| *** keedya has quit IRC | 00:04 | |
| *** baker has joined #openstack-ansible | 00:10 | |
| *** rcarrillocruz has left #openstack-ansible | 00:12 | |
| *** automagically has joined #openstack-ansible | 00:13 | |
| *** baker has quit IRC | 00:18 | |
| *** automagically has quit IRC | 00:27 | |
| *** ntpttr has left #openstack-ansible | 00:41 | |
| *** Bofu2MBP has quit IRC | 00:41 | |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Aodh venv if it mismatches the repo https://review.openstack.org/272079 | 00:48 |
|---|---|---|
| *** automagically has joined #openstack-ansible | 00:57 | |
| *** baker has joined #openstack-ansible | 01:04 | |
| *** baker has quit IRC | 01:09 | |
| openstackgerrit | Merged openstack/openstack-ansible: Reduce keystone cache expiration time https://review.openstack.org/273537 | 01:13 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Ceilometer venv if it mismatches the repo https://review.openstack.org/272984 | 01:25 |
| *** sdake_ has quit IRC | 01:27 | |
| *** weezS has quit IRC | 01:30 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Added conditional templating https://review.openstack.org/273293 | 01:31 |
| *** yarkot has quit IRC | 01:32 | |
| *** hughsaunders has quit IRC | 01:32 | |
| *** mgagne has quit IRC | 01:32 | |
| *** mgagne has joined #openstack-ansible | 01:32 | |
| *** mgagne has quit IRC | 01:32 | |
| *** mgagne has joined #openstack-ansible | 01:32 | |
| *** hughsaunders has joined #openstack-ansible | 01:33 | |
| *** fawadkhaliq has joined #openstack-ansible | 01:38 | |
| *** fawadkhaliq has quit IRC | 01:38 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible-specs: IRR - Create the os_crud_ops role https://review.openstack.org/272764 | 01:54 |
| *** dweaver has quit IRC | 01:55 | |
| *** dweaver has joined #openstack-ansible | 01:56 | |
| jmccrory | cloudnull seeing any slowness with gerrit? my git review has been hanging for about a minute | 01:57 |
| *** klamath has quit IRC | 01:57 | |
| *** automagically has quit IRC | 01:57 | |
| *** automagically has joined #openstack-ansible | 01:58 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Check for existence of /dev/kvm https://review.openstack.org/273844 | 02:02 |
| jmccrory | alright...about time | 02:02 |
| *** daneyon has joined #openstack-ansible | 02:03 | |
| *** baker has joined #openstack-ansible | 02:03 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Glance venv if it mismatches the repo https://review.openstack.org/273800 | 02:03 |
| *** izaakk has quit IRC | 02:03 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Aodh venv if it mismatches the repo https://review.openstack.org/273804 | 02:03 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Swift venv if it mismatches the repo https://review.openstack.org/273799 | 02:04 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Neutron venv if it mismatches the repo https://review.openstack.org/273803 | 02:04 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Ceilometer venv if it mismatches the repo https://review.openstack.org/273806 | 02:04 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Nova venv if it mismatches the repo https://review.openstack.org/273805 | 02:04 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Heat venv if it mismatches the repo https://review.openstack.org/273808 | 02:05 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Cinder venv if it mismatches the repo https://review.openstack.org/273807 | 02:05 |
| *** daneyon_ has joined #openstack-ansible | 02:05 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Re-deploy the Keystone venv if it mismatches the repo https://review.openstack.org/273809 | 02:05 |
| cloudnull | jmccrory: yes... its been dragging ass today | 02:06 |
| *** daneyon has quit IRC | 02:08 | |
| *** daneyon_ has quit IRC | 02:12 | |
| *** baker has quit IRC | 02:20 | |
| *** yarkot has joined #openstack-ansible | 02:48 | |
| *** mgagne has quit IRC | 02:51 | |
| *** mgagne has joined #openstack-ansible | 02:51 | |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Keystone venv if it mismatches the repo https://review.openstack.org/270222 | 03:00 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Nova venv if it mismatches the repo https://review.openstack.org/272027 | 03:00 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Neutron venv if it mismatches the repo https://review.openstack.org/272029 | 03:00 |
| *** tlian has quit IRC | 03:07 | |
| *** baker has joined #openstack-ansible | 03:16 | |
| *** Grey_H has joined #openstack-ansible | 03:18 | |
| *** keedya has joined #openstack-ansible | 03:21 | |
| openstackgerrit | Merged openstack/openstack-ansible: Implement multi-domain LDAP configuration for Keystone https://review.openstack.org/258015 | 03:22 |
| *** Grey_H has left #openstack-ansible | 03:25 | |
| *** Bjoern has joined #openstack-ansible | 03:25 | |
| *** andymccr is now known as meowmeow_ | 03:27 | |
| *** meowmeow_ is now known as imacat_ | 03:27 | |
| *** imacat_ is now known as meowmeow_ | 03:27 | |
| *** meowmeow_ is now known as andymccr | 03:28 | |
| *** baker has quit IRC | 03:33 | |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Heat venv if it mismatches the repo https://review.openstack.org/272032 | 03:39 |
| *** automagically has quit IRC | 03:44 | |
| *** sdake has joined #openstack-ansible | 03:45 | |
| *** keedya has quit IRC | 03:49 | |
| *** allan_h has quit IRC | 03:56 | |
| *** baker has joined #openstack-ansible | 04:04 | |
| *** Bjoern has quit IRC | 04:05 | |
| *** skamithi has joined #openstack-ansible | 04:08 | |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Cinder venv if it mismatches the repo https://review.openstack.org/272021 | 04:18 |
| *** baker_ has joined #openstack-ansible | 04:32 | |
| *** baker has quit IRC | 04:35 | |
| *** woodard has quit IRC | 04:35 | |
| *** markvoelker has quit IRC | 04:38 | |
| *** shausy has joined #openstack-ansible | 04:48 | |
| *** baker_ has quit IRC | 04:57 | |
| *** yatin has joined #openstack-ansible | 05:12 | |
| *** shausy has quit IRC | 05:36 | |
| *** shausy has joined #openstack-ansible | 05:36 | |
| *** markvoelker has joined #openstack-ansible | 05:39 | |
| *** markvoelker has quit IRC | 05:43 | |
| *** javeriak has joined #openstack-ansible | 05:56 | |
| *** javeriak has quit IRC | 05:58 | |
| *** javeriak_ has joined #openstack-ansible | 05:58 | |
| *** yatin has quit IRC | 06:05 | |
| *** yatin has joined #openstack-ansible | 06:06 | |
| *** yatin has quit IRC | 06:06 | |
| *** hybridpollo has quit IRC | 06:08 | |
| *** weezS has joined #openstack-ansible | 06:11 | |
| *** javeriak_ has quit IRC | 06:11 | |
| *** javeriak has joined #openstack-ansible | 06:11 | |
| *** weezS has quit IRC | 06:12 | |
