| *** gouthamr has joined #openstack-ansible | 00:00 | |
| dxiri | http://paste.ubuntu.com/23418570/plain/ | 00:00 |
|---|---|---|
| cloudnull | if you click the link for the full screen does it work ? | 00:00 |
| dxiri | nope | 00:01 |
| dxiri | same thing | 00:01 |
| evrardjp | ssl properly configured? | 00:01 |
| dxiri | and I just redeployed the whole container | 00:01 |
| evrardjp | maybe we do it by default now | 00:01 |
| dxiri | it generated a self signed cert | 00:02 |
| evrardjp | I'll leave that to cloudnull if he has time, I'm not up to date with that issue | 00:02 |
| dxiri | which I accepted | 00:02 |
| dxiri | so its on my exception list | 00:02 |
| cloudnull | lir spice still does not support native ssl, novnc does though. | 00:02 |
| cloudnull | that said it should still work | 00:02 |
| cloudnull | Ima about to run to my last meeting of the day. dxiri i'll be back online in around an hour or so. | 00:03 |
| dxiri | 2016-11-02 18:00:33.433 4028 INFO nova.consoleauth.manager [req-8c2bb043-8f1a-46f0-8c4d-b1916ce5c9bd - - - - -] Checking Token: b3229393-d3aa-409d-bea8-3c16f1440918, False | 00:03 |
| dxiri | token false? | 00:03 |
| cloudnull | dxiri: check memcached and make sure it's up and accessible from the console container. | 00:04 |
| dxiri | cloudnull: no worries :) thanks for the help, if I am not back then we could sync up tomorrow | 00:04 |
| dxiri | on it | 00:04 |
| openstackgerrit | Michael Carden proposed openstack/openstack-ansible: Move storage_address discovery into separate play https://review.openstack.org/392969 | 00:04 |
| cloudnull | thanks mcarden as you can tell, i'm illiterate :P | 00:05 |
| cloudnull | though I do find it amazing folks have learned to read my giberish :) | 00:06 |
| mcarden | Shouldn't you be asleep? :) | 00:06 |
| cloudnull | ok gotta run . | 00:06 |
| mcarden | cya | 00:06 |
| cloudnull | mcarden: sleep is for the weak | 00:06 |
| mcarden | LOL | 00:06 |
| cloudnull | https://cdn.meme.am/instances/500x/34103219.jpg | 00:07 |
| cloudnull | bbl | 00:07 |
| mcarden | Heh | 00:07 |
| *** jlockwood has quit IRC | 00:07 | |
| dxiri | cloudnull: ● memcached.service loaded failed failed memcached daemon | 00:07 |
| dxiri | there ya go | 00:08 |
| dxiri | checking to see if it made any difference | 00:08 |
| dxiri | that did it :) | 00:09 |
| dxiri | what does memcached has to do with this? | 00:09 |
| dxiri | please enlight a noob :) | 00:10 |
| openstackgerrit | Merged openstack/openstack-ansible: Updated from global requirements https://review.openstack.org/392756 | 00:14 |
| *** poopcat has quit IRC | 00:26 | |
| *** adrian_otto has quit IRC | 00:26 | |
| *** dxiri has quit IRC | 00:33 | |
| *** poopcat has joined #openstack-ansible | 00:33 | |
| openstackgerrit | Michael Carden proposed openstack/openstack-ansible-os_magnum: Removing extras/ https://review.openstack.org/384676 | 00:35 |
| *** weezS has quit IRC | 00:36 | |
| *** poopcat has quit IRC | 00:38 | |
| *** askb has quit IRC | 00:41 | |
| *** askb has joined #openstack-ansible | 00:42 | |
| *** poopcat has joined #openstack-ansible | 00:49 | |
| *** thorst has joined #openstack-ansible | 00:51 | |
| *** markvoelker has quit IRC | 00:52 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_tempest: Allow user defined nova flavor resources https://review.openstack.org/393004 | 00:52 |
| *** LiYuenan has joined #openstack-ansible | 00:55 | |
| *** askb has quit IRC | 00:56 | |
| *** askb has joined #openstack-ansible | 00:58 | |
| openstackgerrit | Joshua Hesketh proposed openstack/openstack-ansible: Replace github with git.o.o https://review.openstack.org/392339 | 01:09 |
| *** weezS has joined #openstack-ansible | 01:11 | |
| *** thorst has quit IRC | 01:11 | |
| openstackgerrit | Joshua Hesketh proposed openstack/openstack-ansible: Remove outdated aio heat template https://review.openstack.org/393015 | 01:11 |
| *** askb has quit IRC | 01:11 | |
