| *** asettle has joined #openstack-ansible | 00:00 | |
| *** LiYuenan has joined #openstack-ansible | 00:04 | |
| *** asettle has quit IRC | 00:04 | |
| *** jmckind has joined #openstack-ansible | 00:12 | |
| *** jmckind has quit IRC | 00:20 | |
| *** dxiri has quit IRC | 00:28 | |
| *** wadeholler has joined #openstack-ansible | 00:30 | |
| *** agrebennikov has quit IRC | 00:31 | |
| *** agrebennikov_ has joined #openstack-ansible | 00:31 | |
| *** wadeholler has quit IRC | 00:43 | |
| *** wadeholler has joined #openstack-ansible | 00:44 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Add missing Ironic auth plugin variable https://review.openstack.org/388946 | 00:47 |
|---|---|---|
| *** wadeholler has quit IRC | 00:47 | |
| *** wadeholler has joined #openstack-ansible | 00:47 | |
| *** david-lyle_ has joined #openstack-ansible | 00:51 | |
| *** david-lyle has quit IRC | 00:54 | |
| *** wadeholler has quit IRC | 00:54 | |
| *** gouthamr has joined #openstack-ansible | 00:55 | |
| *** wadeholler has joined #openstack-ansible | 00:55 | |
| *** wadeholler has quit IRC | 00:59 | |
| *** jmckind has joined #openstack-ansible | 01:02 | |
| *** cathrich_ has joined #openstack-ansible | 01:08 | |
| *** cathrichardson has quit IRC | 01:08 | |
| *** agrebennikov_ has quit IRC | 01:15 | |
| *** chandanc has joined #openstack-ansible | 01:17 | |
| *** chandanc has quit IRC | 01:23 | |
| *** thorst_ has quit IRC | 01:24 | |
| *** thorst_ has joined #openstack-ansible | 01:25 | |
| *** thorst_ has quit IRC | 01:33 | |
| *** javeriak has quit IRC | 01:36 | |
| *** maeker has quit IRC | 01:38 | |
| *** hughmFLEXin has joined #openstack-ansible | 01:38 | |
| *** Mudpuppy has joined #openstack-ansible | 01:47 | |
| *** Mudpuppy has quit IRC | 01:49 | |
| *** Mudpuppy has joined #openstack-ansible | 01:49 | |
| *** fops has joined #openstack-ansible | 01:50 | |
| jwitko | Hey all, I have pip installs failing on my galera container because they can't reach the haproxy IP on port 8181 | 01:53 |
| jwitko | I'm wondering what I screwed up... I thought the playbooks installed HA proxy and configured a load balanced setup with the specified IP from the config? | 01:53 |
| *** fops_ has quit IRC | 01:53 | |
| *** Mudpuppy has quit IRC | 01:54 | |
| jwitko | nm i see I'm missing some user_vars | 02:02 |
| cloudnull | jwitko: did you get it to go? | 02:11 |
| *** thorst_ has joined #openstack-ansible | 02:14 | |
| *** thorst_ has quit IRC | 02:14 | |
| *** thorst_ has joined #openstack-ansible | 02:14 | |
| *** thorst_ has quit IRC | 02:23 | |
| *** poopcat has quit IRC | 02:28 | |
| *** poopcat has joined #openstack-ansible | 02:28 | |
| *** chhavi has joined #openstack-ansible | 02:29 | |
| *** thorst_ has joined #openstack-ansible | 02:32 | |
| *** LiYuenan has quit IRC | 02:32 | |
| *** thorst_ has quit IRC | 02:33 | |
| *** poopcat has quit IRC | 02:36 | |
| jwitko | I moved on to a new error :) | 02:38 |
| *** thorst_ has joined #openstack-ansible | 02:39 | |
| *** thorst_ has quit IRC | 02:39 | |
| jwitko | is there a script to generate random entries into user_secrets.yml ? | 02:41 |
| openstackgerrit | Wang Qing wu proposed openstack/openstack-ansible-os_nova: Identify virt type of PowerVM and KVM on Power https://review.openstack.org/388571 | 02:44 |
| *** jcrst has quit IRC | 02:46 | |
| *** chhavi has quit IRC | 02:47 | |
| *** Hosam has quit IRC | 02:48 | |
| *** chandanc has joined #openstack-ansible | 02:49 | |
| *** poopcat has joined #openstack-ansible | 02:49 | |
| coolj | jwitko: is pw-token-gen.py what you're looking for? http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-creds.html | 02:49 |
| jwitko | yes! ty! | 02:51 |
| coolj | np! | 02:51 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Use upper constraints for all tox targets https://review.openstack.org/388448 | 02:53 |
| *** Mudpuppy has joined #openstack-ansible | 02:56 | |
| *** rmelero has quit IRC | 02:59 | |
| *** Mudpuppy has quit IRC | 03:01 | |
| *** david-lyle_ has quit IRC | 03:23 | |
| *** david-lyle has joined #openstack-ansible | 03:23 | |
| *** phalmos has quit IRC | 03:29 | |
| *** jcrst has joined #openstack-ansible | 03:34 | |
| jwitko | ok new error on the last playbook | 03:36 |
| jwitko | setup-openstack.yml | 03:36 |
| jwitko | TASK [os_keystone : Get remote venv checksum] ********************************** | 03:36 |
| jwitko | fatal: [infra3_keystone_container-161bc34a]: FAILED! => {"changed": false, "content": "", "failed": true, "msg": "Status code was not [200]: An unknown error occurred: ''", "redirected": false, "status": -1, "url": "http://10.89.103.25:8181/venvs/14.0.0/ubuntu/keystone-14.0.0-x86_64.checksum"} | 03:36 |
| jwitko | fatal: [infra1_keystone_container-4a9ca527]: FAILED! => {"changed": false, "content": "", "failed": true, "msg": "Status code was not [200]: An unknown error occurred: ''", "redirected": false, "status": -1, "url": "http://10.89.103.25:8181/venvs/14.0.0/ubuntu/keystone-14.0.0-x86_64.checksum"} | 03:36 |
| jwitko | ok: [infra2_keystone_container-be81ce54] | 03:36 |
| jwitko | failed on 2/3 nodes | 03:36 |
| jwitko | looks like it just needed time to propogate? ran it again, 2/3 passed, ran it a 3rd time, 3/3 passed | 03:39 |
| *** thorst_ has joined #openstack-ansible | 03:40 | |
| jwitko | nevermind... its still erroring on 1/3 | 03:41 |
| jwitko | yea every time on the keystone 14.0.0 version | 03:44 |
| *** thorst_ has quit IRC | 03:48 | |
| *** chhavi has joined #openstack-ansible | 03:51 | |
| jwitko | finally got all 3 to pass... just by executing again | 03:52 |
| jwitko | new error though on cache | 03:52 |
| jwitko | fatal: [infra1_keystone_container-4a9ca527]: FAILED! => {"changed": false, "failed": true, "msg": "Source '/var/cache/keystone-14.0.0-x86_64.tgz' does not exist"} | 03:52 |
| *** kamtamtun has joined #openstack-ansible | 04:00 | |
| *** poopcat has quit IRC | 04:04 | |
| *** dxiri has joined #openstack-ansible | 04:09 | |
| *** dxiri has quit IRC | 04:11 | |
| *** dxiri has joined #openstack-ansible | 04:12 | |
| jwitko | now the same error for glance... any ideas whats going on ? | 04:12 |
| jwitko | 2/3 glance containers fail finding /var/cache file | 04:12 |
| jwitko | fatal: [infra2_glance_container-b70a3ad3]: FAILED! => {"changed": false, "failed": true, "msg": "Source '/var/cache/glance-14.0.0-x86_64.tgz' does not exist"} | 04:12 |
| *** maeker has joined #openstack-ansible | 04:16 | |
| *** hj-hpe has quit IRC | 04:17 | |
| *** jmckind_ has joined #openstack-ansible | 04:22 | |
| *** hybridpollo has quit IRC | 04:22 | |
| *** jmckind has quit IRC | 04:23 | |
| *** weezS has joined #openstack-ansible | 04:25 | |
| *** chandanc has quit IRC | 04:25 | |
| *** jmckind_ has quit IRC | 04:27 | |
| *** jmckind has joined #openstack-ansible | 04:30 | |
| *** jmckind has quit IRC | 04:34 | |
| *** chhavi has quit IRC | 04:36 | |
| *** janki has joined #openstack-ansible | 04:41 | |
| *** thorst_ has joined #openstack-ansible | 04:48 | |
| *** markvoelker_ has quit IRC | 04:49 | |
| *** Hosam has joined #openstack-ansible | 04:50 | |
| jwitko | looks like every once in a while my haproxy is responding with no data received | 04:52 |
| jwitko | http://paste.ubuntu.com/23352068/ | 04:52 |
| jwitko | but retries and it works | 04:52 |
| *** javeriak has joined #openstack-ansible | 04:53 | |
| *** thorst_ has quit IRC | 04:54 | |
| *** galstrom is now known as galstrom_zzz | 04:55 | |
| jwitko | i'm seeing a lot of "Oct 20 04:57:46 localhost haproxy[6559]: 10.89.103.166:56794 [20/Oct/2016:04:57:46.174] repo_all-front-1/1: SSL handshake failure | 04:58 |
| jwitko | O" in the logs. is there some setting I should be using to ignore ssl certs? | 04:58 |
| jwitko | all of the /etc/ssl/private/haproxy.pem files have different md5sums ? | 05:09 |
| *** weezS has quit IRC | 05:13 | |
| *** weezS has joined #openstack-ansible | 05:14 | |
| dxiri | jwitko: I am having that same problem :) | 05:17 |
| dxiri | haproxy seems to be intermittent | 05:17 |
| dxiri | are you by any chance using the same vip for internal and external? | 05:17 |
| *** weezS has quit IRC | 05:18 | |
| *** Hosam has quit IRC | 05:19 | |
| *** Hosam has joined #openstack-ansible | 05:19 | |
| *** maeker has quit IRC | 05:20 | |
| *** Hosam_ has joined #openstack-ansible | 05:22 | |
| *** thetrav has joined #openstack-ansible | 05:22 | |
| thetrav | is setup-infrastructure idempotent? | 05:22 |
| thetrav | I'm trying to set up automatic haproxy certificate renewal. Looks like that's the top level task used to make it happen | 05:23 |
| jwitko | dxiri, I am! | 05:24 |
| dxiri | jwitko: then same boat as me :) | 05:24 |
| jwitko | dxiri interesting. should I change them? | 05:25 |
| *** Hosam has quit IRC | 05:25 | |
| dxiri | in my case, haproxy spun up only the https instance first (without any http instance) and the fix was to change the external vip to a hostname (that I set it to resolve to the same ip) | 05:26 |
| jwitko | oh, I already had that | 05:26 |
| jwitko | the external was the hostname | 05:26 |
| dxiri | that got me past the setup-infrastructure.yml | 05:26 |
| jwitko | and the internal was the IP of the resolved hostname | 05:26 |
| dxiri | but now I am stuck with this intermitent stuff | 05:26 |
| dxiri | sometimes works in the first try, sometimes on the 7th try | 05:27 |
| dxiri | yep, that's exactly how I set that up as well | 05:27 |
| jwitko | I am going to change internal to a different subnet | 05:28 |
| jwitko | and re-run infra playbook | 05:28 |
| jwitko | and then setup playbook | 05:28 |
| jwitko | see what happens | 05:28 |
| dxiri | let's wait and see if someone can chime in on this :) I need some sleep but will check on the history and soon as I get up to work | 05:28 |
| dxiri | if you do that, please let me know the result :) | 05:28 |
| *** markvoelker_ has joined #openstack-ansible | 05:30 | |
| *** LiYuenan has joined #openstack-ansible | 05:31 | |
| *** McMurlock1 has joined #openstack-ansible | 05:32 | |
| *** dxiri has quit IRC | 05:33 | |
| *** javeriak has quit IRC | 05:44 | |
| *** markvoelker_ has quit IRC | 05:46 | |
| *** thorst_ has joined #openstack-ansible | 05:52 | |
| thetrav | weird, it decided to skip my certificates | 05:53 |
| *** thorst_ has quit IRC | 05:59 | |
| *** kamtamtun has quit IRC | 06:05 | |
| *** javeriak has joined #openstack-ansible | 06:08 | |
| *** dxiri has joined #openstack-ansible | 06:12 | |
| *** jcrst has quit IRC | 06:14 | |
| *** thetrav has quit IRC | 06:14 | |
| *** jcrst has joined #openstack-ansible | 06:16 | |
| *** dxiri has quit IRC | 06:16 | |
| *** pcaruana has joined #openstack-ansible | 06:18 | |
| *** Hosam_ has quit IRC | 06:37 | |
| *** dxiri has joined #openstack-ansible | 06:40 | |
| *** Hosam has joined #openstack-ansible | 06:40 | |
| *** Hosam has quit IRC | 06:41 | |
| *** Hosam_ has joined #openstack-ansible | 06:41 | |
| *** dxiri has quit IRC | 06:44 | |
| *** omiday has quit IRC | 06:50 | |
| *** jcrst has quit IRC | 06:57 | |
| *** Mudpuppy has joined #openstack-ansible | 06:58 | |
| *** thorst_ has joined #openstack-ansible | 06:58 | |
| *** jcrst has joined #openstack-ansible | 06:59 | |
| *** Mudpuppy has quit IRC | 07:02 | |
| *** cathrich_ has quit IRC | 07:03 | |
| *** cathrichardson has joined #openstack-ansible | 07:03 | |
| *** thorst_ has quit IRC | 07:04 | |
| *** automagically has quit IRC | 07:06 | |
| *** automagically has joined #openstack-ansible | 07:06 | |
| *** fxpester has joined #openstack-ansible | 07:10 | |
| *** Hosam_ has quit IRC | 07:36 | |
| *** haad1 has joined #openstack-ansible | 07:41 | |
| *** haad1 has quit IRC | 07:41 | |
| *** haad1 has joined #openstack-ansible | 07:41 | |
| *** fxpester has quit IRC | 07:51 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Automatically source the rc file for ansible/ansible-playbook https://review.openstack.org/388877 | 07:52 |
| *** javeriak has quit IRC | 07:53 | |
| gaudenz | Matias: I have now reported a bug on ceilometer in launchpad: https://bugs.launchpad.net/ceilometer/+bug/1635148 | 07:56 |
| openstack | Launchpad bug 1635148 in Ceilometer "SSLError in connection from ceilometer-collector to RabbitMQ" [Undecided,New] | 07:56 |
| gaudenz | Matias: Is this the same issue you had? If so, can you add your information so we might find out what fixed it and backport that to mitaka. | 07:57 |
| *** haad1 has quit IRC | 07:59 | |
| *** haad1 has joined #openstack-ansible | 07:59 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Adds support for disks on HP Smart Array Controllers https://review.openstack.org/389053 | 08:03 |
| *** thorst_ has joined #openstack-ansible | 08:03 | |
| *** asettle has joined #openstack-ansible | 08:04 | |
| *** david-lyle_ has joined #openstack-ansible | 08:07 | |
| *** Hosam has joined #openstack-ansible | 08:07 | |
| *** Hosam_ has joined #openstack-ansible | 08:08 | |
| *** karimb has joined #openstack-ansible | 08:08 | |
| *** david-lyle has quit IRC | 08:09 | |
| *** thorst_ has quit IRC | 08:09 | |
| *** Hosam has quit IRC | 08:12 | |
| odyssey4me | andymccr mrda noticed this? https://review.openstack.org/388946 - it seems odd that this would be required in group_vars given that this is the default and should only be used in the role | 08:13 |
| andymccr | odyssey4me: its in the nova role | 08:14 |
| andymccr | i think it was added as part of the auth --> v3 change and we didn't pick up the group_vars change | 08:15 |
| odyssey4me | ah | 08:15 |
| odyssey4me | I don't actually see why we have vars for the auth plugin. | 08:15 |
| odyssey4me | We should just hard set it in the templates, and for the rare use-case where people want to change it - there's config template. | 08:16 |
| *** markvoelker has joined #openstack-ansible | 08:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Work around bad libvirt-python wheel https://review.openstack.org/389064 | 08:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Work around bad libvirt-python wheel https://review.openstack.org/389064 | 08:41 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-galera_server: Fix MYSQLD_STARTUP_TIMEOUT for systemd https://review.openstack.org/389066 | 08:43 |
