| *** miguelgrinberg has quit IRC | 00:04 | |
| *** miguelgrinberg has joined #openstack-ansible | 00:05 | |
| *** tlian has quit IRC | 00:05 | |
| *** tlian has joined #openstack-ansible | 00:11 | |
| *** tlian has quit IRC | 00:16 | |
| *** tlian has joined #openstack-ansible | 00:17 | |
| *** bapalm has joined #openstack-ansible | 00:24 | |
| *** Mudpuppy has quit IRC | 00:28 | |
| *** bapalm has quit IRC | 00:44 | |
| *** bapalm has joined #openstack-ansible | 00:45 | |
| *** bapalm has quit IRC | 00:46 | |
| *** bapalm has joined #openstack-ansible | 00:46 | |
| *** sdake_ has joined #openstack-ansible | 00:47 | |
| *** sdake has quit IRC | 00:50 | |
| *** sdake_ is now known as sdake | 01:12 | |
| *** daneyon has quit IRC | 01:16 | |
| *** bapalm has quit IRC | 01:21 | |
| *** bapalm has joined #openstack-ansible | 01:21 | |
| *** bapalm has quit IRC | 01:26 | |
| *** tlian has quit IRC | 02:23 | |
| *** galstrom_zzz is now known as galstrom | 02:52 | |
| *** fawadkhaliq has joined #openstack-ansible | 03:29 | |
| *** JRobinson__ is now known as JRobinson__afk | 03:30 | |
| *** phalmos has joined #openstack-ansible | 03:35 | |
| *** phalmos has quit IRC | 03:37 | |
| *** britthouser has quit IRC | 03:38 | |
| *** galstrom is now known as galstrom_zzz | 03:38 | |
| *** logan2 has quit IRC | 03:47 | |
| *** logan2 has joined #openstack-ansible | 03:55 | |
| *** jasondotstar has quit IRC | 04:06 | |
| *** galstrom_zzz is now known as galstrom | 04:15 | |
| *** fawadkhaliq has quit IRC | 04:21 | |
| *** daneyon has joined #openstack-ansible | 04:24 | |
| *** JRobinson__afk is now known as JRobinson__ | 04:26 | |
| *** galstrom is now known as galstrom_zzz | 04:34 | |
| *** britthouser has joined #openstack-ansible | 04:41 | |
| *** fawadkhaliq has joined #openstack-ansible | 05:22 | |
| *** fawadk has joined #openstack-ansible | 05:31 | |
| *** fawadkhaliq has quit IRC | 05:31 | |
| *** JRobinson__ has quit IRC | 05:50 | |
| *** javeriak has joined #openstack-ansible | 06:24 | |
| *** cfarquhar has quit IRC | 06:41 | |
| *** cfarquhar has joined #openstack-ansible | 06:53 | |
| *** cfarquhar has quit IRC | 06:53 | |
| *** cfarquhar has joined #openstack-ansible | 06:53 | |
| *** javeriak has quit IRC | 07:21 | |
| *** jasondotstar has joined #openstack-ansible | 07:44 | |
| *** jasondotstar has quit IRC | 07:48 | |
| *** javeriak has joined #openstack-ansible | 08:20 | |
| *** fawadk has quit IRC | 08:25 | |
| *** fawadkhaliq has joined #openstack-ansible | 08:26 | |
| evrardjp | good morning everyone | 08:31 |
|---|---|---|
| evrardjp | Sam-I-Am: FYI, I've cherry picked the change you've asked me to test (reducing neutron config) | 08:32 |
| evrardjp | it looks fine, neutron didn't complain when I used my VXLAN/VLAN | 08:32 |
| evrardjp | however I can't fully test it | 08:32 |
| evrardjp | with the latest changes I can't login on horizon >< | 08:32 |
| evrardjp | it works fine with admin | 08:33 |
| evrardjp | but a user created with the help of the admin account is denied | 08:33 |
| evrardjp | v3 thingy I guess | 08:33 |
| evrardjp | "You're not authorized for any project" (while member of it) | 08:35 |
| evrardjp | Oh I may have found the issue: Managing projects with my dashboard doesn't seem to work either... I'll redeploy horizon and keystone, just to be sure | 08:37 |
| *** javeriak has quit IRC | 08:38 | |
| evrardjp | I'm guessing using the openstack CLI will use the v2 API instead of v3 | 08:41 |
| *** javeriak has joined #openstack-ansible | 08:41 | |
| mattt | anyone got good neutron chops? :) | 08:42 |
| odyssey4me | evrardjp so you may wish to update to the latest tag (assuming you're on kilo) which was released last night: 11.1.0 | 08:47 |
| odyssey4me | that includes fixes to Horizon to handle the v3 API if that's what's in your catalogue | 08:47 |
| odyssey4me | evrardjp the backport for the kilo branch for the reduced neutron config is here: https://review.openstack.org/211171 | 08:49 |
| odyssey4me | mattt why do you ask? | 08:50 |
| evrardjp | odyssey4me: yeah I know, Sam-I-Am asked me to test it | 08:50 |
| evrardjp | I don't need it myself, I was just trying to help | 08:51 |
| evrardjp | congrats for release then odyssey4me | 08:53 |
| evrardjp | 11.1.0 is finally out ! | 08:53 |
| odyssey4me | evrardjp yep, and we have a ton of bug to chase - some of which have reviews in progress in master already | 08:53 |
| odyssey4me | evrardjp feel like helping track some of these down and putting reviews in? https://launchpad.net/openstack-ansible/+milestone/11.1.1 | 08:55 |
| odyssey4me | or reviewing/testing any of the patches in flight if there are any | 08:55 |
| *** sdake has quit IRC | 08:55 | |
| evrardjp | I can't right now, but I'll definitely try this new release, bring bugs and comment/review | 08:56 |
| evrardjp | I've pulled and merged into my kilo | 08:57 |
| *** jasondotstar has joined #openstack-ansible | 08:57 | |
| evrardjp | I don't know if I should have discarded my cherry picks first | 08:57 |
| odyssey4me | what're the differences you're carrying - I'd like to help resolve those so that we're focused on the same code | 08:57 |
| evrardjp | I've cherry picked the ceph and the neutron cleanup | 08:58 |
| evrardjp | I should have reset the head first and pull | 08:58 |
| evrardjp | but now it's too late | 08:58 |
| odyssey4me | evrardjp well, the ceph work cherry-picks cleanly | 08:59 |
| odyssey4me | there's a review in flight now, set to WIP just to hold it back for now: https://review.openstack.org/211151 | 08:59 |
| evrardjp | what do you think I should do? reset to an old commit, reimport kilo branch and cherry pick again? | 09:00 |
| mattt | odyssey4me: fiddling w/ this force_config_drive stuff for the live migrations, when it's disabled instances can't reach the metadata service | 09:00 |
| odyssey4me | it'll probably save you a lot more pain in the future :p | 09:00 |
| evrardjp | that's what I thought | 09:00 |
| odyssey4me | mattt yeah, I think config drive was force enabled to work around the metadata service not working - so now the metadata service not working has to be fixed | 09:01 |
| odyssey4me | mattt I'll trade you - figure out why extraroute is causing a semantic versioning blow up in the logs for https://review.openstack.org/211265 and I'll build an AIO and try to figure out the metadata issue | 09:02 |
| mattt | ok just give me a few mins, testing this out on hardware first | 09:03 |
| mattt | to make sure it's not an AIO thing | 09:04 |
| mattt | odyssey4me: yeah works fine in lab | 09:07 |
| odyssey4me | mattt oh? that's interesting | 09:07 |
| mattt | checking http://169.254.169.254/2009-04-04/instance-id | 09:08 |
| odyssey4me | in that case I think it's probably best to ask Apsu to take a peek | 09:08 |
| mattt | successful after 1/20 tries: up 0.78. iid=i-000000cd | 09:08 |