| openstackgerrit | Joshua Hesketh proposed openstack/openstack-ansible-galera_client: Replace github.com with git.o.o https://review.openstack.org/392340 | 01:12 |
| openstackgerrit | Joshua Hesketh proposed openstack/openstack-ansible-os_ironic: Replace github link with published spec https://review.openstack.org/392345 | 01:13 |
| *** askb has joined #openstack-ansible | 01:15 | |
| *** v1k0d3n has quit IRC | 01:23 | |
| *** v1k0d3n has joined #openstack-ansible | 01:23 | |
| *** v1k0d3n has quit IRC | 01:24 | |
| *** v1k0d3n has joined #openstack-ansible | 01:24 | |
| *** askb has quit IRC | 01:25 | |
| *** askb has joined #openstack-ansible | 01:26 | |
| *** thorst has joined #openstack-ansible | 01:38 | |
| *** thorst has quit IRC | 01:39 | |
| *** markvoelker has joined #openstack-ansible | 01:53 | |
| *** weezS has quit IRC | 01:55 | |
| *** markvoelker has quit IRC | 01:58 | |
| *** adrian_otto has joined #openstack-ansible | 02:14 | |
| *** v1k0d3n has quit IRC | 02:20 | |
| *** weezS has joined #openstack-ansible | 02:30 | |
| *** kjw3 has joined #openstack-ansible | 02:52 | |
| *** weezS has quit IRC | 02:55 | |
| *** schwicht has quit IRC | 02:55 | |
| *** adrian_otto has quit IRC | 03:03 | |
| *** gouthamr has quit IRC | 03:04 | |
| *** weezS has joined #openstack-ansible | 03:14 | |
| *** oanson has joined #openstack-ansible | 03:23 | |
| *** adrian_otto has joined #openstack-ansible | 03:26 | |
| *** adrian_otto has quit IRC | 03:27 | |
| *** adrian_otto has joined #openstack-ansible | 03:28 | |
| *** adrian_otto has quit IRC | 03:32 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Move storage_address discovery into separate play https://review.openstack.org/392969 | 03:34 |
| *** adrian_otto has joined #openstack-ansible | 03:35 | |
| *** hybridpollo has quit IRC | 03:43 | |
| *** Mudpuppy_ has quit IRC | 03:50 | |
| *** Jeffrey4l has joined #openstack-ansible | 03:54 | |
| *** markvoelker has joined #openstack-ansible | 03:54 | |
| *** markvoelker has quit IRC | 03:59 | |
| *** poopcat has quit IRC | 03:59 | |
| *** adrian_otto has quit IRC | 04:15 | |
| *** adrian_otto has joined #openstack-ansible | 04:24 | |
| *** pegmanm_ has joined #openstack-ansible | 04:40 | |
| *** chhavi has joined #openstack-ansible | 04:55 | |
| *** deepak_jon has joined #openstack-ansible | 05:18 | |
| *** weezS has quit IRC | 05:20 | |
| *** v1k0d3n has joined #openstack-ansible | 05:28 | |
| *** adrian_otto has quit IRC | 05:34 | |
| *** adrian_otto has joined #openstack-ansible | 05:35 | |
| *** adrian_otto has quit IRC | 05:35 | |
| *** adrian_otto has joined #openstack-ansible | 05:37 | |
| *** openstackgerrit has quit IRC | 05:48 | |
| *** Jack_Iv has joined #openstack-ansible | 05:48 | |
| *** openstackgerrit has joined #openstack-ansible | 05:48 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible: Move storage_address discovery into a single task https://review.openstack.org/392969 | 05:49 |
| *** v1k0d3n has quit IRC | 05:54 | |
| *** markvoelker has joined #openstack-ansible | 05:55 | |
| *** adrian_otto has quit IRC | 05:56 | |
| *** markvoelker has quit IRC | 05:59 | |
| *** oanson has quit IRC | 06:08 | |
| *** Jack_Iv has quit IRC | 06:11 | |
| *** Jack_Iv has joined #openstack-ansible | 06:11 | |
| *** drifterza has joined #openstack-ansible | 06:50 | |
| *** yolanda has joined #openstack-ansible | 06:51 | |
| *** yolanda_ has joined #openstack-ansible | 06:51 | |
| *** yolanda_ has quit IRC | 06:52 | |
| *** oanson has joined #openstack-ansible | 06:53 | |
| *** markvoelker has joined #openstack-ansible | 06:55 | |
| *** fxpester has joined #openstack-ansible | 06:58 | |
| *** h5t4 has joined #openstack-ansible | 06:58 | |
| *** m4rx has joined #openstack-ansible | 06:59 | |
| *** markvoelker has quit IRC | 07:00 | |
| *** McMurlock1 has joined #openstack-ansible | 07:05 | |
| *** armaan has joined #openstack-ansible | 07:07 | |
| *** pester has joined #openstack-ansible | 07:09 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Updated from global requirements https://review.openstack.org/392780 | 07:10 |