| *** mpotdar has joined #openstack-ansible | 08:44 | |
| *** electrofelix has joined #openstack-ansible | 08:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts: Add ability to set upper constraints file https://review.openstack.org/389067 | 08:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts: Add ability to set upper constraints file https://review.openstack.org/389068 | 08:46 |
| *** karimb has quit IRC | 08:58 | |
| *** McMurlock1 has quit IRC | 09:01 | |
| evrardjp | andymccr: thanks for the bug cleanup :D | 09:02 |
| evrardjp | test-vars I mean | 09:03 |
| andymccr | evrardjp: ahh np! yeah should probably go through some of those bugs and close out ones that no longer apply, but will add it to my todo list | 09:03 |
| evrardjp | remember they are likely to be tagged already | 09:04 |
| evrardjp | I will help you through it if you want | 09:04 |
| odyssey4me | evrardjp FYI for any bugs that are marked fix released, the tags should actually be removed | 09:04 |
| odyssey4me | well, that's the traditional management method that was used in OpenStack. How we decide to use tags is entirely up to us. | 09:05 |
| evrardjp | odyssey4me: agreed | 09:05 |
| evrardjp | but if they are fix released, they are not listed as with tags IIRC | 09:05 |
| evrardjp | I will go over them anyway | 09:05 |
| odyssey4me | mancdaz does this seem sane? perhaps you can verify that this is a good idea? https://review.openstack.org/389066 | 09:05 |
| evrardjp | andymccr: are you busy on this? https://bugs.launchpad.net/openstack-ansible/+bug/1632098 | 09:05 |
| openstack | Launchpad bug 1632098 in openstack-ansible "Liberty to Mitaka upgrade, rabbitmq queues are not cleaned up" [High,Confirmed] - Assigned to Andy McCrae (andrew-mccrae) | 09:05 |
| mancdaz | odyssey4me I was chatting with andymccr about this earlier | 09:06 |
| mancdaz | this is essentially to fix a regression in behaviour | 09:06 |
| mancdaz | it seems we decided to stop placing the defaults file on disk when we are using systemd | 09:06 |
| odyssey4me | sure, my only cause for concern is that systemd can't just 'source' vars like upstart does | 09:06 |
| *** thorst_ has joined #openstack-ansible | 09:07 | |
| mancdaz | odyssey4me that is apparently what it does | 09:07 |
| odyssey4me | well, that's what I believe based on some distant memory of a conversation about how horrible it is | 09:07 |
| mancdaz | odyssey4me on that I don't realy know | 09:07 |
| evrardjp | I thought we said we're gonna trust the startup scripts delivered with packages | 09:07 |
| mancdaz | assuming that it does, this is the correct fix | 09:07 |
| *** jcrst has quit IRC | 09:07 | |
| evrardjp | because it was supposed to be fine | 09:07 |
| odyssey4me | evrardjp sure, until the package-based startups don't work :p | 09:07 |
| evrardjp | should we work with packagers? We are probably not the only ones | 09:08 |
| mancdaz | evrardjp except that we are not placing the defaults file on disk for the package scripts to source | 09:08 |
| odyssey4me | evrardjp IIRC the issue we fix with all that is a known bug fixed in MariaDB 10.1 | 09:08 |
| odyssey4me | but idk for sure | 09:08 |
| evrardjp | yup that's what I recall too odyssey4me | 09:08 |
| mancdaz | odyssey4me what bug? | 09:09 |
| evrardjp | I thought I heard that | 09:09 |
| evrardjp | the startup timing issue | 09:09 |
| mancdaz | no | 09:09 |
| odyssey4me | when it comes to MariaDB, I believe everything that mancdaz tells me because he's awesome :) | 09:09 |
| evrardjp | odyssey4me: +1 | 09:09 |
| mancdaz | the problem in 10.0 was that the package provided initscript was not sourcing /etc/defaults/mysql | 09:09 |
| mancdaz | so we added a oine to make it do that | 09:09 |
| mancdaz | *line | 09:09 |
| evrardjp | mancdaz: agreed | 09:10 |
| odyssey4me | mancdaz ok, if it wasn't sourcing that file then perhaps we shouldn't be putting that file down and we should be implementing config somewhere else? | 09:10 |
| mancdaz | the problem we have with xenial is that someone decided not to place /etc/defaults/mysql on disk becuase we're using systemd | 09:10 |
| mancdaz | the 'fix' in mariadb10.1 is that the initscript does properly source /etc/defaults/mariadb (note, not mysql) | 09:10 |
| odyssey4me | ie instead of forcing it to work the way we work, perhaps we should learn how it's supposed to be used and change how we use it | 09:10 |
| mancdaz | but either way, if you're not putting it down on disk it can't be sourced | 09:11 |
| pjm6 | good morning all | 09:11 |
| mancdaz | odyssey4me this *is* the way it's meant to work | 09:11 |
| mancdaz | the defaults file is a standard mechanism | 09:11 |
| mancdaz | but you have to put it there! | 09:11 |
| evrardjp | mancdaz: standard for ubuntu | 09:11 |
| evrardjp | :p | 09:11 |
| odyssey4me | kylek3h_away when you're in, it'd be great to know if https://review.openstack.org/389066 solves the problem you reported :) | 09:11 |
| mancdaz | evrardjp pretty sure that's what we're using | 09:11 |
| evrardjp | mancdaz: so it works with xenial? | 09:11 |
| mancdaz | evrardjp so I am told | 09:12 |
| evrardjp | just saying there are many standards | 09:12 |
| odyssey4me | evrardjp apparently so - see gate test | 09:12 |
| evrardjp | https://xkcd.com/927/ | 09:12 |
| *** karimb has joined #openstack-ansible | 09:13 | |
| mancdaz | odyssey4me to be fair that gate test is not actually showing you anything | 09:13 |
| evrardjp | maybe I misunderstood | 09:13 |
| odyssey4me | mancdaz yeah I know, I was being facetious | 09:13 |
| evrardjp | so if I understand correctly, per distribution we should define if we use /etc/defaults/mysql or mariadb | 09:13 |
| odyssey4me | no, not as far as I understand it | 09:14 |
| odyssey4me | per mariadb version perhaps | 09:14 |
| odyssey4me | but not per districution | 09:14 |
| *** thorst_ has quit IRC | 09:14 | |
| odyssey4me | bah, need more coffee | 09:14 |
| *** Hosam_ has quit IRC | 09:18 | |
| andymccr | evrardjp: im looking at it now -but if you are already busy with it/have an idea feel free to get it done | 09:18 |
| *** jcrst has joined #openstack-ansible | 09:23 | |
| mancdaz | andymccr https://mariadb.com/kb/en/mariadb/systemd/ | 09:25 |
| mancdaz | https://jira.mariadb.org/browse/MDEV-9202 | 09:25 |
| andymccr | mancdaz: odyssey4me: I just tested using systemctl and adding an "echo x > tmp.file" to ensure it was sourcing/performing the actions in the mysql init.d script - and it is. | 09:26 |
| andymccr | so imo that will work but perhaps there is a better way of doing it | 09:26 |
| mancdaz | andymccr I'd say if it's sourcing that file, then it's sufficient | 09:27 |
| evrardjp | andymccr: I planned to look at it, and I have an upgrade ready | 09:27 |
| odyssey4me | andymccr yeah, if we can get confirmation of the issue being solved from kylek3h_away then great - per we can then implement an improvement later | 09:28 |
| *** karimb has quit IRC | 09:28 | |
| andymccr | evrardjp: honestly it'd be great if you took it off me and did it :) but no pressure - I'll take a look if you can't. | 09:28 |
| evrardjp | I will do it | 09:34 |
| *** markvoelker has quit IRC | 09:39 | |
| *** karimb has joined #openstack-ansible | 09:39 | |
| *** gaudenz has quit IRC | 09:53 | |
| *** jcrst has quit IRC | 09:55 | |
| *** Hosam has joined #openstack-ansible | 10:00 | |
| *** Hosam has quit IRC | 10:05 | |
| *** winggundamth has quit IRC | 10:08 | |
| *** winggundamth has joined #openstack-ansible | 10:10 | |
| *** thorst_ has joined #openstack-ansible | 10:13 | |
| *** thorst_ has quit IRC | 10:19 | |
| *** wadeholler has joined #openstack-ansible | 10:35 | |
| *** markvoelker has joined #openstack-ansible | 10:37 | |
| *** dxiri has joined #openstack-ansible | 10:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that upper constraints are always applied https://review.openstack.org/388220 | 10:41 |
| *** gaudenz has joined #openstack-ansible | 10:42 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that upper constraints are always applied https://review.openstack.org/388220 | 10:43 |
| *** karimb has quit IRC | 10:44 | |
| *** karimb has joined #openstack-ansible | 10:46 | |
| *** dxiri has quit IRC | 10:47 | |
| *** wadeholler has quit IRC | 10:50 | |
| *** janki has quit IRC | 10:52 | |
| *** asettle has quit IRC | 10:54 | |
| *** asettle has joined #openstack-ansible | 10:58 | |
| *** asettle has quit IRC | 10:59 | |
| *** Mudpuppy has joined #openstack-ansible | 11:00 | |
| *** Hosam has joined #openstack-ansible | 11:01 | |
| *** Mudpuppy has quit IRC | 11:04 | |
| *** hughmFLEXin has quit IRC | 11:05 | |
| *** Hosam has quit IRC | 11:06 | |
| *** hughmFLEXin has joined #openstack-ansible | 11:07 | |
| *** markvoelker_ has joined #openstack-ansible | 11:07 | |
| *** markvoelker has quit IRC | 11:11 | |
| *** thorst_ has joined #openstack-ansible | 11:17 | |
| *** thorst_ has quit IRC | 11:17 | |
| *** thorst_ has joined #openstack-ansible | 11:17 | |
| *** Hosam has joined #openstack-ansible | 11:19 | |
| hughsaunders | logan-: this is excellent https://github.com/openstack/openstack-ansible/commit/d651ed70f4d2298ba1a372a6b9906e1942a871e8 | 11:19 |
| odyssey4me | hughsaunders :) | 11:20 |
| *** Hosam_ has joined #openstack-ansible | 11:22 | |
| *** Hosam has quit IRC | 11:25 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that upper constraints are always applied https://review.openstack.org/388220 | 11:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that upper constraints are always applied https://review.openstack.org/388220 | 11:26 |
| *** retreved has joined #openstack-ansible | 11:31 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh: Use centralised Ansible test scripts https://review.openstack.org/389128 | 11:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder: Use centralised Ansible test scripts https://review.openstack.org/389129 | 11:34 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Use centralised Ansible test scripts https://review.openstack.org/389131 | 11:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi: Use centralised Ansible test scripts https://review.openstack.org/389132 | 11:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Use centralised Ansible test scripts https://review.openstack.org/389133 | 11:41 |
| *** gouthamr has quit IRC | 11:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Use centralised Ansible test scripts https://review.openstack.org/389134 | 11:44 |
| *** johnmilton has quit IRC | 11:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic: Use centralised Ansible test scripts https://review.openstack.org/389136 | 11:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Use centralised Ansible test scripts https://review.openstack.org/389137 | 11:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_neutron: Use centralised Ansible test scripts https://review.openstack.org/389139 | 11:51 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova: Use centralised Ansible test scripts https://review.openstack.org/389141 | 11:55 |
| *** kylek3h_away has quit IRC | 11:55 | |
| *** gouthamr has joined #openstack-ansible | 11:55 | |
| odyssey4me | andymccr we need to move os_trove into the combined deliberable now that it's part of the integrated build: https://github.com/openstack/governance/blob/master/reference/projects.yaml#L3390-L3394 | 11:56 |
| odyssey4me | *deliverable | 11:56 |
| odyssey4me | otherwise we won't be able to tag it as part of the milestone 1 release | 11:57 |
| andymccr | odyssey4me: so basically remove it from there and add it to the openstack-ansible list above? | 11:57 |
| odyssey4me | andymccr yep | 11:57 |
| andymccr | any other special thing i ahve to do, or just make the PR? | 11:57 |
| odyssey4me | just push up the review with a commit message explaining that it's now part of the combined release | 11:57 |
| odyssey4me | that's it | 11:57 |
| andymccr | sweet :) thanks | 11:58 |
| andymccr | will do now | 11:58 |
| odyssey4me | I think we'll have to tag a liberty/mitaka release today too thanks to the percona debacle | 11:58 |
| odyssey4me | so a bit later I'll guide you through how to go about doing that :) | 11:59 |
| *** gouthamr has quit IRC | 11:59 | |
| *** gouthamr has joined #openstack-ansible | 12:00 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally: Use centralised Ansible test scripts https://review.openstack.org/389143 | 12:01 |
| andymccr | ok cool yeah that'd be appreciated! | 12:01 |
| andymccr | I assume Trove would only be integrated from Ocata(master onwards) | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_sahara: Use centralised Ansible test scripts https://review.openstack.org/389144 | 12:03 |
| odyssey4me | andymccr yep | 12:03 |
| odyssey4me | I suppose there is the option of pulling it back to Newton if we want to, but until we have scenario testing for the integrated gate, I would be resistant to the backport | 12:03 |
| andymccr | agree | 12:03 |
| andymccr | https://review.openstack.org/#/c/389145/ think its all good | 12:04 |
| odyssey4me | andymccr another thing - I think any non-integrated repo should rather be tagged 'release:none' instead of 'release:cycle-trailing' - the cycle-trailing tag has much more governance around it and results in us not being able to create branches or tags whenever we want to | 12:06 |
| *** johnmilton has joined #openstack-ansible | 12:06 | |
| andymccr | odyssey4me: makes sense | 12:06 |
| odyssey4me | I think we might want to have more flexibility until we integrate them. Once they're integrated then we stick to the stricter governance. | 12:06 |
| andymccr | yeah im not sure why we'd want to have strict rules on repos that arent yet ready at all | 12:07 |
| odyssey4me | exactly | 12:07 |
| odyssey4me | I tagged them with the cycle-trailing tag before really understanding the repurcussions of that | 12:07 |
