| *** markvoelker has quit IRC | 00:01 | |
| *** CheKoLyN has joined #openstack-ansible | 00:14 | |
| *** Mudpuppy has joined #openstack-ansible | 00:14 | |
| *** shoutm has quit IRC | 00:17 | |
| *** shoutm has joined #openstack-ansible | 00:20 | |
| *** Mudpuppy has quit IRC | 00:25 | |
| *** CheKoLyN has quit IRC | 00:29 | |
| *** Mudpuppy has joined #openstack-ansible | 00:32 | |
| *** darrenc_afk is now known as darrenc | 00:32 | |
| *** CheKoLyN has joined #openstack-ansible | 00:42 | |
| *** Mudpuppy has quit IRC | 00:42 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Set the Ubuntu mirror used based on the environment https://review.openstack.org/221666 | 00:43 |
|---|---|---|
| *** Mudpuppy has joined #openstack-ansible | 00:49 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Updated the configuration to current dynamic inventory standards https://review.openstack.org/221854 | 00:52 |
| *** markvoelker has joined #openstack-ansible | 00:58 | |
| *** alop has quit IRC | 01:00 | |
| *** Mudpuppy has quit IRC | 01:01 | |
| *** markvoelker has quit IRC | 01:02 | |
| *** abitha has quit IRC | 01:04 | |
| *** tlian2 has joined #openstack-ansible | 01:09 | |
| *** Mudpuppy has joined #openstack-ansible | 01:12 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Set affinity to 1 if running on hpcloud https://review.openstack.org/221957 | 01:12 |
| *** tlian has quit IRC | 01:13 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Set affinity to 1 if running on hpcloud https://review.openstack.org/221957 | 01:15 |
| *** mrstanwell has quit IRC | 01:16 | |
| *** Mudpuppy has quit IRC | 01:28 | |
| *** Mudpuppy has joined #openstack-ansible | 01:31 | |
| *** Mudpuppy has quit IRC | 01:49 | |
| *** Mudpuppy has joined #openstack-ansible | 01:56 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix keystone shibboleth template to detect ssl usage correctly https://review.openstack.org/221347 | 02:04 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Be more clear about used_ips, mostly in the example file https://review.openstack.org/221713 | 02:11 |
| *** CheKoLyN has quit IRC | 02:24 | |
| *** Mudpuppy has quit IRC | 02:25 | |
| *** tlian2 has quit IRC | 02:26 | |
| *** alejandrito has joined #openstack-ansible | 02:28 | |
| *** Mudpuppy has joined #openstack-ansible | 02:37 | |
| *** arbrandes has quit IRC | 02:43 | |
| *** Mudpuppy has quit IRC | 02:47 | |
| *** arbrandes has joined #openstack-ansible | 02:52 | |
| *** markvoelker has joined #openstack-ansible | 02:55 | |
| *** sdake_ has joined #openstack-ansible | 02:56 | |
| *** Mudpuppy has joined #openstack-ansible | 02:57 | |
| *** sdake__ has joined #openstack-ansible | 02:57 | |
| *** sdake has quit IRC | 02:58 | |
| *** sdake_ has quit IRC | 03:01 | |
| *** Mudpuppy has quit IRC | 03:07 | |
| *** alejandrito has quit IRC | 03:10 | |
| *** alextricity_h has quit IRC | 03:14 | |
| *** dolphm has quit IRC | 03:27 | |
| *** sigmavirus24 has quit IRC | 03:27 | |
| *** eglute has quit IRC | 03:27 | |
| *** d34dh0r53 has quit IRC | 03:27 | |
| *** palendae has quit IRC | 03:27 | |
| *** d34dh0r53 has joined #openstack-ansible | 03:28 | |
| *** Mudpuppy has joined #openstack-ansible | 03:28 | |
| *** eglute has joined #openstack-ansible | 03:28 | |
| *** dolphm has joined #openstack-ansible | 03:28 | |
| *** palendae has joined #openstack-ansible | 03:29 | |
| *** sdake__ is now known as sdake | 03:30 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Adds the ability to provide user certificates to HAProxy https://review.openstack.org/221667 | 03:30 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fixes jinja typo in keystone-httpd.conf.j2 https://review.openstack.org/221977 | 03:30 |
| *** sigmavirus24 has joined #openstack-ansible | 03:32 | |
| *** palendae has quit IRC | 03:36 | |
| *** palendae has joined #openstack-ansible | 03:37 | |
| *** Mudpuppy has quit IRC | 03:38 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 03:39 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Allow testing of cinder multi backends w/ tempest https://review.openstack.org/221851 | 03:51 |
| *** markvoelker has quit IRC | 04:55 | |
| *** sdake_ has joined #openstack-ansible | 05:12 | |
| *** sdake has quit IRC | 05:15 | |
| *** sdake has joined #openstack-ansible | 05:15 | |
| *** sdake_ has quit IRC | 05:16 | |
| *** sdake__ has joined #openstack-ansible | 05:16 | |
| *** sdake__ has quit IRC | 05:17 | |
| *** sdake_ has joined #openstack-ansible | 05:18 | |
| *** sdake has quit IRC | 05:20 | |
| *** sdake_ is now known as sdake | 05:21 | |
| *** shausy has joined #openstack-ansible | 05:28 | |
| *** sdake_ has joined #openstack-ansible | 05:29 | |
| *** sdake has quit IRC | 05:30 | |
| *** shausy has quit IRC | 05:32 | |
| *** markvoelker has joined #openstack-ansible | 05:56 | |
| *** markvoelker has quit IRC | 06:01 | |
| *** sdake has joined #openstack-ansible | 06:08 | |
| *** sdake_ has quit IRC | 06:11 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add galera_address to all playbooks installing galera_client https://review.openstack.org/221301 | 06:27 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Set client versions deployed to use global requirements https://review.openstack.org/220050 | 06:27 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enabling log compression by default https://review.openstack.org/221924 | 06:27 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Install nfs-common with nova-compute https://review.openstack.org/221844 | 06:28 |
| *** sdake has quit IRC | 06:28 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Updated MariaDB to the new release version (10.0) https://review.openstack.org/178259 | 06:28 |
| *** sdake has joined #openstack-ansible | 06:28 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Changed certificate order for pem generation with CA files https://review.openstack.org/221682 | 06:31 |
| *** mcarden has quit IRC | 06:40 | |
| *** javeriak has joined #openstack-ansible | 07:30 | |
| *** javeriak has quit IRC | 07:37 | |
| *** javeriak has joined #openstack-ansible | 07:38 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Shift irqbalance package from lxc_hosts to openstack_hosts https://review.openstack.org/218354 | 07:41 |