| *** fxpester has quit IRC | 07:12 | |
| *** armaan_ has joined #openstack-ansible | 07:18 | |
| *** armaan has quit IRC | 07:18 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Better functional tests for monasca-api. https://review.openstack.org/392515 | 07:25 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic: Replace github link with published spec https://review.openstack.org/392345 | 07:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova: Change protocol of novalink url. https://review.openstack.org/393133 | 07:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops: Replace github with git.o.o https://review.openstack.org/393134 | 07:36 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-plugins: Replace github with git.o.o https://review.openstack.org/393136 | 07:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi: Replace github with git.o.o https://review.openstack.org/393137 | 07:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts: Implement LXC cache timeout https://review.openstack.org/393138 | 07:37 |
| *** gfa has quit IRC | 07:38 | |
| *** gfa has joined #openstack-ansible | 07:38 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392604 | 07:39 |
| *** gfa has quit IRC | 07:40 | |
| *** gfa has joined #openstack-ansible | 07:40 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic: Simplify pip options/constraints mechanism https://review.openstack.org/392654 | 07:45 |
| drifterza | odyssey4me, | 07:45 |
| drifterza | you there ? | 07:46 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Replace github with git.o.o https://review.openstack.org/393143 | 07:47 |
| *** openstackgerrit has quit IRC | 07:48 | |
| *** openstackgerrit has joined #openstack-ansible | 07:48 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Replace github with git.o.o https://review.openstack.org/392347 | 07:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Updated from global requirements https://review.openstack.org/392798 | 07:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392360 | 07:54 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393145 | 07:55 |
| odyssey4me | o/ drifterza - here, barely | 07:56 |
| openstackgerrit | Merged openstack/openstack-ansible-security: Fix stdout_lines check https://review.openstack.org/391934 | 07:56 |
| odyssey4me | bbiab - time to relocate | 07:58 |
| openstackgerrit | Merged openstack/openstack-ansible-security: Updated from global requirements https://review.openstack.org/392798 | 08:07 |
| *** haad1 has joined #openstack-ansible | 08:09 | |
| *** oanson has quit IRC | 08:13 | |
| *** oanson has joined #openstack-ansible | 08:15 | |
| *** jowisz has joined #openstack-ansible | 08:17 | |
| admin0 | morning all \o | 08:17 |
| *** oanson has quit IRC | 08:19 | |
| *** oanson has joined #openstack-ansible | 08:21 | |
| *** oanson has quit IRC | 08:22 | |
| *** oanson has joined #openstack-ansible | 08:23 | |
| *** oanson has quit IRC | 08:25 | |
| *** oanson has joined #openstack-ansible | 08:28 | |
| openstackgerrit | Omer Anson proposed openstack/openstack-ansible-os_neutron: [WIP] Dragonflow https://review.openstack.org/391524 | 08:37 |
| *** mmmddd has joined #openstack-ansible | 08:38 | |
| *** mmmddd has left #openstack-ansible | 08:38 | |
| winggundamth | hi all. I got this problem on swift proxy container log | 08:39 |
| winggundamth | infra2-swift-proxy-container-37399e55 proxy-server: STDERR: ERROR:oslo.messaging._drivers.impl_rabbit:AMQP server on 172.27.5.200:5672 is unreachable: [Errno 104] ECONNRESET. Trying again in 32 seconds. | 08:39 |
| winggundamth | it weird that I attach into that swift proxy container and try both ping and telnet. it works fine.... hmm | 08:40 |
| winggundamth | so it turns out that if you restart rabbitmq container. you need to restart swift proxy container too. | 08:47 |
| *** karimb has joined #openstack-ansible | 08:53 | |
| *** markvoelker has joined #openstack-ansible | 08:56 | |