| *** phiche1 has joined #openstack-ansible | 06:18 | |
| *** sdake has quit IRC | 06:19 | |
| *** phiche has quit IRC | 06:20 | |
| *** adac has joined #openstack-ansible | 06:41 | |
| openstackgerrit | Merged openstack/openstack-ansible: set unique galera server-id https://review.openstack.org/267054 | 06:41 |
| *** targon has joined #openstack-ansible | 06:49 | |
| openstackgerrit | Merged openstack/openstack-ansible: Remove unneeded playbook vars https://review.openstack.org/265665 | 06:59 |
| *** xek_ has quit IRC | 07:00 | |
| *** xek_ has joined #openstack-ansible | 07:01 | |
| *** raddaoui has joined #openstack-ansible | 07:02 | |
| *** adac has quit IRC | 07:05 | |
| *** raddaoui has quit IRC | 07:22 | |
| *** llu has joined #openstack-ansible | 07:34 | |
| *** phiche has joined #openstack-ansible | 07:39 | |
| *** phiche1 has quit IRC | 07:40 | |
| *** markvoelker has joined #openstack-ansible | 07:40 | |
| llu | any guru here can help me identify the problem listed as http://paste.openstack.org/show/485367/, i'm running aio using openstack-ansible 12.0.5 | 07:42 |
| llu | thanks~ | 07:42 |
| odyssey4me | llu oh, that's odd | 07:43 |
| odyssey4me | your kernel is old though, any chance that you can update it? | 07:44 |
| llu | odyssey4me: yes, it's already the latest, i'm using ubuntu 14.04 | 07:44 |
| *** markvoelker has quit IRC | 07:44 | |
| llu | my current kernel is 3.13.0-76-generic, I think it's the latest of ubuntu 14.04 | 07:45 |
| odyssey4me | llu ah, I think you're right - so we need to understand why the version check is failing | 07:51 |
| llu | hmm, I just realized that it's my own repo problem, will switch to official repo to install kernel 4 | 07:51 |
| odyssey4me | llu how do you mean? | 07:51 |
| llu | i'm using a internal ubuntu repo | 07:52 |
| llu | which kernel version do you suggest? | 07:52 |
| odyssey4me | llu your kernel version is fine | 07:53 |
| odyssey4me | it's odd that our check is failing | 07:53 |
| odyssey4me | can you change into /opt/openstack-ansible/playbooks and execute: ansible aio1 -m setup | grep ansible_kernel | 07:54 |
| llu | so do you know where to hack so I can get this running | 07:54 |
| llu | "ansible_kernel": "3.13.0-76-generic", | 07:55 |
| odyssey4me | ok, that's odd - for some reason the version compare thinks that your kernel is older | 07:56 |
| odyssey4me | try adding 'openstack_host_required_kernel: 3.13.0-76-generic' into /etc/openstack_deploy/user_variables.yml | 07:57 |
| llu | odyssey4me: yeah, the weird thing is that the version compare think i'm using kernel 3.13.0-32-generic | 07:57 |
| odyssey4me | hmmm | 07:58 |
| odyssey4me | ok, hang on a sec | 07:58 |
| odyssey4me | this is an AIO, right? | 07:58 |
| llu | yes | 07:58 |
| odyssey4me | very weird | 07:58 |
| llu | I updated the kernel from 32 to 76 when I first saw this message, and reboot the machine | 07:58 |
| llu | but ansible still thinks i'm using the old one | 07:59 |
| odyssey4me | heh, ok I know what happened here | 07:59 |
| odyssey4me | the fact cache still has the old info | 07:59 |
| llu | I also bootstrap-ansible and bootstrap-aio after I upgraded the kernel | 07:59 |
| odyssey4me | remove the fact cache - just execute: rm -rf /etc/openstack_deploy/ansible_facts | 08:00 |
| odyssey4me | we should actually add a kernel version check into the bootstrap | 08:00 |
| llu | and rerun scripts/run-laybooks.sh? | 08:00 |
| odyssey4me | llu yeah, that should do the trick | 08:01 |
| odyssey4me | if you added the openstack_host_required_kernel entry to /etc/openstack_deploy/user_variables.yml then you can remove that entry | 08:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Allow sourcing apt_key from ansible host or URL https://review.openstack.org/268676 | 08:04 |
| llu | odyssey4me: thx, the kernel version passed | 08:11 |
| *** tlian has joined #openstack-ansible | 08:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_client: Added conditional templeting https://review.openstack.org/273292 | 08:16 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Added conditional templating https://review.openstack.org/273293 | 08:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_client: Added conditional templating https://review.openstack.org/273292 | 08:23 |
| *** adac has joined #openstack-ansible | 08:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Interpret the swift_managed_regions variable correctly in with_nested loops https://review.openstack.org/273682 | 08:30 |
| *** mikelk has joined #openstack-ansible | 08:34 | |
| openstackgerrit | git-harry proposed openstack/openstack-ansible: Use Swift venv Python interpreter https://review.openstack.org/273943 | 08:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Consolidate lint tests into common tox environment https://review.openstack.org/272723 | 08:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Consolidate lint tests into common tox environment https://review.openstack.org/272723 | 08:47 |
| *** mikelk has quit IRC | 08:48 | |
| *** mikelk has joined #openstack-ansible | 08:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Consolidate lint tests into common tox environment https://review.openstack.org/272723 | 08:49 |
| openstackgerrit | Merged openstack/openstack-ansible: Don't install openjdk-7-jre into the Utility container https://review.openstack.org/273151 | 08:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Implement multi-domain LDAP configuration for Keystone https://review.openstack.org/273130 | 09:01 |
| odyssey4me | mattt hughsaunders can we get another review on https://review.openstack.org/268812 please? | 09:05 |
| mattt | odyssey4me: looking | 09:08 |
| hughsaunders | odyssey4me: mattt approved | 09:09 |
| odyssey4me | lol thanks guys | 09:09 |
| mattt | hughsaunders: BGBaaS | 09:09 |
| hughsaunders | mattt: I read it throughly.. | 09:09 |
| hughsaunders | and disapproved of cloudnull's early cherrypick | 09:09 |
| mattt | :) | 09:11 |
| odyssey4me | haha, I'm about to update the commit message when it merges. | 09:13 |
| *** skamithi has quit IRC | 09:13 | |
| *** gparaskevas has joined #openstack-ansible | 09:14 | |
| *** phiche has quit IRC | 09:20 | |
| *** admin0 has joined #openstack-ansible | 09:23 | |
| openstackgerrit | Merged openstack/openstack-ansible: Provide logrotate config for rsyncd on Swift storage hosts https://review.openstack.org/273231 | 09:31 |
| *** markvoelker has joined #openstack-ansible | 09:41 | |
| *** markvoelker has quit IRC | 09:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Allow sourcing apt_key from ansible host or URL https://review.openstack.org/268676 | 09:53 |
| openstackgerrit | Merged openstack/openstack-ansible: Ensure that Heat role works with multiple Keystone back-ends https://review.openstack.org/268812 | 10:20 |