| odyssey4me | I'd suggest doing a follow up patch (on top of the previous one) which changes the tags. | 12:07 |
| andymccr | odyssey4me: will do | 12:08 |
| odyssey4me | The governance reviews sit around for a week before merging, so don't worry about chasing them. | 12:08 |
| odyssey4me | I think there's a rule around giving them a week to provide the TC opportunity to review and object. If there are no objections, then it gets merged. | 12:08 |
| *** Hosam has joined #openstack-ansible | 12:08 | |
| *** Hosam__ has joined #openstack-ansible | 12:09 | |
| *** Hosam_ has quit IRC | 12:10 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_swift: Use centralised Ansible test scripts https://review.openstack.org/389147 | 12:12 |
| *** markvoelker has joined #openstack-ansible | 12:12 | |
| *** Hosam has quit IRC | 12:13 | |
| *** Hosam__ has quit IRC | 12:14 | |
| *** markvoelker_ has quit IRC | 12:16 | |
| andymccr | https://review.openstack.org/389150 - odyssey4me done. | 12:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_tempest: Use centralised Ansible test scripts https://review.openstack.org/389151 | 12:18 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install: Use centralised Ansible test scripts https://review.openstack.org/389152 | 12:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-plugins: Use centralised Ansible test scripts https://review.openstack.org/389153 | 12:23 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server: Use centralised Ansible test scripts https://review.openstack.org/389155 | 12:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build: Use centralised Ansible test scripts https://review.openstack.org/389156 | 12:27 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_server: Use centralised Ansible test scripts https://review.openstack.org/389157 | 12:29 |
| *** karimb has quit IRC | 12:29 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_client: Use centralised Ansible test scripts https://review.openstack.org/389162 | 12:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_server: Use centralised Ansible test scripts https://review.openstack.org/389164 | 12:34 |
| *** Jeffrey4l has quit IRC | 12:35 | |
| *** thorst_ has quit IRC | 12:37 | |
| *** fops has quit IRC | 12:43 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Use centralised Ansible test scripts https://review.openstack.org/389166 | 12:43 |
| *** fops has joined #openstack-ansible | 12:44 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Use upper constraints for all tox targets https://review.openstack.org/389167 | 12:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Fix neutron_agent IP to be neutron_server IP https://review.openstack.org/389170 | 12:52 |
| *** schwicht has joined #openstack-ansible | 12:52 | |
| odyssey4me | ok, andymccr with https://review.openstack.org/#/q/topic:bp/central-test-repository+status:open we have all the tests using the central repo for all the bits | 12:53 |
| odyssey4me | we're going to have to be vigilent about keeping the two branches in sync with changes for as long as possible | 12:54 |
| odyssey4me | also, these need attention: https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+topic:use-upper-constraints+status:open | 12:54 |
| *** gaudenz has quit IRC | 12:55 | |
| *** dxiri has joined #openstack-ansible | 12:57 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add missing Ironic auth plugin variable https://review.openstack.org/388946 | 12:58 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron: [WIP] Move to full tempest run for Neutron https://review.openstack.org/387457 | 12:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add missing Ironic auth plugin variable https://review.openstack.org/389174 | 12:58 |
| *** dxiri has quit IRC | 13:02 | |
| *** schwicht has quit IRC | 13:03 | |
| *** schwicht has joined #openstack-ansible | 13:07 | |
| *** klamath has joined #openstack-ansible | 13:08 | |
| odyssey4me | andymccr ok, for the sake of expediency let me handle today's releases - but I'm thinking that perhaps we should move the sources-branch-updater and related release scripts into the ops repo | 13:09 |
| odyssey4me | generally I think it'll be easier to merge changes there and it'll keep it out of the changelog for the production release | 13:09 |
| odyssey4me | thoughts? | 13:09 |
| andymccr | odyssey4me: sounds good to me | 13:09 |
| *** admin0 has joined #openstack-ansible | 13:12 | |
| *** gaudenz has joined #openstack-ansible | 13:12 | |
| *** adrian_otto has joined #openstack-ansible | 13:14 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Remove openstack-ansible.rc sourcing https://review.openstack.org/389185 | 13:15 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Remove openstack-ansible.rc sourcing instruction https://review.openstack.org/389185 | 13:16 |
| *** cathrich_ has joined #openstack-ansible | 13:17 | |
| *** cathrichardson has quit IRC | 13:17 | |
| *** cathrich_ is now known as cathrichardson | 13:19 | |
| * mhayden toots | 13:26 | |
| *** dxiri has joined #openstack-ansible | 13:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 13.3.7 https://review.openstack.org/389198 | 13:27 |
| *** karimb has joined #openstack-ansible | 13:30 | |
| *** dxiri has quit IRC | 13:31 | |
| odyssey4me | andymccr release requests: https://review.openstack.org/389191 / https://review.openstack.org/389202 | 13:32 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 12.2.7 https://review.openstack.org/389204 | 13:33 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Update all SHAs for 12.2.7 https://review.openstack.org/389204 | 13:33 |
| *** schwicht has quit IRC | 13:34 | |
| *** asettle has joined #openstack-ansible | 13:43 | |
| andymccr | odyssey4me: on the mitaka sha bump, the repo_build and lxc_container_create repos seem wrong | 13:45 |
| andymccr | or atleast not head of stable/mitaka | 13:45 |
| *** asettle has quit IRC | 13:47 | |
| odyssey4me | andymccr are you sure you're checking it right? https://github.com/openstack/openstack-ansible-lxc_container_create/commits/stable/mitaka shows d319546 which matches the sha bump? | 13:48 |
| *** schwicht has joined #openstack-ansible | 13:49 | |
| odyssey4me | same goes for the repobuild role | 13:49 |
| *** jheroux has joined #openstack-ansible | 13:51 | |
| *** galstrom_zzz is now known as galstrom | 13:54 | |
| andymccr | - repo: openstack/openstack-ansible-lxc_container_create | 13:54 |
| andymccr | hash: a837705bb20aaddaec56dd3319b8a359c96296e4 | 13:54 |
| andymccr | thats in the PR | 13:54 |
| andymccr | am i reading it wrong? | 13:54 |
| *** galstrom is now known as galstrom_zzz | 13:54 | |
| *** asettle has joined #openstack-ansible | 13:55 | |
| andymccr | - repo: openstack/openstack-ansible-repo_build | 13:55 |
| andymccr | hash: 6e2989d73907d4e01573b1fd6359b146364a75cd | 13:55 |
| andymccr | which i think should be 22c55a72b75c17df84529e09aaa51be9909c5b15 | 13:55 |
| *** kjw3 has joined #openstack-ansible | 14:01 | |
| *** adrian_otto has quit IRC | 14:02 | |
| mgariepy | good morning every one | 14:03 |
| logan- | hughsaunders: thanks. :) | 14:06 |
| odyssey4me | andymccr ah, you're looking at the release request - not the sha bump | 14:08 |
| odyssey4me | the release request is on the current SHA's in https://github.com/openstack/openstack-ansible/blob/stable/mitaka/ansible-role-requirements.yml | 14:09 |
| andymccr | odyssey4me: ahh got you | 14:09 |
| odyssey4me | when we release, we release based on the SHA's bumped immediately after the previous release | 14:09 |
| odyssey4me | that way those SHA's are tested for two weeks before we release | 14:09 |
| andymccr | makes sense | 14:09 |
| odyssey4me | on occasion we do things like https://github.com/openstack/openstack-ansible/commit/2c2cfbb1924ccb15e3a35fc4ccf7181c0fcf3bf4 to bump the role SHA's | 14:10 |
| odyssey4me | but that's not the norm | 14:10 |
| mgariepy | any master on lxb and vxlan here ? i'm having some issue whit the network on centos7 .. | 14:10 |
| mgariepy | with** | 14:11 |
| jwitko | hey odyssey4me, I screwed up last night and deleted /etc/ssl on a controller in a fresh setup. What can I run to re-generate that directory and its contents? | 14:11 |
| odyssey4me | mgariepy well, if jamesdenton or evrardjp are around they could possibly help | 14:11 |
| odyssey4me | jwitko argh, that's not fun | 14:11 |
| jwitko | ;\ | 14:11 |
| jwitko | Do I have to reinstall ubuntu ? | 14:11 |
| odyssey4me | jwitko you should just be able to copy the directory from another controller | 14:12 |
| evrardjp | jwitko: the complete ssl folder? | 14:12 |
| odyssey4me | I think | 14:12 |
| evrardjp | yeah copy from another source I guess | 14:12 |
| mgariepy | evrardjp, do you have some time to help me out a bit ? | 14:12 |
| jwitko | oh, that folder isn't unique to the machine? | 14:12 |
| evrardjp | still there are certs for the host there maybe | 14:12 |
| odyssey4me | our SSL process copies the same private key and public key combo for the services we deploy to all the controller containers | 14:12 |
| evrardjp | sure I can try :D | 14:12 |
| evrardjp | mgariepy: ^ | 14:12 |
| jwitko | odyssey4me, so the reason I deleted it is i was having lots of HA proxy SSL errors in my logs | 14:12 |
| odyssey4me | jwitko there may be some sort of unique cert that ubuntu made, but not OSA | 14:13 |
| jwitko | i noticed that the servers did not share the same haproxy.pem | 14:13 |
| jwitko | it was unique on each server | 14:13 |
| jwitko | and I thought they needed to be identical | 14:13 |
| odyssey4me | oh, that is not right | 14:13 |
| jwitko | my assumption? | 14:13 |
| odyssey4me | hmm, what release? | 14:13 |
| jwitko | newton | 14:13 |
| jwitko | lastest rc | 14:13 |
| odyssey4me | no, your assumption is right - that is a bug if it is repeatable | 14:14 |
| jwitko | so if I copy over the ssl directory and re-run setup-hosts | 14:14 |
| jwitko | it should end up with the same haproxy.pem everywhere? | 14:14 |
| odyssey4me | oh bother - this is a bug | 14:15 |
| jwitko | yay i found a bug :D | 14:15 |
| odyssey4me | we generate unique certs on each server instead of generating on the first and replicating | 14:15 |
| odyssey4me | https://github.com/openstack/openstack-ansible-haproxy_server/blob/master/tasks/haproxy_ssl_configuration.yml#L74 | 14:15 |
| odyssey4me | please report the bug | 14:15 |
| jwitko | is there a quick instruction set on how to do that | 14:16 |
| odyssey4me | report it with the symptoms you saw - while you do that I'll figure out a fix | 14:16 |
| *** agrebennikov_ has joined #openstack-ansible | 14:17 | |
| odyssey4me | jwitko it's not all that complicated - just pop to https://bugs.launchpad.net/openstack-ansible/+filebug | 14:17 |
| jwitko | thanks I'm on it | 14:18 |
| *** Mudpuppy has joined #openstack-ansible | 14:18 | |
| jwitko | would you consider this a security vuln? | 14:18 |
| jwitko | no right ? | 14:18 |
| odyssey4me | add the version of OSA you're using, the symptoms and any information that can help replicate the issue | 14:18 |
| odyssey4me | no, I don't think so personally | 14:18 |
| odyssey4me | mhayden ^ | 14:18 |
| odyssey4me | this is more of a functional bug in my eyes | 14:19 |
| jwitko | https://bugs.launchpad.net/openstack-ansible/+bug/1635274 | 14:19 |
| openstack | Launchpad bug 1635274 in openstack-ansible "Certificates are not being distributed across all controllers and containers" [Undecided,New] | 14:19 |
| odyssey4me | jwitko I'll tweak it a bit, thanks | 14:19 |
| mhayden | odyssey4me: ah, interesting find | 14:20 |
| mhayden | seems more like a bug than a security vulnerability | 14:20 |
| *** haad1 has quit IRC | 14:21 | |
| *** DanyC has joined #openstack-ansible | 14:21 | |
| *** DanyC has left #openstack-ansible | 14:21 | |
| jwitko | odyssey4me, thanks! one more quick question, is it an issue for the external and internal LB vips to be identical ? | 14:22 |
| jwitko | or for one to be a the hostname the other to be the IP the hostname resolves to | 14:22 |
| odyssey4me | jwitko if you use DNS names, the names must be resolvable by the keepalived/haproxy hosts | 14:22 |
| odyssey4me | but yeah, no issue if they're the same | 14:23 |
| *** Mudpuppy_ has joined #openstack-ansible | 14:23 | |
| odyssey4me | generally I'd say use DNS for external (those become the public endpoints) and IP's for internal (internal endpoints are used by services) | 14:23 |
| odyssey4me | but I know that some feel that internal should also use DNS | 14:23 |
| jwitko | ok great thats what I was doing. but the SSL errors made me question things | 14:24 |
| jwitko | I think the SSL errors were a symtpom of each server having different certs though | 14:24 |
| jwitko | but after I matched all the certs haproxy was still throwing shade my way with those SSL errors | 14:24 |
| *** TxGirlGeek has joined #openstack-ansible | 14:26 | |
| *** Mudpuppy has quit IRC | 14:26 | |
| jwitko | is there a way to destroy all containers and start fresh? | 14:28 |
| jwitko | now I'm getting SSL certificate errors eaching out to github ;\ | 14:28 |
| jwitko | fatal: [infra3_repo_container-00f006e4]: FAILED! => {"changed": false, "failed": true, "msg": "Failed to validate the SSL certificate for raw.githubusercontent.com:443. Make sure your managed systems have a valid CA certificate installed. You can use validate_certs=False if you do not need to confirm the servers identity but this is unsafe and not recommended. Paths checked for this | 14:28 |
| jwitko | platform: /etc/ssl/certs, /etc/pki/ca-trust/extracted/pem, /etc/pki/tls/certs, /usr/share/ca-certificates/cacert.org, /etc/ansible"} | 14:28 |
| cloudnull | mornings | 14:31 |
| jwitko | good morning cloudnull | 14:32 |
| cloudnull | o/ jwitko | 14:32 |
| *** Mudpuppy_ is now known as Mudpuppy | 14:32 | |
| *** omiday has joined #openstack-ansible | 14:34 | |
| *** Jeffrey4l has joined #openstack-ansible | 14:35 | |
| *** smatzek has joined #openstack-ansible | 14:35 | |
