| *** thorst has joined #openstack-ansible | 00:40 | |
| *** thorst has quit IRC | 00:47 | |
| *** thorst has joined #openstack-ansible | 00:47 | |
| *** thorst has quit IRC | 00:47 | |
| *** markvoelker has joined #openstack-ansible | 00:53 | |
| *** thorst has joined #openstack-ansible | 00:54 | |
| csmart | mrda: \o | 00:54 |
|---|---|---|
| *** spotz_zzz is now known as spotz | 01:03 | |
| *** thorst has quit IRC | 01:05 | |
| *** matias has quit IRC | 01:11 | |
| *** strobelight has quit IRC | 01:16 | |
| *** severion has joined #openstack-ansible | 01:16 | |
| *** v1k0d3n has quit IRC | 01:16 | |
| *** v1k0d3n has joined #openstack-ansible | 01:26 | |
| *** severion has quit IRC | 01:26 | |
| mrda | hey csmart | 01:33 |
| mhayden | ohai csmart | 01:37 |
| csmart | mhayden morning! | 01:37 |
| *** raddaoui has quit IRC | 01:43 | |
| *** raddaoui has joined #openstack-ansible | 01:44 | |
| *** weezS has joined #openstack-ansible | 01:57 | |
| *** thorst has joined #openstack-ansible | 02:05 | |
| *** thorst has quit IRC | 02:13 | |
| *** kenhui has joined #openstack-ansible | 02:13 | |
| *** agireud has quit IRC | 02:18 | |
| *** agireud has joined #openstack-ansible | 02:25 | |
| openstackgerrit | Darren Chan proposed openstack/openstack-ansible: [docs] Minor edits to the overview chapter https://review.openstack.org/365454 | 02:29 |
| *** agireud has quit IRC | 02:30 | |
| *** kenhui has quit IRC | 02:30 | |
| *** thorst has joined #openstack-ansible | 02:40 | |
| *** agireud has joined #openstack-ansible | 02:40 | |
| *** dkehn_ has joined #openstack-ansible | 02:44 | |
| *** jperry has joined #openstack-ansible | 02:47 | |
| *** markvoelker has quit IRC | 02:53 | |
| openstackgerrit | Drew Thorstensen (thorst) proposed openstack/openstack-ansible-os_neutron: [DOC] Fix the OVS config file in docs https://review.openstack.org/362848 | 03:09 |
| *** janki has joined #openstack-ansible | 03:09 | |
| *** weezS has quit IRC | 03:15 | |
| *** weezS has joined #openstack-ansible | 03:27 | |
| *** weezS has quit IRC | 03:40 | |
| *** dkehn_ has quit IRC | 03:49 | |
| *** dkehn_ has joined #openstack-ansible | 03:51 | |
| *** janki has quit IRC | 03:51 | |
| *** thorst has quit IRC | 03:58 | |
| *** thorst has joined #openstack-ansible | 03:58 | |
| *** thorst has quit IRC | 04:00 | |
| *** thorst has joined #openstack-ansible | 04:03 | |
| *** sdake has joined #openstack-ansible | 04:05 | |
| *** v1k0d3n has quit IRC | 04:06 | |
| *** v1k0d3n has joined #openstack-ansible | 04:06 | |
| *** thorst has quit IRC | 04:08 | |
| *** v1k0d3n has quit IRC | 04:09 | |
| *** v1k0d3n has joined #openstack-ansible | 04:10 | |
| *** thorst has joined #openstack-ansible | 04:10 | |
| *** thorst has quit IRC | 04:15 | |
| *** v1k0d3n has quit IRC | 04:17 | |
| *** sdake has quit IRC | 04:25 | |
| *** sdake has joined #openstack-ansible | 04:50 | |
| *** thorst has joined #openstack-ansible | 04:55 | |
| *** thorst has quit IRC | 04:59 | |
| *** sdake has quit IRC | 05:01 | |
| *** thorst has joined #openstack-ansible | 05:08 | |
| *** sdake has joined #openstack-ansible | 05:10 | |
| *** sdake has quit IRC | 05:17 | |
| *** McMurlock1 has joined #openstack-ansible | 05:18 | |
| *** spotz is now known as spotz_zzz | 05:19 | |
| *** schwicht has joined #openstack-ansible | 05:21 | |
| *** schwicht_at_work has quit IRC | 05:21 | |
| *** thorst has quit IRC | 05:25 | |
| *** thorst has joined #openstack-ansible | 05:29 | |
| *** thorst has quit IRC | 05:33 | |
| *** janki has joined #openstack-ansible | 05:45 | |
| *** McMurlock1 has quit IRC | 05:52 | |
| *** jperry has quit IRC | 05:55 | |
| *** bsv has joined #openstack-ansible | 06:21 | |
| *** berendt has joined #openstack-ansible | 06:40 | |
| *** pcaruana has joined #openstack-ansible | 06:48 | |
| *** markvoelker has joined #openstack-ansible | 06:54 | |
| *** dkehn_ has quit IRC | 06:56 | |
| *** fxpester has joined #openstack-ansible | 06:56 | |
| *** markvoelker has quit IRC | 06:59 | |
| *** dkehn_ has joined #openstack-ansible | 07:01 | |
| *** v1k0d3n has joined #openstack-ansible | 07:01 | |
| *** v1k0d3n has quit IRC | 07:06 | |
| *** chrichip has quit IRC | 07:25 | |
| *** drifterza has joined #openstack-ansible | 07:26 | |
| drifterza | Ehlo all | 07:27 |
| *** jperry has joined #openstack-ansible | 07:35 | |
| *** neilus has joined #openstack-ansible | 07:37 | |
| *** asettle has joined #openstack-ansible | 08:11 | |
| odyssey4me | o/ | 08:15 |
| drifterza | hows things odyssey4me ? | 08:16 |
| *** karimb has joined #openstack-ansible | 08:17 | |
| odyssey4me | drifterza not too bad, working on docs things which I have neglected to do up until now | 08:17 |
| odyssey4me | what're you up to drifterza ? | 08:17 |
| drifterza | getting our lab sorted again, have to have 3+ AIO setups for testing. | 08:18 |
| *** markvoelker has joined #openstack-ansible | 08:19 | |
| *** neilus has quit IRC | 08:22 | |
| drifterza | I hear you tho about docs, I'm really poor in that area myself. | 08:22 |
| *** neilus has joined #openstack-ansible | 08:23 | |
| *** markvoelker has quit IRC | 08:24 | |
| csmart | odyssey4me drifterza \o | 08:27 |
| drifterza | o/ | 08:27 |
| *** Jeffrey4l has quit IRC | 08:28 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: Add a doc example for yaml file overrides https://review.openstack.org/364502 | 08:37 |
| *** evrardjp has quit IRC | 08:42 | |
| *** evrardjp has joined #openstack-ansible | 08:42 | |
| *** hblah has quit IRC | 08:43 | |
| *** karimb has quit IRC | 08:56 | |
| *** karimb has joined #openstack-ansible | 09:00 | |
| *** shausy has joined #openstack-ansible | 09:09 | |
| *** bsv has quit IRC | 09:38 | |
| *** sdake has joined #openstack-ansible | 09:45 | |
| *** v1k0d3n has joined #openstack-ansible | 09:57 | |
| *** v1k0d3n has quit IRC | 10:01 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Add Multi Region Swift gate scenario https://review.openstack.org/356680 | 10:07 |
| *** asettle has quit IRC | 10:26 | |
| *** karimb has quit IRC | 10:27 | |
| *** Oku_OS is now known as Oku_OS-away | 10:30 | |
| *** McMurlock1 has joined #openstack-ansible | 10:30 | |
| *** schwicht has quit IRC | 10:30 | |
| *** schwicht has joined #openstack-ansible | 10:31 | |
| *** Oku_OS-away is now known as Oku_OS | 10:31 | |
| *** karimb has joined #openstack-ansible | 10:33 | |
| *** sdake has quit IRC | 10:33 | |
| *** jperry has quit IRC | 10:39 | |
| *** askb has quit IRC | 10:40 | |
| *** johnmilton has quit IRC | 10:43 | |
| *** johnmilton has joined #openstack-ansible | 10:57 | |
| *** karimb has quit IRC | 10:58 | |
| *** karimb has joined #openstack-ansible | 11:04 | |
| *** Jeffrey4l has joined #openstack-ansible | 11:05 | |
| *** jcannava has quit IRC | 11:11 | |
| *** jcannava has joined #openstack-ansible | 11:12 | |
| *** jcannava has quit IRC | 11:15 | |
| *** eon` has joined #openstack-ansible | 11:21 | |
| *** McMurlock1 has quit IRC | 11:21 | |
| *** karimb has quit IRC | 11:21 | |
| *** dkehn_ has quit IRC | 11:23 | |
| eon` | hello, I'm trying a AIO setup but it seems IPs are not correctly assigned to my containers (http://paste.debian.net/811319/). Any hints ? thanks | 11:25 |
| *** schwicht has quit IRC | 11:26 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Revise Storage Architecture Overview https://review.openstack.org/365711 | 11:31 |
| *** spotz_zzz is now known as spotz | 11:32 | |
| odyssey4me | eon` is that a build using the master branch? if so is that from the head of master? | 11:32 |
| odyssey4me | oh I see it's from the head of stable/mitaka | 11:33 |
| odyssey4me | did you make any adjustments to the AIO configuration in any way after doing the bootstrap? | 11:33 |
| eon` | odyssey4me: yes I've removed some services from the env.d directory, disabled security hardening, pretty much all | 11:35 |
| *** dkehn_ has joined #openstack-ansible | 11:35 | |
| odyssey4me | eon` ok, first of all the env.d directory contains group mappings... so removing things from there will break stuff unless you do it corectly | 11:36 |