| *** MCoLo has joined #openstack-ansible | 10:42 | |
| *** markvoelker has joined #openstack-ansible | 10:56 | |
| openstackgerrit | Merged openstack/openstack-ansible: Interpret the swift_managed_regions variable correctly in with_nested loops https://review.openstack.org/273682 | 10:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Ensure that Heat role works with multiple Keystone back-ends https://review.openstack.org/272714 | 10:59 |
| *** markvoelker has quit IRC | 11:02 | |
| *** adac has quit IRC | 11:16 | |
| *** adac has joined #openstack-ansible | 11:16 | |
| *** adac has quit IRC | 11:20 | |
| *** adac has joined #openstack-ansible | 11:20 | |
| *** adac has quit IRC | 11:21 | |
| *** automagically has joined #openstack-ansible | 11:54 | |
| *** automagically has quit IRC | 11:59 | |
| *** admin0 has quit IRC | 12:06 | |
| *** cvstealt1 has joined #openstack-ansible | 12:12 | |
| openstackgerrit | Merged openstack/openstack-ansible: Allow sourcing apt_key from ansible host or URL https://review.openstack.org/268676 | 12:15 |
| *** admin0 has joined #openstack-ansible | 12:20 | |
| *** retreved_ has quit IRC | 12:38 | |
| *** retreved has joined #openstack-ansible | 12:38 | |
| *** phiche has joined #openstack-ansible | 12:52 | |
| *** markvoelker has joined #openstack-ansible | 12:57 | |
| *** markvoelker has quit IRC | 13:01 | |
| *** baker has joined #openstack-ansible | 13:03 | |
| *** keedya has joined #openstack-ansible | 13:12 | |
| *** woodard has joined #openstack-ansible | 13:14 | |
| *** woodard has quit IRC | 13:19 | |
| *** markvoelker has joined #openstack-ansible | 13:24 | |
| *** electrofelix has joined #openstack-ansible | 13:25 | |
| *** baker has quit IRC | 13:29 | |
| *** baker has joined #openstack-ansible | 13:30 | |
| *** baker has quit IRC | 13:34 | |
| *** automagically has joined #openstack-ansible | 13:47 | |
| *** keedya has quit IRC | 13:49 | |
| *** admin0 has quit IRC | 13:50 | |
| *** admin0 has joined #openstack-ansible | 13:52 | |
| *** javeriak has quit IRC | 13:57 | |
| *** automagically has quit IRC | 13:57 | |
| *** automagically has joined #openstack-ansible | 13:58 | |
| *** admin0 has quit IRC | 13:58 | |
| *** Bofu2U has quit IRC | 13:58 | |
| *** woodard has joined #openstack-ansible | 14:01 | |
| *** woodard has quit IRC | 14:01 | |
| *** automagically has joined #openstack-ansible | 14:01 | |
| *** woodard has joined #openstack-ansible | 14:01 | |
| *** shausy has quit IRC | 14:01 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-rsyslog_client: Added conditional templating https://review.openstack.org/273292 | 14:09 |
| *** sdake has joined #openstack-ansible | 14:11 | |
| *** galstrom_zzz is now known as galstrom | 14:21 | |
| *** keedya has joined #openstack-ansible | 14:22 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Update ceph repo https://review.openstack.org/274081 | 14:22 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Update ceph repo https://review.openstack.org/274082 | 14:22 |
| *** phiche has quit IRC | 14:26 | |
| *** Bjoern has joined #openstack-ansible | 14:33 | |
| *** Bjoern is now known as Bjoern_zZzZzZzZ | 14:33 | |
| *** admin0 has joined #openstack-ansible | 14:40 | |
| *** Bjoern_zZzZzZzZ is now known as Bjoern | 14:44 | |
| *** admin0 has quit IRC | 14:44 | |
| *** admin0 has joined #openstack-ansible | 14:45 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:05 | |
| *** spotz_zzz is now known as spotz | 15:14 | |
| *** Mudpuppy has joined #openstack-ansible | 15:15 | |
| *** admin0 has quit IRC | 15:17 | |
| *** galstrom is now known as galstrom_zzz | 15:18 | |
| *** Mudpuppy has quit IRC | 15:19 | |
| *** Mudpuppy has joined #openstack-ansible | 15:19 | |
| *** sdake has quit IRC | 15:23 | |
| *** galstrom_zzz is now known as galstrom | 15:29 | |
| *** baker has joined #openstack-ansible | 15:37 | |
| Nepoc | Good morning, back with a new problem. It seems the controller_neutron_agents_container is not creating a bridge to connect the router external interface out to the physical external network. I manually created a bridge and added the tap interface from the router to eth1 in the container which connects out to my external network and everything is working. Any suggestions where I should be looking as to why this might be happening? | 15:39 |
| *** admin0 has joined #openstack-ansible | 15:39 | |
| palendae | Morning | 15:40 |
| kysse | eveninks | 15:40 |
| *** admin0 has quit IRC | 15:41 | |
| bgmccollum | odyssey4me im thinking the fact caching is more trouble that its worth...have seen lots of issues on re-runs | 15:42 |
| mattt | bgmccollum: i got burned on that one too | 15:42 |
| mattt | bgmccollum: banged my head trying to figure something out for ages, then realised we fact cache for an eternity | 15:43 |
| bgmccollum | yeah...ive seen it crop up many times too | 15:43 |
| spotz | morning | 15:43 |
| *** admin0 has joined #openstack-ansible | 15:43 | |
| bgmccollum | i dont think the teardown scripts clean the cache, or they didn't the last time i checked... | 15:44 |
| *** gparaskevas has quit IRC | 15:44 | |
| palendae | bgmccollum: Yeah, teardown.sh often misses things | 15:44 |
| palendae | It's not updated in tandem with additions | 15:44 |
| cloudnull | morning | 15:45 |
| automagically | cloudnull: Thanks for the recheck on https://review.openstack.org/#/c/273231 last night | 15:47 |
| cloudnull | thanks for putting through the fix :) | 15:47 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Provide logrotate config for rsyncd on Swift storage hosts https://review.openstack.org/274121 | 15:47 |
| *** admin0 has quit IRC | 15:48 | |
| cloudnull | Nepoc: is it a bridge which neutron should control ? | 15:48 |
| cloudnull | or is it an integration bridge ? | 15:48 |
| openstackgerrit | Byron McCollum proposed openstack/openstack-ansible: Remove cached facts in teardown.sh https://review.openstack.org/274124 | 15:49 |
| Nepoc | As far I know it should be something neutron should control. | 15:50 |
| Nepoc | On new openstack setup, creating the first router from an internal network out to the br-mgmt connected network. | 15:51 |
| *** allan_h has joined #openstack-ansible | 15:51 | |
| cloudnull | is there anything in the logs which looks odd ? | 15:51 |
| cloudnull | i've not see that issue in particular | 15:51 |
| Nepoc | unfortunately no | 15:51 |
| *** MCoLo has quit IRC | 15:51 | |
| Nepoc | The result is the interface is labelled as down to the external network | 15:51 |
| *** MCoLo has joined #openstack-ansible | 15:52 | |
| *** MCoLo has quit IRC | 15:52 | |
| *** MCoLo has joined #openstack-ansible | 15:52 | |
| Nepoc | If I add in the bridge and attach the tap and eth1 interfaces in the agent container ... everything works... though it is still labelled as down | 15:52 |