| *** javeriak has quit IRC | 07:53 | |
| *** markvoelker has joined #openstack-ansible | 07:57 | |
| *** bapalm has quit IRC | 07:58 | |
| *** jmccrory has quit IRC | 07:59 | |
| *** shoutm_ has joined #openstack-ansible | 08:00 | |
| *** bapalm has joined #openstack-ansible | 08:00 | |
| evrardjp | good morning everyone | 08:01 |
| *** javeriak has joined #openstack-ansible | 08:01 | |
| *** shoutm has quit IRC | 08:01 | |
| *** markvoelker has quit IRC | 08:02 | |
| *** jmccrory has joined #openstack-ansible | 08:03 | |
| *** javeriak has quit IRC | 08:09 | |
| *** javeriak has joined #openstack-ansible | 08:10 | |
| *** sdake has quit IRC | 08:17 | |
| *** javeriak_ has joined #openstack-ansible | 08:33 | |
| *** javeriak has quit IRC | 08:35 | |
| odyssey4me | o/ evrardjp | 08:35 |
| *** javeriak has joined #openstack-ansible | 08:38 | |
| *** javeriak_ has quit IRC | 08:39 | |
| evrardjp | quick question: I'd like to discuss about the inventory approach for keepalived component, what's the best way to do it? Have a point in open topics in the meeting today? | 08:39 |
| odyssey4me | evrardjp yes, that would be a good way to do it - perhaps add it to the reviews agenda item | 08:42 |
| *** javeriak has quit IRC | 08:42 | |
| odyssey4me | mattt hughsaunders git-harry there was quite a bit of discussion in-channel regarding https://review.openstack.org/202822 yesterday, but none of those views are recorded in the review - please can you review the patch and add your thoughts accordingly | 08:43 |
| *** javeriak_ has joined #openstack-ansible | 08:46 | |
| *** gparaskevas has joined #openstack-ansible | 08:48 | |
| evrardjp | odyssey4me: done | 08:52 |
| *** markvoelker has joined #openstack-ansible | 08:58 | |
| *** markvoelker has quit IRC | 09:02 | |
| *** jmccrory has quit IRC | 09:04 | |
| *** shoutm has joined #openstack-ansible | 09:05 | |
| *** jmccrory has joined #openstack-ansible | 09:06 | |
| *** javeriak has joined #openstack-ansible | 09:07 | |
| *** shoutm_ has quit IRC | 09:07 | |
| *** javeriak_ has quit IRC | 09:09 | |
| hughsaunders | git-harry: you should comment on https://review.openstack.org/202822 | 09:12 |
| *** jmccrory has quit IRC | 09:12 | |
| *** jmccrory has joined #openstack-ansible | 09:15 | |
| *** javeriak_ has joined #openstack-ansible | 09:24 | |
| *** javeriak has quit IRC | 09:24 | |
| *** javeriak_ has quit IRC | 09:32 | |
| odyssey4me | hughsaunders I agree with what you've said, but I think that perhaps a revisit of what's being done here should be a proposal for a new role in a subsequent patch/ | 09:36 |
| odyssey4me | actually, this is a perfect opportunity for a new blueprint/spec to create something like a rabbitmq-admin role which the service playbooks can consume | 09:37 |
| *** javeriak has joined #openstack-ansible | 09:39 | |
| hughsaunders | I'll see what cloudnull says first | 09:43 |
| *** javeriak has quit IRC | 09:49 | |
| *** javeriak has joined #openstack-ansible | 09:50 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Shift irqbalance package from lxc_hosts to openstack_hosts https://review.openstack.org/218354 | 09:52 |
| odyssey4me | evrardjp can you please backport https://review.openstack.org/221713 to kilo? | 10:01 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fixes jinja typo in keystone-httpd.conf.j2 https://review.openstack.org/221974 | 10:03 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fixes jinja typo in keystone-httpd.conf.j2 https://review.openstack.org/221974 | 10:04 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fixes jinja typo in keystone-httpd.conf.j2 https://review.openstack.org/221974 | 10:05 |
| odyssey4me | mattt can you backport https://review.openstack.org/221851 to kilo please | 10:06 |
| odyssey4me | unless that's not an intention for the patch? | 10:06 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Shift irqbalance package from lxc_hosts to openstack_hosts https://review.openstack.org/222104 | 10:07 |
| mattt | odyssey4me: i can, i wasn't sure if that was necessary tho since it's effectively introducing a new feature | 10:15 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Set affinity to 1 if running on hpcloud https://review.openstack.org/221957 | 10:17 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Set affinity to 1 if running on hpcloud https://review.openstack.org/221957 | 10:19 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Set affinity to 1 if running on hpcloud https://review.openstack.org/221957 | 10:19 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: More changes to neutron db migrations https://review.openstack.org/219759 | 10:20 |
| odyssey4me | mattt I think it should be backported - it's a new testing feature and ideally I'd like master and kilo to have the same testing capabilities | 10:22 |
| odyssey4me | it's not a customer facing feature | 10:22 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Allow testing of cinder multi backends w/ tempest https://review.openstack.org/222119 | 10:27 |
| openstackgerrit | Jean-Philippe Evrard proposed stackforge/os-ansible-deployment: Be more clear about used_ips, mostly in the example file https://review.openstack.org/222121 | 10:29 |
| evrardjp | what's the total scale of the solution? I have the feeling it's starting to be a pain to deploy when you start to have more than 4 controllers | 10:31 |
| evrardjp | sorry, not really intended to be here | 10:32 |
| *** javeriak has quit IRC | 10:43 | |
| odyssey4me | evrardjp so the primary design it was built around was with 3 controllers | 10:52 |
| odyssey4me | as I recall, Rabbit is the biggest bottleneck due to the way the clustering works and the way we're using it | 10:52 |
| odyssey4me | evrardjp https://review.openstack.org/202822 is set to improve that | 10:53 |
| odyssey4me | however, it'd be good to know what your experience is and how we could possibly improve things | 10:53 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: More changes to neutron db migrations https://review.openstack.org/219759 | 10:53 |
| *** markvoelker has joined #openstack-ansible | 10:59 | |
| *** markvoelker has quit IRC | 11:04 | |