| *** asettle has joined #openstack-ansible | 11:37 | |
| odyssey4me | eon` what you've changed in there has clearly broken things, so you'll have to work back to figure out how to fix it up | 11:37 |
| odyssey4me | what're you trying to achieve? | 11:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Revise Storage Architecture Overview https://review.openstack.org/365711 | 11:38 |
| eon` | odyssey4me: I wanted to not install services I do not need like horizon or heat | 11:38 |
| *** retreved has joined #openstack-ansible | 11:40 | |
| odyssey4me | eon` seeing as those are part of the os-infra_hosts group, to remove those you can either use the affinity ability documented in http://docs.openstack.org/developer/openstack-ansible/mitaka/install-guide/configure-initial.html#affinity | 11:40 |
| *** v1k0d3n has joined #openstack-ansible | 11:41 | |
| odyssey4me | or you can change the openstack_user_config.yml to not use the os-infra_hosts group and instead to use the particular service host group names | 11:42 |
| odyssey4me | or you could remove/edit some of the env.d files... but you'll need to be careful not to remove core files | 11:42 |
| *** strobelight has joined #openstack-ansible | 11:43 | |
| *** v1k0d3n has quit IRC | 11:46 | |
| eon` | odyssey4me: ok, thanks, I'll look into it | 11:47 |
| *** asettle has quit IRC | 11:52 | |
| *** asettle has joined #openstack-ansible | 11:58 | |
| *** Matias has joined #openstack-ansible | 11:59 | |
| andymccr | ping lbragstad whenever you get in! | 12:01 |
| *** jperry has joined #openstack-ansible | 12:01 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Add Multi Region Swift gate scenario https://review.openstack.org/356680 | 12:07 |
| *** spotz is now known as spotz_zzz | 12:09 | |
| *** schwicht has joined #openstack-ansible | 12:11 | |
| *** karimb has joined #openstack-ansible | 12:14 | |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: [DOCS] IA movement of the install guide https://review.openstack.org/366088 | 12:18 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: [WIP] Add tempest to keystone role tests https://review.openstack.org/365716 | 12:21 |
| *** neillc has quit IRC | 12:21 | |
| *** schwicht has quit IRC | 12:21 | |
| mhayden | morning | 12:21 |
| *** neillc has joined #openstack-ansible | 12:22 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 12:23 |
| Matias | morning | 12:23 |
| *** psilvad has joined #openstack-ansible | 12:24 | |
| Matias | there seems to be no simple way of filling filter_project for dispatcher_gnocchi automatically in the os_ceilometer role (http://docs.openstack.org/developer/ceilometer/install/manual.html) | 12:24 |
| Matias | perhaps the solution by now is to use ceilometer_ceilometer_conf_overrides? | 12:24 |
| Matias | (this when using swift for storage) | 12:25 |
| odyssey4me | Matias yes, anything that's not directly filfilled by vars via ansible can be facilitated using conf overrides | 12:25 |
| *** schwicht has joined #openstack-ansible | 12:26 | |
| Matias | maybe it would be nice to document this? I don't know how much I/O and CPU it wastes if this filter is not set, though | 12:28 |
| *** thorst has joined #openstack-ansible | 12:30 | |
| odyssey4me | Matias that may be something that stevelle was still working on? not sure | 12:36 |
| stevelle | Matias: best is to filter it in the ceilometer middleware | 12:38 |
| openstackgerrit | Merged openstack/openstack-ansible: [DOCS] IA movement of the install guide https://review.openstack.org/366088 | 12:38 |
| stevelle | Matias: this specifically: https://github.com/openstack/ceilometermiddleware/blob/master/ceilometermiddleware/swift.py#L39 | 12:39 |
| stevelle | I'll make a note to get that set up properly with group vars | 12:40 |
| andymccr | stevelle: unrelated note, but on the nginx work for keystone: https://review.openstack.org/#/c/365708/ i am getting errors when using openstack client (or in swift testing with nginx keystone) | 12:40 |
| andymccr | although the tempest tests seem to pass - was wondering your thoughts | 12:40 |
| *** klamath has joined #openstack-ansible | 12:42 | |
| *** woodard has joined #openstack-ansible | 12:43 | |
| *** dmsimard|pto is now known as dmsimard | 12:43 | |
| stevelle | andymccr: I hadn't seen that yet. Wondering if it's some header issue? | 12:43 |
| andymccr | perhaps | 12:43 |
| *** klamath has quit IRC | 12:43 | |
| stevelle | might need to snarf traffic to work that out | 12:43 |
| *** klamath has joined #openstack-ansible | 12:43 | |
| andymccr | stevelle: http://paste.openstack.org/show/567148/ you can see the href | 12:44 |
| stevelle | oh wow, thats wrong | 12:44 |
| andymccr | yeh hahaha | 12:45 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 12:45 |
| stevelle | andymccr: do you get that if you hit the uwsgi port direct? | 12:47 |
| andymccr | stevelle: im gonna guess no - since i can fix that up by changing the "SCRIPT_NAME" var in nginx | 12:47 |
| stevelle | seems most likely a uwsgi issue, but odd I didn't notice that before | 12:47 |
| stevelle | ok | 12:48 |
| *** furlongm_ is now known as furlongm | 12:48 | |
| odyssey4me | andymccr stevelle perhaps dstanek can help get it right? | 12:48 |
| *** v1k0d3n has joined #openstack-ansible | 12:48 | |
| stevelle | thought I didn't need a SCRIPT_NAME originally. | 12:48 |
| andymccr | stevelle: i think it complained about it when i was running it - so i had to add one | 12:49 |
| andymccr | i find it weird that tempest tests pass in that state :P but hey | 12:49 |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: [DOCS] Renaming sections for install guide https://review.openstack.org/366100 | 12:49 |
| *** pcaruana has quit IRC | 12:50 | |
| stevelle | andymccr: if you throw in a fix I'll look it over in about 4 hours or so :) | 12:50 |
| andymccr | stevelle: already done (the one i linked you) but i wasnt too sure if is the correct approach - i found a guide on nginx + uwsgi for keystone and it uses a script_name. | 12:51 |
| andymccr | but i guess if it works, it works! so lets get rid of it. | 12:51 |
| *** schwicht has quit IRC | 12:54 | |
| automagically | Morning all | 12:55 |
| *** psilvad has quit IRC | 12:58 | |
| Matias | stevelle: thanks :D meanwhile i will test overriding filter_service_activity and filter_project | 13:01 |
| odyssey4me | o/ automagically | 13:01 |
| Matias | stevelle: what about ceph support in gnocchi? that's only for octavia? | 13:01 |
| *** javis has quit IRC | 13:02 | |
| stevelle | Matias: set gnocchi_storage_driver and use overrides and set the ceph vars, manageing ceph is external | 13:03 |
| stevelle | there really is nothing I can do to integrate it | 13:03 |
| Matias | stevelle: I tried it, but the ceph_client role is missing | 13:04 |
| Matias | to setup the rados lib etc. | 13:04 |
| stevelle | you can override gnocchi_distro_packages | 13:04 |
| openstackgerrit | Merged openstack/openstack-ansible-ops: Removed file name which should not have been included https://review.openstack.org/364595 | 13:05 |
| stevelle | I have a bug open around our requirements management that is really hard to solve, and didn't want to require the libs for that by default | 13:05 |
| stevelle | or you could jiggle with inventory to apply the ceph_client role there? | 13:06 |
| stevelle | I don't have a good solution for that yet tbh | 13:06 |
| *** kenhui has joined #openstack-ansible | 13:08 | |
| *** schwicht has joined #openstack-ansible | 13:12 | |
| odyssey4me | stevelle yeah, I think we're going to have to review many of the os_ roles and try and apply the dict based selection model for settings that we've been using for nova/neutron | 13:12 |
| odyssey4me | ie you change a var, and a whole bunch of settings pivot from that | 13:12 |
| odyssey4me | an example is when you're using cinder-backup with ceph... right now a ton of vars need to be set, whereas we could pre-set most of them in a dict and just have one pivot var | 13:13 |