| *** alhu_ has joined #openstack-ansible | 15:55 | |
| *** allan_h has quit IRC | 15:57 | |
| *** phiche has joined #openstack-ansible | 15:58 | |
| *** targon has left #openstack-ansible | 16:00 | |
| mattt | bgmccollum: hey, i know you've committed to ceph-ansible, any thoughts on this change? https://github.com/mattt416/ceph-ansible/commit/0dab4084467a0194abaa2cd20496962c00f0d49c | 16:02 |
| *** alextricity_r has joined #openstack-ansible | 16:02 | |
| *** alhu_ has quit IRC | 16:02 | |
| bgmccollum | mattt looking | 16:02 |
| cloudnull | Apsu: might you have any thoughts on the issue Nepoc is describing ? | 16:03 |
| *** raddaoui has joined #openstack-ansible | 16:03 | |
| Apsu | cloudnull: I'm off today, but lemme read real quick | 16:03 |
| cloudnull | my bad | 16:04 |
| Apsu | No worries, didn't announce it. hehe | 16:04 |
| *** alextricity_r has quit IRC | 16:04 | |
| Apsu | Nepoc: You said you're creating a router from an internal Neutron network to a br-mgmt connected network | 16:05 |
| Apsu | But br-mgmt shouldn't participate in Neutron networks. It's the container control plane | 16:05 |
| Apsu | Sounds like a good way to have a bad time. | 16:05 |
| Nepoc | then there is my issue... apparently I like having a bad time | 16:05 |
| cloudnull | ok that makes sense. | 16:06 |
| Apsu | But that aside... which interfaces in which containers/namespaces are showing as down? And what do you mean "if I add in the bridge"? You're creating your own bridge? | 16:06 |
| *** phiche has quit IRC | 16:07 | |
| Apsu | The way this is supposed to work: inside the agents container, in its root namespace, Neutron should create a bridge and veth pair for a Neutron network | 16:07 |
| cloudnull | Nepoc: are you doing the deployment on a single interface for management and your data plain ? If so can you do vlan tagged interfaces ? | 16:07 |
| Nepoc | So in horizon the router I created which I didn't realize I couldn't piggyback accross the br-mgmt connection was showing down for that interface on the router | 16:07 |
| Apsu | One end of the veth pair stays in the root namespace and is put into the bridge; it names these tap$UUID | 16:08 |
| *** Mudpuppy_ has joined #openstack-ansible | 16:08 | |
| Apsu | Nepoc: Ahh. So you're looking at Neutron's status for the ports/networks/routers. | 16:08 |
| Apsu | Not actual interfaces on the box. | 16:08 |
| Nepoc | correct | 16:08 |
| bgmccollum | mattt: the only potential issue i see is the change of default behavior to not upgrade...it *should not* upgrade by default, but your change goes against their established default behavior. | 16:09 |
| Nepoc | We had a non ansible setup that was working with a single external network/ management combined | 16:09 |
| Apsu | Neutron's status for a network thingy isn't directly equal to an interface status. I believe for things like routers and networks it may show down when you don't have any ports using it yet, for instance. | 16:09 |
| Apsu | Like when you haven't created a VM yet. | 16:09 |
| *** alextricity_r has joined #openstack-ansible | 16:09 | |
| mattt | bgmccollum: but that behaviour was only introduced to allow rolling_upgrades.yml to work :( | 16:09 |
| bgmccollum | mattt true...if we can get a better default through...then thats fantastic! | 16:10 |
| bgmccollum | let it ride ;) | 16:10 |
| Nepoc | Apsu: thank you for shedding some light on this for me :) now to fix my network | 16:10 |
| mattt | bgmccollum: cool! maybe i'll add some |defaults() to be sure | 16:10 |
| *** Mudpuppy has quit IRC | 16:11 | |
| mattt | bgmccollum: not sure if that pattern is the greatest, but i'll submit it and see what they say | 16:11 |
| *** galstrom is now known as galstrom_zzz | 16:11 | |
| Apsu | Nepoc: Sure. Hope it helps. Basically... don't use br-mgmt in openstack, and don't create your own bridges or manually touch tap* interfaces. Or any of the inside container interfaces. | 16:11 |
| Apsu | Nepoc: You shouldn't have to ever. | 16:11 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible-specs: IRR - Create the os_crud_ops role https://review.openstack.org/272764 | 16:12 |
| Nepoc | Yeah I didn't think I would have to... just was trying to debug my issue | 16:12 |
| Apsu | Understood. | 16:12 |
| Apsu | On an interface level, just look at "ip link" to see if it's truly DOWN or UP | 16:12 |
| Apsu | Well... DOWN or $other. There's a few !DOWN statuses | 16:12 |
| Nepoc | Okay sounds good | 16:13 |
| Apsu | If it's !DOWN, then you can debug elsewhere :) | 16:13 |
| Nepoc | Thank you kindly | 16:13 |
| Apsu | Np | 16:13 |
| bgmccollum | mattt: don't forget to update `group_vars/all.sample` as well | 16:13 |
| cloudnull | Nepoc: idk if this helps at all, but Bofu2MBP put together this interface to run everything on a single bond (or single interface) using bridges with vlan tagged interfaces https://gist.github.com/cloudnull/e81115119dddee5e2a06 | 16:14 |
| mattt | bgmccollum: ah yeah, will do | 16:14 |
| openstackgerrit | Hugh Saunders proposed openstack/openstack-ansible: Update Defcore test list function https://review.openstack.org/270138 | 16:15 |
| Nepoc | cloudnull: nice, that may actually come in handy | 16:15 |
| cloudnull | our normal config is vlan tagged interfaces in bridges but pulling the interfaces in bridges and then creating the vlan tagged devices from the bridge worked quite well and lessened the need for more physical devices. | 16:17 |
| cloudnull | i tested that in a lab and it worked really well. | 16:17 |
| *** woodard has quit IRC | 16:17 | |
| Apsu | cloudnull: That's how we used to configure RPC in the v4.x days | 16:17 |
| *** woodard has joined #openstack-ansible | 16:17 | |
| cloudnull | idk right :) | 16:17 |
| cloudnull | whats old is new again | 16:18 |
| Apsu | It's full circle time again | 16:18 |
| Apsu | brb, driving to my local Amazon Bookstore | 16:18 |
| cloudnull | now we just have to start using chef agai n | 16:18 |
| Nepoc | hehe | 16:18 |
| * cloudnull not funny i know | 16:18 | |
| Nepoc | no... don't do that | 16:18 |
| mattt | someone say chef? | 16:18 |
| * mattt gets his bundler ready | 16:18 | |
| Apsu | Careful, you'll get mattt excited | 16:18 |
| Nepoc | I had enough of that setting up my last company in rackspace | 16:19 |
| Apsu | cloudnull: I bet my embedded Chef Server VM from v1.x still works. | 16:19 |
| cloudnull | test kitchen for life ! | 16:19 |
| Apsu | I'll build a new Alamo ISO right meow | 16:19 |
| cloudnull | I know it does in at least one env :| | 16:19 |
| Apsu | haha | 16:19 |
| *** alejandrito has joined #openstack-ansible | 16:19 | |
| *** galstrom_zzz is now known as galstrom | 16:24 | |
