorst_ has quit IRC | 20:57 | |
| admin0 | am in vacation, but wanted to test and comment on the SSL thing .. i could not get it to work .. so was wondering if you have a working config example somewhere that i can copy/paste and try | 20:57 |
| cloudnull | admin0: my test env was just default and it seemed to work. | 21:01 |
| cloudnull | i built out a 22 node lab to test. | 21:02 |
| admin0 | yay ! | 21:02 |
| cloudnull | but if there are issues with it we should fix them :) | 21:02 |
| admin0 | i might be overdoing on the configs maybe .. so what is the minimum i should have ? | 21:02 |
| *** johnmilton has quit IRC | 21:04 | |
| *** fawadkhaliq has joined #openstack-ansible | 21:04 | |
| *** weezS has joined #openstack-ansible | 21:04 | |
| admin0 | cloudnull: also imp .. what branch should it be tested on | 21:05 |
| cloudnull | master is where i've been testing | 21:05 |
| *** saneax is now known as saneax_AFK | 21:05 | |
| admin0 | ok | 21:05 |
| admin0 | check! | 21:05 |
| admin0 | master branch, horizon_user(ssl,ca,key_cert) path | 21:06 |
| cloudnull | i've changed nothing in config | 21:06 |
| cloudnull | and let it do the self signed ssl | 21:06 |
| cloudnull | I've also done it w/ letsencrypt now too | 21:06 |
| admin0 | could it be different for self signed vs user_provided ? | 21:06 |
| cloudnull | it should not | 21:06 |
| cloudnull | i've not tried a user provided cert tbh so that may be where some of the issues are. im not sure. | 21:07 |
| admin0 | if it works iwth letsencrypt, that is also fine | 21:07 |
| admin0 | does it auto generate letsencrypt based on the hostname provided ? | 21:07 |
| admin0 | my internal points to a local IP .. i think it cannot use letsencrypt for that | 21:09 |
| *** raddaoui has joined #openstack-ansible | 21:09 | |
| *** psilvad has quit IRC | 21:11 | |
| *** thorst_ has joined #openstack-ansible | 21:12 | |
| *** smatzek has joined #openstack-ansible | 21:13 | |
| *** Ger-chervyak has quit IRC | 21:14 | |
| *** thorst_ has quit IRC | 21:17 | |
| cloudnull | it does not "work" with lets encrypt | 21:20 |
| admin0 | ok | 21:20 |
| cloudnull | but i did the deployment | 21:20 |
| cloudnull | and then replaced the cert with lets encrypt | 21:20 |
| admin0 | aha | 21:20 |
| admin0 | ok | 21:20 |
| cloudnull | so not quite the same test | 21:20 |
| admin0 | i don’t mind doing that also | 21:20 |
| admin0 | is it an easy process ? | 21:20 |
| admin0 | just change it once in haproxy right ? | 21:20 |
| admin0 | hmm.. so use master branch, and the only setting that goes to config is apply_security_hardening: false .. rest = nothing @ all ( except my ceph integration ) | 21:21 |
| cloudnull | yup | 21:22 |
| admin0 | will give it a try | 21:22 |
| cloudnull | then install lets encrypt and replace the certs | 21:22 |
| cloudnull | works well | 21:23 |
| cloudnull | use the staging certs from lets encrypt | 21:23 |
| cloudnull | then you dont have to worry about rate limits | 21:23 |
| admin0 | i have a real wildcard cert | 21:23 |
| cloudnull | ok | 21:23 |
| cloudnull | that works too | 21:23 |
| cloudnull | and likley better :) | 21:23 |
| admin0 | will give it a try | 21:23 |
| admin0 | to see if the self-signed path helps | 21:23 |
| admin0 | so what won’t be in SSL ? | 21:24 |
| cloudnull | everything public facing will use ssl termination | 21:24 |
| admin0 | ok | 21:25 |
| *** weezS has quit IRC | 21:25 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Switch from dict to individual variables https://review.openstack.org/312980 | 21:27 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Removing equal signs in docs https://review.openstack.org/312982 | 21:27 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: iniital commit to enable mitaka>newton upgrades https://review.openstack.org/312292 | 21:29 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-os_horizon: Enable ha router support in Horizon https://review.openstack.org/313031 | 21:31 |
| *** fawadkhaliq has quit IRC | 21:33 | |