| cloudnull | jwitko: did you get the ssl to github going? | 14:37 |
| jwitko | cloudnull, i destroyed all my containers and am running setup-hosts.yml again | 14:37 |
| cloudnull | ok | 14:37 |
| openstackgerrit | Merged openstack/openstack-ansible: [docs] Update all links to refer to Newton published docs https://review.openstack.org/388624 | 14:37 |
| jwitko | will keep you updated | 14:38 |
| *** karimb has quit IRC | 14:38 | |
| cloudnull | i'll be around today . | 14:38 |
| odyssey4me | andymccr here's hoping this one will go through :) https://review.openstack.org/389232 | 14:39 |
| *** chris_hultin|AWA is now known as chris_hultin | 14:39 | |
| *** karimb has joined #openstack-ansible | 14:40 | |
| *** dxiri has joined #openstack-ansible | 14:48 | |
| *** smatzek has quit IRC | 14:51 | |
| openstackgerrit | Merged openstack/openstack-ansible: [docs] Remove openstack-ansible.rc sourcing instruction https://review.openstack.org/389185 | 14:54 |
| *** phalmos has joined #openstack-ansible | 14:55 | |
| jwitko | cloudnull, setup-hosts.yml completed successfully. moving on to setup-infrastructure.yml | 14:56 |
| mgariepy | jwitko, I saw a similar issue with github certs, at the end it was a mtu issue. | 14:57 |
| *** marst has quit IRC | 14:58 | |
| *** Jack_Iv has joined #openstack-ansible | 15:00 | |
| mhayden | odyssey4me / andymccr: i need to leave about halfway through our weekly meeting today -- would one of y'all be able to finish it out? | 15:01 |
| *** galstrom_zzz is now known as galstrom | 15:01 | |
| *** weezS has joined #openstack-ansible | 15:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server: Ensure self-signed certificates are distributed https://review.openstack.org/389240 | 15:02 |
| andymccr | mhayden: sometimes i doubt your commitment to sparkle motion | 15:02 |
| odyssey4me | jwitko ^ patch to test out | 15:02 |
| mhayden | andymccr: WAT | 15:02 |
| andymccr | mhayden: its a Donnie Darko ref! | 15:04 |
| *** smatzek has joined #openstack-ansible | 15:06 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server: Ensure self-signed certificates are distributed https://review.openstack.org/389240 | 15:06 |
| palendae | andymccr, I thought that was Little Miss Sunshine | 15:07 |
| *** smatzek has quit IRC | 15:07 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 15:09 |
| andymccr | palendae: now you've got me double checking haha | 15:09 |
| *** marst has joined #openstack-ansible | 15:10 | |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install: Add ability to set upper constraints file https://review.openstack.org/388097 | 15:10 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install: Add ability to set upper constraints file https://review.openstack.org/388109 | 15:11 |
| *** thorst_ has joined #openstack-ansible | 15:11 | |
| *** thorst_ has quit IRC | 15:11 | |
| *** thorst_ has joined #openstack-ansible | 15:13 | |
| jwitko | odyssey4me, thanks! | 15:15 |
| odyssey4me | jwitko if it works for you, please comment and vote in the review | 15:16 |
| jwitko | odyssey4me can you help me apply it to my repo and test? do i need to delete anything or do something to test it ? | 15:16 |
| odyssey4me | jwitko on the deployment host do this: | 15:17 |
| odyssey4me | cd /etc/ansible/roles; rm -rf haproxy_server; git clone https://github.com/openstack/openstack-ansible-haproxy_server.git haproxy_server | 15:17 |
| odyssey4me | then: git fetch https://git.openstack.org/openstack/openstack-ansible-haproxy_server refs/changes/40/389240/2 && git cherry-pick FETCH_HEAD | 15:18 |
| odyssey4me | then: cd /opt/openstack-ansible/playbooks; openstack-ansible haproxy-install.yml | 15:18 |
| odyssey4me | then check if it worked | 15:19 |
| odyssey4me | you may need to do this to regen the keys | 15:19 |
| odyssey4me | openstack-ansible haproxy-install.yml -e 'haproxy_ssl_self_signed_regen=yes' | 15:20 |
| jwitko | should I do that before running the haproxy-install.yml playbook ? | 15:21 |
| *** hughmFLEXin has quit IRC | 15:21 | |
| odyssey4me | jwitko that will run the playbook and regen the keys at the same time | 15:21 |
| odyssey4me | but the other bits to update the role, yes do them in that order | 15:21 |
| jwitko | oh, doh. | 15:22 |
| jwitko | ty! will report back | 15:22 |
| *** deepak_jon has joined #openstack-ansible | 15:22 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-tests: Fix post-up and post-down script on CentOS https://review.openstack.org/389247 | 15:23 |
| mgariepy | andymccr, ^^ ;) | 15:23 |
| andymccr | mgariepy: ahh nice - i thought i had tested that one :P but obviously not that well | 15:24 |
| *** kylek3h has joined #openstack-ansible | 15:25 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_neutron: [WIP] Fix Calico testing https://review.openstack.org/386526 | 15:27 |
| *** admin0 has left #openstack-ansible | 15:28 | |
| *** admin0 has quit IRC | 15:28 | |
| *** deepak_jon1 has joined #openstack-ansible | 15:31 | |
| *** deepak_jon has quit IRC | 15:32 | |
| *** deepak_jon1 is now known as deepak_jon | 15:32 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Ensure that pip_install uses upper-constraints https://review.openstack.org/388700 | 15:33 |
| *** weezS has quit IRC | 15:35 | |
| *** hj-hpe has joined #openstack-ansible | 15:37 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Use centralised Ansible test scripts https://review.openstack.org/389133 | 15:38 |
| *** karimb has quit IRC | 15:43 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010010 https://review.openstack.org/383943 | 15:44 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010020 https://review.openstack.org/385583 | 15:44 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Security: Add tasks for RHEL-07-010260 https://review.openstack.org/385587 | 15:44 |
| *** deepak_jon1 has joined #openstack-ansible | 15:45 | |
| *** karimb has joined #openstack-ansible | 15:46 | |
| jwitko | ok odyssey4me, setup-infrastructure has completed successfully. currently all servers hold haproxy.pem that I generated and distributed. I will execute your commands now and see if the md5 sum is new and matches on all servers afterwards. | 15:47 |
| *** deepak_jon has quit IRC | 15:47 | |
| *** deepak_jon1 is now known as deepak_jon | 15:47 | |
| kylek3h | andymccr: Hey thanks for patch for galera_server. I'm tried to recreate the issue but hit other problems. Now the host system is being reloaded. I probably won't be able to get to another recreate until at least Monday. | 15:48 |
| jwitko | odyssey4me, looks like it worked | 15:49 |
| jwitko | keys are new and identical | 15:49 |
| andymccr | kylek3h: no problem! hopefully that solves it - I believe it should, we hit the same bug in trusty a while back, but i think we thought the systemd files worked slightly differently - it turns out it's still using the old mysql init file | 15:51 |
| *** kylek3h has quit IRC | 15:51 | |
| *** DanyC has joined #openstack-ansible | 15:53 | |
| *** DanyC has quit IRC | 15:55 | |
| jwitko | odyssey4me, I left a comment on the ticket | 15:55 |
| mhayden | OpenStack-Ansible weekly meeting coming up soon in #openstack-meeting-4! :) | 15:55 |
| mhayden | ^^ 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, admin0, michaelgugino, ametts, v1k0d3n, severion, | 15:55 |
| mhayden | bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397, qwang,nishpatwa_ | 15:55 |
| *** DanyC has joined #openstack-ansible | 15:56 | |
| palendae | mhayden, I'm gonna edit the page real quick | 15:56 |
| mhayden | palendae: THE AGENDA HAS BEEN SET FIRMLY IN STONE | 15:56 |
| mhayden | IT IS IMMUTABLE | 15:56 |
| * mhayden cackles | 15:56 | |
| *** karimb has quit IRC | 15:56 | |
| palendae | It sure is - that's been the agenda for the last 3 weeks :) | 15:56 |
| asettle | mhayden: yeah yeah yeah | 15:56 |
| mhayden | oh burn | 15:57 |
| andymccr | hahaha | 15:57 |
| palendae | Ok my edit's done | 15:57 |
| jwitko | damn, even after implementing the SSL cert fix I am still seeing SSL handshake failures come across haproxy | 15:58 |
| jwitko | Oct 20 15:57:46 localhost haproxy[20172]: 10.89.103.147:46118 [20/Oct/2016:15:57:46.447] repo_cache-front-1/1: SSL handshake failure | 15:58 |
| * asettle raises hand | 15:58 | |
| asettle | mhayden: | 15:58 |
| asettle | is it going to be a logn one or a short one? | 15:58 |
| *** DanyC has quit IRC | 15:58 | |
| asettle | I have a craving for a chicken schnitzel | 15:58 |
| palendae | Looked short to me | 15:58 |
| mhayden | asettle: i have to leave at half past | 15:59 |
| asettle | Excellent. I c'mon for you schnitty | 15:59 |
| mhayden | so i hope it's short! | 15:59 |
| asettle | comin'* | 15:59 |
| asettle | mhayden: that's what she said? | 15:59 |
| palendae | >.> | 15:59 |
| mhayden | asettle: i am so offended | 15:59 |
| * asettle exist stage left | 15:59 | |
| asettle | exits* | 15:59 |
| asettle | omg I can't even type | 15:59 |
| asettle | Who hired me? | 15:59 |
| * mhayden toots | 15:59 | |
| *** david-lyle_ is now known as david-lyle | 16:01 | |
| *** BjoernT has joined #openstack-ansible | 16:02 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_ironic: [WIP] Enable virtualBMC testing https://review.openstack.org/381049 | 16:02 |
| mgariepy | can I get some review on this please ? :D https://review.openstack.org/#/c/389247/ | 16:03 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh: Use centralised Ansible test scripts https://review.openstack.org/389128 | 16:05 |
| *** TxGirlGeek has quit IRC | 16:07 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:07 | |
| *** pjm6 has quit IRC | 16:07 | |
| *** kjw3 has quit IRC | 16:10 | |
| *** pjm6 has joined #openstack-ansible | 16:11 | |
| *** TxGirlGeek has quit IRC | 16:12 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:13 | |
| jwitko | hey guys, I can not get passed the setup-openstack.yml playbook. It looks like a couple steps that are dependant on the haproxy VIP for checking venvs are consistently getting SSL errors in HAProxy and the ansible step fails. I've created http://paste.ubuntu.com/23354425/ to show you | 16:14 |
| jwitko | this is 14.0.0.rc4 | 16:14 |
| *** weezS has joined #openstack-ansible | 16:15 | |
| *** kjw3 has joined #openstack-ansible | 16:15 | |
| *** TxGirlGeek has quit IRC | 16:15 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:15 | |
| stevelle | jwitko: using any overrides for the keystone service (nginx or uwsgi enabled?) or providing your own cert in overrides? | 16:16 |
| jwitko | stevelle, I am using no overrides and not providing my own certs. | 16:17 |
| jwitko | I had an issue where the certs were not being distributed but odyssey4me submitted a patch that fixed that issue | 16:17 |
| jwitko | I can also see the issue in a simple wget, although wget retries a bunch of times | 16:18 |
| *** TxGirlGeek has quit IRC | 16:18 | |
| *** openstackgerrit has quit IRC | 16:18 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:19 | |
| *** maeker has joined #openstack-ansible | 16:19 | |
| *** openstackgerrit has joined #openstack-ansible | 16:19 | |
| Adri2000 | trying to deploy ceilometer with gnocchi as a backend in newton, it seems ceilometer still wants to connect to a mongo database, am I correct? | 16:19 |
| openstackgerrit | chandanc proposed openstack/openstack-ansible-os_neutron: Add support for Neutron FWaaS v2 https://review.openstack.org/388548 | 16:21 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-pip_install: Fix same-host offline install race https://review.openstack.org/389269 | 16:21 |
| *** phalmos has quit IRC | 16:24 | |
| openstackgerrit | Merged openstack/openstack-ansible: Automatically source the rc file for ansible/ansible-playbook https://review.openstack.org/388877 | 16:26 |
| openstackgerrit | Merged openstack/openstack-ansible: Work around bad libvirt-python wheel https://review.openstack.org/389064 | 16:26 |
| openstackgerrit | Merged openstack/openstack-ansible: Passing eth0 to br-vlan https://review.openstack.org/388932 | 16:26 |
| openstackgerrit | Merged openstack/openstack-ansible: Do not extend lists in AIO config https://review.openstack.org/388933 | 16:26 |
| *** TxGirlGeek has quit IRC | 16:27 | |
| *** TxGirlGeek has joined #openstack-ansible | 16:27 | |
| *** asettle has quit IRC | 16:29 | |
| *** TxGirlGeek has quit IRC | 16:30 | |
| palendae | If people have the time, https://review.openstack.org/#/c/383920/ adds some more checking to the inventory check mode | 16:33 |
| *** hughmFLEXin has joined #openstack-ansible | 16:34 | |
| Adri2000 | bug #1628952 - that seems to be what I was talking about | 16:38 |
| openstack | bug 1628952 in openstack-ansible "os_ceilometer role missing functional gate testing of gnocchi storage option" [Medium,In progress] https://launchpad.net/bugs/1628952 - Assigned to Travis Truman (travis-truman) | 16:38 |
| *** TxGirlGeek has joined #openstack-ansible | 16:39 | |
| jwitko | damn I can not get past this, I even attempted to add retries but its not retrying on failure with the uri module | 16:39 |
| *** hughmFLEXin has quit IRC | 16:40 | |
| *** hughmFLEXin has joined #openstack-ansible | 16:41 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Work around bad libvirt-python wheel https://review.openstack.org/388934 | 16:45 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Remove openstack-ansible.rc sourcing instruction https://review.openstack.org/389274 | 16:47 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Use centralised Ansible test scripts https://review.openstack.org/388792 | 16:48 |
| *** kjw3 has quit IRC | 16:59 | |
| *** mgariepy has quit IRC | 16:59 | |
| *** TxGirlGeek has quit IRC | 17:00 | |
| *** TxGirlGeek has joined #openstack-ansible | 17:00 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts: Restore configurability of container resolvers https://review.openstack.org/389276 | 17:00 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_server: Remove duplicate when conditions https://review.openstack.org/389278 | 17:00 |