| cloudnull | any chance i can get some core reviews on https://review.openstack.org/#/c/272689/ | 16:25 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: [WIP] Converted all OS_* roles to use os_crud_ops https://review.openstack.org/272837 | 16:28 |
| mattt | cloudnull: in that review, does that mean you'd have git cruft in /etc/ansible ? | 16:31 |
| palendae | mattt: If it's an ansible role requirement but defined with git, no | 16:32 |
| mattt | ok great | 16:32 |
| cloudnull | no. it creates a directory in /etc/ansible called plugins | 16:32 |
| palendae | Loading the review now to see if that's the actual thing | 16:32 |
| palendae | It's kind of like pip installing from a version control source | 16:32 |
| *** daneyon has joined #openstack-ansible | 16:42 | |
| *** daneyon_ has joined #openstack-ansible | 16:43 | |
| *** mgoddard_ has joined #openstack-ansible | 16:45 | |
| *** daneyon has quit IRC | 16:47 | |
| *** mgoddard has quit IRC | 16:49 | |
| *** weezS has joined #openstack-ansible | 16:49 | |
| *** skamithi has joined #openstack-ansible | 16:50 | |
| *** raddaoui has quit IRC | 16:50 | |
| *** ntpttr has joined #openstack-ansible | 16:51 | |
| *** mikelk has quit IRC | 16:52 | |
| ntpttr | Hi #openstack-ansible, I'm trying to run 'openstack-ansible setup-hosts.yml' to start my openstack deployment on 10 nodes, and it's able to connect to all of the nodes just fine but on the step where it waits for SSH to be available in the containers it's timing out | 16:52 |
| ntpttr | Would this have to do with the way the bridges are set up? | 16:53 |
| *** izaakk has joined #openstack-ansible | 16:57 | |
| ntpttr | It looks like I'm having the same problem as seen here - http://www.gossamer-threads.com/lists/openstack/operators/51317 I can also ping the containers just fine | 17:01 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: [WIP] Add support for upgrading from Kilo to Liberty https://review.openstack.org/272182 | 17:02 |
| *** mgoddard_ has quit IRC | 17:02 | |
| Bjoern | ntpttr : can you even reach the containers if you manually ssh into them ? your bridge just need to use the same networks as specified as container_network inside the openstack_user_config | 17:02 |
| *** mgoddard has joined #openstack-ansible | 17:02 | |
| Bjoern | Also I do always a ip neigh flush dev <br-mgmt if the bridge is named that way> | 17:03 |
| *** skamithi has quit IRC | 17:03 | |
| ntpttr | Bjoern: If I ssh into one of the hosts and then try and ssh to a container from there, it does connect and then says it couldn't authenticate due to the public key | 17:04 |
| ntpttr | The bridge is set up that way, so I don't think it's that | 17:04 |
| ntpttr | I just tried adding an ssh delay to see if that works | 17:05 |
| Bjoern | can you do a ansible <container name> -mshell -a'uptime' even ? | 17:07 |
| ntpttr | from the deployment host? | 17:07 |
| Bjoern | yeah from the playbook directory | 17:08 |
| ntpttr | looks like that command might be hanging | 17:09 |
| ntpttr | but all the containers are pingable | 17:10 |
| Bjoern | don't think timeout will fix anything it's not that many nodes. add a -vvvv and see what ssh command ansible is doing | 17:10 |
| ntpttr | <172.29.238.77> ESTABLISH CONNECTION FOR USER: root | 17:11 |
| ntpttr | <172.29.238.77> REMOTE_MODULE command uptime #USE_SHELL | 17:11 |
| ntpttr | <172.29.238.77> EXEC ssh -C -v -o ControlMaster=auto -o ControlPersist=60s -o ControlPath="/home/vagrant/.ansible/cp/ansible-ssh-%h-%p-%r" -o StrictHostKeyChecking=no -o KbdInteractiveAuthentication=no -o PreferredAuthentications=gssapi-with-mic,gssapi-keyex,hostbased,publickey -o PasswordAuthentication=no -o ConnectTimeout=120 172.29.238.77 /bin/sh -c 'LANG=en_US.UTF-8 LC_CTYPE=en_US.UTF-8 /usr/bin/python' | 17:11 |
| ntpttr | that looks like it's as far as it gets | 17:11 |
| ntpttr | or is it just stopping there because I put ssh-delay 100 in user variables :-X | 17:12 |
| Bjoern | yeah usually you get the same issues when you connect manually via SSH to this IP | 17:13 |
| ntpttr | lab3_heat_apis_container-87f7d7af | FAILED => SSH Error: data could not be sent to the remote host. Make sure this host can be reached over ssh | 17:14 |
| ntpttr | just got that output | 17:14 |
| ntpttr | can't ping the IP it's showing either, the 172.29.238.77 | 17:15 |
| ntpttr | from the deployment host anyways, I can from the host the container is on | 17:16 |
| Bjoern | network mask is correct on your management bridge (perhaps br-mgmt) ? | 17:16 |
| ntpttr | do you need to set up the bridges and everything on the deployment host as well? I haven't done that | 17:16 |
| Bjoern | depends how you route internally | 17:16 |
| ntpttr | there's a management network that it connects to all of the target hosts to with | 17:17 |
| Bjoern | we usually do not to have any routing component in between the deployment hosts and the containers | 17:17 |
| ntpttr | and I can ssh to them all from there | 17:17 |
| Bjoern | if the deployment host is not in the same networks, then there's a routing issue between the deployment host and the management network (container_network) | 17:18 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Fix failing tests in defcore test set https://review.openstack.org/274171 | 17:19 |
| Bjoern | that's why I usually install from the first infra nodes and use that as the deployment host | 17:20 |
| ntpttr | fair enough, maybe I'll try just copying the config files over there and doing it that way | 17:20 |
| ntpttr | so deploying from one of the target hosts doesn't mess anything up? | 17:21 |
| Bjoern | no, most stuff happens inside containers | 17:21 |
| *** weezS has quit IRC | 17:26 | |
| Nepoc | This is new... I just did a teardown and lost all the vhost_*.ko modules. | 17:29 |
| *** coolg has joined #openstack-ansible | 17:34 | |
| *** galstrom is now known as galstrom_zzz | 17:40 | |
| *** coolg has quit IRC | 17:46 | |
| bgmccollum | anyone seeing this in the gate? -- MissingSchema: Invalid URL 'mirror.IAD.rax.openstack.org/pypi/simple/jinja2/': No schema supplied. Perhaps you meant http://mirror.IAD.rax.openstack.org/pypi/simple/jinja2/? | 17:46 |
| *** coolg has joined #openstack-ansible | 17:46 | |
| *** skamithi has joined #openstack-ansible | 17:47 | |
| Nepoc | Disregard my comment... I magically had the wrong kernel installed | 17:49 |
| *** skamithi has quit IRC | 17:49 | |
| Nepoc | or someone is playing tricks on me | 17:49 |
| *** javeriak has joined #openstack-ansible | 17:53 | |
| *** javeriak has quit IRC | 17:59 | |
| *** javeriak has joined #openstack-ansible | 18:00 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 18:02 | |
| *** javeriak_ has joined #openstack-ansible | 18:04 | |
| *** javeriak has quit IRC | 18:05 | |
| *** daneyon_ has quit IRC | 18:07 | |
| *** galstrom_zzz is now known as galstrom | 18:34 | |
| *** woodard has quit IRC | 18:40 | |