| mattt | aeh | 09:08 |
| mattt | *yeah | 09:08 |
| mattt | i'm going to put force_config_drive back to always/True | 09:08 |
| odyssey4me | there's probably an interfering rule somewhere | 09:08 |
| mattt | and move on | 09:08 |
| mattt | i was a bit on the fence about flipping that anyway | 09:08 |
| mattt | maybe in liberty we can revert back to the upstream default | 09:09 |
| odyssey4me | I'll take a look into doing a multi-node gate check today - maybe just simply one controller, one compute to start with and go from there | 09:09 |
| evrardjp | odyssey4me: mattt: there is an issue with metadata service? | 09:10 |
| mattt | evrardjp: apparantly only in the AIO ? | 09:10 |
| evrardjp | I'll check that later | 09:10 |
| evrardjp | I need to have it working ;) | 09:11 |
| evrardjp | I'll test on non-AIO | 09:11 |
| odyssey4me | mattt if it's only the AIO then perhaps we should default it to whatever the upstream default is, but add a setting to enable it in the AIO bootstrap | 09:11 |
| mattt | odyssey4me: well i'd rather not flip it mid way through kilo | 09:11 |
| mattt | i think it makes sense to flip it when we release liberty, i've updated the main.yml defaults file to remind us | 09:11 |
| odyssey4me | mattt fair point - perhaps log a bug for it and add it to 11.2.0 for now | 09:11 |
| mattt | and in the mean time we can ping Apsu / rackertom to see if they can help us out | 09:12 |
| odyssey4me | we can move it to the liberty milestone later when it's been created | 09:12 |
| odyssey4me | perhaps also link the bug to the liberty blueprint | 09:12 |
| odyssey4me | hughsaunders https://review.openstack.org/209426 | 09:19 |
| mattt | evrardjp: yeah, flip force_config_drive to False and give it a test if you don't mind | 09:20 |
| evrardjp | I'll do that | 09:25 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Remove hardcoded config drive enforcement https://review.openstack.org/195403 | 09:32 |
| mattt | odyssey4me: where do i link the bug in the blueprint? i just paste it in work items? | 09:34 |
| odyssey4me | mattt hang on - you can link a bug by number - let me find where it is | 09:36 |
| odyssey4me | mattt so from https://blueprints.launchpad.net/openstack-ansible/+spec/liberty on the right there is a 'Link a bug report' | 09:37 |
| odyssey4me | between related branches and sprints | 09:37 |
| mattt | ah derp, done! | 09:39 |
| odyssey4me | thanks mattt :) | 09:41 |
| *** javeriak has quit IRC | 09:48 | |
| *** sura8257_ has joined #openstack-ansible | 09:52 | |
| *** sura8257_ has quit IRC | 10:05 | |
| *** javeriak has joined #openstack-ansible | 10:10 | |
| *** javeriak_ has joined #openstack-ansible | 10:11 | |
| *** sura8257 has joined #openstack-ansible | 10:13 | |
| odyssey4me | mattt andymccr can we perhaps get another vote for the backport https://review.openstack.org/209425 | 10:14 |
| odyssey4me | that'll help get the forks patch into kilo, which improves the success of both deployments and gating | 10:14 |
| *** javeriak_ has quit IRC | 10:14 | |
| *** javeriak has quit IRC | 10:14 | |
| *** javeriak has joined #openstack-ansible | 10:15 | |
| mattt | odyssey4me: looking | 10:15 |
| odyssey4me | thanks mattt | 10:15 |
| *** javeriak has quit IRC | 10:15 | |
| odyssey4me | mattt andymccr we also have https://review.openstack.org/209047 and https://review.openstack.org/210593 needed to get master up to liberty-2 - please review, they're pretty simple patches | 10:16 |
| *** javeriak has joined #openstack-ansible | 10:16 | |
| *** javeriak has quit IRC | 10:16 | |
| *** fawadkhaliq has quit IRC | 10:22 | |
| *** fawadkhaliq has joined #openstack-ansible | 10:22 | |
| *** javeriak has joined #openstack-ansible | 10:23 | |
| mattt | odyssey4me: can the latter patch be gated alongside the neutron sha bump that introduces the neutron issues so we can validate that patch does indeed work? | 10:23 |
| odyssey4me | mattt that's not a bad idea, but introducing that sha bump without bumping the others may cause a non-working gate | 10:25 |
| odyssey4me | I was trying to do focused patches instead of a mega patch | 10:25 |
| odyssey4me | the liberty-2 sha bump patch does gat properly with this one as a dep | 10:26 |
| odyssey4me | *gate | 10:26 |
| odyssey4me | before these changes, it did not | 10:26 |
| mattt | ah, ok | 10:28 |
| mattt | that works for me | 10:28 |
| *** javeriak has quit IRC | 10:31 | |
| odyssey4me | mattt that's why I set https://review.openstack.org/209047 and https://review.openstack.org/210593 as dependencies for https://review.openstack.org/199126 as without those two patches, liberty-2 will not work. | 10:31 |
| *** javeriak has joined #openstack-ansible | 10:31 | |
| odyssey4me | but I could split those fixes out into their own patches, allowing them to be more focused :) | 10:31 |
| mattt | odyssey4me: better link https://bugs.launchpad.net/openstack-ansible/+bug/1482756 to the liberty blueprint also | 10:33 |
| openstack | Launchpad bug 1482756 in openstack-ansible trunk "networking fails due to new ebtables-based arp spoofing protection" [High,In progress] - Assigned to Jesse Pretorius (jesse-pretorius) | 10:33 |
| mattt | as we don't want to forget about that TODO | 10:33 |
| odyssey4me | mattt good plan, I linked it to the 11.2.0 milestone as a reminder - but to the liberty blueprint is better :) | 10:34 |
| mattt | odyssey4me: is there an outstanding upstream bug for the prevent_arp_spoofing bug ? | 10:34 |
| odyssey4me | mattt yep, it's linked in https://bugs.launchpad.net/openstack-ansible/+bug/1482756 in the comments | 10:34 |
| openstack | Launchpad bug 1482756 in openstack-ansible trunk "networking fails due to new ebtables-based arp spoofing protection" [High,In progress] - Assigned to Jesse Pretorius (jesse-pretorius) | 10:34 |
| odyssey4me | https://bugs.launchpad.net/neutron/+bug/1483315 | 10:35 |
| openstack | Launchpad bug 1483315 in neutron "ebtables ARP rules don't account for floating IPs on LinuxBridge" [Undecided,New] | 10:35 |
| mattt | odyssey4me: this is presumably not going into 11.2 tho (the ebtables stuff) | 10:35 |
| mattt | not sure why this is flagged for 11.2 | 10:35 |
| odyssey4me | mattt nope, it was just a way to track it and not lose it | 10:35 |
| odyssey4me | until we have a liberty milestone - which I'll arrange to have created later today | 10:35 |
| mattt | ah ok | 10:36 |
| *** javeriak has quit IRC | 10:49 | |
| *** javeriak has joined #openstack-ansible | 10:59 | |
| odyssey4me | mattt did you take a look at https://review.openstack.org/209047 ? | 11:00 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Set lxc.autodev=0 for cinder_volumes in containers https://review.openstack.org/211502 | 11:00 |