| *** markvoelker has quit IRC | 09:00 | |
| *** karimb has quit IRC | 09:13 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add supported Operating System check https://review.openstack.org/392091 | 09:19 |
| *** adogrant has joined #openstack-ansible | 09:20 | |
| admin0 | rabbit restarting has a lot of “undesired” consenquences | 09:21 |
| *** deepak_jon has quit IRC | 09:21 | |
| *** xek has quit IRC | 09:24 | |
| *** f13o_ has quit IRC | 09:26 | |
| adogrant | Hi all, Just wondered if anyone has had any issues with rbd backed nova-compute using venvs? It appears the rbd packages are not installing and I cannot install them: | 09:27 |
| *** haad1 has quit IRC | 09:27 | |
| adogrant | (nova-13.3.7) root@compute-1:~# pip install rbd | 09:27 |
| adogrant | Ignoring indexes: https://pypi.python.org/simple | 09:27 |
| adogrant | Collecting rbd | 09:27 |
| adogrant | Could not find a version that satisfies the requirement rbd (from versions: ) | 09:27 |
| adogrant | No matching distribution found for rbd | 09:27 |
| odyssey4me | adogrant the rbd libs are not pip installable, so we link them from the system packages into the venv | 09:28 |
| *** armaan_ has quit IRC | 09:28 | |
| odyssey4me | this is how it's done: https://github.com/openstack/openstack-ansible-ceph_client/blob/stable/newton/tasks/ceph_install_python_libs.yml | 09:29 |
| *** haad1 has joined #openstack-ansible | 09:29 | |
| evrardjp | logan-: bad news, the apt module still stucks and doesn't update its cache, even if you put cache_update_time to something as low as 1. | 09:29 |
| adogrant | thanks odyssey4me | 09:29 |
| odyssey4me | evrardjp is that with ansible 2.2 ? | 09:29 |
| evrardjp | yup | 09:30 |
| odyssey4me | evrardjp argh, is cloudnull's patch in there? | 09:30 |
| evrardjp | the mtime of /var/cache/apt doesn't change and even that, it output itself cache_updated: false | 09:31 |
| evrardjp | I will have a look | 09:31 |
| *** oanson has quit IRC | 09:31 | |
| evrardjp | doesn't look like it https://github.com/ansible/ansible-modules-core/blob/devel/packaging/os/apt.py | 09:31 |
| odyssey4me | it's there: https://github.com/ansible/ansible-modules-core/commit/9b2ee881d9470e1db0c3c7eef4cbb2fa28002db2#diff-9a59e06d4a12ceb05b4e5d01c271f1b3 | 09:32 |
| neith | should I Use git.openstack.org or the github repo? | 09:32 |
| pjm6 | good morning all | 09:32 |
| odyssey4me | neith whichever works best for you | 09:32 |
| *** oanson has joined #openstack-ansible | 09:32 | |
| neith | odyssey4me: I think its easier to pull patches from git.openstack.org | 09:33 |
| odyssey4me | neith well, considering that github is often the target of DDOS attacks, and git.o.o is not - probably | 09:33 |
| odyssey4me | evrardjp it's also in stable-2.2: https://github.com/ansible/ansible-modules-core/commits/stable-2.2/packaging/os/apt.py | 09:34 |
| evrardjp | odyssey4me: I will have a double check but it doesn't seem to work | 09:34 |
| neith | odyssey4me: :D | 09:34 |
| *** oanson has quit IRC | 09:34 | |
| *** karimb has joined #openstack-ansible | 09:35 | |
| *** oanson has joined #openstack-ansible | 09:35 | |
| odyssey4me | evrardjp maybe there's an issue on the system itself? ie the file system parms? | 09:36 |
| evrardjp | it's default rax ubuntu image :p | 09:36 |
| evrardjp | ext4 | 09:36 |
| odyssey4me | sure, but check whether mtime is disabled or something | 09:36 |
| evrardjp | noatime | 09:36 |
| evrardjp | not something about mtime | 09:37 |
| *** haad1 has quit IRC | 09:37 | |
| *** oanson has quit IRC | 09:37 | |
| *** oanson has joined #openstack-ansible | 09:38 | |
| *** armaan has joined #openstack-ansible | 09:39 | |
| *** haad1 has joined #openstack-ansible | 09:40 | |
| *** oanson has quit IRC | 09:40 | |
| *** oanson has joined #openstack-ansible | 09:41 | |
| *** electrofelix has joined #openstack-ansible | 09:42 | |
| odyssey4me | evrardjp so it appears that the mtime change is dependent on an apt.conf.d entry - see http://askubuntu.com/questions/410247/how-to-know-last-time-apt-get-was-executed | 09:43 |