| *** hughmFLE_ has joined #openstack-ansible | 17:01 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-galera_server: Remove duplicate when conditions https://review.openstack.org/389278 | 17:01 |
| *** hughmFLEXin has quit IRC | 17:03 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_ceilometer: Roll-up of multiple fixes for os_ceilometer role https://review.openstack.org/385697 | 17:05 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_server: [DOCS] Added release-name as a watermark to Docs. https://review.openstack.org/388163 | 17:09 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Simplify pip options/constraints mechanism https://review.openstack.org/388230 | 17:09 |
| *** karimb has joined #openstack-ansible | 17:14 | |
| *** karimb has quit IRC | 17:15 | |
| *** karimb has joined #openstack-ansible | 17:16 | |
| *** hughmFLE_ has quit IRC | 17:16 | |
| *** karimb has quit IRC | 17:17 | |
| *** karimb has joined #openstack-ansible | 17:17 | |
| *** phalmos has joined #openstack-ansible | 17:18 | |
| *** hughmFLEXin has joined #openstack-ansible | 17:19 | |
| *** hughmFLEXin has quit IRC | 17:22 | |
| *** hughmFLEXin has joined #openstack-ansible | 17:23 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts: Restore configurability of container resolvers https://review.openstack.org/389276 | 17:30 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins: Remove deprecated first_available_file https://review.openstack.org/388845 | 17:38 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [TEST] Update to Ansible v2.1.3.0-0.1.rc1 https://review.openstack.org/382001 | 17:40 |
| *** omiday has quit IRC | 17:43 | |
| jwitko | Hey if your guys meeting is over and anyone is around I'm a bit stuck on these intermittent SSL handshake failures coming across my HAProxy logs and stopping setup-openstack.yml from succeeding | 17:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [TEST] Update to Ansible v2.2.0.0-0.2.rc2 https://review.openstack.org/389292 | 17:44 |
| jwitko | there was a guy last night experiencing the same issue as well | 17:44 |
| jwitko | thought the SSL cert fix would get rid of them but it doesn't look like that is true | 17:45 |
| openstackgerrit | weezer su proposed openstack/openstack-ansible-ops: Enhanced the setup-cobbler.sh https://review.openstack.org/389296 | 17:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: [DOCS] Added release-name as a watermark to Docs. https://review.openstack.org/388134 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386802 | 17:48 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Use upper constraints for all tox targets https://review.openstack.org/388429 | 17:48 |
| *** alij has joined #openstack-ansible | 17:49 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Changed the home-page link https://review.openstack.org/382220 | 17:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Make ceilometer sample interval configurable https://review.openstack.org/382057 | 17:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385543 | 17:50 |
| odyssey4me | jmccrory if you have a little free time, https://review.openstack.org/#/q/topic:bp/central-test-repository+status:open needs an extra eye :) | 17:52 |
| odyssey4me | also https://review.openstack.org/#/q/project:%255Eopenstack/openstack-ansible.*+topic:use-upper-constraints+status:open | 17:52 |
| jmccrory | sure, i'll start going through them | 17:53 |
| dxiri | guys, I am having a problem with haproxy where its kind of intermitent and making the playbook run fail cause it fails to retrieve some files | 17:56 |
| *** markvoelker has quit IRC | 17:56 | |
| dxiri | reading trought history, its the same exact problem as jwitko | 17:57 |
| dxiri | *trough | 17:57 |
| cloudnull | ssl related? | 17:57 |
| evrardjp | jwitko: dxiri is it an horizon+haproxy issue? | 18:00 |
| dxiri | cloudnull: I haven't done anything regarding ssl, its just the default config | 18:00 |
| dxiri | and it seems the task tries over plain http and not on https | 18:00 |
| dxiri | if I do a wget it works, but sometimes on the first try, other times upon retrying | 18:01 |
| jwitko | here is an example of the playbook failure and the resutling /var/log/haproxy.log entry | 18:01 |
| jwitko | http://paste.ubuntu.com/23354425/plain/ | 18:01 |
| jwitko | at first I thought it was because all the haproxy.pem files were different, but we fixed that here https://review.openstack.org/#/c/389240/2 | 18:02 |
| jwitko | and it is still happening | 18:02 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts: Restore configurability of container resolvers https://review.openstack.org/389276 | 18:02 |
| dxiri | evrardjp: I don't believe horizon has anything to do here yet, is it even configured at this point? | 18:02 |
| *** karimb has quit IRC | 18:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Roll-up of multiple fixes for os_ceilometer role https://review.openstack.org/385697 | 18:03 |
| dxiri | its the repo http://172.25.190.251:8181/venvs/14.0.0/ubuntu/glance-14.0.0-x86_64.checksum | 18:03 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Add ability to set upper constraints file https://review.openstack.org/389067 | 18:03 |
| jwitko | Aye I do not believe this is horizon related, most errors are coming at the repo level on port 8181 | 18:03 |
| dxiri | above its a example url I took from the task | 18:03 |
| evrardjp | dxiri: just asking, I don't know where your problem is yet :p | 18:03 |
| dxiri | evrardjp: sure, I am just giving as much info as possible :D | 18:04 |
| evrardjp | can one of you show the issue? | 18:04 |
| dxiri | jwitko pasted it above | 18:04 |
| evrardjp | ok | 18:04 |
| evrardjp | could you drop your haproxy config file somewhere? | 18:05 |
| *** jcrst has joined #openstack-ansible | 18:05 | |
| evrardjp | I'm surprised it worked for the first and not for the other ones | 18:05 |
| evrardjp | dxiri: you got the same issue? | 18:05 |
| dxiri | yes | 18:06 |
| evrardjp | is lsyncd running? | 18:06 |
| evrardjp | (just checking everything to make sure the most standard errors are verified) | 18:06 |
| jwitko | evrardjp, it is intermittent (the SSL handshake failure) it will happen in a wget attempt as well | 18:06 |
| jwitko | but wget handles failures beter than uri module | 18:07 |
| jwitko | and retries | 18:07 |
| dxiri | lsyncd running where? | 18:07 |
| dxiri | jwitko: why SSL handshake? url is fetched using plain http | 18:07 |
| -openstackstatus- NOTICE: The Gerrit service on review.openstack.org is being restarted now in an attempt to resolve some mismatched merge states on a few changes, but should return momentarily. | 18:07 | |
| jwitko | evrardjp the haproxy config would be the default install of a 5 node cluster | 18:07 |
| evrardjp | dxiri: on the repo container nodes | 18:07 |
| jwitko | oh wow... | 18:07 |
| jwitko | thats a good point | 18:07 |
| dxiri | :P | 18:07 |
| jwitko | i mean you can literally see though | 18:07 |
| evrardjp | jwitko: I'd be happy if you could show it to me | 18:07 |
| jwitko | the failure is directly related | 18:08 |
| evrardjp | jwitko: well not 100% | 18:08 |
| evrardjp | depending on how haproxy is configured it has to (or doesn't have to) terminate connections | 18:09 |
| dxiri | evrardjp: here: http://pastebin.com/xNnqgYZr | 18:09 |
| dxiri | haproxy config | 18:09 |
| evrardjp | thanks | 18:09 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Add ability to set upper constraints file https://review.openstack.org/389068 | 18:09 |
| dxiri | lsyncd: Active: active (exited) since Wed 2016-10-19 23:44:40 UTC; 18h ago | 18:10 |
| dxiri | looks running ok | 18:10 |
| evrardjp | dxiri: thanks | 18:10 |
| *** javeriak has joined #openstack-ansible | 18:10 | |
| evrardjp | well 18h ago is far, but let's say it's running for the example | 18:10 |
| evrardjp | could you curl into 172.25.190.222:8181/venvs/14.0.0/ubuntu/glance-14.0.0-x86_64.checksum correctly ? | 18:11 |
| *** phalmos has quit IRC | 18:11 | |
| *** electrofelix has quit IRC | 18:11 | |
| evrardjp | or just 172.25.190.222:8181/venvs/ | 18:11 |
| dxiri | you mean 251? | 18:12 |
| evrardjp | from the inside | 18:12 |
| dxiri | 172.25.190.251 | 18:12 |
| evrardjp | nope | 18:12 |
| dxiri | where did you got 222 from? | 18:12 |
| evrardjp | from your paste :p | 18:12 |
| dxiri | lol | 18:12 |
| dxiri | haven't seen it | 18:12 |
| evrardjp | I told you I was doing on a step by step | 18:13 |
| *** alij has quit IRC | 18:13 | |
| *** alij has joined #openstack-ansible | 18:13 | |
| jwitko | evrardjp, my config is the same as dxiri | 18:13 |
| jwitko | just fyi | 18:13 |
| jwitko | looked it over | 18:13 |
| jwitko | except IPs of course | 18:13 |
| evrardjp | good 2 birds 1 stone | 18:13 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance: Add os-brick to glance_pip_packages https://review.openstack.org/388014 | 18:13 |
| evrardjp | nobody found the solution yet, right? | 18:13 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Add os-brick to glance_pip_packages https://review.openstack.org/389305 | 18:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Add os-brick to glance_pip_packages https://review.openstack.org/389306 | 18:14 |
| dxiri | (I ran the playbook again before you told me to curl, will do as soon as it fails) | 18:14 |
| evrardjp | OMG I just realized for one work I'm doing on mitaka I used a module rabbitmq_queue that is available starting on ansible 2.0 | 18:15 |
| jwitko | evrardjp, please see here | 18:15 |
| jwitko | http://paste.ubuntu.com/23354996/ | 18:15 |
| evrardjp | FAIL | 18:15 |
| *** karimb has joined #openstack-ansible | 18:15 | |
| evrardjp | jwitko: could you show which ip does that refer to? | 18:16 |
| evrardjp | jwitko: is 10.89.103.25 your repo container ip? | 18:16 |
| jwitko | that is my VIP | 18:16 |
| *** phalmos has joined #openstack-ansible | 18:16 | |
| jwitko | haproxy hosted VIP | 18:16 |
| evrardjp | then please do on the container :D | 18:16 |
| evrardjp | first container IP, then internal VIP then external VIP | 18:17 |
| evrardjp | progressive :p | 18:17 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-lxc_hosts: Restore configurability of container resolvers https://review.openstack.org/389276 | 18:17 |
| evrardjp | is there someone here familiar with rabbitmqadmin ? | 18:17 |
| dxiri | root@sjociara03:/opt/openstack-ansible/playbooks# curl 172.25.190.222:8181/venvs/14.0.0/ubuntu/glance-14.0.0-x86_64.checksum | 18:18 |
| dxiri | a39a1e2ea6281d9c86a39f275149dcf72c382d8f | 18:18 |
| dxiri | oh look a that...it got past that part this time...I got really lucky...haha | 18:18 |
| evrardjp | dxiri: so it works, and you never have problems when accessing directly on the repo container | 18:18 |
| *** phalmos has quit IRC | 18:18 | |
| dxiri | works everytime yes, no retries | 18:19 |
| *** phalmos has joined #openstack-ansible | 18:19 | |
| dxiri | but when passing trough haproxy | 18:19 |
| evrardjp | dxiri: ok | 18:19 |
| dxiri | root@sjociara03:/opt/openstack-ansible/playbooks# curl 172.25.190.251:8181/venvs/14.0.0/ubuntu/glance-14.0.0-x86_64.checksum | 18:19 |
| dxiri | curl: (52) Empty reply from server | 18:19 |
| dxiri | root@sjociara03:/opt/openstack-ansible/playbooks# curl 172.25.190.251:8181/venvs/14.0.0/ubuntu/glance-14.0.0-x86_64.checksum | 18:19 |
| dxiri | a39a1e2ea6281d9c86a39f275149dcf72c382d8f | 18:19 |
| dxiri | and the next time works | 18:19 |
| evrardjp | hehe | 18:19 |
| dxiri | sometimes its more than 1 try, sometimes works right away, sometimes its about 7 tries :P | 18:20 |
| evrardjp | dxiri: yup not surprised | 18:20 |
| evrardjp | let's do some haproxy then | 18:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-tests: Use upper constraints for all tox targets https://review.openstack.org/389167 | 18:21 |
| evrardjp | well I'm gonna pollute the chan if I continue | 18:21 |
| jwitko | same results here | 18:21 |
| *** phalmos_ has joined #openstack-ansible | 18:24 | |
| *** phalmos has quit IRC | 18:27 | |
| openstackgerrit | Merged openstack/openstack-ansible: Implement upper constraints for Ansible bootstrap https://review.openstack.org/388051 | 18:31 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that upper constraints are always applied https://review.openstack.org/388220 | 18:36 |
| evrardjp | jwitko: dxiri | 18:41 |
| evrardjp | I've got your solution | 18:42 |
| jwitko | :D! | 18:42 |
| jwitko | !!! | 18:42 |
| openstack | jwitko: Error: "!!" is not a valid command. | 18:42 |
| evrardjp | :D | 18:42 |
| *** karimb has quit IRC | 18:42 | |
| evrardjp | your external vip and your internal vip are identical | 18:42 |
| jwitko | yes | 18:42 |
| jwitko | I was told this is not an issue | 18:42 |
| evrardjp | well it is | 18:42 |
| jwitko | :O | 18:42 |
| evrardjp | you cannot bind two times on the same ip/port | 18:43 |
| evrardjp | and then haproxy doesn't know what to do | 18:43 |
| evrardjp | :p | 18:43 |
| evrardjp | have two different ips | 18:43 |
| evrardjp | OR | 18:43 |
| evrardjp | better | 18:43 |
| *** karimb has joined #openstack-ansible | 18:43 | |
| evrardjp | have a real external net | 18:43 |
| evrardjp | and have boundaries | 18:44 |
| dxiri | well...shit :) | 18:44 |
| dxiri | can I use an ip from the lxcbridge? | 18:44 |
| dxiri | inet 10.0.3.1/24 brd 10.0.3.255 scope global lxcbr0 | 18:44 |
| evrardjp | nope | 18:44 |
| dxiri | say...that one? as internal? | 18:44 |
| evrardjp | for the ease of use | 18:44 |
| evrardjp | internal is part of the br-mgmt | 18:44 |
| evrardjp | easier | 18:44 |
| *** deepak_jon has quit IRC | 18:45 | |
| evrardjp | it avoids many redirections | 18:45 |
| evrardjp | external you could use an ip on the same net if you want | 18:45 |
| evrardjp | it doesn't really make sense in terms of security, but I'm not your CSO, right :p | 18:45 |
| dxiri | mmmm, and can I just leave the external_vip blank? I can reach the IP I set on internal from anywhere on my network | 18:45 |
| evrardjp | so you could 2 2 different ips of mgmt net | 18:46 |