| mattt | odyssey4me: nope not yet | 11:06 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Added gate process and more logs https://review.openstack.org/209425 | 11:18 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Change ansible forks used https://review.openstack.org/209426 | 11:20 |
| odyssey4me | mattt shouldn't https://review.openstack.org/195403 include passing some of those options to tempest? | 11:38 |
| mattt | odyssey4me: to tempest? | 11:39 |
| odyssey4me | yeah, if we end up disabling config drive, then surely that should reflect in the tempest config too? | 11:39 |
| *** sura8257 has quit IRC | 11:39 | |
| odyssey4me | mattt never mind, it appears that tempest has no such option | 11:42 |
| odyssey4me | unless disk_config relates | 11:42 |
| mattt | odyssey4me: there is some password changing related options in tempest, which cloudnull_afk originally set in that review | 11:42 |
| odyssey4me | but that seems more to do with doing different disk partitioning | 11:42 |
| mattt | but that ws because he actually enabled all the inject password stuff etc. | 11:43 |
| odyssey4me | ah, let me look into those | 11:43 |
| odyssey4me | let me try those out in a follow-up patch | 11:44 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add support for additional nova.conf options https://review.openstack.org/210492 | 11:45 |
| *** sdake has joined #openstack-ansible | 11:53 | |
| *** sdake_ has joined #openstack-ansible | 11:56 | |
| evrardjp | mattt: sorry for the late answer but, I can flip force_config_drive in the template, but not in a variable (as it's not set like that in the template by default in kilo). That's what you're looking for ? | 11:58 |
| *** sdake has quit IRC | 12:00 | |
| odyssey4me | evrardjp that's what https://review.openstack.org/195403 does | 12:01 |
| *** openstackgerrit_ has joined #openstack-ansible | 12:02 | |
| odyssey4me | mattt ok, after some inspection - change_password is false by default, and enable_instance_password is true by default, but I think it should stay that way until we find a condition that it should be set to false | 12:08 |
| *** woodard has joined #openstack-ansible | 12:09 | |
| evrardjp | for the force config drive, I don't know, but for the some of the pass injection: http://ceph.com/docs/master/rbd/rbd-openstack/#id2 | 12:12 |
| evrardjp | I'll test with False | 12:13 |
| odyssey4me | evrardjp there's a follow up patch for the nova.conf bits - https://review.openstack.org/210492 | 12:13 |
| evrardjp | I've seen that one odyssey4me, but thanks for pointing: I forgot to cherry pick it again ;) | 12:14 |
| mattt | odyssey4me: derp, i hadn't seen that other bug (allow password injection) | 12:16 |
| mattt | i was wondering why cloudnull_afk enabled all that stuff, but i think this explains it | 12:17 |
| odyssey4me | mattt yeah, not exactly a duplicate bug but perhaps handled in the same patch | 12:17 |
| odyssey4me | That's why I unduplicated it, so that we can list it in the list of bugs fixed in the next milestone | 12:18 |
| mattt | odyssey4me: i'm going to re-add some of the stuff cloudnull_afk initially added and also link to that bug from the review | 12:20 |
| odyssey4me | mattt great! | 12:21 |
| odyssey4me | alternatively, perhaps do a subsequent dependant patch? | 12:22 |
| odyssey4me | that might be cleaner and more focused | 12:22 |
| mattt | sure | 12:23 |
| odyssey4me | it's up to you | 12:23 |
| mattt | yeah if i want to do it properly i probably have to rip out some of the stuff in the original review | 12:26 |
| mattt | so i'm probably going to leave it for the time being | 12:26 |
| *** bapalm has joined #openstack-ansible | 12:26 | |
| *** bapalm has quit IRC | 12:26 | |
| mattt | guess i may as well do it right | 12:27 |
| *** bapalm has joined #openstack-ansible | 12:27 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Remove hardcoded config drive enforcement https://review.openstack.org/195403 | 12:38 |
| *** shausy has joined #openstack-ansible | 12:42 | |
| *** jaypipes has joined #openstack-ansible | 12:44 | |
| *** bapalm_ has joined #openstack-ansible | 12:45 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Allow nova key/partition/password injection https://review.openstack.org/211543 | 12:48 |
| *** bapalm has quit IRC | 12:49 | |
| *** openstackgerrit_ has quit IRC | 12:51 | |
| *** sdake_ is now known as sdake | 12:54 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Allow nova key/partition/password injection https://review.openstack.org/211543 | 12:54 |
| *** tlian has joined #openstack-ansible | 12:54 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Set lxc.autodev=0 for cinder_volumes in containers https://review.openstack.org/211502 | 12:56 |
| *** openstackgerrit_ has joined #openstack-ansible | 12:56 | |
| Sam-I-Am | odyssey4me: i was messing with that aio of yours on 231.17 - what branch/patch is that running? | 13:04 |
| Sam-I-Am | i tried doing a git review on the ebtables patch, but it didnt grab the sha update patch | 13:04 |
| odyssey4me | Sam-I-Am that's running a frankenstein, but includes the sha patch already | 13:07 |
| odyssey4me | it's running master with the liberty-2 sha updates | 13:07 |
| Sam-I-Am | ok | 13:09 |
| Sam-I-Am | well, either way, i confirmed the problem | 13:09 |
| Sam-I-Am | just couldnt get my own lab to look that way | 13:09 |
| odyssey4me | Sam-I-Am so if you want your own lab to look that way, you have to cherry-pick each of the patches in the chain in succession | 13:09 |
| odyssey4me | I can fix that one to get the patches in if you like? | 13:10 |
| Sam-I-Am | odyssey4me: probably not a big deal now | 13:11 |
| *** fawadkhaliq has quit IRC | 13:20 | |
| *** jwagner_away is now known as jwagner | 13:40 | |
| *** KLevenstein has joined #openstack-ansible | 13:43 | |
| *** sdake_ has joined #openstack-ansible | 13:55 | |
| *** openstackgerrit_ has quit IRC | 13:55 | |
| *** openstackgerrit_ has joined #openstack-ansible | 13:56 | |
| *** sdake__ has joined #openstack-ansible | 13:58 | |
| *** sdake has quit IRC | 13:58 | |
| *** spotz_zzz is now known as spotz | 14:00 | |
| *** fawadkhaliq has joined #openstack-ansible | 14:01 | |
| *** sdake_ has quit IRC | 14:02 | |
| *** Mudpuppy has joined #openstack-ansible | 14:04 | |
| Apsu | odyssey4me: What am I helping you out with? | 14:07 |
| *** vdo_ has joined #openstack-ansible | 14:09 | |
| bgmccollum | following upgrade procedure says to run `nova-manage db null_instance_uuid_scan`, however that sub command doesn't seem to exist... | 14:09 |
| *** bapalm_ has quit IRC | 14:11 | |
| *** bapalm has joined #openstack-ansible | 14:11 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:15 | |
| *** bapalm has quit IRC | 14:15 | |
| *** bapalm has joined #openstack-ansible | 14:15 | |
| *** bapalm has quit IRC | 14:16 | |
| *** javeriak has quit IRC | 14:17 | |
| *** bapalm has joined #openstack-ansible | 14:17 | |
| *** openstackgerrit_ has quit IRC | 14:17 | |