| odyssey4me | look at the second answer | 09:43 |
| *** oanson has quit IRC | 09:43 | |
| *** oanson has joined #openstack-ansible | 09:44 | |
| *** oanson has quit IRC | 09:46 | |
| *** oanson has joined #openstack-ansible | 09:47 | |
| evrardjp | I was already checking /var/lib/apt/periodic | 09:48 |
| *** chhavi has quit IRC | 09:49 | |
| *** oanson has quit IRC | 09:49 | |
| odyssey4me | periodic? no - I mean look at the update about the first file needing 'update-notifier-common' installed, and that /var/cache/apt/pkgcache.bin is probably a better file to stat | 09:50 |
| *** oanson has joined #openstack-ansible | 09:50 | |
| odyssey4me | these are the two things currently checked: https://github.com/ansible/ansible-modules-core/blob/devel/packaging/os/apt.py#L251-L252 | 09:51 |
| *** oanson has quit IRC | 09:52 | |
| *** oanson has joined #openstack-ansible | 09:54 | |
| openstackgerrit | Omer Anson proposed openstack/openstack-ansible-os_neutron: [WIP] Dragonflow https://review.openstack.org/391524 | 09:55 |
| *** oanson has quit IRC | 09:55 | |
| *** oanson has joined #openstack-ansible | 09:56 | |
| *** oanson has quit IRC | 09:58 | |
| *** oanson has joined #openstack-ansible | 09:59 | |
| neith | guys, I must admit, deploying Openstack with OSA is really the best method, kudos! | 10:01 |
| *** oanson has quit IRC | 10:01 | |
| evrardjp | haha good to hear neith | 10:01 |
| neith | I tried with the canonical tools -> not prod ready | 10:02 |
| neith | I tried Suse -> too far from my knowledge | 10:02 |
| *** oanson has joined #openstack-ansible | 10:02 | |
| neith | I tried manually -> give me a gun | 10:02 |
| neith | OSA is <3 | 10:02 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone: Simplify pip options/constraints mechanism https://review.openstack.org/393203 | 10:02 |
| evrardjp | odyssey4me: I'm not sure to understand what you mean | 10:03 |
| evrardjp | On my system I have /var/lib/apt/periodic | 10:03 |
| odyssey4me | neith :) I'm glad you like it | 10:03 |
| odyssey4me | evrardjp do you have the file /var/lib/apt/periodic/update-success-stamp ? | 10:03 |
| evrardjp | so https://github.com/ansible/ansible-modules-core/blob/devel/packaging/os/apt.py#L738 | 10:03 |
| evrardjp | yes I'm lazy at typing | 10:04 |
| odyssey4me | ok, and is the package 'update-notifier-common' installed? | 10:04 |
| evrardjp | so L738 exists | 10:04 |
| evrardjp | why does it matter? | 10:04 |
| *** oanson has quit IRC | 10:04 | |
| odyssey4me | does the file '/var/cache/apt/pkgcache.bin' exist, and does its timestamp match that of /var/lib/apt/periodic/update-success-stamp ? | 10:05 |
| *** chhavi has joined #openstack-ansible | 10:05 | |
| odyssey4me | I think there is actually a timestamp in the content of /var/lib/apt/periodic/update-success-stamp - I wonder if that matches the file's timestamp | 10:05 |
| *** f13o_ has joined #openstack-ansible | 10:05 | |
| evrardjp | my pkgcache.bin exist | 10:06 |
| odyssey4me | it matters because in order for this to work, the most reliable way of telling when it was last updated needs to be used - and it appears that the current checks are not the most reliable | 10:06 |
| evrardjp | I'll compare the timestamp, but I see no dependency on the package update-notifier-common in the code | 10:06 |
| evrardjp | odyssey4me: it looks to me it just compare timestamps | 10:06 |
| odyssey4me | apparently /var/lib/apt/periodic/update-success-stamp is updated by an apt.conf.d config entry | 10:07 |
| *** oanson has joined #openstack-ansible | 10:07 | |
| odyssey4me | yes, the module compares the timestamps of one of two options... neither of which appear to be very reliable | 10:07 |
| *** oanson has quit IRC | 10:07 | |
| evrardjp | I'll check that in details later today | 10:09 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client: Ensure upper constraints are respected https://review.openstack.org/393204 | 10:09 |
| evrardjp | brain melting right now :p | 10:10 |