| *** Guest75 has joined #openstack-ansible | 18:40 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 18:41 | |
| *** bumba_clyde has joined #openstack-ansible | 18:41 | |
| *** woodard has joined #openstack-ansible | 18:46 | |
| *** woodard has quit IRC | 18:48 | |
| *** woodard has joined #openstack-ansible | 18:49 | |
| *** sdake has joined #openstack-ansible | 18:51 | |
| *** weezS has joined #openstack-ansible | 18:54 | |
| *** Guest75 has quit IRC | 18:54 | |
| *** pcaruana has joined #openstack-ansible | 18:57 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: Cleanup syntax issues for Ansible 2 compatibility https://review.openstack.org/267842 | 19:02 |
| Nepoc | When adding a static_route to the openstack_user_config how can I push that change out to all containers? | 19:15 |
| Nepoc | I guess with the assumption that you are unable to ssh into the containers. | 19:16 |
| bgmccollum | Nepoc well, if you can't SSH to the containers, then Ansible can't do any configuration. | 19:21 |
| *** pcaruana has quit IRC | 19:24 | |
| coolg | Hi All | 19:31 |
| coolg | I am trying to install openstack liberty using openstack-ansible and followed the installation guide. http://docs.openstack.org/developer/openstack-ansible/install-guide after i ran openstack-ansible setup-hosts.yml , target containers are created. I am not able to do ssh to the container. It is getting failed in this task failed: [target_keystone_container-5032a112 -> 127.0.0.1] => {"elapsed": 301, "failed": true} msg: T | 19:31 |
| *** raddaoui has joined #openstack-ansible | 19:31 | |
| coolg | It is failing in this task TASK: [Wait for ssh to be available] ****************************************** | 19:32 |
| coolg | Can anybody help me on this | 19:33 |
| Nepoc | bgmccollum: true but you have access to the root filessytem on the host they are on. | 19:33 |
| *** sdake has quit IRC | 19:33 | |
| Nepoc | coolg: you probably need to do what I just did, create a static route in the containers | 19:34 |
| *** pcaruana has joined #openstack-ansible | 19:37 | |
| coolg | can you pls elaborate Nepoc. I am trying to installation on the target node from the deployment node and I followed the installation guide | 19:37 |
| stevelle | Nepoc: I wonder if coolg has something else going on ... target_keystone_container-5032a112 -> 127.0.0.1 | 19:38 |
| cloudnull | coolg: checkout http://docs.openstack.org/developer/openstack-ansible/install-guide/configure-networking.html?highlight=static_routes | 19:38 |
| cloudnull | on creating static routes | 19:38 |
| cloudnull | at the bottom | 19:38 |
| Nepoc | I have the exact same error | 19:38 |
| Nepoc | or had | 19:39 |
| stevelle | Nepoc: ah, I might have missed the loopback when I scrolled up | 19:39 |
| *** electrofelix has quit IRC | 19:40 | |
| *** eil397 has joined #openstack-ansible | 19:41 | |
| *** admin0 has joined #openstack-ansible | 19:43 | |
| coolg | m checking the link you have provided for static_routes | 19:43 |
| *** sdake has joined #openstack-ansible | 19:44 | |
| *** hybridpollo has joined #openstack-ansible | 20:01 | |
| *** daneyon has joined #openstack-ansible | 20:02 | |
| *** retreved has quit IRC | 20:03 | |
| coolg | I feel i have configured something wrong in opestack_user_config.yml | 20:03 |
| coolg | I will paste my openstack_uses_config.yml..can anybody help me to figure out what i have wrongly | 20:04 |
| coolg | --- cidr_networks: container: 10.10.1.0/22 tunnel: 10.10.2.0/22 storage: 10.10.3.0/22 global_overrides: internal_lb_vip_address: 192.100.100.225 external_lb_vip_address: x.x.x.x(target node ip) tunnel_bridge: "br-vxlan" management_bridge: "br-mgmt" provider_networks: - network: container_bridge: "br-mgmt" container_type: "veth" container_interface: "eth1" ip_from_q: "conta | 20:04 |
| mgariepy | coolg, pastebin | 20:05 |
| *** pcaruana has quit IRC | 20:06 | |
| coolg | --- cidr_networks: container: 10.10.1.0/22 tunnel: 10.10.2.0/22 storage: 10.10.3.0/22 global_overrides: internal_lb_vip_address: 192.100.100.225 external_lb_vip_address: x.x.x.x(target node ip) tunnel_bridge: "br-vxlan" management_bridge: "br-mgmt" provider_networks: - network: container_bridge: "br-mgmt" container_type: "veth" container_interface: "eth1" ip_from_q: "conta | 20:07 |
| mgariepy | coolg, http://pastebin.com/ | 20:08 |
| coolg | is it possible to attach filehere | 20:08 |
| coolg | ? | 20:08 |
| mgariepy | you can paste yaml there and just post the link in the channel. | 20:08 |
| palendae | coolg: A pastebin service like paste.openstack.org lets you put arbitrary text in and provide a link | 20:09 |
| coolg | http://pastebin.com/Hwstd67N | 20:09 |
| *** pcaruana has joined #openstack-ansible | 20:09 | |
| coolg | thank you.. pls check whether you are able to access | 20:09 |
| cloudnull | coolg: not that its related to the connectivity issues you're seeing but do you need a flat network ? | 20:11 |
| cloudnull | and do you have eth12 on your physical hosts? | 20:11 |
| coolg | M trying to install in a vm | 20:11 |
| cloudnull | If not you can remove http://cdn.pasteraw.com/g4jrufjdsg4219oxa3hf64abw5mylzg | 20:11 |
| cloudnull | coolg: so its an AIO ? | 20:12 |
| coolg | no | 20:12 |
| cloudnull | multi-node using VMs ? | 20:12 |
| coolg | I am trying to install on target node from deployment node | 20:12 |
| coolg | yeah | 20:13 |
| coolg | multinode using vms | 20:13 |
| cloudnull | does your VM do mac filtering or have ACLs in place to prevent you from spoofing an address? | 20:13 |
| cloudnull | we do testing like that using pulic cloud providers and have to create vxlan tunnels between nodes to enable networking because of filtering | 20:14 |
| coolg | we may need that while going for neutron networking.. now all m trying is just executed openstack-ansible setup-hosts.yml | 20:15 |
| *** retreved has joined #openstack-ansible | 20:15 | |
| coolg | where it throws the ssh error | 20:15 |
| cloudnull | ssh to the container right? | 20:15 |
| coolg | yeah | 20:15 |
| coolg | if i want to do AIO from deployment node to target node | 20:16 |
| coolg | this will throw error | 20:16 |
| cloudnull | and both nodes are VMs ? | 20:16 |
| coolg | Mainly I am concentrating on installation on target from deployment | 20:16 |
| coolg | ya | 20:17 |
| coolg | both are VM | 20:17 |
| cloudnull | it sounds like the filtering issues | 20:17 |
| cloudnull | https://github.com/rcbops/rpc-heat/blob/master/openstack_multi_node.yml#L259-L272 | 20:17 |
| cloudnull | we have a heat template to deploy openstack on openstack | 20:17 |
| cloudnull | which has those same symptoms | 20:17 |
| *** pcaruana has quit IRC | 20:19 | |
| coolg | i think its not available for liberty | 20:19 |
| palendae | Probably not | 20:19 |
| palendae | I don't think anyone's updated it yet | 20:19 |