| odyssey4me | the whole way ceph is currently integrated could probably do with an overhaul - it was done in kilo | 13:14 |
| stevelle | odyssey4me: I believe for ceph and gnocchi it's just {pool, user, key, conf file} | 13:14 |
| stevelle | the harder part is the rados lib etc, | 13:15 |
| openstackgerrit | Alexandra Settle proposed openstack/openstack-ansible: [DOCS] Renaming sections for install guide https://review.openstack.org/366100 | 13:15 |
| odyssey4me | stevelle that's pretty common for most roles, but our current integration for ceph into cinder, nova, etc is very messy and could be radically simplified I think | 13:15 |
| Matias | yep | 13:15 |
| *** jperry has quit IRC | 13:15 | |
| odyssey4me | the rados lib itself is easy enough - we have a pattern already which just needs to be applied | 13:15 |
| stevelle | insert image macro about openstack | 13:16 |
| odyssey4me | :) | 13:16 |
| odyssey4me | matias you are welcome to submit a patch if you want to, but there are no guarantees that it will make it into Newton before release | 13:16 |
| odyssey4me | but the sooner the patch is in, the better :) | 13:16 |
| *** npawelek has joined #openstack-ansible | 13:17 | |
| Matias | odyssey4me: i think i'm more on the safe side using swift by now :P | 13:19 |
| odyssey4me | it would be marvellous if our consumers of ceph could put together a spec to overhaul the ceph integration and perhaps even include playbooks for deploying ceph. Ideally we should get that in to try and get it done by Ocata 2. | 13:20 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Adding V-38438 (auditd during boot) https://review.openstack.org/354363 | 13:21 |
| Matias | integrating ceph-ansible is really easy, we only did not integrate it with OSA inventory because we already have a separate cobbler inventory for doing the metal prerequisites / networking setup | 13:22 |
| Matias | so we just added some mgmtclasses in cobbler for the MONs and OSDs and used to cobbler inventory to feed ceph-ansible | 13:23 |
| odyssey4me | Matias neat :) would love to see a blog post or something about how you did it :) | 13:23 |
| Matias | s/used to/used the/ | 13:23 |
| Matias | nice, i'll try to write something up after newton is released :D | 13:24 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Only configure systat on debian based system. https://review.openstack.org/364903 | 13:26 |
| *** woodard has quit IRC | 13:28 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Update librabbit package for novnc in 16.04 https://review.openstack.org/364975 | 13:30 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_swift: Add Multi Region Swift gate scenario https://review.openstack.org/356680 | 13:31 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create: Check if lxc_container_backing_method is defined https://review.openstack.org/365119 | 13:31 |
| *** jmckind has joined #openstack-ansible | 13:33 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests: Only override UCA repo url on OpenStack CI https://review.openstack.org/364525 | 13:34 |
| andymccr | race to rebase on my 2 keystone patches nooooo | 13:34 |
| *** ametts has joined #openstack-ansible | 13:36 | |
| *** markvoelker has joined #openstack-ansible | 13:36 | |
| asettle | mhayden odyssey4me and automagically - email out for testing. | 13:38 |
| *** jperry has joined #openstack-ansible | 13:41 | |
| odyssey4me | asettle great, thanks | 13:41 |
| mhayden | 4OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOORLY | 13:41 |
| cloudnull | mornings | 13:42 |
| odyssey4me | mhayden automagically I'd appreciate it if you could read through my ramblings to see if you think it makes sense and is a good fit: https://review.openstack.org/365711 | 13:42 |
| * asettle reads | 13:43 | |
| asettle | (again) | 13:43 |
| asettle | mhayden: ya rly | 13:44 |
| asettle | Get ready for this jelly | 13:44 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Use the central test repository for Keystone https://review.openstack.org/364796 | 13:44 |
| *** shausy has quit IRC | 13:47 | |
| asettle | odyssey4me: it just clicked. You're the offender who always uses 'which' :p | 13:47 |
| *** aludwar has joined #openstack-ansible | 13:48 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Add config template for qemu.conf https://review.openstack.org/365095 | 13:49 |
| odyssey4me | which offense is that which does not please you? | 13:49 |
| * asettle inwardly screams | 13:49 | |
| odyssey4me | :) | 13:49 |
| odyssey4me | get ready for more of it :p | 13:49 |
| asettle | odyssey4me: yeah got a few more edits on your patch :) | 13:51 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 13:51 |
| *** johnmilton has quit IRC | 13:52 | |
| *** jperry has quit IRC | 13:55 | |
| *** jperry has joined #openstack-ansible | 13:56 | |
| *** javis has joined #openstack-ansible | 13:58 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-openstack_hosts: Fix module loading on centos hosts. https://review.openstack.org/362881 | 13:59 |
| mrhillsman | g'morning | 14:00 |
| *** ametts has quit IRC | 14:03 | |
| *** johnmilton has joined #openstack-ansible | 14:05 | |
| *** thorst has quit IRC | 14:05 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Revise Storage Architecture Overview https://review.openstack.org/365711 | 14:05 |
| *** thorst has joined #openstack-ansible | 14:05 | |
| *** ametts has joined #openstack-ansible | 14:06 | |
| odyssey4me | asettle ^ fixed up - one sentence restructure in there | 14:07 |
| evrardjp | cloudnull: do you know if there is an ansible variable I can use that holds the value of a inventory_hostname of a delegated node? | 14:07 |
| evrardjp | this way I can use something like hostvars[delegated_node]['blablablafact'] | 14:08 |
| *** retreved_ has joined #openstack-ansible | 14:08 | |
| *** retreved_ has quit IRC | 14:08 | |
| *** retreved_ has joined #openstack-ansible | 14:09 | |
| asettle | odyssey4me: on it :) | 14:09 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 14:11 |
| *** retreved has quit IRC | 14:12 | |
| cloudnull | evrardjp: no. sadly . | 14:12 |
| cloudnull | delegate_to is a task var | 14:12 |
| cloudnull | and its only present when the task gets put through the work queue | 14:12 |
| lbragstad | andymccr o/ | 14:12 |
| *** thorst has quit IRC | 14:14 | |
| andymccr | lbragstad: hey! I added tempest api tests for keystone in OSA (you can see the PR above actually), but 2 sets of test failed - ec2_credentials (which i kind of expect) but also the 2 v3 credentials checks: https://github.com/openstack/tempest/blob/master/tempest/api/identity/admin/v3/test_credentials.py | 14:14 |
| andymccr | do you reckon its all good if we skip those ones, or do you think we should fix that up within OSA | 14:15 |
| andymccr | or perhaps it works already and im not setting the appropriate setting | 14:15 |
| lbragstad | andymccr we *just* merged a patch that encrypted credentials at rest | 14:15 |
| lbragstad | andymccr what are the tests failing with/ | 14:16 |
| andymccr | 500 | 14:16 |
| evrardjp | cloudnull: makes sense | 14:16 |
| evrardjp | thanks | 14:16 |
| lbragstad | andymccr they are failing in your patch, right? | 14:16 |
| andymccr | lbragstad: nope im excluding them for now | 14:16 |
| *** hj-hpe has joined #openstack-ansible | 14:16 | |
| *** michauds has joined #openstack-ansible | 14:17 | |
| andymccr | lbragstad: gimme 2 seconds i have an instance up i was playing around with - i'll get an output for you. | 14:18 |
| lbragstad | andymccr for Newton and onward, in order to use the credential API, you'll need to run `keystone-manage credential_setup` | 14:18 |
| lbragstad | which will do pretty much the same thing as `keystone-manage fernet_setup` but to a different place on disk - and those keys are used to encrypt credentials before writing them to the backend | 14:19 |
| *** scarlisle has joined #openstack-ansible | 14:20 | |