| *** shoutm has quit IRC | 11:24 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Install logrotate with ryslog https://review.openstack.org/222143 | 11:25 |
| *** shoutm has joined #openstack-ansible | 11:26 | |
| evrardjp | odyssey4me: I agree, but if I want to go in production soon, I'd avoid messing up with these "not tested" ways ;) | 11:37 |
| evrardjp | I'll downscale then | 11:37 |
| evrardjp | I'll test the 4 controllers later | 11:39 |
| cloudnull | morning | 11:54 |
| *** markvoelker has joined #openstack-ansible | 11:55 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: More changes to neutron db migrations https://review.openstack.org/219759 | 12:00 |
| mgariepy | good morning everyone | 12:02 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: More changes to neutron db migrations https://review.openstack.org/219759 | 12:07 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: More changes to neutron db migrations https://review.openstack.org/219759 | 12:14 |
| mhayden | morning | 12:31 |
| mhayden | and good afternoon mattt and odyssey4me | 12:31 |
| mattt | mhayden: morning2u :) | 12:37 |
| *** woodard has joined #openstack-ansible | 12:48 | |
| *** woodard has quit IRC | 12:49 | |
| *** woodard has joined #openstack-ansible | 12:50 | |
| evrardjp | good morning cloudnull and mhayden | 12:50 |
| cloudnull | o/ evrardjp | 12:52 |
| cloudnull | odyssey4me: https://review.openstack.org/#/q/status:open+project:stackforge/os-ansible-deployment+branch:kilo,n,z looks like kilo is having a wheel build issue. im testing now will see whats what in a min | 12:53 |
| *** javeriak has joined #openstack-ansible | 12:57 | |
| cloudnull | git-harry hughsaunders odyssey4me - RE: https://review.openstack.org/#/c/202822/ - My stance remains unchanged. However, If the "community" wants to seperate these tasks into other roles, or just tasks within a given play, I'm fine with that. You guys let me know and I'll update the review accordingly. | 12:59 |
| hughsaunders | morning cloudnull | 13:01 |
| odyssey4me | o/ mgariepy cloudnull mhayden | 13:03 |
| odyssey4me | cloudnull yeah, I was about to figure out the wheel build issue, but happy that you're on it | 13:04 |
| hughsaunders | I think it would be useful to keep the ability to use an external rabbit. As you and git-harry have said, duplication is a wider issue and can't be addressed in a review discussion. | 13:04 |
| odyssey4me | hughsaunders cloudnull I'd like us to table a discussion at the summit around opportunities for managing technical debt which should include figuring out a 'standard' or 'policy' regarding where another role should or shouldn't be used | 13:05 |
| odyssey4me | personally I think the change is fine for now as it keeps with the current conventions, but I think we do need to discuss and agree on whether the existing conventions are good | 13:06 |
| mattt | agreed, this change is in line w/ how we're doing things now, but i do think it begs some discussion as to how we do this going forward | 13:09 |
| odyssey4me | cloudnull fyi https://review.openstack.org/221957 will not change the hpcloud-b4 build failures... those are not resource related, they appear to be networking related as the issue is that the apt get upgrade/installs can't successfully complete | 13:09 |
| cloudnull | odyssey4me: on https://review.openstack.org/221957 last night I was seeing lots of build failures in hpb.* so i made that, its not quite right but I wanted to see resource usage w/ affinity=1 vs not | 13:13 |
| odyssey4me | mattt hughsaunders I've added my feedback in the review and encourage you to do the same. | 13:13 |
| odyssey4me | cloudnull yeah - I adjusted it this morning, but figured that it might be a tester :) I've done this before as well, but didn't find it to make a significant difference at the time. | 13:14 |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Remove unused Juno-specific groups from inventory https://review.openstack.org/222186 | 13:15 |
| cloudnull | on https://review.openstack.org/#/c/202822 would it be better for this discussion to move the tasks into the play indstead of the role ? would that garner more agreement in implementation? | 13:16 |
| odyssey4me | cloudnull where I am seeing it make a difference is in the success of the tempest tests | 13:16 |
| odyssey4me | cloudnull I think that'd be a bad idea, personally | 13:16 |
| evrardjp | good morning mgariepy! I've changes for you, and I guess a possible discussion to bring room for improvements :) | 13:17 |
| odyssey4me | personally I think that there is room to have a role which handles the rabbitmq administration bits separately from the rabbitmq setup | 13:17 |
| odyssey4me | cloudnull ie the openstack service roles should be built with the assumption that mysql and rabbitmq databases/vhosts/users/etc are already setup when they run | 13:18 |
| odyssey4me | the playbooks, however, should orchestrate the creation of the vhosts, users, databases, etc through the use of a role that sets them up after setting up the database | 13:19 |
| odyssey4me | does that make sense? | 13:19 |
| *** pradk has joined #openstack-ansible | 13:19 | |
| evrardjp | ideally it's not even about that... it's about letting your customers to work the way they want... | 13:19 |
| odyssey4me | in other words, it's not a role dependancy - it's a playbook orchestration | 13:19 |
| evrardjp | because they should know better what they need | 13:19 |
| odyssey4me | evrardjp exactly | 13:19 |
| odyssey4me | this way we are more loosely coupled, and the playbooks do the coupling | 13:20 |
| mgariepy | evrardjp, I agree with not stopping haproxy | 13:20 |
| odyssey4me | this would also mean that the rabbitmq cluster could be externally setup and managed | 13:20 |
| odyssey4me | same with galera | 13:20 |
| evrardjp | mgariepy: ok. stopping backends or not? | 13:20 |
| odyssey4me | and I'm sure there are other opportunities for this | 13:20 |
| evrardjp | I don't think that would be smart | 13:20 |
| mgariepy | we just need to make sure it's running, if not mark keepalive as faulted. | 13:21 |
| evrardjp | it's kinda error prone, and easy to miss for the lambda administrator | 13:21 |
| evrardjp | mgariepy: I agree | 13:21 |
| evrardjp | I've done my change on the galaxy role | 13:21 |