| *** ametts has quit IRC | 21:34 | |
| *** tiagogomes has joined #openstack-ansible | 21:37 | |
| *** Mudpuppy has quit IRC | 21:41 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible-rabbitmq_server: install rabbitmq-server in serial https://review.openstack.org/310542 | 21:48 |
| *** jamesdenton has quit IRC | 21:49 | |
| *** sdake has quit IRC | 21:50 | |
| *** jang has quit IRC | 21:50 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible: install rabbitmq-server in serial https://review.openstack.org/313168 | 21:50 |
| admin0 | cloudnull: running again .. will report/update tomorrow | 21:50 |
| admin0 | now late | 21:50 |
| admin0 | will let it run over the night | 21:51 |
| *** sdake has joined #openstack-ansible | 21:51 | |
| *** tiagogomes has quit IRC | 21:52 | |
| *** woodard has quit IRC | 21:56 | |
| *** sdake has quit IRC | 21:56 | |
| *** woodard has joined #openstack-ansible | 21:56 | |
| *** ametts has joined #openstack-ansible | 21:56 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: Doc: Configuring the network on target refactor https://review.openstack.org/291310 | 21:58 |
| *** kstev has left #openstack-ansible | 22:00 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible-rabbitmq_server: install rabbitmq-server in serial https://review.openstack.org/310542 | 22:05 |
| cloudnull | thanks admin0 | 22:08 |
| cloudnull | let me know how it goes | 22:08 |
| admin0 | will do | 22:08 |
| *** admin0 has quit IRC | 22:11 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:14 | |
| *** mageshgv has quit IRC | 22:15 | |
| *** kiranv_ has joined #openstack-ansible | 22:15 | |
| *** kiranv has quit IRC | 22:17 | |
| *** asettle has joined #openstack-ansible | 22:18 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically increment the patch ver with sources-branch-updater https://review.openstack.org/313118 | 22:18 |
| *** galstrom is now known as galstrom_zzz | 22:20 | |
| *** phalmos has quit IRC | 22:21 | |
| openstackgerrit | Merged openstack/openstack-ansible: Enable SSL termination for all services https://review.openstack.org/277199 | 22:23 |
| *** alextricity25_ has joined #openstack-ansible | 22:27 | |
| cloudnull | boom , | 22:29 |
| cloudnull | my work is done here. | 22:30 |
| cloudnull | for today ... | 22:30 |
| *** ametts has quit IRC | 22:35 | |
| cloudnull | im off cheers | 22:38 |
| coolj_ | later cloudnull | 22:38 |
| *** daneyon has joined #openstack-ansible | 22:40 | |
| *** Brew has quit IRC | 22:40 | |
| *** thorst_ has joined #openstack-ansible | 22:43 | |
| *** sacharya has quit IRC | 22:45 | |
| *** david-lyle has quit IRC | 22:51 | |
| *** thorst_ has quit IRC | 22:52 | |
| *** david-lyle has joined #openstack-ansible | 22:53 | |
| *** david-lyle has quit IRC | 22:58 | |
| *** alextricity25_ has quit IRC | 23:00 | |
| openstackgerrit | Merged openstack/openstack-ansible: Ensure that the sources-branch-updater updates the Ironic role https://review.openstack.org/313139 | 23:02 |
| *** saneax_AFK is now known as saneax | 23:03 | |
| *** daneyon has quit IRC | 23:11 | |
| *** spotz is now known as spotz_zzz | 23:15 | |
| *** jaypipes has quit IRC | 23:16 | |
| *** retreved has quit IRC | 23:19 | |
| *** jayc_ has quit IRC | 23:19 | |
| *** Tebro has quit IRC | 23:31 | |
| *** Tebro has joined #openstack-ansible | 23:31 | |
| *** krotscheck has quit IRC | 23:31 | |
| *** krotscheck has joined #openstack-ansible | 23:31 | |
| *** jayc_ has joined #openstack-ansible | 23:32 | |
| *** klamath has quit IRC | 23:38 | |
| *** jthorne has quit IRC | 23:42 | |
| *** sacharya has joined #openstack-ansible | 23:45 | |
| *** johnmilton has joined #openstack-ansible | 23:48 | |
| *** sacharya has quit IRC | 23:50 | |
| *** johnmilton has quit IRC | 23:52 | |
| *** eil397 has quit IRC | 23:56 | |
| *** mummer has quit IRC | 23:57 | |
| *** karimb has joined #openstack-ansible | 23:58 | |
| *** woodard has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!