| andymccr | lbragstad: do you have to do that per user? | 14:22 |
| lbragstad | andymccr nope it's per keystone node | 14:23 |
| andymccr | ahh so is keystone-user keystone-group the on disk perms? | 14:23 |
| lbragstad | once your setup your key repository - it can be copied to the rest of the keystone nodes in your deployment (this they all need to be able to encrypt and decrypt credentials) | 14:23 |
| lbragstad | andymccr yep | 14:23 |
| lbragstad | exactly | 14:23 |
| andymccr | oh boom | 14:24 |
| andymccr | thanks lbragstad! | 14:24 |
| *** Mudpuppy has joined #openstack-ansible | 14:24 | |
| *** ametts has quit IRC | 14:24 | |
| lbragstad | andymccr does that make more sense? | 14:24 |
| andymccr | lbragstad: another question - is this something we should do on all deploys? or is this specific to certain situations | 14:24 |
| andymccr | or something we wouldnt want to do in all cases | 14:24 |
| *** BjoernT has joined #openstack-ansible | 14:25 | |
| *** spotz_zzz is now known as spotz | 14:25 | |
| lbragstad | andymccr it won't bother deployments that don't use the credential api | 14:25 |
| lbragstad | but for everything Newton on, if the credential API is used and a key repository doesn't exist, keystone will 500 | 14:25 |
| *** janki has quit IRC | 14:26 | |
| lbragstad | we're pretty much making it so that if you're going to put secret stuff in keystone, it has to be encrypted | 14:26 |
| andymccr | makes sense - perhaps we should do that by default on all deploys from now on. additionally do you need to setup the [credential] stanza in keystone.conf or no? | 14:27 |
| lbragstad | andymccr good question | 14:27 |
| andymccr | i tested without and it worked | 14:28 |
| lbragstad | andymccr there is only one config options for the feature and that's `CONF [credential] key_repository` | 14:28 |
| lbragstad | and it defaults to `/etc/keystone/credential_keys` | 14:28 |
| lbragstad | andymccr we removed the ability to specify `max_active_keys` for credential encryption | 14:29 |
| *** ametts has joined #openstack-ansible | 14:29 | |
| lbragstad | it is hardcoded to 3 in keystone, since there really isn't a need to have more than that | 14:29 |
| andymccr | lbragstad: ok cool so just that one option. or maybe i just add the stanza so you can use conf_overrides | 14:30 |
| lbragstad | andymccr yeah - I documented a rolling upgrade (with limited service downtime) from Mitaka to Newton here - https://gist.github.com/lbragstad/ddfb10f9f9048414d1f781ba006e95d1#file-migration-md | 14:31 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops: implement minimal metric collection https://review.openstack.org/366167 | 14:31 |
| cloudnull | mrhillsman: ^ | 14:31 |
| cloudnull | https://cloud1.osic.org:8443/grafana/ <- that repo is active now. | 14:32 |
| cloudnull | the dash is not complete and there are a few checks i need to add to telegraf but its a WIP and is working quite well | 14:34 |
| *** jamesdenton has joined #openstack-ansible | 14:35 | |
| *** jcannava has joined #openstack-ansible | 14:35 | |
| *** neilus has quit IRC | 14:36 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 14:37 |
| openstackgerrit | Merged openstack/openstack-ansible: Fixed assumed utility pip install for specific clients https://review.openstack.org/364849 | 14:38 |
| openstackgerrit | Merged openstack/openstack-ansible: [DOCS] Renaming sections for install guide https://review.openstack.org/366100 | 14:38 |
| *** galstrom_zzz is now known as galstrom | 14:41 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: [WIP] Disable ml2 ports registered with new agent hostnames https://review.openstack.org/366169 | 14:44 |
| *** sdake has joined #openstack-ansible | 14:44 | |
| *** kjw3 has joined #openstack-ansible | 14:45 | |
| evrardjp | dear cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, | 14:46 |
| evrardjp | spotz, vdo, jmccrory, alextricity25, jasondotstar, KLevenstein, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397 : | 14:46 |
| evrardjp | the osa bug triage will start in 1h15’. Please have a look at the bug list before starting: https://etherpad.openstack.org/p/osa-bugtriage | 14:46 |
| evrardjp | asettle: you're gonna have fun :) | 14:47 |
| mhayden | asettle: i'm so ready for that jelly | 14:47 |
| asettle | I might just low key triage those now :/ hahahaha | 14:47 |
| asettle | evrardjp: I'm not going to triage yours because I actually want a discussion around that. | 14:48 |
| asettle | Specifically teh linking | 14:48 |
| cloudnull | o/ | 14:49 |
| cloudnull | oh. an hour :) | 14:49 |
| *** admin0 has joined #openstack-ansible | 14:50 | |
| *** ddaskal has joined #openstack-ansible | 14:52 | |
| *** TxGirlGeek has joined #openstack-ansible | 14:55 | |
| *** michaelgugino has joined #openstack-ansible | 14:56 | |
| *** dkehn_ has quit IRC | 14:57 | |
| *** KLevenstein has joined #openstack-ansible | 14:59 | |
| *** spedione|AWAY is now known as spedione | 15:05 | |
| *** karimb has quit IRC | 15:07 | |
| *** dkehn_ has joined #openstack-ansible | 15:10 | |
| *** gouthamr has joined #openstack-ansible | 15:10 | |
| *** thorst has joined #openstack-ansible | 15:11 | |
| *** gouthamr_ has joined #openstack-ansible | 15:12 | |
| *** pester has joined #openstack-ansible | 15:13 | |
| *** janki has joined #openstack-ansible | 15:13 | |
| *** fxpester has quit IRC | 15:15 | |
| *** gouthamr has quit IRC | 15:15 | |
| *** weezS has joined #openstack-ansible | 15:17 | |
| openstackgerrit | Merged openstack/openstack-ansible: [docs] Revise Storage Architecture Overview https://review.openstack.org/365711 | 15:18 |
| *** thorst has quit IRC | 15:19 | |
| openstackgerrit | Nish Patwa(IRC: nishpatwa) proposed openstack/openstack-ansible: [DOC]Added Xenial support in install guide https://review.openstack.org/364572 | 15:23 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Disable martian logging by default https://review.openstack.org/366203 | 15:25 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_keystone: Add tempest to keystone role tests https://review.openstack.org/365716 | 15:25 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Disable martian logging by default https://review.openstack.org/366204 | 15:25 |
| openstackgerrit | Nish Patwa(IRC: nishpatwa) proposed openstack/openstack-ansible: [DOC]Added Xenial support in install guide https://review.openstack.org/364572 | 15:25 |
| *** FrankZha- is now known as FrankZhang | 15:27 | |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to highlight changes between stock env.d and /etc/openstack_deploy/env.d for Mitaka -> Newton upgrade https://review.openstack.org/366208 | 15:29 |
| *** thorst has joined #openstack-ansible | 15:30 | |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to highlight changes between stock env.d and /etc/openstack_deploy/env.d for Mitaka -> Newton upgrade https://review.openstack.org/366208 | 15:31 |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 15:33 |
| mhayden | odyssey4me: the functional testing fell off of openstack-ansible-security -> https://review.openstack.org/#/c/366203/ | 15:34 |
| odyssey4me | mhayden eh? | 15:35 |
| *** Jeffrey4l has quit IRC | 15:35 | |
| palendae | How'd it fall off? | 15:35 |
| mhayden | odyssey4me: i'm looking through projects.yaml to see if something changed | 15:35 |
| odyssey4me | mhayden oh - I see | 15:36 |
| mhayden | it appears to still be running on newton, just not liberty/mitaka | 15:36 |
| odyssey4me | mhayden oh neat, it appears that we're getting checks for it in some places... but not gate checks: https://review.openstack.org/#/c/364356/ | 15:37 |
| odyssey4me | I suspect that they've modified a filter again | 15:37 |
| mhayden | that's peculiar | 15:38 |
| odyssey4me | thanks, let me hunt this down | 15:38 |
| mhayden | odyssey4me: thank you sir | 15:38 |
| odyssey4me | I suspect that something happened between Sep 1 and Sep 4 | 15:38 |
| *** pester has quit IRC | 15:38 | |