| dxiri | so I don't really have a need for the external | 18:46 |
| dxiri | you mean have 2 ips on br-mgmt? | 18:46 |
| evrardjp | yup | 18:46 |
| evrardjp | one for the internal facing | 18:46 |
| evrardjp | and one for the external facing | 18:46 |
| evrardjp | and then reach from the external facing for external facing operations | 18:46 |
| *** markvoelker has joined #openstack-ansible | 18:46 | |
| dxiri | and when VMs need to use the bridge, which IP will they use to go out the bridge? | 18:46 |
| evrardjp | that's entirely up to you | 18:47 |
| evrardjp | that's how you define your networks | 18:47 |
| *** markvoelker has quit IRC | 18:47 | |
| *** markvoelker has joined #openstack-ansible | 18:47 | |
| evrardjp | I'd suggest you to isolate tenant traffic and openstack control plane traffic | 18:47 |
| jwitko | evrardjp, so do you know what I have to do to make this work now? I've edited the config file with the internal vip | 18:47 |
| jwitko | do I need to create all containers? or just run the infra setup again ? | 18:48 |
| evrardjp | well there is one big problem jwitko dxiri of the change of internal_lb_vip_address or external_lb_vip_address | 18:48 |
| evrardjp | we don't force a change of endpoints | 18:48 |
| evrardjp | so the easier for you | 18:49 |
| evrardjp | is to redeploy | 18:49 |
| evrardjp | and redestroy every container :p | 18:49 |
| evrardjp | you could do it smarter, but it's tedious | 18:49 |
| evrardjp | :p | 18:49 |
| jwitko | so | 18:49 |
| jwitko | openstack-ansible lxc-containers-destroy.yml, followed by setup-hosts, setup-infra, setup-openstack ? | 18:49 |
| evrardjp | jwitko: could you show me your o_u_c ? | 18:49 |
| evrardjp | yup | 18:50 |
| jwitko | sure one moment | 18:50 |
| evrardjp | oh also | 18:50 |
| evrardjp | delete your /etc/haproxy/conf.d folder | 18:50 |
| jwitko | on the bare metal controllers right ? | 18:51 |
| jwitko | the haproxy_infra_hosts ? | 18:51 |
| evrardjp | on tha haproxy hosts yes | 18:51 |
| evrardjp | ansible -m shell -a "rm -rf /etc/haproxy/conf.d" haproxy_all | 18:51 |
| evrardjp | if I'm not mistaken | 18:51 |
| jwitko | that is correct | 18:52 |
| *** tobias_ has joined #openstack-ansible | 18:52 | |
| odyssey4me | mhayden with https://review.openstack.org/388799 merged it might be time to do an osa-differ and tweet it :) ? | 18:53 |
| odyssey4me | unfortunately not all the emails have come through to the announce list - they're working on it | 18:54 |
| openstackgerrit | Merged openstack/openstack-ansible-security: Use centralised Ansible test scripts https://review.openstack.org/389166 | 18:54 |
| dxiri | wait up...in my case | 18:55 |
| dxiri | I had a hostname, what If I just change the record to point to the new IP and restart haproxy? | 18:55 |
| dxiri | shouldn't that do it? | 18:55 |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_client: Use centralised Ansible test scripts https://review.openstack.org/389162 | 18:56 |
| openstackgerrit | Merged openstack/openstack-ansible-rsyslog_server: Use centralised Ansible test scripts https://review.openstack.org/389164 | 18:56 |
| *** hybridpollo has joined #openstack-ansible | 18:56 | |
| dxiri | so ansible -m shell -a "systemctl restart haproxy" haproxy_all | 18:56 |
| evrardjp | odyssey4me: cool | 18:57 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins: Use centralised Ansible test scripts https://review.openstack.org/389153 | 18:57 |
| evrardjp | dxiri: yes | 18:58 |
| evrardjp | that would be fine | 18:58 |
| evrardjp | don't forget to change your hosts file etc | 18:58 |
| dxiri | I have my infra hosts pointing to a DNS server on the same subnet | 18:59 |
| *** tobias_ has quit IRC | 18:59 | |
| evrardjp | I have to go guys | 18:59 |
| evrardjp | but I think you're closer to a solution | 18:59 |
| dxiri | evrardjp: thanks a lot dude! | 18:59 |
| jwitko | ty, i'm redeploying now with new IP | 18:59 |
| jwitko | in the same subnet | 18:59 |
| evrardjp | drop message on the channel here, I'll read that tomorrow morning | 18:59 |
| jwitko | will let you know how it goes | 19:00 |
| evrardjp | jwitko: cool | 19:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Use centralised Ansible test scripts https://review.openstack.org/389147 | 19:01 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_server: Use centralised Ansible test scripts https://review.openstack.org/389157 | 19:02 |
| dxiri | [WARNING]: Host file not found: /etc/ansible/hosts | 19:02 |
| dxiri | [WARNING]: provided hosts list is empty, only localhost is available | 19:02 |
| dxiri | what am I missing here? | 19:02 |
| dxiri | trying to run the ansible command above | 19:02 |
| openstackgerrit | weezer su proposed openstack/openstack-ansible-ops: Remove the expired key and opensuse repo after cobbler being installed https://review.openstack.org/388906 | 19:05 |
| *** asettle has joined #openstack-ansible | 19:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Use centralised Ansible test scripts https://review.openstack.org/389156 | 19:06 |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally: Use centralised Ansible test scripts https://review.openstack.org/389143 | 19:08 |
| odyssey4me | dxiri that is cosmetic for newton onwards - all config is done via environment variables instead of an ansible.cfg file | 19:08 |
| odyssey4me | hmm, if there's a way to turn those off then we should include the env vars to do that | 19:09 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server: Use centralised Ansible test scripts https://review.openstack.org/389155 | 19:09 |
| *** jcrst has quit IRC | 19:10 | |
| dxiri | odyssey4me: so the command actually went trough? | 19:10 |
| *** TxGirlGeek has quit IRC | 19:10 | |
| dxiri | cause I got no ansible output back | 19:10 |
| dxiri | no success/failure...not nothing | 19:10 |
| *** TxGirlGeek has joined #openstack-ansible | 19:10 | |
| odyssey4me | dxiri that sounds very odd | 19:11 |
| dxiri | I am on the deployment host trying to run ansible -m shell -a "systemctl restart haproxy" haproxy_all | 19:11 |
| dxiri | and I am inside /opt/openstack-ansible/playbooks | 19:11 |
| Matias | dxiri: add -i inventory to the command line | 19:12 |
| mhayden | odyssey4me: oh nice | 19:12 |
| odyssey4me | dxiri before executing that, execute: . /usr/local/bin/openstack-ansible.rc | 19:13 |
| odyssey4me | we have a patch coming in to resolve that | 19:13 |
| mhayden | odyssey4me: a diff from 13.0.0 to 14.0.0? that'd be, uh, large | 19:13 |
| odyssey4me | mhayden yeah, apparently that's why the release announcement got held ;) | 19:13 |
| dxiri | infra1 | SUCCESS | rc=0 >> | 19:13 |
| dxiri | there we go! | 19:13 |
| odyssey4me | mebbe there's a way to slim it down to some essentials | 19:13 |
| mhayden | odyssey4me: i'll give it a try :) | 19:13 |
| *** TxGirlGeek has quit IRC | 19:13 | |
| mhayden | let's hope i don't start raiding swap memory on my system | 19:14 |
| *** TxGirlGeek has joined #openstack-ansible | 19:14 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi: Use centralised Ansible test scripts https://review.openstack.org/389132 | 19:14 |
| openstackgerrit | Merged openstack/openstack-ansible-ops: Enhanced the setup-cobbler.sh https://review.openstack.org/389296 | 19:14 |
| openstackgerrit | Merged openstack/openstack-ansible-ops: Add remove known_hosts in the instruction for Multi-Node AIO https://review.openstack.org/388898 | 19:14 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance: Use centralised Ansible test scripts https://review.openstack.org/389131 | 19:15 |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Use upper constraints for all tox targets https://review.openstack.org/389167 | 19:15 |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Fix neutron_agent IP to be neutron_server IP https://review.openstack.org/389170 | 19:15 |
| *** TxGirlGeek has quit IRC | 19:16 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:16 | |
| odyssey4me | mhayden w00t! http://lists.openstack.org/pipermail/openstack-announce/2016-October/001855.html | 19:16 |
| mhayden | wow | 19:17 |
| mhayden | lengthy | 19:17 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Use centralised Ansible test scripts https://review.openstack.org/389137 | 19:18 |
| *** asettle has quit IRC | 19:18 | |
| mhayden | odyssey4me: done tweeted it | 19:18 |
| odyssey4me | mhayden done retweeted your tweet | 19:18 |
| mhayden | gonna run down my Pebble battery | 19:19 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron: Use centralised Ansible test scripts https://review.openstack.org/389139 | 19:19 |
| *** johnmilton has quit IRC | 19:21 | |
| *** alij has quit IRC | 19:21 | |
| mhayden | odyssey4me: oh, got goosed with the openstack_other.yml not being there :) | 19:22 |
| odyssey4me | mhayden eh? | 19:22 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder: Use centralised Ansible test scripts https://review.openstack.org/389129 | 19:23 |
| mhayden | i rely on playbooks/defaults/repo_packages/openstack_other.yml | 19:23 |
| mhayden | but now that's split up a bit | 19:23 |
| odyssey4me | mhayden ah, yes - you should just iterate through playbooks/defaults/repo_packages/*.yml ? | 19:23 |
| * odyssey4me is going to stop computering now | 19:23 | |
| odyssey4me | release done, mic dropped - andymccr it's your baby now :p | 19:23 |
| *** asettle has joined #openstack-ansible | 19:24 | |
| mhayden | awww | 19:24 |
| mhayden | time for some scotch, odyssey4me? :) | 19:24 |
| odyssey4me | mhayden most definitely | 19:24 |
| *** jheroux has quit IRC | 19:24 | |
| odyssey4me | night all - catch y'all tomorrow | 19:24 |
| *** TxGirlGeek has quit IRC | 19:24 | |
| dxiri | odyssey4me: see ya | 19:24 |
| jwitko | are direct pull requests to github ignored for sauce-ansible ? | 19:25 |
| *** TxGirlGeek has joined #openstack-ansible | 19:25 | |
| mhayden | i heard andymccr talking about converting this project into openstack-cfengine | 19:26 |
| andymccr | mhayden: i trusted you | 19:26 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Changed the home-page link https://review.openstack.org/382220 | 19:26 |
| mhayden | andymccr: first mistake | 19:26 |
| andymccr | now i know :( | 19:26 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Use centralised Ansible test scripts https://review.openstack.org/389151 | 19:27 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Use centralised Ansible test scripts https://review.openstack.org/389141 | 19:32 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic: Use centralised Ansible test scripts https://review.openstack.org/389136 | 19:35 |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Fix post-up and post-down script on CentOS https://review.openstack.org/389247 | 19:35 |
| *** phalmos_ has quit IRC | 19:36 | |
| *** jcrst has joined #openstack-ansible | 19:37 | |
| jwitko | Hey guys, does HA proxy not set up a floating IP for the internal VIP address as it does for the external? | 19:38 |
| baz | So what's the desired default for Horizon (Newton) image creation access? I ask because the out-of-box OSA has missing config in /etc/ansible/roles/os_horizon/templates/horizon_local_settings.py.j2 that prevents image creation. | 19:38 |
| *** Jack_Iv has quit IRC | 19:38 | |
| *** johnmilton has joined #openstack-ansible | 19:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: [DOCS] Added release-name as a watermark to Docs. https://review.openstack.org/388111 | 19:40 |
| baz | jwitko you need to define both in internal and external in your global_overrides and add the config mentioned in the openstack_user_config example to your user_variables.yml | 19:41 |
| jwitko | wait... I need the global_overrides config in two places? | 19:41 |
| baz | and of course they need to be in your excluded address range since a container doesn't try to pick it. | 19:41 |
| baz | global_overrides is in openstack_user_config.yml | 19:42 |
| jwitko | so I have this in user_variables.yml | 19:43 |
| jwitko | haproxy_keepalived_external_vip_cidr: "{{external_lb_vip_address}}/32" | 19:43 |
| jwitko | haproxy_keepalived_internal_vip_cidr: "{{internal_lb_vip_address}}/32" | 19:43 |
| jwitko | haproxy_keepalived_external_interface: br-mgmt | 19:43 |
| jwitko | haproxy_keepalived_internal_interface: br-mgmt | 19:43 |
| jwitko | keepalived_use_latest_stable: True | 19:43 |
| baz | that's what I have in my testing deploy as well. | 19:44 |
| baz | however you need in openstack_user_config.yml you need: | 19:44 |
| baz | global_overrides: | 19:44 |
| baz | internal_lb_vip_address: 1.2.3.4 | 19:45 |
| *** admin0 has joined #openstack-ansible | 19:45 | |
| baz | external_lb_vip_address: 5.6.7.8 | 19:45 |
| jwitko | yes | 19:45 |
| jwitko | i have that as well | 19:45 |
| jwitko | global_overrides: | 19:46 |
| jwitko | internal_lb_vip_address: 10.89.103.24 | 19:46 |
| jwitko | external_lb_vip_address: openstack.lab.mgmt.domain.net | 19:46 |
| jwitko | and that hostname resolves to 10.89.103.25 | 19:46 |
| jwitko | however haproxy only creates and listens on the floating .25 IP, there is no IP on the servers listening for 10.89.103.24 | 19:46 |
| baz | it should be VRRP'ing both of those addresses with keepalived | 19:46 |
| baz | and there should be bind statements in /etc/haproxy.cfg for each of those addresses. | 19:47 |
| jwitko | there are bind statements in haproxy.cfg | 19:47 |
| *** spotz_zzz is now known as spotz | 19:47 | |
| jwitko | frontend repo_all-front-2 | 19:48 |
| jwitko | bind 10.89.103.24:8181 | 19:48 |
| baz | and keepalived is alive? and do you see VRRP for both vips with tcpdump? | 19:48 |
| jwitko | keepalived is running | 19:49 |
| jwitko | status is happy | 19:49 |
| jwitko | tcp 0 0 10.89.103.24:8181 0.0.0.0:* LISTEN 1803/haproxy | 19:50 |
| jwitko | tcp 0 0 10.89.103.25:8181 0.0.0.0:* LISTEN 1803/haproxy | 19:50 |
| jwitko | yet I can't telnet to .24, i can telnet to .25 though | 19:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance: Add os-brick to glance_pip_packages https://review.openstack.org/389305 | 19:50 |