| *** oanson has joined #openstack-ansible | 10:10 | |
| *** oanson has quit IRC | 10:10 | |
| *** oanson has joined #openstack-ansible | 10:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_heat: Simplify pip options/constraints mechanism https://review.openstack.org/393211 | 10:13 |
| *** oanson has quit IRC | 10:13 | |
| *** oanson has joined #openstack-ansible | 10:14 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_horizon: Simplify pip options/constraints mechanism https://review.openstack.org/393214 | 10:15 |
| *** oanson has quit IRC | 10:16 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Simplify pip options/constraints mechanism https://review.openstack.org/393216 | 10:17 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance: Simplify pip options/constraints mechanism https://review.openstack.org/393217 | 10:18 |
| *** oanson has joined #openstack-ansible | 10:19 | |
| *** oanson has quit IRC | 10:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_gnocchi: Simplify pip options/constraints mechanism https://review.openstack.org/393218 | 10:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_aodh: Simplify pip options/constraints mechanism https://review.openstack.org/393219 | 10:21 |
| *** oanson has joined #openstack-ansible | 10:22 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder: Simplify pip options/constraints mechanism https://review.openstack.org/393220 | 10:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Simplify pip options/constraints mechanism https://review.openstack.org/393222 | 10:24 |
| *** oanson has quit IRC | 10:25 | |
| *** oanson has joined #openstack-ansible | 10:27 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_magnum: Use centralised Ansible test scripts https://review.openstack.org/393224 | 10:28 |
| *** oanson has quit IRC | 10:28 | |
| *** kjw3 has quit IRC | 10:29 | |
| *** oanson has joined #openstack-ansible | 10:31 | |
| *** oanson has quit IRC | 10:34 | |
| *** haasn has quit IRC | 10:36 | |
| *** oanson has joined #openstack-ansible | 10:37 | |
| *** oanson has quit IRC | 10:37 | |
| *** haasn has joined #openstack-ansible | 10:38 | |
| *** oanson has joined #openstack-ansible | 10:42 | |
| *** oanson has quit IRC | 10:42 | |
| *** oanson has joined #openstack-ansible | 10:43 | |
| *** oanson has quit IRC | 10:43 | |
| *** oanson has joined #openstack-ansible | 10:45 | |
| *** oanson has quit IRC | 10:46 | |
| odyssey4me | FYI - not sure if someone can pick this up, but https://bugs.launchpad.net/openstack-ansible/+bug/1638883 | 10:48 |
| openstack | Launchpad bug 1638883 in openstack-ansible "Sahara testing no longer included in tempest" [High,New] | 10:48 |
| *** oanson has joined #openstack-ansible | 10:49 | |
| *** oanson has quit IRC | 10:49 | |
| *** oanson has joined #openstack-ansible | 10:50 | |
| *** oanson has quit IRC | 10:55 | |
| *** oanson has joined #openstack-ansible | 10:56 | |
| *** markvoelker has joined #openstack-ansible | 10:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Execute tempest using updated method https://review.openstack.org/371753 | 10:58 |
| *** oanson has quit IRC | 10:59 | |
| *** oanson has joined #openstack-ansible | 10:59 | |
| *** oanson has quit IRC | 11:02 | |
| *** markvoelker has quit IRC | 11:02 | |
| *** chhavi has quit IRC | 11:02 | |
| *** askb has quit IRC | 11:02 | |
| pjm6 | odyssey4me, in order to run the (./run_tests.sh) of each individual playbook, we first need to have the AIO bootstraped, right? | 11:04 |
| odyssey4me | pjm6 when role testing you do not bootstrap anything - you simply git clone the role, then execute run_tests.sh | 11:05 |
| pjm6 | ok odyssey4me, so i'm just doing something wrong, at least the os_horizon role fails | 11:05 |
| odyssey4me | pjm6 how does it fail? | 11:06 |
| pjm6 | odyssey4me, please let me rerun again and I will paste bin | 11:06 |
| pjm6 | but IIRC it fails when executing | 11:06 |
| pjm6 | ERROR: InvocationError: '/bin/bash -c /root/osa/openstack-ansible-os_horizon/tests/common/test-ansible-functional.sh' | 11:06 |