| automagically | odyssey4me: Am I correct in the belief that any patch submitted at the moment, should be a bug fix if the intent is for it to merge in time for the newton release? If so, should every patch fitting that description be accompanied by a LaunchPad bug? | 15:39 |
| odyssey4me | automagically I'm not a fan of implementing a strict policy on bug registration, but if the reason for the patch is unclear then I would request a bug to be registered. | 15:40 |
| odyssey4me | (with more details) | 15:41 |
| automagically | Gotcha, okay, makes sense | 15:41 |
| odyssey4me | and yes, we're now curbing new functionality inclusion | 15:41 |
| odyssey4me | the focus needs to be on test improvements, bug fixing and closing out any patches that are already in review - assuming they don't change functionality in any radical way | 15:41 |
| odyssey4me | we want to avoid risky functional changes | 15:41 |
| *** pcaruana has joined #openstack-ansible | 15:42 | |
| *** admin0 has quit IRC | 15:42 | |
| *** sdake has quit IRC | 15:44 | |
| *** berendt has quit IRC | 15:45 | |
| *** gtracy200 has joined #openstack-ansible | 15:48 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Move network_entry function to top level https://review.openstack.org/354396 | 15:50 |
| *** dkehn__ has joined #openstack-ansible | 15:53 | |
| sigmavirus | automagically: to be fair, what you suggest is what pretty much every other openstack project requires, so it's not unreasonable or unheard of | 15:55 |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 15:59 |
| *** woodard has joined #openstack-ansible | 16:00 | |
| *** adrian_otto has joined #openstack-ansible | 16:00 | |
| automagically | sigmavirus: After reading https://etherpad.openstack.org/p/osa-newton-rc1 I think I have a pretty good idea of the current focus for RC1. I missed last week’s meeting so doing some catching up. | 16:00 |
| evrardjp | Bug triage cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, | 16:01 |
| evrardjp | spotz, vdo, jmccrory, alextricity25, jasondotstar, KLevenstein, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397 | 16:01 |
| evrardjp | Here is our bug list for today https://etherpad.openstack.org/p/osa-bugtriage | 16:01 |
| evrardjp | sorry to interrupt your conversations | 16:01 |
| *** gfa has quit IRC | 16:01 | |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 16:01 |
| evrardjp | the list has probably been updated | 16:01 |
| evrardjp | so | 16:02 |
| openstackgerrit | Merged openstack/openstack-ansible: [DOC]Added Xenial support in install guide https://review.openstack.org/364572 | 16:02 |
| evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1620233 | 16:02 |
| openstack | Launchpad bug 1620233 in openstack-ansible "Doc: There is no branch for role documentation" [Undecided,New] | 16:02 |
| openstackgerrit | SHASHANK TAVILDAR (shasha_tavil) proposed openstack/openstack-ansible: Added the option to copy changes between stock env.d and repo env.d[M->N] https://review.openstack.org/366208 | 16:03 |
| evrardjp | odyssey4me: should we abandon this? | 16:04 |
| *** javis has quit IRC | 16:04 | |
| *** tomjosekal has joined #openstack-ansible | 16:04 | |
| odyssey4me | well, it's a wishlist item for sure - I'm not sure if it makes sense to provide an index to the branches available as suggested - or whether we should rather just ensure that the main docs refer to the role docs in a branch-applicable way | 16:05 |
| *** david-lyle_ has joined #openstack-ansible | 16:05 | |
| odyssey4me | I'm open to ideas | 16:05 |
| *** david-lyle_ has quit IRC | 16:05 | |
| odyssey4me | it can be a bit misleading for someone to come to the role docs as it stands right now | 16:06 |
| odyssey4me | automagically thoughts? | 16:06 |
| openstackgerrit | Qin Wang (qwang) proposed openstack/openstack-ansible: [docs] added a necessary arg for ansible command after removal of ansible.cfg https://review.openstack.org/363965 | 16:06 |
| evrardjp | If we expect ppl to go directly to role docs, I think we should have a proper landing page | 16:06 |
| automagically | I think its a wishlist item for now | 16:07 |
| evrardjp | if we include it, I think we should not care | 16:07 |
| evrardjp | that's for sure -- let's move this to confirmed + wishlist | 16:07 |
| openstackgerrit | Qin Wang (qwang) proposed openstack/openstack-ansible: [docs] added a necessary arg for ansible command after removal of ansible.cfg https://review.openstack.org/363965 | 16:07 |
| odyssey4me | yeah, some sort of simple index which points to the applicable branch docs would be good if it could be automated | 16:07 |
| evrardjp | I don't see a problem to have a simple index + link directly into the proper branch in the "main docs" | 16:08 |
| evrardjp | because you already use an index there | 16:08 |
| evrardjp | ok let's move on | 16:08 |
| evrardjp | https://bugs.launchpad.net/openstack-ansible/+bug/1619730 | 16:08 |
| openstack | Launchpad bug 1619730 in openstack-ansible "Instances created in AIO can not reach APIs on host" [Undecided,New] | 16:08 |
| *** dkehn__ has quit IRC | 16:09 | |
| evrardjp | spotz: are you there? | 16:10 |
| spotz | evrardjp: Yeah | 16:10 |
| *** bfg9000 has joined #openstack-ansible | 16:10 | |
| evrardjp | I think br-vxlan should be set to 1500 and all the traffic encapsulated will have an mtu of 1450 | 16:10 |
| evrardjp | that mtu should be (mitaka forward) auto calculated | 16:11 |
| evrardjp | so question: | 16:11 |
| evrardjp | on which branch did that happen? | 16:11 |
| evrardjp | I think the checksum is a good hint of the issue, but I don't see why the mtu is wrong | 16:12 |
| spotz | evrardjp: It was mitaka, cloudnull was the one who thought the mtu might have been an issue | 16:12 |
| evrardjp | was it after an upgrade? | 16:12 |
| spotz | so he said open the bug with both issues. Fresh install | 16:12 |
| cloudnull | evrardjp: this was an AIO. | 16:12 |
| spotz | had the person install twice | 16:12 |
| evrardjp | AIO and checksum fail :) | 16:12 |
| *** eil397 has joined #openstack-ansible | 16:13 | |
| *** eil397_ has joined #openstack-ansible | 16:13 | |
| evrardjp | so brand new mitaka AIO had this issue with checksums | 16:13 |
| cloudnull | evrardjp: it looked like a checksum fail and there were fragmenting issues getting out of the VM | 16:13 |
| evrardjp | well that's interesting | 16:13 |
| evrardjp | cloudnull: yup that's what I suspect when I see that | 16:14 |
| *** kenhui has quit IRC | 16:14 | |
| evrardjp | but how come? | 16:14 |
| evrardjp | mitaka should be properly calculating its mtu | 16:14 |
| evrardjp | based on ethernet 1500 | 16:14 |
| evrardjp | so except if you're running on something that's already encapsulated and not running with an mtu of 1500 | 16:14 |
| evrardjp | would that be the case? | 16:15 |
| *** schwicht has quit IRC | 16:15 | |
| cloudnull | so i created an integration veth with on the br-vxlan device, set the mtu to 9000 which made the bridge have an MTU of 9000, then I changed the neutron network to allow 9000 MTUs, the tenant networks then got an MTU of 8950 which solved the fragmenting issues. | 16:15 |
| cloudnull | The test was to send traffic to RAX auth:5000 and exepect JSON in return | 16:15 |
| evrardjp | makes sense | 16:15 |
| cloudnull | w/out the MTU fix it didn't work | 16:16 |
| cloudnull | then we did a test internally and that wasn't working | 16:16 |
| cloudnull | so we added the checksum and viola | 16:16 |
| evrardjp | We had checksum rules in the past | 16:16 |
| evrardjp | that are supposed to run on AIO | 16:17 |
| cloudnull | because the AIO is using bridges w/out an interface it can't fragment and checksum the traffic | 16:17 |
| Mudpuppy | did you have to touch sysctl? | 16:17 |
| cloudnull | Mudpuppy: no | 16:17 |
| cloudnull | so larger MTU means no fragmentation and the checksum rule forces the interaction a real network would handle | 16:18 |
| evrardjp | I don't see the issue here | 16:18 |