| coolg | I want to do liberty installation using openstack-ansible | 20:19 |
| palendae | Though I do think you can specify the branch you want to run | 20:19 |
| cloudnull | also you can recreate the setup by simply creating a tunnel between the nodes | 20:20 |
| coolg | is it possible to use openstack-ansible project itslef | 20:20 |
| *** daneyon_ has joined #openstack-ansible | 20:21 | |
| *** skamithi has joined #openstack-ansible | 20:21 | |
| cloudnull | to create the tunnels? | 20:23 |
| *** daneyon has quit IRC | 20:24 | |
| cloudnull | the tunnel interface between VMs can be created like so | 20:25 |
| cloudnull | https://github.com/rcbops/rpc-heat/blob/master/build/controller_all.sh#L7-L11 | 20:25 |
| *** Nepoc has quit IRC | 20:25 | |
| *** Nepoc has joined #openstack-ansible | 20:26 | |
| coolg | not for tunnels.. i wanted to use this http://docs.openstack.org/developer/openstack-ansible/install-guide/install-foundation.html | 20:26 |
| coolg | since i have already started and I am in the middle | 20:26 |
| coolg | any clue why I am not able to take ssh? | 20:26 |
| *** jthorne has joined #openstack-ansible | 20:29 | |
| cloudnull | my only suspicion is because your VM network provider is filtering the traffic | 20:29 |
| cloudnull | which is why I'm recommending you try and create a vxlan tunnel between the instances for the management traffic. | 20:30 |
| cloudnull | the only time i've seen the issue your describing is when running a multi-node deployment on VMs (this is assuming your network is properly setup) | 20:31 |
| palendae | openstack-ansible needs SSH working between the hosts you're going to set up before it can do any of the install | 20:31 |
| *** skamithi has left #openstack-ansible | 20:32 | |
| coolg | i am able to do ssh.. containers have been created in the target host | 20:32 |
| *** javeriak_ has quit IRC | 20:32 | |
| coolg | to install the openstack service inside container | 20:32 |
| coolg | m not ablet to do ssh | 20:33 |
| cloudnull | but you cant ssh to the container from deployment host . | 20:33 |
| cloudnull | can you ping it | 20:33 |
| cloudnull | and do you see the traffic on both sides ? | 20:33 |
| coolg | after adding static route my vm gone down | 20:34 |
| coolg | i remember i was able to do the ping | 20:34 |
| coolg | but not able to do ssh | 20:34 |
| coolg | in target node i check with sudo lxc-ls | 20:35 |
| coolg | which listed the containers | 20:35 |
| coolg | from my understanding if we configure the /etc/network/interfaces in the target node correctly and opestack_user_config.yml in /etc/openstack_deploy | 20:37 |
| coolg | instllation should go smoothly | 20:37 |
| coolg | anybody have correct working configuration files for ref | 20:38 |
| coolg | ? | 20:38 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Horizon venv if it mismatches the repo https://review.openstack.org/273802 | 20:38 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Glance venv if it mismatches the repo https://review.openstack.org/273800 | 20:39 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Neutron venv if it mismatches the repo https://review.openstack.org/273803 | 20:39 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Aodh venv if it mismatches the repo https://review.openstack.org/273804 | 20:39 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Nova venv if it mismatches the repo https://review.openstack.org/273805 | 20:39 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Cinder venv if it mismatches the repo https://review.openstack.org/273807 | 20:39 |
| coolg | ?? | 20:42 |
| *** coolg has quit IRC | 20:45 | |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Heat venv if it mismatches the repo https://review.openstack.org/273808 | 20:46 |
| openstackgerrit | Merged openstack/openstack-ansible: Re-deploy the Keystone venv if it mismatches the repo https://review.openstack.org/273809 | 20:46 |
| *** skamithi has joined #openstack-ansible | 20:47 | |
| cloudnull | sorry coolg was afk | 20:49 |
| cloudnull | yes if you have your network configured correctly the installation goes smooth | 20:49 |
| cloudnull | that said VM multi node installs can be quirky | 20:49 |
| cloudnull | im my experience | 20:50 |
| *** andyhky has joined #openstack-ansible | 20:59 | |
| *** galstrom is now known as galstrom_zzz | 21:00 | |
| admin0 | how to handle multi-avaibality zone in the playbooks ? | 21:10 |
| admin0 | recommended way | 21:10 |
| *** skamithi has quit IRC | 21:16 | |
| *** skamithi has joined #openstack-ansible | 21:18 | |
| cloudnull | admin0: idk that there is a recommended way . i typically do the full deployment then carve up the compute nodes into different AZs post deployment. | 21:19 |
| cloudnull | would be nice to define that process for depoyers that know they'll be doing multi-az upfront however i dont think there is one currently. | 21:20 |
| admin0 | manual overrides for every node is a pain | 21:22 |
| admin0 | maybe where i can make a official feature request ? | 21:22 |
| cloudnull | ++ agreed | 21:22 |
| cloudnull | that | 21:22 |
| cloudnull | *that'd be a good one | 21:23 |
| *** weezS has quit IRC | 21:24 | |
| *** skamithi has quit IRC | 21:27 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Fixed discoverable-apis in tempest for object https://review.openstack.org/274248 | 21:28 |
| *** sdake has quit IRC | 21:29 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: DEFCORE: Fixed discoverable-apis in tempest object https://review.openstack.org/274248 | 21:29 |
| admin0 | cloudnull: do you know where to make a feature request ? | 21:30 |
| cloudnull | I'd recommend creating a bug in LP https://bugs.launchpad.net/openstack-ansible marking it a wishlist item . I'd then add it to the https://wiki.openstack.org/wiki/Meetings/openstack-ansible#Agenda_for_next_meeting to bring up and have it targetted for the next milestone . | 21:32 |
| *** admin0 has quit IRC | 21:33 | |
| *** sdake has joined #openstack-ansible | 21:43 | |
| *** alejandrito has quit IRC | 21:54 | |
| logan- | the AZ chop up could be done with group_vars I guess. id be curious in some docs or example configs for multi-region too. | 21:58 |
| jmccrory | there's currently a task that does AZ assignment? | 22:01 |
| logan- | with compute I usually do it via host aggregates, but with cinder you could use group_vars to carve out AZs | 22:02 |
| jmccrory | creating the compute host aggregates post deployment though, right? not through some playbook? | 22:04 |
| logan- | right | 22:04 |
| logan- | yea nothing in OSA for that yet afai | 22:04 |
| jmccrory | ok, thought i was missing something already included with overrides | 22:05 |
| *** vCloudernBeer has joined #openstack-ansible | 22:05 | |
| vCloudernBeer | ping llu | 22:05 |