| mgariepy | for the first iteration i guess kiss is better. | 13:21 |
| evrardjp | mgariepy: I agree | 13:21 |
| odyssey4me | cloudnull does that make sense? | 13:22 |
| odyssey4me | cloudnull note, though, that I don't think this method refactor needs to be done in that patch - it's probably better suited to a follow on patch | 13:22 |
| mgariepy | would be nice to have email sent when state change also haha | 13:23 |
| evrardjp | that's something you could do | 13:23 |
| evrardjp | without even changing the notify script | 13:23 |
| mgariepy | yeah it' | 13:24 |
| mgariepy | s keepalived config. | 13:24 |
| evrardjp | oh btw, I removed the logging part | 13:24 |
| evrardjp | for the next iteration | 13:24 |
| evrardjp | because most of the important is already logged in syslog | 13:24 |
| evrardjp | most of the important events* | 13:25 |
| * cloudnull was breaking the fast | 13:26 | |
| cloudnull | odyssey4me: makes sense. just wnated to give everyone a chance to weigh in with the hope that more folks would also comment on the review . | 13:27 |
| cloudnull | also on https://review.openstack.org/#/c/221957/ resource reduction seems to make a difference - HPB2. SUCCESS in 1h 30m 07s which is just under the build timeout wire :) | 13:28 |
| cloudnull | but I do agree with mattt kinda in that we may not catch some things especially in cases where there are code paths that execute on N+1 | 13:29 |
| *** tlian has joined #openstack-ansible | 13:32 | |
| *** vdo has joined #openstack-ansible | 13:35 | |
| mattt | cloudnull: i'm +2, except had a question before I actually +2 :) | 13:38 |
| *** shoutm has quit IRC | 13:55 | |
| *** Mudpuppy has joined #openstack-ansible | 14:01 | |
| openstackgerrit | Christopher H. Laco proposed stackforge/os-ansible-deployment: Enable log_slave_updates https://review.openstack.org/222198 | 14:02 |
| *** cloudtrainme has joined #openstack-ansible | 14:03 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:03 | |
| *** KLevenstein has joined #openstack-ansible | 14:04 | |
| *** shoutm has joined #openstack-ansible | 14:07 | |
| evrardjp | hughsaunders: I tried to answer your questions on https://review.openstack.org/#/c/214089/ | 14:08 |
| evrardjp | if it's still not clear, I'll rephrase | 14:08 |
| hughsaunders | evrardjp: thanks, will have a look | 14:08 |
| evrardjp | I'm not sure if I'm clear ;) | 14:08 |
| evrardjp | understandable* | 14:09 |
| *** javeriak has quit IRC | 14:10 | |
| *** spotz_zzz is now known as spotz | 14:12 | |
| evrardjp | cloudnull: if you don't mind I added a point to the agenda for one review, but it's more a discussion about already discussed topic, so I guess you can push me later if required... this item shouldn't last more than 5 minutes | 14:12 |
| cloudnull | i dont mind at all. thanks evrardjp :) | 14:13 |
| *** javeriak has joined #openstack-ansible | 14:17 | |
| javeriak | hello folks, need a little help trying to deploy from master, so I can't figure out where the repo-hosts are set up. The configure-hostlist.rst doc refers to a '/etc/openstack_deploy/user_group_vars.yml' which is actually 'playbooks/inventory/group_vars/hosts.yml', am I missing something here? | 14:23 |
| javeriak | and the 'openstack_repo_url' parameter in that file points to my deploy node, which isn't hosting anything, which brings me back to my original question, where is the repo hosting installation done in the plays? | 14:25 |
| *** Mudpuppy has quit IRC | 14:25 | |
| mattt | javeriak: repo-install.yml gets called in setup-infrastructure.yml | 14:27 |
| *** Mudpuppy has joined #openstack-ansible | 14:29 | |
| *** agireud has joined #openstack-ansible | 14:29 | |
| *** Mudpuppy has quit IRC | 14:30 | |
| mattt | javeriak: does that answer your question or did it side-step entirely? :) | 14:30 |
| *** Mudpuppy has joined #openstack-ansible | 14:30 | |
| javeriak | mattt yes and no :). problem is the pip_install task fails during setup-host.yml because the repo its looking for doesn't exist, | 14:31 |
| palendae | javeriak: Ah, yeah...I don't know that the repo servers are set up by then. | 14:33 |
| evrardjp | they aren't | 14:33 |
| evrardjp | oh wait | 14:33 |
| palendae | The openstack_repo_url option can be left out if you specify the repo servers | 14:34 |
| *** alextricity has joined #openstack-ansible | 14:34 | |
| javeriak | Will it work if I point it to "https://rpc-repo.rackspace.com/" ? | 14:34 |
| evrardjp | the repo doesn't exist or some packages are not yet up to date? | 14:34 |
| javeriak | evrardjp, it doesnt exist at all, i mean nothings being hosted | 14:35 |
| palendae | if you specify repo-infra_hosts, the repos will be made, populated from rpc-repe.rackspace.com | 14:35 |
| palendae | I don't remember exactly, but I think it needs to be the os-releases diir | 14:36 |
| palendae | https://rpc-repo.rackspace.com/os-releases/ | 14:36 |
| evrardjp | javeriak: I'm not sure why you need that at that point | 14:36 |
| javeriak | palendae, yes but if that happens in the setup-infrastructure.yml, how are the previous installations supposed to work since nothing is up yet | 14:36 |
| evrardjp | during setup hosts you setup lxc IIRC | 14:36 |
| evrardjp | and other stuff on your hosts | 14:36 |
| evrardjp | yeah, I'm not sure you need that... what's the part of the play that fails? | 14:37 |
| palendae | javeriak: That's a great question...I'm not sure I have an answer. Maybe cloudnull does, but I think you need a repo server already up if you're specifying openstack_repo_url | 14:37 |
| palendae | If you want repo servers to be built, define the repo-infra_hosts in user_config | 14:38 |
| evrardjp | I'm not sure, but I don't think it should read it during the setup-hosts | 14:38 |
| palendae | And leave openstack_repo_url undefined | 14:38 |
| evrardjp | I'll check | 14:38 |
| javeriak | evrardjp, it tries to install pip during setup hosts, this one fails [pip_install | Install PIP (fall back mode)] | 14:38 |
| mattt | javeriak: now that you mention it, i thought i did see a failure when i ran a deploy last, but it didn't fail the run | 14:39 |
| javeriak | matt: it wont fail if it finds pip already there from a previous install, but as it happens one of my machines doesnt have it, so it tries to contact the repo | 14:40 |