| evrardjp | everything you say makes perfect sense | 16:18 |
| evrardjp | is there a problem with expectations? | 16:19 |
| cloudnull | user wanted to spin a vm and hit the API | 16:19 |
| cloudnull | which it could not do | 16:19 |
| cloudnull | spedione: you around ^ | 16:20 |
| spedione | cloudnull: Yep | 16:20 |
| cloudnull | talking about the MTU/checksum fixes from the other day | 16:20 |
| *** gouthamr_ has quit IRC | 16:20 | |
| *** thorst has quit IRC | 16:21 | |
| spedione | cloudnull: Ah, OK. I didn't get all the details from spotz | 16:21 |
| cloudnull | evrardjp: ^ spedione was the user. | 16:21 |
| cloudnull | so idk | 16:21 |
| evrardjp | my understanding is that IF you have a NIC of mtu 1500 on the AIO host with br-vxlan mtu set to 1500 | 16:21 |
| *** thorst has joined #openstack-ansible | 16:21 | |
| evrardjp | that's fine | 16:21 |
| evrardjp | so the instances on this AIO cloud would get mtu 1450 | 16:21 |
| evrardjp | that's fine | 16:21 |
| evrardjp | if they don't, that's not | 16:21 |
| cloudnull | all is good. | 16:22 |
| cloudnull | however the traffic isn't fragmenting | 16:22 |
| cloudnull | because it never leaves a nic | 16:22 |
| evrardjp | doesn't refrag because it doesn't go through a router? | 16:22 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-tests: Add tempest tests to test-repository https://review.openstack.org/366243 | 16:22 |
| cloudnull | so when you get a payload that is >1450 the connection just hangs. | 16:22 |
| cloudnull | if the bridge was plugged into a nic it would fragment | 16:23 |
| evrardjp | but we have this checksumming and refrag on veths IIRC | 16:23 |
| cloudnull | and while slow it'd work | 16:23 |
| evrardjp | indeed | 16:23 |
| cloudnull | as it stands now it just hangs because nothing is doing the fragmenting | 16:23 |
| cloudnull | IMO checksum rule is needed regardless. | 16:24 |
| cloudnull | for AIOs that is | 16:24 |
| cloudnull | and MTU/fragmentation may need some investigation | 16:25 |
| cloudnull | for AIOs that is. | 16:25 |
| cloudnull | in the real world all of this works as expected | 16:25 |
| evrardjp | I agree with you | 16:25 |
| evrardjp | so | 16:25 |
| evrardjp | question | 16:25 |
| evrardjp | MTU/frag could be investigated separately | 16:25 |
| spotz | So in spedione (nice new stealth name:)) case the checksum fixed it but the mtu issue may still be reneeded | 16:25 |
| evrardjp | let's talk about the checksum | 16:25 |
| *** drifterza has quit IRC | 16:26 | |
| spedione | spotz: Not so much stealth as 'My bouncer is misconfigured and I'm too lazy to fix it ATM' | 16:26 |
| spotz | heheh | 16:26 |
| *** spedione is now known as chris_hultin | 16:26 | |
| evrardjp | there is in neutron_post_install a rule to hangle the metadata iptables checksum fix | 16:26 |
| evrardjp | handle* | 16:27 |
| cloudnull | yes. | 16:27 |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-tests: Add tempest tests to test-repository https://review.openstack.org/366243 | 16:28 |
| evrardjp | was that the rule applied to fix the checksuming issues? | 16:29 |
| *** dkehn__ has joined #openstack-ansible | 16:29 | |
| cloudnull | evrardjp: https://github.com/openstack/openstack-ansible-os_neutron/blob/master/files/post-up-metadata-checksum | 16:29 |
| cloudnull | we added -- ${IPTABLES} -A POSTROUTING -t mangle -p tcp -j CHECKSUM --checksum-fill | 16:29 |
| cloudnull | which is just a blanket apply | 16:29 |
| *** thorst has quit IRC | 16:30 | |
| evrardjp | it all sounds to me we need to reproduce this first | 16:30 |
| evrardjp | I didn't get this kind of issues on my mitaka's and I can ensure you I run api calls :p | 16:31 |
| evrardjp | I'm just surprised | 16:32 |
| cloudnull | API calls from within a VM ? | 16:32 |
| evrardjp | well not that much but it already happened :p | 16:32 |
| evrardjp | I can check with a brand new AIO | 16:33 |
| *** schwicht has joined #openstack-ansible | 16:33 | |
| *** gouthamr has joined #openstack-ansible | 16:33 | |
| cloudnull | ok. | 16:33 |
| cloudnull | chris_hultin: do you have the API calls that you were running? | 16:33 |
| chris_hultin | cloudnull: I don't have them offhand, but even attempting to CURL the Identity endpoint was just hanging. | 16:34 |
| cloudnull | ok | 16:34 |
| evrardjp | That was all for today | 16:37 |
| evrardjp | thanks for you time everyone, I'll document our conversations in the appropriate bugs | 16:37 |
| *** alokj has joined #openstack-ansible | 16:37 | |
| *** jascott1 has joined #openstack-ansible | 16:38 | |
| *** gtracy200 has quit IRC | 16:39 | |
| *** jamesdenton has quit IRC | 16:48 | |
| *** javis has joined #openstack-ansible | 16:51 | |
| *** KLevenstein has quit IRC | 16:51 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-rabbitmq_server: Allow configuration of rabbitmq plugins https://review.openstack.org/366252 | 16:52 |
| *** jperry has quit IRC | 16:53 | |
| *** sdake has joined #openstack-ansible | 16:57 | |
| *** asettle has quit IRC | 16:57 | |
| *** andymccr has quit IRC | 17:04 | |
| *** evrardjp has quit IRC | 17:05 | |
| *** AnarchyAo has joined #openstack-ansible | 17:06 | |
| *** andymccr has joined #openstack-ansible | 17:07 | |
| *** janki has quit IRC | 17:08 | |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Fix module loading on centos hosts. https://review.openstack.org/362881 | 17:08 |
| *** alokj has quit IRC | 17:09 | |
| *** evrardjp has joined #openstack-ansible | 17:11 | |
| *** sdake_ has joined #openstack-ansible | 17:12 | |
| *** sdake has quit IRC | 17:15 | |
| *** admin0 has joined #openstack-ansible | 17:26 | |
| *** thorst has joined #openstack-ansible | 17:26 | |
| *** michauds has quit IRC | 17:28 | |
| *** sdake_ has quit IRC | 17:28 | |
| *** sdake has joined #openstack-ansible | 17:29 | |
| *** jheroux has joined #openstack-ansible | 17:34 | |
| *** thorst has quit IRC | 17:35 | |
| *** michauds has joined #openstack-ansible | 17:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP][docs] Revise host layout and example configurations https://review.openstack.org/366271 | 17:48 |
| *** tomjosekal has quit IRC | 17:48 | |
| *** tomjosekal has joined #openstack-ansible | 17:49 | |
| *** sdake has quit IRC | 17:49 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_gnocchi: Only use dedicated Project for Gnocchi w/ Swift https://review.openstack.org/366274 | 17:55 |
| *** jamesdenton has joined #openstack-ansible | 17:55 | |
| *** sdake has joined #openstack-ansible | 17:56 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: [Docs] Update security appendix https://review.openstack.org/366276 | 17:58 |
| *** persia has quit IRC | 18:01 | |
| *** mrjk has quit IRC | 18:01 | |
| *** poopcat has joined #openstack-ansible | 18:02 | |
| *** dkehn__ has quit IRC | 18:03 | |
| *** dkehn_ has quit IRC | 18:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server: Allow configuration of rabbitmq plugins https://review.openstack.org/366252 | 18:05 |
| *** kenhui has joined #openstack-ansible | 18:10 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible-os_swift: Filter Gnocchi traffic optionally https://review.openstack.org/366278 | 18:12 |
| *** mrjk has joined #openstack-ansible | 18:13 | |
| openstackgerrit | Steve Lewis (stevelle) proposed openstack/openstack-ansible: Configure Ceilometer middleware for Gnocchi-Swift https://review.openstack.org/366279 | 18:13 |
| openstackgerrit | Paulo Matias proposed openstack/openstack-ansible-repo_build: Drop privileges before running the git clone script https://review.openstack.org/366280 | 18:15 |
| *** dkehn__ has joined #openstack-ansible | 18:15 | |
| *** dkehn_ has joined #openstack-ansible | 18:16 | |
| Matias | stevelle: nice ^ thank you, but were the variable is used inside the configuration files? | 18:19 |
| Matias | were is the variable* | 18:19 |
| Matias | where is the variable* | 18:19 |