| baz | what about the acl lines for the frontends in haproxy.cfg? | 19:51 |
| baz | and one of them should be ssl'd | 19:52 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Update paste, policy and rootwrap configurations 2016-10-12 https://review.openstack.org/385543 | 19:53 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: [DOCS] Added release-name as a watermark to Docs. https://review.openstack.org/388134 | 19:53 |
| baz | so you'd need to use openssl s_client -connect host-or-ip:8181 | 19:53 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Use upper constraints for all tox targets https://review.openstack.org/388429 | 19:53 |
| jwitko | baz, there should still be an open connection via telnet with or without ssl | 19:57 |
| *** omiday has joined #openstack-ansible | 20:00 | |
| baz | and both 24 and 25 show up in 'ip a show br-mgmt'? | 20:03 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386802 | 20:08 |
| jwitko | baz, no neither does | 20:18 |
| *** thorst_ has quit IRC | 20:22 | |
| *** phalmos has joined #openstack-ansible | 20:25 | |
| *** spotz is now known as spotz_zzz | 20:25 | |
| dxiri | guys, my openstack is running! yay! | 20:27 |
| dxiri | installing some images now | 20:27 |
| dxiri | but I noticed this: The current Horizon settings indicate no valid image creation methods are available. Providing an image location and/or uploading from the local file system must be allowed to support image creation. | 20:27 |
| dxiri | where can I change that? | 20:27 |
| *** galstrom is now known as galstrom_zzz | 20:33 | |
| *** weezS has quit IRC | 20:36 | |
| *** phalmos has quit IRC | 20:36 | |
| openstackgerrit | Merged openstack/openstack-ansible: [docs] Add os_trove to Advanced Config section https://review.openstack.org/388627 | 20:40 |
| *** askb has joined #openstack-ansible | 20:47 | |
| jwitko | so I'm changing the internal VIP to an IP on my vxlan network | 20:49 |
| jwitko | to be on a different network than my external VIP | 20:49 |
| jwitko | do I need to do anything to make sure the internal VIP uses the correct br-vxlan interface? | 20:49 |
| jwitko | I think just set haproxy_keepalived_internal_interface right? | 20:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Remove 'ignore_errors: true' in favor of 'failed_when: false' https://review.openstack.org/386802 | 20:53 |
| *** weezS_ has joined #openstack-ansible | 20:54 | |
| *** thorst_ has joined #openstack-ansible | 20:56 | |
| mrda | Morning OSA | 21:00 |
| *** jeh has joined #openstack-ansible | 21:06 | |
| *** jeh has quit IRC | 21:07 | |
| *** jeh has joined #openstack-ansible | 21:12 | |
| cloudnull | baz: RE: Image create. personally I think that's an oversight | 21:13 |
| cloudnull | do you know which sections are needed? | 21:13 |
| openstackgerrit | weezer su proposed openstack/openstack-ansible-ops: Remove the expired key and opensuse repo after cobbler being installed https://review.openstack.org/388906 | 21:15 |
| cloudnull | dxiri: -- baz mentioned that horizon is not allowing image creation. | 21:15 |
| cloudnull | in general I use the CLI to do all of that. | 21:16 |
| cloudnull | which is likely why the horizon config is missing. | 21:16 |
| cloudnull | as I imagine that most of us use the CLI for day to day ops | 21:17 |
| cloudnull | I'm looking into that now. | 21:17 |
| dxiri | cloudnull: awesome! thank you :) | 21:19 |
| dxiri | also found another problem, I can't create instances | 21:19 |
| dxiri | Error: Failed to perform requested operation on instance "inst1", the instance has an error status: Please try again later [Error: No valid host was found. There are not enough hosts available.]. | 21:19 |
| dxiri | this is using a cirros test image...with 1gb ram and 5gb disk | 21:19 |
| dxiri | really small | 21:19 |
| cloudnull | are the compute-nodes checking in ? | 21:20 |
| *** jeh has quit IRC | 21:20 | |
| cloudnull | baz: dxiri: it would seem we need to set https://github.com/openstack/horizon/blob/master/openstack_dashboard/local/local_settings.py.example#L369-L378 | 21:22 |
| c-mart | dxiri, check nova-compute.log for your compute host(s) | 21:23 |
| *** retreved has quit IRC | 21:23 | |
| dxiri | if I go to system-information, I see all compute services enabled and ip, last updated 0 minutes | 21:23 |
| c-mart | i had the "no valid host was found" issue and it was due to something completely unrelated to the compute service -- Nova couldn't talk to another API, and that was apparent in nova-compute.log | 21:24 |
| dxiri | here is the error from nova-compute.log | 21:28 |
| dxiri | http://pastebin.com/WgnM4aSP | 21:28 |
| *** hybridpollo has quit IRC | 21:31 | |
| *** schwicht has quit IRC | 21:33 | |
| *** Jeffrey4l has quit IRC | 21:34 | |
| admin0 | dxiri: check your libvirt logs | 21:34 |
| admin0 | 2016-10-20 16:26:52.637 3573 ERROR nova.compute.manager PortBindingFailed: Binding failed for port 5bc482f5-7c8b-4496-bdee-230165294653, please check neutron logs for more information. | 21:34 |
| admin0 | evrardjp: around ? | 21:34 |
| admin0 | evrardjp: when you read through the logs next, tomorrow i will have 3 new environments iwth 3 controlles each .. will test/validate the haproxy/keepalive issue and confirm | 21:36 |
| c-mart | is anyone here familiar with using Ceph as a storage back-end for cinder and Glance? I have a question about how OSA is configured to talk to Ceph | 21:36 |
| cloudnull | baz: dxiri: so nevermind . https://bugs.launchpad.net/horizon/+bug/1632383 | 21:37 |
| openstack | Launchpad bug 1632383 in OpenStack Dashboard (Horizon) "The current Horizon settings indicate no valid image creation methods are available" [Undecided,New] | 21:37 |
| cloudnull | the image create via horizon seems to be an upstream bug | 21:37 |
| admin0 | c-mart it works :) | 21:39 |
| dxiri | cloudnull: so all I need to do is wait ? | 21:39 |
| cloudnull | yup | 21:39 |
| cloudnull | sadly | 21:39 |
| dxiri | admin0: checking neutro logs now | 21:39 |
| *** weezS_ has quit IRC | 21:39 | |
| admin0 | check libvrit as well | 21:40 |
| admin0 | sometimes no logs give any error, you still get host not found, but real reason in libvirt log | 21:40 |
| *** spotz_zzz is now known as spotz | 21:40 | |
| c-mart | admin0, do you know why we need to define storage_hosts in openstack_user_config.yml, if we'll just be talking to Ceph as the storage back-end? what do the storage_hosts do in that case? | 21:41 |
| cloudnull | dxiri: baz: i've poked a few folks and ill let you know what shakes out. | 21:42 |
| admin0 | c-mart, i actually wrote that part i the ansible docs | 21:42 |
| admin0 | they do nothing | 21:42 |
| admin0 | so you need to set the is_metal: false | 21:42 |
| admin0 | so that the api’s run in the controller nodes themselves | 21:42 |
| admin0 | c-mart: i wrote a blog on this: http://www.openstackfaq.com/openstack-ansible-ceph/ | 21:43 |
| admin0 | i am working to expand the current blog to be with newton and 3 ceph backends/pools: SSD, SATA and Hybrid | 21:43 |
| *** hughmFLEXin has quit IRC | 21:43 | |
| c-mart | I found that blog post admin0 :) | 21:44 |
| admin0 | i already tested 3 backends for volumes so when you create volume, in dropdown you can select sata vs ssd | 21:44 |
| admin0 | that will go up .. maybe will write after the summit | 21:44 |
| c-mart | cool. so in that blog post, when you define three storage hosts, you're saying those containers are just created on the controller nodes?? | 21:45 |
| *** hybridpollo has joined #openstack-ansible | 21:46 | |
| admin0 | yes | 21:48 |
| admin0 | in that way, the storage-api/containers are load balanced | 21:48 |
| admin0 | and de-coupled from your ceph | 21:48 |
| admin0 | they are just like any other api | 21:48 |
| c-mart | ah, now that makes so much more sense. thank you | 21:48 |
| *** schwicht has joined #openstack-ansible | 21:49 | |
| admin0 | right now i am doing something complex .. i have OSA as region2, but on a keystone db that is from region1 and on juno .. so its upgraded to newton, region 2 uses the v3, while i am chaning the v2 to this new newton keystone so that my existing region1 (still on icehoise) works fine as usual with 1 single keystone serving region2 and region1 :D | 21:50 |
| admin0 | \o/ | 21:50 |
| admin0 | chaning/fixing* | 21:51 |
| dxiri | so, this may have something to do with my problem...I created a new network on horizon (made up the subnet) and set it shared | 21:51 |
| admin0 | dxiri: it works :) | 21:52 |
| *** chris_hultin is now known as chris_hultin|AWA | 21:52 | |
| dxiri | does that subnet needs to be the subnet for the br-vxlan or br-vlan I configured when setting openstack up | 21:52 |
| admin0 | dxiri: something like this: http://www.openstackfaq.com/openstack-add-direct-attached-dhcp-ip/ | 21:52 |
| admin0 | no | 21:52 |
| admin0 | it needs to be different | 21:52 |
| admin0 | there http://www.openstackfaq.com/openstack-add-direct-attached-dhcp-ip/ , i add a different range as shared, and a different range for floating ip | 21:53 |
| admin0 | br-vxlan is just one subnet you need for the vtep and on its own vlan/tag/isolation | 21:53 |
| admin0 | br-vlan has no ips | 21:54 |
| admin0 | so whatever ips you add as floating or shared , and if its under a vlan tag, it carries/forwards that | 21:54 |
| *** spotz is now known as spotz_zzz | 21:55 | |
| dxiri | ok so If I want an internal network between the vms only (for now) how should I create the network? I just want the VM to have an internal IP, it doesn't matter if it can't reach anything else for now | 21:56 |
| *** thorst_ has quit IRC | 21:56 | |
| *** thorst_ has joined #openstack-ansible | 21:57 | |
| *** hughmFLEXin has joined #openstack-ansible | 21:57 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add missing Ironic auth plugin variable https://review.openstack.org/389174 | 21:58 |
| *** spotz_zzz is now known as spotz | 22:01 | |
| *** weezS has joined #openstack-ansible | 22:02 | |
| admin0 | dxiri: in that case, create a private network .. and just spawn those vms int he private net | 22:02 |
| cloudnull | ^ admin0 beat me too it :) | 22:02 |
| admin0 | if you are using osa and defaults, they will use some X vxlan | 22:02 |
| * admin0 goes back to meditation :D | 22:03 | |
| admin0 | cloudnull: i suddenly have 4 multi-node environments to test stuff out :D | 22:03 |
| admin0 | so been very occupied | 22:03 |
| c-mart | *bow* thank you admin0 | 22:03 |
| admin0 | will you be in the summit ? | 22:03 |
| cloudnull | sadly I will not be at the summit. | 22:04 |
| cloudnull | :'( | 22:04 |
| dxiri | admin0: I am using OSA and defaults yes, but *how* do I create that private network? that's my question, do I just make stuff up and create a network with whatever range I want and whatever vlan ID I want? | 22:04 |
| admin0 | skype link for people to meetup in openestack summit barcelona: https://join.skype.com/fnzd1JCc4Vxp | 22:04 |
| admin0 | dxiri: | 22:04 |
| admin0 | you do not worry about the id | 22:04 |
| admin0 | please feel free to create whatever valid IP you want to have | 22:05 |
| *** spotz is now known as spotz_zzz | 22:05 | |
| admin0 | 1.1.1.0/8 is perfectly valid as well :) | 22:05 |
| *** thorst_ has quit IRC | 22:05 | |
| *** asettle has quit IRC | 22:05 | |
| admin0 | because its internal and tied to some vxlan | 22:05 |
| admin0 | i use 1.1.1.0/8; 2.2.2.0/24 — those are nice :D | 22:06 |
| admin0 | 11.11.11.0/24 :D | 22:06 |
| admin0 | if you create a router, then you need to enter the router namespace before you can ssh in or reach it internally | 22:06 |
| admin0 | if you don’t then console is your friend :) | 22:06 |
| baz | cloudnull the "missing" config is https://github.com/openstack/horizon/blob/master/openstack_dashboard/local/local_settings.py.example#L754-L756 | 22:10 |
| baz | an additional problem relates to haproxy if you set 'haproxy_stats_enabled: true' in user_variables.yml | 22:11 |
| cloudnull | baz: so adding that list fixed it ? | 22:13 |
| c-mart | admin0 a suggested change to your Ceph blog post: "disable is_metal from /etc/openstack_deploy/env.d/cinder.yml" is now a different folder path, should be /opt/openstack-ansible/env.d/cinder.yml, I believe. | 22:13 |
| admin0 | yes | 22:13 |
| admin0 | that was for the old release | 22:13 |
| admin0 | newton one is coming up | 22:13 |
| baz | https://github.com/openstack/openstack-ansible-haproxy_server/blob/master/templates/haproxy.cfg.j2#L26 | 22:14 |
| cloudnull | baz: also what's the other issue when stats are enabled? | 22:14 |
| cloudnull | ha. | 22:14 |
| cloudnull | is that not working? | 22:14 |
| baz | the part after listen stats needs to be on a separate line as a 'bind' | 22:14 |
| baz | just like normal services | 22:14 |
| baz | additionally might want to add 'stats refresh 60s' to have it auto-reload | 22:15 |
| baz | as for the horizon thing, yes that's what was ultimately needed to fix the Horizon image upload issue after digging through the code. | 22:15 |
| cloudnull | ok | 22:16 |
| cloudnull | do you have a moment to submit a PR? | 22:17 |
| dxiri | k I created the net...same problem :( | 22:19 |
| *** thorst_ has joined #openstack-ansible | 22:20 | |
| baz | those required REST APIs were added either in commits on Feb 10th and Apr 15th of this year. | 22:20 |
| dxiri | mmm | 22:20 |
| dxiri | 2016-10-20 22:16:20.832 5199 WARNING stevedore.named [req-41fc2818-9a7a-4e91-ba89-2f7fc7f11941 1ead964a48754da2822f0fafd8be3890 98b68cdb2f834a06b56733f57db7f8e5 - - -] Could not load neutron.agent.linux.interface.BridgeInterfaceDriver | 22:20 |
| cloudnull | baz: regatding haproxy -- http://104.130.20.56:1936/ its working for me as is | 22:20 |
| * cloudnull removed auth just FYI | 22:20 | |
| cloudnull | but the horizon bits seem to fix it | 22:20 |
| *** admin0 has quit IRC | 22:20 | |
| baz | cloudnull which version of haproxy? | 22:21 |
| cloudnull | 1.5.18 | 22:21 |
| baz | HA-Proxy version 1.6.3 2015/12/25 fails if the line is 'listen stats 1.2.3.4:1936' | 22:21 |