| mattt | javeriak: ah i see | 14:40 |
| palendae | javeriak: Are you trying to avoid building local repos? | 14:40 |
| evrardjp | lxc_hosts dependencies pip install >< | 14:40 |
| javeriak | palendae, nope I defined the repo-infra_hosts, same as my infra_hosts | 14:41 |
| palendae | Ok | 14:41 |
| cloudnull | palendae: whats the issue ? | 14:41 |
| palendae | Make sure you also undefine openstack_repo_url | 14:41 |
| javeriak | was i supposed to remove "openstack_repo_url" too? | 14:41 |
| cloudnull | o/ javeriak :) | 14:41 |
| palendae | javeriak: Yeah, that takes precedence | 14:41 |
| * cloudnull reading | 14:41 | |
| *** shoutm has quit IRC | 14:42 | |
| cloudnull | pip should attempt to install using the repo infra for a locked down version of pip however if its not there the options are removed and the pip install simply installs the upstream version. | 14:44 |
| cloudnull | which on a greenfield deployment should be ok as what is used from the upstream install is also the thing used within the initial repo build process. | 14:45 |
| palendae | cloudnull: But if you define openstack_repo_url to a place that doesn't exist or doesn't contain the repo dir structure, it'll fail, no? | 14:45 |
| evrardjp | I think it should work fine | 14:45 |
| cloudnull | it should be fine | 14:46 |
| cloudnull | that variable is used in the following two places | 14:46 |
| cloudnull | http://cdn.pasteraw.com/hekgwt1uge4csanckyrktgnzx38vnkf | 14:46 |
| javeriak | cloudnull palendae, yes its trying to install through fall back mode, no options, but the error I get is still: RetryError: HTTPConnectionPool(host='10.22.37.149', port=8181): Max retries exceeded with url: /os-releases/11.2.2/ (Caused by ResponseError('too many 503 error responses',)) | 14:47 |
| javeriak | nothings being hosted on 10.22.37.149 (my deploy) | 14:47 |
| javeriak | its picking that up from pip.conf I believe | 14:48 |
| cloudnull | do your hosts have the file ~/.pip/pip.conf that is locking pip down to that adress. | 14:48 |
| palendae | Ah, your pip.conf probably didn't get updated after removing the openstack_repo_url var | 14:49 |
| cloudnull | so maybe the Install PIP (fall back mode) task needs to run with the --isolated flag to ignore pip.conf ? | 14:50 |
| javeriak | cloudnull yep its there, i guessing it doesnt get overwritten on a new install, I see an entry to https://mirror.rackspace.com/rackspaceprivatecloud/python_packages/10.1.11/ from my prev install | 14:50 |
| evrardjp | question, not related at all... Your rackspace maas is updated for kilo? Is it upstreamed somewhere? | 14:50 |
| cloudnull | javeriak: previous meaning Juno ? | 14:50 |
| javeriak | cloudnull, yes juno | 14:50 |
| cloudnull | evrardjp: https://github.com/rcbops/rpc-openstack | 14:51 |
| javeriak | 10.1.11 to be precise | 14:51 |
| evrardjp | thanks | 14:51 |
| cloudnull | the plugins are here: https://github.com/rcbops/rpc-openstack/tree/master/maas | 14:51 |
| cloudnull | maas roles are here https://github.com/rcbops/rpc-openstack/tree/master/rpcd/playbooks/roles/rpc_maas | 14:51 |
| cloudnull | javeriak: this is an upgrade issue sadly. | 14:52 |
| cloudnull | we have an upgrade script that does the translation from juno (rax technical debt) to kilo (community osad) | 14:52 |
| cloudnull | https://github.com/stackforge/os-ansible-deployment/blob/master/scripts/run-upgrade.sh#L600-L604 | 14:52 |
| cloudnull | thats the line where thats resolved. | 14:52 |
| cloudnull | structurally there are quite a few changes between juno and kilo (master as well) and so we've taken on the job of translating all of those changes for upgrade. | 14:54 |
| javeriak | cloudnull, ah okay, yes so it works by specifying --isolated or taking out the local repo entry in pip.conf, so we do need to un-specify the openstack_repo_url while specifying repo-hosts ? shouldnt that be put in the doc somewhere? | 14:54 |
| cloudnull | you shouldnt need to set openstack_repo_url | 14:55 |
| cloudnull | i think it may be a good idea to add the --isolated flag to the fall back mode for pip install. | 14:55 |
| javeriak | also this: https://github.com/stackforge/os-ansible-deployment/blob/master/doc/source/install-guide/configure-hostlist.rst needs to be updated with the correct path for this variable | 14:55 |
| cloudnull | that too | 14:56 |
| javeriak | no the variable is already there by default | 14:56 |
| javeriak | would you like me to take care of it? I can push a patch | 14:56 |
| cloudnull | thatd be awesome ! | 14:56 |
| cloudnull | if you have the cycles to do it , I would appreciate it :) | 14:56 |
| javeriak | sure, we should comment out that variable by default then as well, or ask the deployer to do it explicitly in the example user config | 14:58 |
| palendae | Is it not commented? | 14:58 |
| javeriak | nope: https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/inventory/group_vars/hosts.yml | 14:59 |
| palendae | I guess stepping even further back | 15:00 |
| palendae | What exactly are you trying to do? Upgrade Juno to Kilo? | 15:00 |
| javeriak | palendae, not exactly, but I didnt remove all of the juno config before starting, only the previous containers and some main directories | 15:04 |
| cloudnull | so its a rekick but there was an existing deployment | 15:08 |
| cloudnull | which was torn down ? | 15:08 |
| cloudnull | i assume ? | 15:09 |
| javeriak | yes | 15:10 |
| *** jwagner_away is now known as jwagner | 15:10 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment-specs: Move specs to appropriate cycle folders https://review.openstack.org/220929 | 15:10 |
| javeriak | so I should've run the upgrade script then I think? What exactly is the use case for it, if you have an up and running juno setup, and you want to move to kilo? You run the script and then run all new playbooks | 15:11 |
| cloudnull | the upgrade script will take you from juno to kilo wholesale | 15:24 |
| cloudnull | which also does the clean up | 15:24 |
| cloudnull | but in your case it seems that you were doing a greenfield deployment on hosts that were used in the past. | 15:25 |
| cloudnull | which just needed a bit of clean up before running the new deployment. | 15:25 |