| *** openstackgerrit_ has joined #openstack-ansible | 14:18 | |
| *** javeriak has joined #openstack-ansible | 14:19 | |
| evrardjp | ok so | 14:20 |
| evrardjp | according to your advise odyssey4me, I've updated and redeployed (the 11.1.0) | 14:21 |
| bgmccollum | `null_instance_uuid_scan` wasn't added until kilo, so having it as a pre-req is kinda not exactly possible... | 14:21 |
| evrardjp | however the create project doesnt seem to work in horizon | 14:22 |
| bgmccollum | evrardjp: you need to create the default role _member_ | 14:23 |
| *** prad_ has joined #openstack-ansible | 14:24 | |
| evrardjp | indeed that's missing | 14:25 |
| evrardjp | but that wasn't in the previous release | 14:25 |
| bgmccollum | evrardjp: it used to get created automatically | 14:25 |
| bgmccollum | something something about keystone v3 | 14:25 |
| *** prad_ is now known as pradk | 14:25 | |
| evrardjp | is it something that I should map attributes to? | 14:26 |
| evrardjp | or just create an empty _member_ role ? | 14:26 |
| bgmccollum | there is an open bug report...let me locate | 14:26 |
| bgmccollum | just create it | 14:26 |
| bgmccollum | openstack role create _member_ | 14:26 |
| bgmccollum | evrardjp: https://bugs.launchpad.net/openstack-ansible/+bug/1474916 | 14:27 |
| openstack | Launchpad bug 1474916 in openstack-ansible trunk "Keystone v3 does not automatically create _member_ role" [High,In progress] - Assigned to Ian Cordasco (icordasc) | 14:27 |
| evrardjp | that's just what I did | 14:27 |
| Sam-I-Am | the role is arbitrary | 14:27 |
| Sam-I-Am | but defaults to _member_ | 14:27 |
| Sam-I-Am | what horizon thinks exists needs to exist | 14:27 |
| bgmccollum | Sam-I-Am: and it used to exist automatically...and everybody was none the wiser | 14:28 |
| bgmccollum | :) | 14:28 |
| Sam-I-Am | yeah, keystone v2 used to Do Some Magic to create it | 14:29 |
| Sam-I-Am | and that stopped with v3, because there's no concept of a default role iirc | 14:29 |
| bgmccollum | Sam-I-Am: funny thing is, all horizon is doing is selecting a default role in a dropdown. when it can't find that role name in the list of roles, it throws its hands up... | 14:30 |
| Sam-I-Am | bgmccollum: which is crap, imo | 14:30 |
| *** openstackgerrit_ has quit IRC | 14:31 | |
| *** yaya has joined #openstack-ansible | 14:34 | |
| *** sdake__ is now known as sdake | 14:37 | |
| odyssey4me | Apsu: so if we disable config_drive we find that instances don't get metadata in the AIO - it seems that it works in a multi-node deployment | 14:37 |
| odyssey4me | evrardjp yes bgmccollum is right - we have a bug for that and a review that needs some work: https://bugs.launchpad.net/openstack-ansible/+bug/1474916 | 14:40 |
| openstack | Launchpad bug 1474916 in openstack-ansible trunk "Keystone v3 does not automatically create _member_ role" [High,In progress] - Assigned to Ian Cordasco (icordasc) | 14:40 |
| Apsu | odyssey4me: This sounds like something we ran into and fixed like a year+ ago | 14:41 |
| odyssey4me | Apsu it seems that the fix was to just permanently enable config_drive :p | 14:41 |
| Apsu | Probably that was chosen to get around the metadata hack I came up with | 14:42 |
| odyssey4me | but that's changing as we've had the request to make it optional | 14:42 |
| Apsu | You recall that? | 14:42 |
| odyssey4me | Apsu oh yeah, most definitely :p | 14:42 |
| odyssey4me | it's probably something quite simple - I just haven't had the time to look into it, it was just picked up by mattt this morning | 14:43 |
| Apsu | A long time ago I submitted a patch to fix it, and they said I didn't submit any "tests". Yet my patch reduced the number of code paths, and didn't change any combinations of things.... nothing new to test | 14:43 |
| Apsu | Well, what was previously the issue and has been since Neutron was introduced, is that if you have a gateway on an instance network, it won't automatically inject metadata routes | 14:44 |
| openstackgerrit | Christopher H. Laco proposed stackforge/os-ansible-deployment: Add resume_guests_state_on_host_boot to nova.conf https://review.openstack.org/211600 | 14:44 |
| mattt | step 1, just add tets. | 14:44 |
| Apsu | So you have to fool it | 14:44 |
| evrardjp | odyssey4me: I can confirm the bug | 14:44 |
| evrardjp | but it has been done already | 14:44 |
| bgmccollum | Apsu: there are two settings i recall...proxy metadata requests, and enable isolated metadata access or something like that...ive gotten it to work before on AIOs... | 14:45 |
| mattt | bgmccollum: yeah we have isolated_metadata enabled, i'll have to look at proxy metadata requests | 14:45 |
| Apsu | bgmccollum: All that's set already and has been forever | 14:45 |
| Apsu | What you have to do is disable the gateway field for the subnet, then add in host-routes that do the same thing. | 14:45 |
| odyssey4me | it sounds like it needs a revisit | 14:45 |
| Apsu | odyssey4me: This on master/liberty-2? | 14:46 |
| mattt | Apsu: but this all works on a multi-node, so i'm not sure it's that bug | 14:46 |
| odyssey4me | this is master (liberty), so it's possible that things have changed upstream | 14:46 |
| *** javeriak has quit IRC | 14:46 | |
| Apsu | mattt: Ah. Right, good point. I recall an issue with That specifically too before | 14:47 |
| Apsu | I think it was split horizon routing going on, but I might be remembering wrong. | 14:47 |
| Apsu | Due to the AIO plumbing | 14:47 |
| mattt | i need to dig into it a bit more, but w/ the AIO i couldn't ssh into the instance although the instance did get an IP | 14:47 |
| mattt | i could ping the instance's IP from the qdhcp ns, but that was abou tit | 14:47 |
| Apsu | Got a box up? | 14:48 |
| Apsu | I can find out pretty quick what's going on | 14:48 |
| mattt | Apsu: sure, let me grab your key | 14:48 |
| Apsu | kk | 14:48 |
| *** galstrom_zzz is now known as galstrom | 14:48 | |
| *** shausy has quit IRC | 14:49 | |
| odyssey4me | morning palendae | 14:49 |
| odyssey4me | would you mind working through https://review.openstack.org/#/q/starredby:%22Jesse+Pretorius%22+project:stackforge/os-ansible-deployment,n,z today? | 14:50 |
| palendae | Jeebus | 14:50 |
| palendae | Don't we have more cores around today too? | 14:50 |
| *** javeriak has joined #openstack-ansible | 14:50 | |
| odyssey4me | miguelgrinberg when you're online, please let me know - I'd like your assistance determining how best to resolve the extraroutes failure in https://review.openstack.org/211265 | 14:51 |
| odyssey4me | palendae nope, all focused on feature work :/ | 14:51 |
| palendae | ... | 14:51 |
| palendae | As am I... | 14:51 |
| Apsu | Just say no! | 14:53 |
| palendae | That's my no | 14:53 |
| bgmccollum | Apsu, odyssey4me, mattt: you also need to set `enable_metadata_network = True` in dhcp_agent.ini ... that was the last piece of the magic | 14:54 |