| Matias | lol | 18:19 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-memcached_server: Ansible 2.1.1 role testing https://review.openstack.org/354401 | 18:20 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_glance: Ansible 2.1.1 role testing https://review.openstack.org/353589 | 18:21 |
| stevelle | Matias: see the note I put on, it needs to be converted to an ID and I have to figure out how to do that part still :) | 18:21 |
| stevelle | that's why I didn't solve this 2 months ago too | 18:22 |
| *** berendt has joined #openstack-ansible | 18:23 | |
| Matias | hm, this is tough | 18:23 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_cinder: Ansible 2.1.1 role testing https://review.openstack.org/353569 | 18:26 |
| stevelle | Matias: I suspect I need to add something to https://github.com/openstack/openstack-ansible-plugins/blob/master/library/keystone to allow us to get the ID out | 18:30 |
| stevelle | or you could if you feel up to it, and it might happen sooner | 18:30 |
| Matias | so it could be in the keystone module itself? no need to create another module? | 18:31 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-rabbitmq_server: Adding Vagrantfile for local testing https://review.openstack.org/366284 | 18:31 |
| *** KLevenstein has joined #openstack-ansible | 18:31 | |
| stevelle | Matias: no need for a new module, there is even an existing command actually https://github.com/openstack/openstack-ansible-plugins/blob/master/library/keystone#L620 so maybe it's just about getting that id out and registering a local var. | 18:32 |
| *** thorst has joined #openstack-ansible | 18:32 | |
| *** jamesdenton has quit IRC | 18:32 | |
| stevelle | I'm thinking add a task to lookup the project when swift_gnocchi_enabled and register a var named swift_gnocchi_service_project_id and that will end up used in the template | 18:33 |
| *** jperry has joined #openstack-ansible | 18:33 | |
| stevelle | and it looks like line 636 has the function getting the id back so we can probably fix up that patch set in 20 minutes to make it correct now that I bothered to look :) | 18:34 |
| Matias | stevelle: https://github.com/openstack/openstack-ansible-os_tempest/blob/ebe329597980eeb330827ee38c525f4869e96aa1/tasks/tempest_post_install.yml#L34 | 18:36 |
| Matias | yep, perhaps something like this | 18:36 |
| stevelle | something like that. yes | 18:36 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-rabbitmq_server: Enable HiPE compile configuration https://review.openstack.org/366288 | 18:37 |
| Matias | os_swift does not use keystone_facts for nothing yet, so no risk of overwriting something the role already uses | 18:37 |
| *** asettle has joined #openstack-ansible | 18:37 | |
| stevelle | the unresolved issue is that the gnocchi playbook may not have run yet when the swift playbook runs... so the project may not exist yet. | 18:37 |
| stevelle | you may need to run the swift playbook, then gnocchi, then swift again. | 18:37 |
| Matias | circular dependencies suck | 18:38 |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Updated to use the openstack-ansible-tests repo https://review.openstack.org/360130 | 18:38 |
| stevelle | Yes, odyssey4me proposed a change to the workflow that might give us a way to unwind that in Ocata, but in the mean time running the Swift setup again is possible and would work. | 18:39 |
| *** thorst has quit IRC | 18:39 | |
| Matias | so let's go this way and document it | 18:40 |
| Matias | just out of curiosity.. | 18:40 |
| Matias | what are the advantages of filtering directly in the swift middleware instead of filtering in ceilometer? | 18:41 |
| stevelle | Less load placed on Rabbit | 18:41 |
| Matias | ah, nice | 18:41 |
| stevelle | If you never send the message it doesn't need to be processed and filtered before being dropped | 18:42 |
| *** asettle has quit IRC | 18:42 | |
| Matias | i'm redeploying OSA here and os_swift's "Ensure services are started" is taking forever :( | 18:44 |
| *** asettle has joined #openstack-ansible | 18:45 | |
| *** asettle has quit IRC | 18:46 | |
| Matias | stevelle: do you know if aodh in OSA is currently working with a pure-gnocchi deployment (collecting data directly from the gnocchi API instead of trying to retrieve from ceilometer API) | 18:55 |
| *** michauds has quit IRC | 18:56 | |
| Matias | it was not clear for me from aodh docs if some special setup is required for that | 18:56 |
| stevelle | Matias: aodh should be checking the service catalog for the presence of gnocchi and switching behavior internally based on that. | 18:56 |
| Matias | nice, thanks | 18:56 |
| stevelle | please let us know if it doesn't work right in OSA but shouldn't need special handling | 18:57 |
| Matias | stevelle: I don't know if it is innocuous, but I was getting some complaints in ceilometer api log about failure to connect to mongodb, and the ceilometer endpoint was not becoming active (haproxy would fail to connect to ceilometer API) | 19:00 |
| Matias | I solved by setting ceilometer_connection_string: "log://" | 19:00 |
| *** TxGirlGeek has quit IRC | 19:00 | |
| Matias | I tried to remove the [database] section entirely from the config, but it didn't work | 19:00 |
| stevelle | Matias: does your deploy have this fix? https://github.com/openstack/openstack-ansible/commit/91032efcf1f891ec774a7de67e9704824e6f0954 | 19:01 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift: Add Multi Region Swift gate scenario https://review.openstack.org/356680 | 19:01 |
| Matias | stevelle: now it has (just updated to latest master), but before that I was setting ceilometer_gnocchi_enabled: yes manually | 19:01 |
| stevelle | sounds like something still not quite right if ceilo api wants to connect to a mongodb | 19:02 |
| Matias | yep, maybe it is a ceilometer's bug | 19:02 |
| stevelle | could be osa, could be ceilo | 19:02 |
| Matias | well, OSA just redeployed here | 19:03 |
| Matias | let's see if rabbitmq will fry this time | 19:03 |
| *** jascott1 has quit IRC | 19:04 | |
| Matias | Error on AMQP connection ... user: 'swift', state: opening): access to vhost '/swifttopic = notifications' refused for user 'swift' | 19:06 |
| Matias | grumpft | 19:07 |
| *** javis has quit IRC | 19:07 | |
| *** michauds has joined #openstack-ansible | 19:08 | |
| *** pcaruana has quit IRC | 19:11 | |
| cloudnull | Matias: is ceilometer in your swift-proxy.conf? | 19:14 |
| cloudnull | and trying to connect over ssl ? | 19:14 |
| Matias | swift-proxy.conf? is that deployed by OSA? | 19:16 |
| Matias | # find /etc/ansible/roles/ -iname '*swift-proxy*' | 19:16 |
| Matias | /etc/ansible/roles/os_swift/tests/host_vars/swift-proxy.yml | 19:16 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_cinder: Ansible 2.1.1 update https://review.openstack.org/353569 | 19:17 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-plugins: Update glance image properties example https://review.openstack.org/366299 | 19:18 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Update image_properties in image upload task https://review.openstack.org/366300 | 19:18 |
| stevelle | Matias: in the swift proxy container it will be under /etc/swift/proxy-server | 19:19 |
| Matias | stevelle: [filter:ceilometer] is there | 19:20 |
| Matias | url = rabbit://swift:password@host1:port,swift:password@host2:port,swift:password@host3:port//swifttopic = notifications | 19:21 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Add basic functional tests to role https://review.openstack.org/355155 | 19:21 |
| Matias | appears to be correct.. but i'm not very proficient in rabbitmq | 19:21 |
| stevelle | the port number is what matters there Matias, it should be :5671 iirc to Not use SSL | 19:21 |
| Matias | ah | 19:22 |
| Matias | it is 5672 | 19:22 |
| *** rromans has quit IRC | 19:22 | |
| stevelle | swift_rabbitmq_telemetry_use_ssl should be False | 19:22 |
| Matias | /etc/ansible/roles/os_swift/defaults/main.yml:swift_rabbitmq_telemetry_use_ssl: "False" | 19:23 |
| Matias | and I didn't override it | 19:23 |
| stevelle | I might have the port number wrong | 19:23 |