| baz | the config changed in 1.6 and stats looks just like services in that it takes a bind statement | 22:22 |
| baz | so there'll need to be ansible conditionals based on which version of haproxy is used. | 22:22 |
| *** admin0 has joined #openstack-ansible | 22:23 | |
| cloudnull | it looks like the 1.6 syntax format works in 1.5 | 22:23 |
| cloudnull | I changed my config to http://cdn.pasteraw.com/nol2eyn1mld7nod3wslw4sahqlahf8s | 22:24 |
| cloudnull | and the stats dash is still working | 22:24 |
| *** thorst_ has quit IRC | 22:24 | |
| *** schwicht has quit IRC | 22:24 | |
| baz | it likely works on 1.5.x but fails in 1.4.x, but does anything OSA supports as the host distribution ship 1.4.x? | 22:25 |
| jwitko | hey cloudnull, can you help me understand how to work the internal/external vip where I don't have a separate network? | 22:28 |
| jwitko | it doesn't seem to work with both ips on the same subnet | 22:28 |
| jwitko | and definitely doesn't work where both IPs are identical | 22:28 |
| admin0 | jwitko: its supposed to be different on production setting | 22:28 |
| admin0 | i use that a lot | 22:28 |
| admin0 | the vip addresses | 22:29 |
| jwitko | although it seems like it would work in the case of identical IPs if there was a conditional that only generated HAProxy output once instead of twice | 22:29 |
| jwitko | admin0, so can you help me ? | 22:29 |
| cloudnull | baz: i think we're always using 1.5 and above in newton + | 22:29 |
| admin0 | use different ips and endpoints jwitko | 22:29 |
| cloudnull | jwitko: it should work with different IPs on the same subnet | 22:30 |
| jwitko | admin0, "ips and endpoints" isn't the IP the endpoint ? | 22:30 |
| jwitko | cloudnull, only one of them is ever pingable | 22:30 |
| admin0 | nope | 22:30 |
| baz | cloudnull then it's a safe change. 1.6 just made it a requirement | 22:30 |
| admin0 | endpoint traditionally refers to URL :) | 22:30 |
| admin0 | URL is something that points to IP | 22:30 |
| cloudnull | baz: agreeded. | 22:30 |
| jwitko | like right now I have an IP for internal and a hostname in a different subnet for external | 22:30 |
| jwitko | i can ping the IP and telnet to the ports | 22:30 |
| jwitko | but the external hostname I can not connect to any of the ports or ping it | 22:31 |
| jwitko | and it is on the same network as br-mgmt | 22:31 |
| jwitko | i can see the servers listening on that IP with the ports open | 22:31 |
| cloudnull | i think if you bind to two ips on the same subnet you'll need to add an alias IP on the network interface of the host. | 22:31 |
| jwitko | but can't connect | 22:31 |
| cloudnull | baz: do you have a moment to make the change to hap ? | 22:32 |
| admin0 | jwitko: i have this on config: internal_lb_vip_address: cloud101int.stack31.com external_lb_vip_address: cloud101.stack31.com && and on variables haproxy_keepalived_external_vip_cidr: "10.11.12.3/22" haproxy_keepalived_internal_vip_cidr: "172.29.236.3/22" haproxy_keepalived_external_interface: ens2 haproxy_keepalived_internal_interface: br-mgmt | 22:32 |
| cloudnull | or horizon. if not I'd be happy to, I just don't want to take your commit. | 22:32 |
| baz | cloudnull give me a second to create a pull | 22:33 |
| admin0 | so the VIP addresses gets used as endpoints | 22:33 |
| admin0 | from DNS i map the IPs to the addresses | 22:33 |
| *** TxGirlGeek has quit IRC | 22:33 | |
| jwitko | admin0, cloudnull, ok thanks I think that helps. if I want to change the IP/endpoints is there any easy way or do i have to destroy containers? | 22:33 |
| admin0 | truncate table keystone :D | 22:33 |
| admin0 | if its new one | 22:33 |
| admin0 | sorry | 22:33 |
| cloudnull | jwitko: change the IPs of all containers ? | 22:34 |
| *** weezS has quit IRC | 22:34 | |
| cloudnull | or just the endpoints? | 22:34 |
| admin0 | no .. its just the VIP/endpoint | 22:34 |
| admin0 | truncate the endpoints | 22:34 |
| jwitko | cloudnull, if I'm changing my VIPs and CIDRs for haproxy | 22:34 |
| baz | since there's already a launchpad bug for the Horizon issue I'll just attach a patch to that. | 22:34 |
| jwitko | just to reset those throughout the environment | 22:34 |
| cloudnull | cool thanks baz | 22:34 |
| admin0 | jwitko: you need to re-do the endpoints | 22:35 |
| admin0 | delete the existing one from d | 22:35 |
| admin0 | db | 22:35 |
| admin0 | and then re-run the os-keystone | 22:35 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Add upgrade playbook for ceilometer-api https://review.openstack.org/389378 | 22:35 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Add upgrade playbook for ceilometer-api https://review.openstack.org/389378 | 22:36 |
| cloudnull | jwitko: i'd make the change to config and then change the DB like so : update endpoint set url = REPLACE(url, '172.19.8.100', '172.19.0.21') where interface='public'; | 22:37 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Add upgrade playbook for ceilometer-api https://review.openstack.org/389378 | 22:37 |
| jwitko | i'd have to change the interface it binds too as well | 22:37 |
| jwitko | I think i'll just destroy the containers and the haproxy configs | 22:38 |
| jwitko | that would work yes? | 22:38 |
| admin0 | yesp | 22:38 |
| admin0 | for i in `lxc-ls`; do lxc-destroy -f -n $i ; done :D | 22:38 |
| cloudnull | the nuclear option works too. | 22:38 |
| cloudnull | the lxc-container-destroy.yml play will do it to all the things | 22:39 |
| admin0 | :D | 22:39 |
| admin0 | that too | 22:39 |
| cloudnull | https://www.youtube.com/watch?v=aCbfMkh940Q | 22:39 |
| admin0 | those coming to barcelona and want to join in https://join.skype.com/fnzd1JCc4Vxp | 22:39 |
| admin0 | its a skype chat group | 22:39 |
| admin0 | to arrange a meeting | 22:39 |
| dxiri | admin0: libvirt logs are blank | 22:45 |
| admin0 | :( | 22:45 |
| admin0 | strace :D | 22:45 |
| dxiri | unless I'm looking at the wrong place | 22:45 |
| admin0 | well, figure out why the VM does not create | 22:45 |
| admin0 | could be neutron, if not neutron, libvirt | 22:45 |
| dxiri | looking in /var/log/libvirt on the compute node | 22:46 |
| dxiri | /var/log/libvirt/libxl: | 22:46 |
| dxiri | total 0 | 22:46 |
| dxiri | /var/log/libvirt/lxc: | 22:46 |
| dxiri | total 0 | 22:46 |
| dxiri | /var/log/libvirt/qemu: | 22:46 |
| dxiri | total 0 | 22:46 |
| dxiri | /var/log/libvirt/uml: | 22:46 |
| dxiri | total 0 | 22:46 |
| dxiri | ah! | 22:47 |
| dxiri | 2016-10-20 17:47:14.742 5433 ERROR neutron.plugins.ml2.drivers.linuxbridge.agent.linuxbridge_neutron_agent [-] Interface eth12 for physical network flat does not exist. Agent terminated! | 22:47 |
| dxiri | but I don't recall creating any flat networks | 22:48 |
| dxiri | so wtf | 22:48 |
| admin0 | yeah | 22:49 |
| admin0 | see :D | 22:49 |
| BjoernT | it's probably listed inside the provider networks section | 22:49 |
| admin0 | well | 22:49 |
| admin0 | i know this | 22:49 |
| admin0 | its the defaults | 22:49 |
| admin0 | they list eth12 | 22:49 |
| admin0 | you have to remove the eth12 to yoru br-pxe or your pxe/standard interface and retry | 22:50 |
| BjoernT | https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L47 | 22:50 |
| BjoernT | remove that whole flat block | 22:50 |
| admin0 | dunno who ( yes developers, i am pointing to you) .. thought eth12 will exist in a normal box | 22:50 |
| admin0 | - network: | 22:51 |
| admin0 | group_binds: | 22:51 |
| admin0 | - neutron_linuxbridge_agent | 22:51 |
| admin0 | container_bridge: "br-vlan" | 22:51 |
| admin0 | container_type: "veth" | 22:51 |
| admin0 | container_interface: "eth12" | 22:51 |
| admin0 | host_bind_override: "ens2" | 22:51 |
| admin0 | type: "flat" | 22:51 |
| admin0 | net_name: "flat" | 22:51 |
| admin0 | boo | 22:51 |
| admin0 | sorry | 22:51 |
| BjoernT | Lol you ad admin need to make sure that you have proper config, OSA doesn't deploy clean without that. Unless you have AIO | 22:51 |
| admin0 | host_bind_override: "ens2" in my case | 22:51 |
| BjoernT | yepp | 22:51 |
| admin0 | dxiri: its a consipracy from the devs to ensure we operators are awake .. so need to change host_bind_override: "eth12” to whatver you have in reality | 22:52 |
| admin0 | else the first deployment fails | 22:52 |
| admin0 | due to this | 22:52 |
| admin0 | #nastydevs :D | 22:53 |
| * admin0 is an operator | 22:53 | |
| jwitko | looks like i need to reset bare metal pip configs as well.. they're still trying to reach the old vip | 22:53 |
| *** marst has quit IRC | 22:53 | |
| admin0 | well, you have 00apt-proxy everywhere which you need to rm -f as well | 22:54 |
| admin0 | else the playbooks fail | 22:54 |
| dxiri | is that the physical interface itself? not the br-vlan bridge? | 22:55 |
| dxiri | In my case I am using a single nic (eth0) | 22:55 |
| admin0 | put eth0 and yuo are done :) | 22:56 |
| jwitko | admin0, any others I should know about ? | 22:56 |
| admin0 | lots :) | 22:56 |
| admin0 | depends on your deployment | 22:57 |
| admin0 | greenfield or brown ? | 22:57 |
| admin0 | cinder, ceph, | 22:57 |
| jwitko | greenfield, cinder | 22:57 |
| admin0 | well, apart from this change, and if hte VIps are correctly defined, it should work without issues | 22:58 |
| dxiri | admin0: k changed...now...do I have to destroy everything and start over? | 22:59 |
| admin0 | yes please | 22:59 |
| dxiri | that is great! | 22:59 |
| * admin0 loves destruction and chaos | 22:59 | |
| dxiri | :P | 22:59 |
| dxiri | by the way I am following appendix A of the documentation | 23:00 |
| dxiri | would be good if this stuff get mentioned there | 23:00 |
| admin0 | i have 3 virtual and 2 phiysical multi-node envs.. i think i have destroyed and ran 500+ times already | 23:00 |
| dxiri | I also see yet another eth11 for neutron | 23:00 |
| dxiri | - network: | 23:01 |
| dxiri | container_bridge: "br-vlan" | 23:01 |
| dxiri | container_type: "veth" | 23:01 |
| dxiri | container_interface: "eth11" | 23:01 |
| dxiri | type: "vlan" | 23:01 |
| dxiri | range: "1:1" | 23:01 |
| dxiri | net_name: "vlan" | 23:01 |
| dxiri | group_binds: | 23:01 |
| dxiri | - neutron_linuxbridge_agent | 23:01 |
| dxiri | do I leave that alone? | 23:01 |
| admin0 | its for the container | 23:01 |
| admin0 | yes | 23:01 |
| admin0 | leave that alone please | 23:01 |
| dxiri | cool | 23:01 |
| dxiri | guess I will have to redownload and recreate the images after I destroy everything right? | 23:02 |
| admin0 | https://gist.github.com/a1git/af22cce0d39e389d689835e4a0d6aa4c - this is for a normal one | 23:02 |
| dxiri | Are you sure you want to destroy the LXC container data? [no]: | 23:04 |
| dxiri | what happens if I don't destroy data? | 23:04 |
| dxiri | bad thing will happen? | 23:04 |
| admin0 | and the variables https://gist.github.com/a1git/9831617c5d0933a53c8e346482e0f935 | 23:04 |
| admin0 | its greenfield :) and new | 23:04 |
| admin0 | you can destroy everything | 23:04 |
| dxiri | openstack-ansible lxc-containers-destroy.yml | 23:06 |
| dxiri | just that one right? | 23:06 |
| admin0 | yes | 23:06 |
| admin0 | and the ansible facts as well | 23:06 |
| dxiri | destroy the ansible facts? | 23:07 |
| *** schwicht has joined #openstack-ansible | 23:07 | |
| admin0 | remove the ansible facts if you remove the containers | 23:07 |
| dxiri | where? how? | 23:07 |
| dxiri | sry for being this noob :P | 23:07 |
| admin0 | rm -rf /etc/openstack_deploy/ansible_facts/ ## on the deploy hosts | 23:07 |
| admin0 | i am a noob as well | 23:07 |
| dxiri | k thanks! | 23:07 |
| admin0 | there is no shame in asking what you dont know | 23:08 |
| admin0 | dxiri: rnage 1:1 in vxlan | 23:10 |
| admin0 | it means you cannot have more than 1 private network | 23:10 |
| admin0 | look at mine: range: "10000:100000" | 23:11 |
| admin0 | br-vlan | 23:12 |
| admin0 | is 1:! | 23:12 |
| admin0 | 1:1 | 23:12 |
| admin0 | means you can have just 1 network | 23:13 |
| admin0 | thati is wrong | 23:13 |
| admin0 | you need at least a few | 23:13 |
| admin0 | one for floating, one for test, one for dev, one for prod — all share | 23:13 |
| admin0 | shared | 23:13 |
| * admin0 needs sleep | 23:17 | |
| admin0 | good night all | 23:17 |
| BjoernT | rnage 1:1 in vxlan: only means that you can't create networks over horizon beyond one vlan | 23:17 |
| admin0 | 1:17 AM | 23:17 |
| admin0 | its in his vlan | 23:17 |
| BjoernT | in neutron cli you can just override the segmentation_id | 23:17 |
| admin0 | i have at least 3 vlans for floating, 3 vlans for shared , all /22 each | 23:18 |
| jwitko | damn my deploy is still trying to go through the old VIPs | 23:19 |
| jwitko | something with pip is still telling it to go to the old IP | 23:19 |
| *** admin0 has left #openstack-ansible | 23:19 | |
| *** admin0 has quit IRC | 23:20 | |
| *** c-mart has quit IRC | 23:21 | |
| *** weezS has joined #openstack-ansible | 23:21 | |
| jwitko | "/root/.pip " | 23:24 |
| jwitko | haha | 23:24 |
| *** thorst_ has joined #openstack-ansible | 23:24 | |
| *** gouthamr has quit IRC | 23:27 | |
| *** hughmFLEXin has quit IRC | 23:27 | |
| *** maeker has quit IRC | 23:28 | |
| *** fops_ has joined #openstack-ansible | 23:30 | |
| *** Jeffrey4l has joined #openstack-ansible | 23:31 | |
| *** fops has quit IRC | 23:32 | |
| *** thorst_ has quit IRC | 23:33 | |
| openstackgerrit | Merged openstack/openstack-ansible: Work around bad libvirt-python wheel https://review.openstack.org/388934 | 23:35 |
| *** weezS has quit IRC | 23:35 | |
| *** BjoernT has quit IRC | 23:36 | |
| *** markvoelker has quit IRC | 23:37 | |
| *** agrebennikov_ has quit IRC | 23:46 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Add upgrade playbook for ceilometer-api https://review.openstack.org/389378 | 23:46 |
| *** schwicht has quit IRC | 23:47 | |
| openstackgerrit | Jameson Finney proposed openstack/openstack-ansible: Adds support for disks on HP Smart Array Controllers https://review.openstack.org/389053 | 23:56 |
| *** javeriak has quit IRC | 23:58 | |
| dxiri | having some problem with apt now, trying to run setup-hosts again | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!