| javeriak | okay cool, I'll just push a fix for the doc paths, and comment out the openstack_repo_url in another, | 15:31 |
| javeriak | should we add the isolated flag to the fall back pip install as well? | 15:31 |
| *** jwagner is now known as jwagner_away | 15:33 | |
| cloudnull | i think that might be best. | 15:33 |
| odyssey4me | javeriak you may ba better off archiving your old config and starting fresh | 15:33 |
| odyssey4me | several of your old configs will still work, but you'll need to consider whether each of the items you had should be there now, and whether they have been renamed or not | 15:34 |
| odyssey4me | the upgrade script took care of all that for you, but you blew your environment away :p | 15:34 |
| palendae | Hm | 15:36 |
| palendae | Actually, I don't think openstack_repo_url should be commented out by default | 15:36 |
| palendae | I'm probably misremembering a bunch of this | 15:36 |
| javeriak | odyssey4me: true :), I should probably clean it up properly. I don't require any of the old configs, I'm interested in a fresh install anyway. Shouldnt have taken the quicker way | 15:36 |
| palendae | But now that I see it's the internal LB by default, I'm pretty sure that's necessary | 15:36 |
| palendae | javeriak: Yeah, I thought you were doing a 100% clean install | 15:37 |
| palendae | I think I was wrong; if you completely remove openstack_repo_url, nothing will be able to find the repo servers you define | 15:38 |
| *** alop has joined #openstack-ansible | 15:38 | |
| javeriak | palendae, yes but why is it the internal LB, where are we setting that up as a repo? I havent looked through the infrastructure setup plays yet, but I assume they overwrite this to the ones you define as repo hosts | 15:38 |
| palendae | So should probably clarify that pointing to the internal LB VIP is needed for the repo-shared_infra hosts | 15:38 |
| javeriak | oh okay I see, it points to the LB which redirects it to the actual repo hosts.. | 15:39 |
| palendae | Right | 15:39 |
| palendae | We don't overwrite the variable | 15:39 |
| palendae | The repo hosts themselves ignore tha var | 15:39 |
| palendae | And pull from upstream | 15:40 |
| palendae | Everything else points to the LB | 15:40 |
| javeriak | its not really being used anywhere except for the pip-lock down file | 15:40 |
| palendae | Right, which basically locks down everything to the internal repo | 15:40 |
| javeriak | okay I'll leave the variable to what it is, just add --isolated to the pip install so that initial installs dont fail | 15:41 |
| palendae | I've not seen this behavior in pure greenfield deploys | 15:41 |
| *** mrstanwell has joined #openstack-ansible | 15:41 | |
| javeriak | well it failing makes sense, the repo isnt there yet, so it cant work. But dunno why it hasnt been seen before. Let me check what happens on a new install and let you guys know if anything happens | 15:43 |
| *** arbrandes has quit IRC | 15:44 | |
| cloudnull | thanks javeriak | 15:46 |
| odyssey4me | javeriak if you're doing a fresh install on hosts that have been used for previous builds then you'll need to do some clean-up | 15:49 |
| odyssey4me | rm -rf the /root/.pip directory, for one | 15:49 |
| odyssey4me | I'm not sure if the teardown script might be useful for you? | 15:50 |
| javeriak | odyssey4me, i had a local cleanup playbook thats probably missing some items, now im just going to do a fresh ubuntu | 15:51 |
| javeriak | yes i used the teardown script as source :) | 15:51 |
| odyssey4me | javeriak it takes longer, but is probably the best action to save some of your hair :p | 15:51 |
| palendae | The teardown script has missed stuff too | 15:52 |
| javeriak | yes hair is important :p, i need to go configure my cluster with maas now first | 15:52 |
| palendae | Things get added to the project, but not to that | 15:52 |
| *** gparaskevas has quit IRC | 15:53 | |
| *** jwagner_away is now known as jwagner | 15:53 | |
| cloudnull | meeting in 7 min in #openstack-meeting-4 -- cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden | 15:54 |
| *** vdo has quit IRC | 16:04 | |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment: Fix doc paths for openstack_repo_url https://review.openstack.org/222262 | 16:07 |
| *** Bjoern_ has joined #openstack-ansible | 16:11 | |
| *** jwagner is now known as jwagner_away | 16:11 | |
| *** Bjoern_ is now known as BjoernT | 16:11 | |
| *** jwagner_away is now known as jwagner | 16:15 | |
| *** tlian has quit IRC | 16:19 | |
| *** tlian has joined #openstack-ansible | 16:26 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix of haproxy ssl misconfiguration with nova_console https://review.openstack.org/221853 | 16:29 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Update Ansible to v1.9.3-1 https://review.openstack.org/221899 | 16:30 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Create config_template plugin https://review.openstack.org/220212 | 16:36 |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment: Add isolated flag to pip fall back installation https://review.openstack.org/222285 | 16:38 |
| prometheanfire | mhayden: what return do you get if that log is overloaded? | 16:41 |
| mhayden | LXC says everything's cool | 16:42 |
| mhayden | no errors, even with debug | 16:42 |
| mhayden | in its mind, it told the kernel to do something and it hopes the kernel actually does it | 16:42 |
| prometheanfire | I mean from the kernel | 16:42 |
| mhayden | but when it floods the netlink message queue with deletions, bad things happen | 16:42 |
| mhayden | nothing from the kernel | 16:42 |
| prometheanfire | huh, sounds like a kernel bug | 16:42 |
| prometheanfire | kernel should say that the queue is full | 16:43 |
| prometheanfire | so anything using it can try again | 16:43 |
| mhayden | prometheanfire: i poked rlove and got https://twitter.com/rlove/status/641286548286115840 | 16:46 |
| mhayden | haven't gone about working with that yet | 16:46 |
| Apsu | prometheanfire: Netlink is not a reliable transport with retransmissions and such like TCP | 16:48 |
| prometheanfire | mhayden: I'm saying that we should get a return code if overloaded, we probably shouldn't care about the size | 16:48 |
| Apsu | Netlink is more like UDP, and the kernel doesn't tell you jack over the netlink socket. | 16:48 |