| Matias | proxy-server.conf.j2 uses {{ swift_rabbitmq_telemetry_port }} | 19:24 |
| Matias | https://github.com/openstack/openstack-ansible-os_swift/search?utf8=%E2%9C%93&q=swift_rabbitmq_telemetry_port | 19:24 |
| Matias | which is not overriden? | 19:24 |
| stevelle | override swift_rabbitmq_telemetry_port to 5671 though and rerun os-swift-install playbook | 19:24 |
| Matias | ok, let me try | 19:24 |
| *** persia has joined #openstack-ansible | 19:27 | |
| Matias | it's running, i'll check it later, need to go out for a meeting now :| | 19:27 |
| Matias | stevelle: thank you | 19:27 |
| stevelle | I might be totally wrong on that port bit, so we might not be done. off to lunch | 19:27 |
| *** alwaysatthenoc has joined #openstack-ansible | 19:31 | |
| *** johnmilton has quit IRC | 19:32 | |
| *** thorst has joined #openstack-ansible | 19:36 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-galera_server: Allow galera to listen on v4/v6 https://review.openstack.org/366307 | 19:37 |
| *** thorst has quit IRC | 19:45 | |
| *** sdake has quit IRC | 19:46 | |
| *** johnmilton has joined #openstack-ansible | 19:49 | |
| openstackgerrit | Merged openstack/openstack-ansible: Move pip_lock_to_internal_repo to group_vars https://review.openstack.org/344388 | 19:53 |
| *** AnarchyAo has quit IRC | 19:58 | |
| *** asettle has joined #openstack-ansible | 19:58 | |
| *** asettle has quit IRC | 19:58 | |
| cloudnull | Matias: ping whne you around | 20:00 |
| cloudnull | *when | 20:00 |
| *** sdake has joined #openstack-ansible | 20:03 | |
| *** TxGirlGeek has joined #openstack-ansible | 20:04 | |
| *** AnarchyAo has joined #openstack-ansible | 20:04 | |
| *** rromans has joined #openstack-ansible | 20:04 | |
| stevelle | cloudnull: related to that swift middleware, possible he is seeing "AMQP server 172.29.239.252:5671 closed the connection. Check login credentials: Socket closed" in the proxy-error.log | 20:18 |
| stevelle | or something like that | 20:18 |
| stevelle | I get that sometimes on both ports oddly | 20:18 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Move network_entry function to top level https://review.openstack.org/354396 | 20:24 |
| *** thorst has joined #openstack-ansible | 20:33 | |
| automagically | stevelle and Matias - 5672 is the RabbitMQ port without SSL | 20:36 |
| automagically | https://www.rabbitmq.com/networking.html | 20:36 |
| *** thorst has quit IRC | 20:39 | |
| *** javis has joined #openstack-ansible | 20:40 | |
| *** hughmFLEXin has joined #openstack-ansible | 20:42 | |
| *** adib_ has joined #openstack-ansible | 20:42 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible: [WIP] Fix ml2 ports after an upgrade https://review.openstack.org/366169 | 20:46 |
| *** adib_ has quit IRC | 20:49 | |
| *** adib_ has joined #openstack-ansible | 20:50 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_tempest: Ensure tempest runs independently in testing https://review.openstack.org/366335 | 20:51 |
| *** agireud has quit IRC | 20:52 | |
| *** oneswig has joined #openstack-ansible | 20:55 | |
| *** Mudpuppy_ has joined #openstack-ansible | 20:56 | |
| *** Mudpuppy has quit IRC | 20:58 | |
| *** agireud has joined #openstack-ansible | 20:59 | |
| *** retreved_ has quit IRC | 21:00 | |
| *** Mudpuppy_ has quit IRC | 21:00 | |
| *** hughmFLEXin has quit IRC | 21:01 | |
| *** sdake has quit IRC | 21:02 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-plugins: Deserialize glance image_properties https://review.openstack.org/366299 | 21:03 |
| *** kylek3h has quit IRC | 21:04 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Update image_properties in image upload task https://review.openstack.org/366300 | 21:06 |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Add basic functional tests to role https://review.openstack.org/355155 | 21:06 |
| *** chris_hultin is now known as spedione|AWAY | 21:07 | |
| *** woodard_ has joined #openstack-ansible | 21:10 | |
| openstackgerrit | Andy McCrae proposed openstack/openstack-ansible-os_tempest: Ensure tempest runs independently in testing https://review.openstack.org/366335 | 21:11 |
| *** woodard has quit IRC | 21:13 | |
| *** adib_ has left #openstack-ansible | 21:13 | |
| *** woodard_ has quit IRC | 21:14 | |
| mrda | Morning OSA | 21:19 |
| *** ametts has quit IRC | 21:30 | |
| *** schwicht has quit IRC | 21:33 | |
| *** thorst has joined #openstack-ansible | 21:37 | |
| *** adrian_otto has quit IRC | 21:38 | |
| *** michaelgugino has quit IRC | 21:40 | |
| *** thorst has quit IRC | 21:45 | |
| *** strobelight_ has joined #openstack-ansible | 21:48 | |
| *** mrjk has quit IRC | 21:51 | |
| *** schwicht has joined #openstack-ansible | 21:51 | |
| *** strobelight has quit IRC | 21:52 | |
| *** kenhui1 has joined #openstack-ansible | 21:52 | |
| *** strobelight_ has quit IRC | 21:52 | |
| *** kenhui has quit IRC | 21:52 | |
| *** spedione|AWAY is now known as spedione | 21:53 | |
| *** alwaysatthenoc has quit IRC | 21:54 | |
| *** mrjk has joined #openstack-ansible | 21:56 | |
| *** schwicht has quit IRC | 21:59 | |
| *** ddaskal has quit IRC | 21:59 | |
| *** ddaskal has joined #openstack-ansible | 21:59 | |
| *** schwicht has joined #openstack-ansible | 21:59 | |
| *** askb has joined #openstack-ansible | 22:00 | |
| *** TxGirlGeek has quit IRC | 22:00 | |
| *** admin0 has quit IRC | 22:00 | |
| *** jperry has quit IRC | 22:01 | |
| *** oneswig has quit IRC | 22:02 | |
| openstackgerrit | Merged openstack/openstack-ansible: Ensure that repo build arch grouping always runs https://review.openstack.org/365448 | 22:03 |
| *** ddaskal has quit IRC | 22:04 | |
| *** asettle has joined #openstack-ansible | 22:05 | |
| *** BjoernT has quit IRC | 22:05 | |
| *** asettle has quit IRC | 22:06 | |
| *** kylek3h has joined #openstack-ansible | 22:06 | |
| *** asettle has joined #openstack-ansible | 22:07 | |
| *** spotz is now known as spotz_zzz | 22:09 | |
| *** javis has quit IRC | 22:18 | |
| *** kenhui has joined #openstack-ansible | 22:20 | |
| *** kenhui1 has quit IRC | 22:21 | |
| *** jmckind has quit IRC | 22:21 | |
| *** schwicht has quit IRC | 22:23 | |
| *** berendt has quit IRC | 22:35 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone: Fix nginx SCRIPT_NAME uwsgi_param https://review.openstack.org/365708 | 22:40 |
| *** KLevenstein has quit IRC | 22:41 | |
| *** michauds has quit IRC | 22:42 | |
| *** thorst has joined #openstack-ansible | 22:43 | |
| *** weezS has quit IRC | 22:44 | |
| *** klamath has quit IRC | 22:46 | |
| *** thorst has quit IRC | 22:49 | |
| *** thorst has joined #openstack-ansible | 22:49 | |
| *** M00nr41n has quit IRC | 22:53 | |
| *** asettle has quit IRC | 22:54 | |
| *** asettle has joined #openstack-ansible | 22:57 | |
| *** asettle has quit IRC | 22:57 | |
| *** schwicht has joined #openstack-ansible | 23:02 | |
| *** eil397 has quit IRC | 23:03 | |
| *** eil397_ has quit IRC | 23:04 | |
| *** strobelight has joined #openstack-ansible | 23:07 | |
| *** hj-hpe has quit IRC | 23:08 | |
| *** asettle has joined #openstack-ansible | 23:09 | |
| *** asettle has quit IRC | 23:09 | |
| *** sdake has joined #openstack-ansible | 23:10 | |
| *** spedione is now known as spedione|AWAY | 23:13 | |
| *** adrian_otto has joined #openstack-ansible | 23:15 | |
| *** kenhui has quit IRC | 23:17 | |
| *** markvoelker has quit IRC | 23:19 | |
| *** jamesdenton has joined #openstack-ansible | 23:21 | |
| *** javis has joined #openstack-ansible | 23:22 | |
| *** jamesdenton has quit IRC | 23:26 | |
| *** bapalm has quit IRC | 23:35 | |
| *** bapalm has joined #openstack-ansible | 23:41 | |
| *** strobelight has quit IRC | 23:43 | |
| *** strobelight has joined #openstack-ansible | 23:45 | |
| *** Jeffrey4l has joined #openstack-ansible | 23:49 | |
| *** kenhui has joined #openstack-ansible | 23:49 | |
| *** Jeffrey4l has quit IRC | 23:55 | |
| *** markvoelker has joined #openstack-ansible | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!