| bgmccollum | then you could freely switch between config drive, metadata service, completely isolated network, or let the router handle it...it seemed to work in all situations... | 14:55 |
| Apsu | bgmccollum: No that only worked if you had a 169.254/16 network also connected to the instance | 14:56 |
| Apsu | That's what enable_metadata_network does -- uses a dedicated metadata network if you have one | 14:56 |
| mattt | bgmccollum: 99% sure i tried that today | 14:56 |
| bgmccollum | Apsu: it injects the routes automatically | 14:56 |
| mattt | my instances already have a route for the metadata stuff without that | 14:56 |
| Apsu | bgmccollum: Only if you have a 169.254/16 neutron network :) | 14:56 |
| bgmccollum | i seem to recall it plumbed everything... | 14:57 |
| Sam-I-Am | do you also have to specifically configure dnsmasq to push routes? | 14:58 |
| mattt | like i said this is only an issue on the AIO | 14:59 |
| Apsu | mattt: Connecting and taking a look | 14:59 |
| Apsu | Started up tmux | 14:59 |
| mattt | Apsu: party! | 15:00 |
| bgmccollum | Sam-I-Am: no | 15:00 |
| Apsu | mattt: I take it you deleted all the instances and whatnot. Just verifying I'm looking at the right place :P | 15:02 |
| mattt | Apsu: yeah let me spin one up | 15:03 |
| mattt | Apsu: 444b9f42-a3c5-4d4e-9776-c4295eac2bce|test | 15:03 |
| Apsu | mattt: Wait, you're using cirros. | 15:05 |
| Apsu | Are you sure that cirros fixed their udhcpc options to request extra host routes? | 15:06 |
| mattt | yeah the same cirros image works on the hardware install | 15:06 |
| mattt | i also tried trusty, same diff | 15:06 |
| Apsu | kk | 15:06 |
| Apsu | Just verificating! | 15:06 |
| *** galstrom is now known as galstrom_zzz | 15:07 | |
| odyssey4me | Apsu yeah that was fixed in 0.3.1 as I recall - or maybe 0.3.2 | 15:07 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Retain single hashed comments in secrets file https://review.openstack.org/211609 | 15:08 |
| Apsu | Got the multi-node install up? Want to see the router-show for its router and subnet-show for private-subnet | 15:08 |
| *** sdake_ has joined #openstack-ansible | 15:09 | |
| Apsu | @ mattt | 15:10 |
| mattt | Apsu: should be identical, as all neutron configs were created by tempest, but sec | 15:12 |
| *** sdake has quit IRC | 15:13 | |
| bgmccollum | neutron behaves differently depending on if you have a router port or not: see https://github.com/openstack/neutron/blob/stable/kilo/etc/dhcp_agent.ini#L39-L52 | 15:23 |
| bgmccollum | this is why i was saying you need to have both set to true | 15:23 |
| *** bapalm has quit IRC | 15:27 | |
| *** bapalm has joined #openstack-ansible | 15:28 | |
| Apsu | bgmccollum: Line 47 is the important caveat there | 15:29 |
| Apsu | Has to have a CIDR of 169.254/16, the APIPA range, like I said :) | 15:29 |
| Apsu | So it doesn't automatically work -- it lets you create a metadata-only network | 15:30 |
| Apsu | Also, the way the code *used* to work, and I'd be surprised if it changed, is that it didn't matter if the network was on a router port or not -- it mattered if gateway_ip was set or not | 15:30 |
| bgmccollum | Apsu: its completely isolated in the namespace...take a look, things change. you might be surprised. | 15:34 |
| bgmccollum | or not...your choice | 15:34 |
| Apsu | mattt: FYI, the metadata route is injected into the DHCP options for the instances on both the AIO and multi-node. | 15:34 |
| Apsu | Why it's not being picked up is the question | 15:35 |
| Apsu | root@aio_neutron_agents_container-d9f0e958:~# cat /var/lib/neutron/dhcp/33ef7b72-7239-41b9-8cbf-5e2a4d4f0b8a/opts | 15:35 |
| Apsu | tag:tag0,option:classless-static-route,169.254.169.254/32,192.168.74.1,0.0.0.0/0,192.168.74.1 | 15:35 |
| *** javeriak has quit IRC | 15:37 | |
| *** shausy has joined #openstack-ansible | 15:38 | |
| Apsu | mattt: I see the DHCP server offering the static-classless-route for metadata, but not the cirros image picking them up. Going to try trusty | 15:44 |
| Apsu | Just because | 15:44 |
| mattt | Apsu: do what you need to do ! | 15:45 |
| Apsu | I'll do what I want! | 15:45 |
| Apsu | Got the route. | 15:45 |
| Apsu | ci-info: | 1 | 169.254.169.254 | 172.29.248.3 | 255.255.255.255 | eth0 | UGH | | 15:45 |
| Apsu | Comparing cirros images | 15:45 |
| Apsu | Identical. How odd | 15:46 |
| Sam-I-Am | Apsu: cirros version? | 15:49 |
| Apsu | Don't know, but the checksum is the same. | 15:50 |
| Sam-I-Am | iirc you need 0.34 for extra dhcp options (including mtu) | 15:50 |
| Apsu | Poking more closely | 15:50 |
| Apsu | You can poke with me if you like | 15:50 |
| Sam-I-Am | ee1eca47dc88f4879d8a229cc70a07c6 is the glance checksum i have for my 0.3.4 | 15:51 |
| Apsu | That's not what this checksum is | 15:51 |
| *** sdake_ is now known as sdake | 15:51 | |
| Apsu | 133eae9fb1c98f45894a4e60d8736619 | 15:52 |
| Sam-I-Am | the stock tempest image is not 0.3.4 iirc | 15:52 |
| Apsu | size | 13200896 | 15:52 |
| odyssey4me | yup, that's in review to be updated: https://review.openstack.org/210107 | 15:52 |
| Apsu | === cirros: current=0.3.3 uptime=1.59 === | 15:53 |
| Apsu | But, that got the extra route | 15:53 |
| Apsu | ip-route:169.254.169.254 via 192.168.74.1 dev eth0 | 15:53 |
| Sam-I-Am | interesting | 15:53 |
| Apsu | tcpdumping to compare what the udhcpc clients say | 15:54 |
| *** sdake_ has joined #openstack-ansible | 15:54 | |
| *** sdake_ has quit IRC | 15:54 | |
| *** sdake has quit IRC | 15:57 | |
| *** sdake has joined #openstack-ansible | 15:57 | |
| *** daneyon has quit IRC | 15:58 | |
| *** daneyon has joined #openstack-ansible | 15:59 | |
| *** Bjoern_ has joined #openstack-ansible | 16:00 | |
| *** Bjoern_ is now known as BjoernT | 16:00 | |
| *** jasondotstar has quit IRC | 16:00 | |
| odyssey4me | cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire bug triage meeting here | 16:01 |
| prometheanfire | yes | 16:01 |
| hughsaunders | word | 16:01 |
| dstanek | howdy odyssey4me | 16:01 |
| odyssey4me | howdy all :) | 16:01 |
| mattt | \o | 16:01 |
| odyssey4me | it looks like a short list today, so we can make it quick | 16:02 |
| odyssey4me | the list is https://bugs.launchpad.net/openstack-ansible/+bugs?search=Search&field.status=New | 16:02 |
| odyssey4me | I see several docimpact bugs though, I'll handle those | 16:02 |
| odyssey4me | right, first bug: https://bugs.launchpad.net/openstack-ansible/+bug/1482265 | 16:02 |
| openstack | Launchpad bug 1482265 in openstack-ansible "Nova-computes need nfs-common installed if NFS cinder backend" [Undecided,New] | 16:02 |
| *** pfire has joined #openstack-ansible | 16:03 | |
| Sam-I-Am | howdy | 16:03 |
| odyssey4me | ouch, git-harry is this a sensible requirement? | 16:03 |
| pfire | Yo | 16:03 |
| odyssey4me | it looks pretty legit to me :) | 16:03 |