| mhayden | the size was mainly a curiosity ;) | 16:48 |
| prometheanfire | Apsu: ah, better option? | 16:48 |
| Apsu | You are responsible for making sure what you asked happened. | 16:48 |
| mhayden | fedora kernel devs said "don't use netlink, it's old" | 16:49 |
| prometheanfire | don't blame them | 16:49 |
| mhayden | but that sounds like a big fat patch for lxc :P | 16:49 |
| palendae | prometheanfire: gus (I think) mentioned you basically have to keep pinging with back offs to communicate over netlink | 16:49 |
| mhayden | i'll toss what i found into a big fat LXC issue later today on the githubs | 16:49 |
| palendae | mhayden: LXc needs some big fat patches from what I hear | 16:49 |
| palendae | Some just being style consistency | 16:49 |
| mhayden | yeah | 16:50 |
| mhayden | as soon as i mention lxc, the kernel devs seem to hide under their desks | 16:50 |
| mhayden | then again, i do the same | 16:50 |
| palendae | lol | 16:52 |
| palendae | What about when you mention docker? | 16:52 |
| prometheanfire | docker? | 16:52 |
| prometheanfire | dockerdocker? | 16:52 |
| mhayden | dockerxc? | 16:55 |
| *** jmckind has joined #openstack-ansible | 16:56 | |
| alextricity | Hey. I'm probably late to the party, but does anyone know what's going on with the InsecurePlatform warning while running the repo plays? | 16:57 |
| evrardjp | curl https? | 17:01 |
| prometheanfire | paste? | 17:01 |
| evrardjp | I guess an update from curl that now mentions certificate warnings | 17:01 |
| odyssey4me | in case y'all are wondering why the kilo gate is broken, you can thank a library: https://bugs.launchpad.net/openstack-gate/+bug/1494347 | 17:01 |
| openstack | Launchpad bug 1494347 in OpenStack-Gate "PyECLib 1.0.9 released on 9/10 blowing up grenade jobs due to "src/c/pyeclib_c/pyeclib_c.c:32:25: fatal error: erasurecode.h: No such file or directory"" [Undecided,New] | 17:01 |
| prometheanfire | lol, it's because they aren't doing it right | 17:02 |
| evrardjp | :) | 17:02 |
| evrardjp | I'm off for today | 17:02 |
| alextricity | odyssey4me: thanks | 17:03 |
| prometheanfire | odyssey4me: https://gitweb.gentoo.org/repo/gentoo.git/tree/dev-python/PyECLib/files/1.0.7-erasurecode_locations.patch | 17:03 |
| odyssey4me | we'll have to bump the sha for juno / kilo to accommodate the global-requirements updates to work around the issue while upstream sorts their stuff out | 17:04 |
| odyssey4me | lol prometheanfire is that patch sent upstream? | 17:04 |
| odyssey4me | upstream bug: https://bitbucket.org/kmgreen2/pyeclib/issues/76/pyeclib-109-fails-to-install-if-not | 17:05 |
| prometheanfire | I could | 17:06 |
| prometheanfire | suppose I should too | 17:06 |
| cloudnull | o/ evrardjp have a good one. | 17:06 |
| prometheanfire | but I'm busy buying a domain | 17:06 |
| odyssey4me | prometheanfire there's mention of what you've done in the log | 17:07 |
| prometheanfire | neat | 17:07 |
| odyssey4me | not what you've done, but the paths you've changed are mentioned | 17:07 |
| prometheanfire | ya, this might be a packaging problem on their end | 17:09 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Replaced the copy_update module https://review.openstack.org/216790 | 17:14 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Allow testing of cinder multi backends w/ tempest https://review.openstack.org/222119 | 17:15 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fixes jinja typo in keystone-httpd.conf.j2 https://review.openstack.org/221974 | 17:17 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Used named veth pairs that match container https://review.openstack.org/222295 | 17:23 |
| cloudnull | odyssey4me: hughsaunders: other: on https://review.openstack.org/#/c/217030/ < do we want to have the config template be applied in one patch or many patches to the various openstack services ? | 17:27 |
| odyssey4me | cloudnull I would suggest one patch per role | 17:27 |
| odyssey4me | it makes it quicker and simpler to review | 17:27 |
| cloudnull | it also makes it a backport mess but i agree its simpler to review . | 17:28 |
| odyssey4me | a backport mess? once the first patch is in - all the others dep only on the module being there, surely? | 17:29 |
| cloudnull | yes its just lots of patches is all . | 17:29 |
| cloudnull | but if thats better for folks im game . | 17:29 |
| odyssey4me | yeah, it's a bit of a pain - but big patches are scary :p | 17:30 |
| * cloudnull shakes :) | 17:30 | |
| * cloudnull lunching afk a bit | 17:31 | |
| *** jwagner is now known as jwagner_away | 17:55 | |
| *** phschwartz has joined #openstack-ansible | 17:59 | |
| *** arbrandes has joined #openstack-ansible | 18:10 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Set client versions deployed to use global requirements https://review.openstack.org/220050 | 18:11 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add galera_address to all playbooks installing galera_client https://review.openstack.org/221301 | 18:11 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Changed certificate order for pem generation with CA files https://review.openstack.org/221682 | 18:12 |
| *** arbrandes has quit IRC | 18:25 | |
| *** ChanServ sets mode: +o palendae | 18:29 | |
| cloudnull | look at that, three merges in a row :) | 18:29 |
| palendae | Topic: Launchpad: https://launchpad.net/openstack-ansible Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Repo rename from stackforge/os-ansible-deployment to openstack/openstack-ansible happens Sept 11 2015 23:00 to 23:30 | 18:30 |
| *** palendae changes topic to "Topic: Launchpad: https://launchpad.net/openstack-ansible Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Repo rename from stackforge/os-ansible-deployment to openstack/openstack-ansible happens Sept 11 2015 23:00 to 23:30" | 18:30 | |
| svg | There goes 'osad'. What's the new short name for the project? | 18:30 |
| palendae | OA? | 18:31 |
| svg | OSA? | 18:31 |
| cloudnull | OSAD 4 life! | 18:31 |
| mrstanwell | ohappy? | 18:31 |
| *** palendae changes topic to "Topic: Launchpad: https://launchpad.net/openstack-ansible Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Repo rename from stackforge/os-ansible-deployment to openstack/openstack-ansible happens Sept 11 2015 23:00 to 23:30. See https://review.openstack.org/#/c/200730/" | 18:31 | |