| logan- | gotcha :) it would be handy to have some post-install plays for sure, host aggs, upload some images, etc | 22:06 |
| logan- | probably something like that out there already that you could run against the utility containers | 22:06 |
| *** hybridpollo has quit IRC | 22:12 | |
| *** vCloudernBeer has quit IRC | 22:12 | |
| *** raddaoui has quit IRC | 22:15 | |
| *** bumba_clyde has quit IRC | 22:19 | |
| *** dmsimard has quit IRC | 22:19 | |
| *** sdake has quit IRC | 22:20 | |
| *** Mudpuppy_ has quit IRC | 22:21 | |
| *** dmsimard has joined #openstack-ansible | 22:22 | |
| *** admin0 has joined #openstack-ansible | 22:22 | |
| *** Mudpuppy has joined #openstack-ansible | 22:22 | |
| *** retreved_ has joined #openstack-ansible | 22:24 | |
| admin0 | task: lxc_hosts create apt repos in the cached container .. failing .. it means the haproxy address is bad or not working ? | 22:24 |
| admin0 | unable to reach rpc-repo url | 22:24 |
| *** retreved has quit IRC | 22:27 | |
| admin0 | figured it out :) was using a wrong ip | 22:28 |
| admin0 | hmm.. no .. | 22:32 |
| *** automagically has quit IRC | 22:40 | |
| *** spotz is now known as spotz_zzz | 22:41 | |
| *** alextricity_r has quit IRC | 22:42 | |
| admin0 | anyone can guide me on why the step “ lxc_hosts create apt repos in the cached container” fails ? | 22:43 |
| *** skamithi has joined #openstack-ansible | 22:53 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:00 | |
| *** alejandrito has joined #openstack-ansible | 23:01 | |
| admin0 | cloudnull: does this look OK ? https://bugs.launchpad.net/openstack-ansible/+bug/1539803 | 23:02 |
| openstack | Launchpad bug 1539803 in openstack-ansible "enable support for availability zones for compute and services" [Undecided,New] | 23:02 |
| *** admin0 has quit IRC | 23:11 | |
| *** admin0 has joined #openstack-ansible | 23:11 | |
| *** bryan_att has joined #openstack-ansible | 23:12 | |
| admin0 | hello | 23:12 |
| cloudnull | admin0: was afk in meetings. | 23:15 |
| cloudnull | looking at it now | 23:15 |
| admin0 | ok | 23:15 |
| cloudnull | looks sensible . | 23:16 |
| cloudnull | I'll have a gander over the weekend to see how this might be implemented. | 23:16 |
| admin0 | ok | 23:16 |
| cloudnull | thanks for putting that in . sorry its not there already | 23:17 |
| admin0 | hopefully it will be implemented some day | 23:18 |
| admin0 | suddenly ansible fails at create apt repos in the cached container .. and says /var/cache/lxc/trusty/rootfs/etc/apt does not exist | 23:20 |
| admin0 | is it just me or should i file it as a bug as well | 23:20 |
| admin0 | the folder does not exist and had to do -vvvv to get the exact reason | 23:21 |
| admin0 | cloudnull: i will try to put that in the meeting points | 23:22 |
| admin0 | the bug | 23:22 |
| cloudnull | tyvm | 23:22 |
| cloudnull | as for the bug , you can remove the cached image are rerun the play which will create new cache | 23:23 |
| admin0 | task: lxc-hosts : update container resolvers . fails @ dest dir /var/cache/lxc/trusty/rootfs-amd64/run/resolvconf does not exist — huh ? shouln’t something create this file first from within ansible ? | 23:23 |
| cloudnull | no this is from the cached image . idk how you dont have it? | 23:24 |
| cloudnull | the file is assumed to have already been there. | 23:24 |
| admin0 | in which step is this created ? | 23:24 |
| openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-rsyslog_server: [WIP] Multi-distro support for rsyslog-server role https://review.openstack.org/274290 | 23:24 |
| cloudnull | lxc-hosts.yml | 23:24 |
| admin0 | now resolv.conf.d does not exist :( | 23:26 |
| admin0 | let me try to run just lxc-hosts.yml playbook | 23:26 |
| cloudnull | you can remove /var/cache/lxc/trusty and /var/cache/lxc_trusty.tgz and run the lxc-host play which will repopulate and update all cache | 23:27 |
| admin0 | removed from all .. running setup-hosts again with -vvv | 23:29 |
| admin0 | doing this in a fresh new ubuntu 14.04 .. | 23:29 |
| admin0 | not aio but with multi nodes | 23:29 |
| admin0 | that worked i think .. removing the cached files | 23:31 |
| admin0 | now its in downloading lxc_cache(s) step | 23:31 |
| cloudnull | nice! | 23:31 |
| *** baker has quit IRC | 23:31 | |
| *** baker has joined #openstack-ansible | 23:32 | |
| admin0 | are there plans to move to ansible 2 ? | 23:33 |
| admin0 | looks like it might work this time :) thanks cloudnull | 23:34 |
| admin0 | unless stuck again | 23:34 |
| *** sshen has quit IRC | 23:34 | |
| cloudnull | we've been prep'ing / talking about ansible 2 however idk if well make the move until after the mitaka release. | 23:35 |
| admin0 | its moving forward . thanks cloudnull | 23:35 |
| admin0 | finally creating the containers | 23:36 |
| cloudnull | sweet! | 23:36 |
| admin0 | not sure what happened/went wrong this time | 23:36 |
| admin0 | that it was stuck there | 23:36 |
| admin0 | but i know what to do now :D | 23:36 |
| *** baker has quit IRC | 23:36 | |
| *** baker has joined #openstack-ansible | 23:37 | |
| *** sshen has joined #openstack-ansible | 23:37 | |
| cloudnull | yea. if the cache is bad or if a task was interupted while building the cache it can cause a very bad time. | 23:38 |
| admin0 | oh | 23:39 |
| *** skamithi has quit IRC | 23:39 | |
| *** baker has quit IRC | 23:41 | |
| *** alejandrito has quit IRC | 23:41 | |
| admin0 | horizon is someting that has public access, and lxc might not provide 100% security to not allow break-out to host if its compromized for any reasons .. is it a good idea to be able to define horizon_hosts as well ? | 23:41 |
| admin0 | or an idea for a feature request ? | 23:42 |
| cloudnull | you can modify the environment.yml and env.d/*yml files to move things around. | 23:43 |
| *** sshen has quit IRC | 23:43 | |
| cloudnull | you can redefine the groups in https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/env.d/horizon.yml | 23:44 |
| cloudnull | which is primarily https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/env.d/horizon.yml#L24-L26 | 23:44 |
| admin0 | \o/ :) | 23:44 |
| *** sshen has joined #openstack-ansible | 23:45 | |
| cloudnull | http://cdn.pasteraw.com/hc6hcr1zibv4fr2fhwapeiwtyd5npki <- that should be all you need to redefine the hosts that horizon is on | 23:47 |
| cloudnull | in openstack_user_config.yml you would set horizon-infra_hosts: and the containers would be deployed at that host | 23:47 |
| admin0 | got it :) | 23:48 |
| cloudnull | while the groups we create are perfectly functional, they can be taylored to fit just about any scenario | 23:52 |
| admin0 | yeah .. just need to figure out where to edit and how to structure stuff | 23:53 |
| *** hughsaunders has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!