| *** pfire has quit IRC | 16:03 | |
| *** pfire has joined #openstack-ansible | 16:03 | |
| odyssey4me | ok, nfs-common is already installed in kilo: https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/os_cinder/defaults/main.yml#L189 | 16:05 |
| odyssey4me | hmm, but not compute | 16:07 |
| BjoernT | you need nfs-common if you do cinder on nfs | 16:08 |
| Sam-I-Am | i thought cinder nfs meant cinder mounted nfs filesystems and handed them out using another method to instances | 16:08 |
| BjoernT | on the compute | 16:08 |
| Sam-I-Am | because you can't make nfs a block device | 16:08 |
| *** openstackgerrit_ has joined #openstack-ansible | 16:08 | |
| pfire | Yes you can iirc | 16:08 |
| Sam-I-Am | in other words, do we support the cinder backend drivers with nfs, or nfs directly from compute nodes? | 16:08 |
| odyssey4me | BjoernT so nfs-common would need to be added to compute hosts? | 16:09 |
| BjoernT | yes | 16:09 |
| Sam-I-Am | BjoernT: i'd like to see how this is configured in the deployment config file | 16:09 |
| odyssey4me | BjoernT and this is an issue with juno I take it? | 16:09 |
| BjoernT | yes juno | 16:09 |
| Sam-I-Am | because i'm guessing its undocumented | 16:09 |
| odyssey4me | perhaps also with kilo - no confirmation for that yet | 16:09 |
| BjoernT | if it is part of kilo it's ok | 16:09 |
| BjoernT | we have internal documentation about this how to configure nfs on cinder | 16:10 |
| Sam-I-Am | BjoernT: might be something you want to pass to docs | 16:10 |
| odyssey4me | alright, I'll look into that seeing as I'm also looking into prepping a gate check for nfs-backed glance/cinder | 16:10 |
| Sam-I-Am | we don't like getting blind-sided | 16:10 |
| Sam-I-Am | i know we have glance backed with nfs | 16:11 |
| odyssey4me | alright, next one: https://bugs.launchpad.net/openstack-ansible/+bug/1482045 | 16:11 |
| openstack | Launchpad bug 1482045 in openstack-ansible "Allow configuration of dhcp_domain in nova.conf" [Undecided,New] | 16:11 |
| Sam-I-Am | weird that it would use a value from nova.conf | 16:13 |
| Sam-I-Am | rather than a neutrony thing | 16:13 |
| Sam-I-Am | but i suppose it makes sense, since this is for md | 16:13 |
| pfire | Nova is the metadata source iirx | 16:13 |
| odyssey4me | yup, we have this in neutron already - so it seems odd | 16:13 |
| odyssey4me | in both juno and kilo | 16:14 |
| *** jasondotstar has joined #openstack-ansible | 16:14 | |
| pfire | Doesn't Neutron forward metadata to nova? | 16:14 |
| Sam-I-Am | it does simply proxy it | 16:14 |
| pfire | The requests | 16:14 |
| odyssey4me | ah, dhcp_domain is used from nova.conf when using nova network I imagine | 16:15 |
| Sam-I-Am | odyssey4me: thats what i'd think too | 16:15 |
| Sam-I-Am | but if jordan set it, and it fixed an issue... | 16:15 |
| Sam-I-Am | there's obviously a difference between whats handed out via dhcp and what ends up in metadata? | 16:15 |
| Sam-I-Am | we need more support folks in here :/ | 16:16 |
| miguelgrinberg | odyssey4me: looking at the extraroute thing now | 16:16 |
| odyssey4me | ok, going to mark this as more info needed for now | 16:17 |
| odyssey4me | I've added a comment | 16:17 |
| odyssey4me | miguelgrinberg thanks | 16:17 |
| Apsu | pfire: Yes, once it gets them. | 16:17 |
| odyssey4me | BjoernT perhaps you can ask Jordan to pop into the next meeting or provide more info | 16:17 |
| BjoernT | sure | 16:18 |
| odyssey4me | thanks | 16:18 |
| odyssey4me | and last one: https://bugs.launchpad.net/openstack-ansible/+bug/1481787 | 16:18 |
| openstack | Launchpad bug 1481787 in openstack-ansible trunk "cinder-backup doesn't work with ceph" [Medium,New] - Assigned to Matt Thompson (mattt416) | 16:18 |
| BjoernT | I think we have it implemented it already | 16:18 |
| odyssey4me | mattt I think you've already got a patch in review for this? | 16:18 |
| mattt | odyssey4me: yeah i believe it's still in flight | 16:18 |
| odyssey4me | yep: https://review.openstack.org/209537 | 16:18 |
| mattt | let me chase up that gate failure | 16:19 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: Allow cinder-backup to use ceph https://review.openstack.org/209537 | 16:19 |
| odyssey4me | mattt I think the change-id needs to be lower, and docimpact higher - not sure if that affects the link to launchpad... odd | 16:19 |
| odyssey4me | that's all folks - the rest are docimpact bugs which I'll transfer to the docs team | 16:20 |
| Sam-I-Am | i dont think the order matters | 16:20 |
| *** ig0r_ has joined #openstack-ansible | 16:20 | |
| odyssey4me | does anyone want to raise any specific bugs for discussion? | 16:20 |
| odyssey4me | note that we have quite a few bugs targeted at 10.1.12 which is scheduled for release on friday - so please look through them and see if you can assist | 16:23 |
| odyssey4me | https://launchpad.net/openstack-ansible/+milestone/10.1.12 | 16:23 |
| *** shausy has quit IRC | 16:24 | |
| *** daneyon has quit IRC | 16:26 | |
| *** pfire has quit IRC | 16:30 | |
| *** bapalm_ has joined #openstack-ansible | 16:35 | |
| Apsu | Sam-I-Am: odyssey4me: solved the problem, btw. | 16:36 |
| Apsu | Same problem I solved like a year+ ago. | 16:36 |
| odyssey4me | Apsu oh? | 16:36 |
| *** javeriak has joined #openstack-ansible | 16:36 | |
| Apsu | Yeah. So... on an AIO, there's no physical interfaces involved in the instance -> metadata request | 16:37 |
| Sam-I-Am | Apsu: what do you mean? | 16:37 |
| Apsu | And without that, the checksums are incorrect. Which is sometimes okay, but often not depending on the path taken | 16:37 |
| Apsu | Sam-I-Am: I mean what I said. There's no physical interface being traversed. It's vm -> tap -> bridge -> veth -> container -> veth -> bridge -> veth -> netns -> veth | 16:38 |
| Sam-I-Am | oh, right AIO | 16:38 |
| * Sam-I-Am didnt parse that part | 16:38 | |
| Apsu | And it turns out, the checksums don't automagically work, which is why neutron inserts a CHECKSUM mangle rule for DHCP | 16:38 |
| Sam-I-Am | dont mind me | 16:38 |
| Apsu | But that doesn't help the metadata request :P | 16:38 |
| Apsu | The solution, as it was when I first solved it, is to insert a blanket mangle in the neutron-agents container. | 16:38 |
| Apsu | iptables -t mangle -A POSTROUTING -j CHECKSUM --checksum-fill | 16:39 |
| Apsu | et voila | 16:39 |
| Apsu | Cloud-in-a-box is neat and fun | 16:39 |
| miguelgrinberg | odyssey4me: do we need to be on the greatest and latest version of pbr? Pinning it to the latest 0.10.x release will address the extraroute problem | 16:39 |
| *** bapalm has quit IRC | 16:39 | |
| odyssey4me | miguelgrinberg we don't, but we're deriving the pbr version from openstack's global requirements | 16:41 |