| mrstanwell | (because you can actually install ostack with it...) | 18:31 |
| *** ChanServ sets mode: +o palendae | 18:32 | |
| palendae | Un-op! | 18:32 |
| *** ChanServ sets mode: -o palendae | 18:33 | |
| cloudnull | mrstanwell: related https://twitter.com/cloudnull/status/642043518978912256 | 18:34 |
| mrstanwell | cloudnull: Excellent! | 18:35 |
| *** jwagner_away is now known as jwagner | 18:36 | |
| *** arbrandes has joined #openstack-ansible | 18:42 | |
| *** phalmos has joined #openstack-ansible | 18:52 | |
| *** cloudtrainme has quit IRC | 19:14 | |
| *** cloudtrainme has joined #openstack-ansible | 19:25 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add profiling for Ansible tasks https://review.openstack.org/216849 | 19:34 |
| *** javeriak has quit IRC | 19:37 | |
| *** javeriak has joined #openstack-ansible | 19:48 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment-specs: Add spec for Independent Role Repositories https://review.openstack.org/213779 | 19:53 |
| odyssey4me | mhayden can you backport https://review.openstack.org/216849 please ? | 19:54 |
| mhayden | odyssey4me: of course -- now how do i do that? ;) | 19:54 |
| *** sdake has joined #openstack-ansible | 19:54 | |
| odyssey4me | mhayden seeing as it's already merged and doesn't rely on anything else, it's easy | 19:55 |
| odyssey4me | even a windows engineer could do it ;) | 19:55 |
| odyssey4me | clickety on the link | 19:55 |
| odyssey4me | clickety on the cherry-pick button ont he left | 19:55 |
| odyssey4me | typey 'kilo' as the branch to cherry-pick to | 19:56 |
| odyssey4me | clickety 'cherry pick change' | 19:56 |
| odyssey4me | et voila! | 19:56 |
| * mhayden prepares for clickeyu | 19:56 | |
| openstackgerrit | Major Hayden proposed stackforge/os-ansible-deployment: Add profiling for Ansible tasks https://review.openstack.org/222348 | 19:57 |
| odyssey4me | et voila! | 19:57 |
| mhayden | odyssey4me: damn, never have done that before -- thanks! | 19:57 |
| odyssey4me | mhayden thanks :) we'll do the more complex backports another time :) | 19:58 |
| *** sdake_ has joined #openstack-ansible | 20:02 | |
| javeriak | odyssey4me should I rebase https://review.openstack.org/#/c/222262/ ? kinda hard to make out why its failing from the logs | 20:03 |
| odyssey4me | javeriak looking | 20:03 |
| odyssey4me | ah, that's a build timeout which is exactly 90 minutes | 20:04 |
| odyssey4me | you can either rebase, or recheck | 20:04 |
| javeriak | oh i see it now | 20:04 |
| javeriak | recheck? | 20:04 |
| *** sdake has quit IRC | 20:05 | |
| odyssey4me | javeriak yeah, just comment on it with 'recheck build timeout' | 20:06 |
| odyssey4me | which is of the format 'recheck <reason why>' | 20:06 |
| odyssey4me | :) | 20:06 |
| javeriak | cool thanks, I assume this will have retriggered the dsvm pipeline | 20:08 |
| *** cloudtrainme has quit IRC | 20:09 | |
| *** KLevenstein has quit IRC | 20:09 | |
| odyssey4me | javeriak yup - you can see progress at http://status.openstack.org/zuul/ | 20:09 |
| odyssey4me | filter by 'stackforge/os-ansible-deployment' and expand by default | 20:10 |
| *** kukacz has joined #openstack-ansible | 20:10 | |
| *** cloudtrainme has joined #openstack-ansible | 20:10 | |
| odyssey4me | it's busy waiting for nodes from nodepool, and I think there's some trouble with nodepool which -infra is working out | 20:10 |
| odyssey4me | so the nodes are trickling through | 20:11 |
| javeriak | yep i see it now :) | 20:11 |
| javeriak | btw what happens when the repo changes to openstack/ do we need to reclone our local copies? | 20:11 |
| javeriak | what about pending commits | 20:11 |
| odyssey4me | javeriak the pending commits should be fine | 20:12 |
| odyssey4me | you will need to update your remotes, or reclone, yes | 20:12 |
| javeriak | ah okay, will you guys be sending out an official statement or something? | 20:13 |
| odyssey4me | javeriak we'll send a note out to the ML | 20:13 |
| *** KLevenstein has joined #openstack-ansible | 20:14 | |
| javeriak | great, I'll have something share with my folks | 20:15 |
| *** cloudtra_ has joined #openstack-ansible | 20:24 | |
| *** cloudtrainme has quit IRC | 20:27 | |
| *** javeriak has quit IRC | 20:30 | |
| openstackgerrit | Bjoern Teipel proposed stackforge/os-ansible-deployment: Implement Neutron LBAAS using haproxy https://review.openstack.org/220365 | 20:59 |
| *** sdake_ has quit IRC | 21:13 | |
| *** jbweber_ has quit IRC | 21:18 | |
| *** sdake has joined #openstack-ansible | 21:26 | |
| *** jmckind has quit IRC | 21:33 | |
| *** jwagner is now known as jwagner_away | 21:36 | |
| *** kukacz has quit IRC | 21:36 | |
| openstackgerrit | Bjoern Teipel proposed stackforge/os-ansible-deployment: Implement Neutron LBAAS using haproxy https://review.openstack.org/220365 | 21:40 |
| *** phalmos has quit IRC | 21:43 | |
| *** mcarden has joined #openstack-ansible | 21:52 | |
| * cloudnull out | 21:55 | |
| cloudnull | have a good one | 21:55 |
| BjoernT | by bye | 21:56 |
| *** KLevenstein has quit IRC | 21:56 | |
| *** Mudpuppy_ has joined #openstack-ansible | 21:58 | |
| *** agireud has quit IRC | 21:58 | |
| *** Mudpuppy has quit IRC | 22:02 | |
| *** Mudpuppy_ has quit IRC | 22:02 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:18 | |
| *** BjoernT has quit IRC | 22:37 | |
| *** alop has quit IRC | 22:45 | |
| *** woodard_ has joined #openstack-ansible | 23:22 | |
| *** woodard has quit IRC | 23:26 | |
| *** cloudtra_ has quit IRC | 23:28 | |
| *** woodard_ has quit IRC | 23:33 | |
| *** spotz is now known as spotz_zzz | 23:36 | |
| *** arbrandes has quit IRC | 23:40 | |
| openstackgerrit | Bjoern Teipel proposed stackforge/os-ansible-deployment: Implement Neutron LBAAS using haproxy https://review.openstack.org/220365 | 23:46 |
| openstackgerrit | Bjoern Teipel proposed stackforge/os-ansible-deployment: Implement Neutron LBAAS using haproxy https://review.openstack.org/220365 | 23:47 |
| *** shoutm has joined #openstack-ansible | 23:51 | |
| *** arbrandes has joined #openstack-ansible | 23:56 | |
| *** rromans has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!