| Apsu | Sam-I-Am: For reference, the symptom you'll see in checksum/window/other "INVALID" packet issues is you'll see the packet come in on tcpdump, on a bridge port, and show checksum: (incorrect), but you won't see it On the bridge it's plugged into | 16:41 |
| miguelgrinberg | odyssey4me: the other possibility is to fix it upstream, if we can wait to move the SHA | 16:41 |
| Apsu | Sam-I-Am: That confused the shit out of me at first, how a bridge port's traffic could somehow not hit the bridge. But it's the checksum :P | 16:42 |
| odyssey4me | Apsu I thought we already had that as a post-up in the network interfaces file? | 16:42 |
| Apsu | odyssey4me: No idea. Wasn't on this container. Just the DHCP checksum | 16:43 |
| odyssey4me | miguelgrinberg we can implement a workaround for now to get us to libery-2, but perhaps it's already fixed upstream in a later sha? | 16:43 |
| miguelgrinberg | odyssey4me: doesn't look fixed for kilo, it is fixed for liberty. This used to work, pbr changes caused the problem | 16:43 |
| odyssey4me | Apsu https://github.com/stackforge/os-ansible-deployment/blob/kilo/etc/network/interfaces.d/aio_interfaces.cfg#L26-L27 ? | 16:44 |
| *** daneyon has joined #openstack-ansible | 16:44 | |
| Apsu | odyssey4me: That's for the root of the host, not in a container, and is for ssh to work from the root. | 16:44 |
| odyssey4me | miguelgrinberg ok so there's a stable/kilo branch issue which will need a patch :) meanwhile we'll work around it - what's the max pbr version we'd prefer to have? | 16:44 |
| Apsu | Neutron handles DHCP already by inserting rules for port 68 | 16:45 |
| miguelgrinberg | odyssey4me: 0.10.8 will work I think | 16:45 |
| Apsu | If we wanted to be more 'precise', we could just handle port 80 | 16:45 |
| Apsu | To manage metadata | 16:45 |
| miguelgrinberg | odyssey4me: 0.11.x and 1.x will not | 16:45 |
| *** galstrom_zzz is now known as galstrom | 16:45 | |
| odyssey4me | miguelgrinberg https://github.com/openstack/heat/blob/stable/kilo/requirements.txt#L5 ? | 16:46 |
| *** vdo_ has quit IRC | 16:46 | |
| miguelgrinberg | how come we ended up with 1.4 then? | 16:46 |
| odyssey4me | it may just be the repo container that has the issue here - let me look into that :) | 16:47 |
| odyssey4me | thanks for following up with it | 16:47 |
| Apsu | Testing just port 80 | 16:47 |
| miguelgrinberg | odyssey4me: np, I think I'll patch this upstream anyway | 16:47 |
| odyssey4me | miguelgrinberg it seems that is is already patched upstream, no? | 16:47 |
| odyssey4me | unless you mean upstream as in pbr? | 16:47 |
| *** ig0r_ has quit IRC | 16:48 | |
| miguelgrinberg | odyssey4me: I mean upstream heat in kilo | 16:48 |
| miguelgrinberg | the requirements allow for 0.11 version of pbr, which will break | 16:48 |
| odyssey4me | ah ok | 16:48 |
| Apsu | odyssey4me: Sam-I-Am: mattt: Yep, that works. iptables -t mangle -A POSTROUTING -p tcp -m tcp --sport 80 -j CHECKSUM --checksum-fill | 16:49 |
| Apsu | (it's the replies that get eaten, fyi) | 16:49 |
| odyssey4me | Apsu in the neutron-agents container? | 16:50 |
| Apsu | Same as with DHCP, which is why the rule mangles udp port 68 only, not 67 | 16:50 |
| Apsu | odyssey4me: Yep | 16:50 |
| *** daneyon_ has joined #openstack-ansible | 16:51 | |
| *** daneyon has quit IRC | 16:53 | |
| odyssey4me | thanks Apsu :) | 16:53 |
| Apsu | You betcha | 16:54 |
| * Apsu 's a helper | 16:54 | |
| *** bapalm_ has quit IRC | 16:54 | |
| *** bapalm has joined #openstack-ansible | 16:54 | |
| *** galstrom is now known as galstrom_zzz | 16:55 | |
| *** bapalm_ has joined #openstack-ansible | 16:55 | |
| *** bapalm has quit IRC | 16:59 | |
| *** yaya has quit IRC | 17:00 | |
| *** bapalm_ has quit IRC | 17:00 | |
| *** woodard has quit IRC | 17:03 | |
| *** woodard has joined #openstack-ansible | 17:06 | |
| *** yaya has joined #openstack-ansible | 17:16 | |
| *** abitha has joined #openstack-ansible | 17:20 | |
| *** blackmonk has joined #openstack-ansible | 17:44 | |
| *** harlowja has quit IRC | 17:51 | |
| *** harlowja has joined #openstack-ansible | 17:54 | |
| *** bapalm has joined #openstack-ansible | 17:56 | |
| stevelle | I think these patches need workflow refreshed. They seem to have slipped through the cracks. https://review.openstack.org/#/c/207105/ | 17:56 |
| *** galstrom_zzz is now known as galstrom | 17:58 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Adds a pep8 target to tox.ini https://review.openstack.org/207662 | 17:59 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Removes trailing whitespace for bashate https://review.openstack.org/207663 | 18:00 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Removes tab indents for bashate https://review.openstack.org/207664 | 18:00 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Fixes deprecated arithmetic expansion for bashate https://review.openstack.org/207665 | 18:00 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Fixes function declarations for bashate https://review.openstack.org/207666 | 18:00 |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Fixes loops for bashate https://review.openstack.org/207667 | 18:00 |
| odyssey4me | stevelle don't bother yet - that's an issue I'm resolving with -infra around our docs not hitting any gate jobs | 18:00 |
| stevelle | odyssey4me: ok, thought that got resolved, my mistake. | 18:01 |
| *** bapalm has quit IRC | 18:01 | |
| odyssey4me | nope, -infra wants us to use standard docs jobs - so I'll have to re-roll the patch to make it do that | 18:02 |
| stevelle | I would like to understand more about this but not right now | 18:03 |
| *** bapalm has joined #openstack-ansible | 18:05 | |
| *** openstackgerrit_ has quit IRC | 18:06 | |
| *** galstrom is now known as galstrom_zzz | 18:10 | |
| *** phalmos has joined #openstack-ansible | 18:11 | |
| *** blackmonk has quit IRC | 18:14 | |
| *** yaya has quit IRC | 18:16 | |
| mgariepy | hello, is there anyway to run lxc-container create in parallels easily with OSAD ? | 18:21 |
| *** yaya has joined #openstack-ansible | 18:51 | |
| *** bapalm has quit IRC | 19:04 | |
| *** bapalm has joined #openstack-ansible | 19:05 | |
| *** KLevenstein has quit IRC | 19:05 | |
| *** bapalm_ has joined #openstack-ansible | 19:08 | |
| *** KLevenstein has joined #openstack-ansible | 19:09 | |
| *** bapalm has quit IRC | 19:09 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add swift service region to group vars https://review.openstack.org/209326 | 19:17 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Re-order setup-openstack play to match the gate check order https://review.openstack.org/209071 | 19:17 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Allow Swift middleware to be set via a variable https://review.openstack.org/210297 | 19:17 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add resume_guests_state_on_host_boot to nova.conf https://review.openstack.org/211600 | 19:19 |
| *** bapalm_ has quit IRC | 19:24 | |
| *** bapalm has joined #openstack-ansible | 19:24 | |
| *** alejandrito has joined #openstack-ansible | 19:24 | |
| *** bapalm has quit IRC | 19:25 | |
| *** bapalm has joined #openstack-ansible | 19:26 | |
| openstackgerrit | Christopher H. Laco proposed stackforge/os-ansible-deployment: Add resume_guests_state_on_host_boot to nova.conf https://review.openstack.org/211721 | 19:30 |
| *** bapalm has quit IRC | 19:31 | |
| *** bapalm has joined #openstack-ansible | 19:31 | |
| *** bapalm has quit IRC | 19:33 | |
| *** bapalm has joined #openstack-ansible | 19:34 | |
| *** bapalm has quit IRC | 19:38 | |
| *** phalmos has quit IRC | 19:40 | |
| *** bapalm has joined #openstack-ansible | 19:41 | |
| *** jwagner is now known as jwagner_away | 19:48 | |
| *** bapalm has quit IRC | 19:51 | |
| *** bapalm has joined #openstack-ansible | 19:51 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add resume_guests_state_on_host_boot to nova.conf https://review.openstack.org/211721 | 19:53 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Move ansible logging to gate-check script only https://review.openstack.org/207502 | 19:54 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add swift service region to group vars https://review.openstack.org/211729 | 19:55 |
| *** bapalm has quit IRC | 19:55 | |
| *** bapalm has joined #openstack-ansible | 19:55 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable debug logging for gate checks https://review.openstack.org/210278 | 19:56 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Move ansible logging to gate-check script only https://review.openstack.org/211730 | 19:57 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Re-order setup-openstack play to match the gate check order https://review.openstack.org/211732 | 20:00 |
| *** bapalm has quit IRC | 20:01 | |
| *** bapalm has joined #openstack-ansible | 20:01 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Document some of the common scripts https://review.openstack.org/207105 | 20:03 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Adds ceilometer link in configure.rst https://review.openstack.org/207576 | 20:04 |
| *** bapalm_ has joined #openstack-ansible | 20:04 | |
| *** bapalm has quit IRC | 20:05 | |
| *** javeriak has quit IRC | 20:10 | |
| *** fawadkhaliq has quit IRC | 20:10 | |
| openstackgerrit | Christopher H. Laco proposed stackforge/os-ansible-deployment: Add resume_guests_state_on_host_boot to nova.conf https://review.openstack.org/211737 | 20:18 |
| *** bapalm_ has quit IRC | 20:25 | |
| *** bapalm has joined #openstack-ansible | 20:25 | |
| *** phalmos has joined #openstack-ansible | 20:26 | |
| *** bapalm has quit IRC | 20:30 | |
| *** bapalm has joined #openstack-ansible | 20:31 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Move ansible logging to gate-check script only https://review.openstack.org/211730 | 20:34 |
| *** bapalm has quit IRC | 20:35 | |
| *** yaya has quit IRC | 20:38 | |
| openstackgerrit | Erik Wilson proposed stackforge/os-ansible-deployment: Archive Keystone to Keystone Federation rst content https://review.openstack.org/211747 | 20:39 |
| *** yaya has joined #openstack-ansible | 20:53 | |
| errr | eglute: hi | 21:08 |
| * eglute waves to errr | 21:08 | |
| *** pradk has quit IRC | 21:16 | |
| *** wmlynch has quit IRC | 21:18 | |
| *** phalmos has quit IRC | 21:19 | |
| *** phalmos has joined #openstack-ansible | 21:21 | |
| *** yaya has quit IRC | 21:24 | |
| *** tlian has quit IRC | 21:24 | |
| *** woodard has quit IRC | 21:37 | |
| *** alejandrito has quit IRC | 21:41 | |
| *** bapalm has joined #openstack-ansible | 21:42 | |
| *** jwagner_away is now known as jwagner | 21:45 | |
| openstackgerrit | Erik Wilson proposed stackforge/os-ansible-deployment: Archive Keystone to Keystone Federation rst content https://review.openstack.org/211747 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Uses tox for automating documentation builds https://review.openstack.org/207577 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Removes tab indents for bashate https://review.openstack.org/207664 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Fixes deprecated arithmetic expansion for bashate https://review.openstack.org/207665 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Fixes function declarations for bashate https://review.openstack.org/207666 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Fixes loops for bashate https://review.openstack.org/207667 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Removes trailing whitespace for bashate https://review.openstack.org/207663 | 21:46 |
| openstackgerrit | David Stanek proposed stackforge/os-ansible-deployment: Adds a pep8 target to tox.ini https://review.openstack.org/207662 | 21:46 |
| *** bapalm has quit IRC | 21:49 | |
| *** bapalm has joined #openstack-ansible | 21:49 | |
| *** Mudpuppy has quit IRC | 21:50 | |
| *** bapalm has quit IRC | 21:56 | |
| *** bapalm has joined #openstack-ansible | 21:57 | |
| openstackgerrit | Erik Wilson proposed stackforge/os-ansible-deployment: Archive Keystone to Keystone Federation rst content https://review.openstack.org/211747 | 21:58 |
| openstackgerrit | Erik Wilson proposed stackforge/os-ansible-deployment: Archive Keystone to Keystone Federation rst content https://review.openstack.org/211747 | 21:58 |
| *** jwagner is now known as jwagner_away | 21:58 | |
| *** daneyon_ has left #openstack-ansible | 21:59 | |
| *** bapalm has quit IRC | 22:11 | |
| *** bapalm has joined #openstack-ansible | 22:11 | |
| *** bapalm has quit IRC | 22:16 | |
| *** KLevenstein has quit IRC | 22:24 | |
| *** bapalm has joined #openstack-ansible | 22:24 | |
| *** phalmos has quit IRC | 22:34 | |
| *** spotz is now known as spotz_zzz | 22:35 | |
| *** bapalm has quit IRC | 22:42 | |
| *** bapalm has joined #openstack-ansible | 22:42 | |
| *** bapalm has quit IRC | 22:47 | |
| *** jaypipes has quit IRC | 22:50 | |
| *** abitha has quit IRC | 22:50 | |
| dstanek | i think your Jenkins hates me | 23:02 |
| dstanek | or else he's just really mean | 23:02 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:15 | |
| openstackgerrit | Yoshitaka Tahara proposed stackforge/os-ansible-deployment: Invalid link "deployment-os.html" in Install Guide https://review.openstack.org/211186 | 23:49 |
| *** alop_ has joined #openstack-ansible | 23:56 | |
| *** alop has quit IRC | 23:56 | |
| *** alop_ is now known as alop | 23:56 | |
| *** alejandrito has joined #openstack-ansible | 23:57 | |
| openstackgerrit | Yoshitaka Tahara proposed stackforge/os-ansible-deployment: Invalid link "deployment-os.html" in Install Guide https://review.openstack.org/211186 | 23:57 |
| *** alop has quit IRC | 23:58 | |
| *** BjoernT has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!