| pjm6 | and when I run that command, it says that the host is unreachable | 11:07 |
| odyssey4me | ok, so it's already running tox then | 11:08 |
| odyssey4me | if you execute 'tox -e functional' does it work? | 11:08 |
| odyssey4me | and can you pastebin the entire output | 11:09 |
| pjm6 | of tox -e functional? ok let me try that | 11:09 |
| odyssey4me | of either of them | 11:09 |
| pjm6 | just waiting for the results, HDD its so slow x.x | 11:13 |
| *** chhavi has joined #openstack-ansible | 11:15 | |
| *** haad1 has quit IRC | 11:22 | |
| *** haad1 has joined #openstack-ansible | 11:26 | |
| *** smatzek has joined #openstack-ansible | 11:27 | |
| pjm6 | odyssey4me, http://pastebin.com/5mHqdX9e | 11:28 |
| *** retreved has joined #openstack-ansible | 11:28 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Implement JIT container creation https://review.openstack.org/387675 | 11:32 |
| *** armaan has quit IRC | 11:35 | |
| *** hw_wutianwei has joined #openstack-ansible | 11:36 | |
| *** thorst has joined #openstack-ansible | 11:45 | |
| *** kjw3 has joined #openstack-ansible | 11:48 | |
| *** armaan has joined #openstack-ansible | 11:48 | |
| openstackgerrit | Joshua Hesketh proposed openstack/openstack-ansible-os_keystone: Add support for setting the role driver https://review.openstack.org/393259 | 11:51 |
| *** haad1 has quit IRC | 11:52 | |
| admin0 | anyone here users cache tiering on ceph ? | 11:53 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server: Remove dynamic includes from package installations https://review.openstack.org/392980 | 11:58 |
| openstackgerrit | Merged openstack/openstack-ansible: Add supported Operating System check https://review.openstack.org/392091 | 11:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] Implement JIT container creation https://review.openstack.org/387675 | 12:04 |
| *** LiYuenan has quit IRC | 12:07 | |
| *** johnmilton has joined #openstack-ansible | 12:07 | |
| *** oanson has joined #openstack-ansible | 12:12 | |
| oanson | evrardjp, hi. Are you in? | 12:13 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-haproxy_server: Remove dynamic includes from package installations https://review.openstack.org/393270 | 12:14 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ceilometer: Replace github with git.o.o https://review.openstack.org/393273 | 12:16 |
| oanson | logan-, hi, are you here? | 12:19 |
| mhayden | morning | 12:19 |
| oanson | mhayden, morning | 12:19 |
| oanson | Are you free for a quick question? | 12:19 |
| mhayden | sure -- not sure if i can answer it, but i'll try | 12:20 |
| oanson | I am trying to have dragonflow use etcd. I understood that etcd proxy is installed and available at 127.0.0.1:4001 | 12:21 |
| oanson | On the host, I see the etcd process running. | 12:21 |
| oanson | But I can;t connect. (even with wget). netstat doesn't show anyone listening on the relevant ports | 12:21 |
| oanson | I'm wondering if there's a config option I'm missing? | 12:21 |
| *** markvoelker has joined #openstack-ansible | 12:22 | |
| oanson | (I may have lied when I said quick (: ) | 12:22 |
| *** kjw3 has quit IRC | 12:24 | |
| pjm6 | mhayden, o/ | 12:26 |
| mgariepy | good morning everyone | 12:29 |
| mhayden | i didn't know we deployed etcd! :) | 12:31 |
| mhayden | which role deploys etcd? | 12:31 |
| odyssey4me | mhayden we don't usually - it gets deployed if you use calico | 12:31 |
| mhayden | ah, that's why i haven't seen it | 12:31 |
| oanson | mhayden, I don't know. I just thought it does. And I see the process running | 12:31 |
| mhayden | oanson: is etcd running on the host or within a container? | 12:32 |
| oanson | mhayden, on the host. | 12:32 |
| mhayden | hmm | 12:32 |
| oanson | Without a container. | 12:32 |
| oanson | And I'm trying to access it from the host, not within a container | 12:32 |
| mhayden | if you do "lsof -p <pid>" on the etcd pid -- is it listening anywhere? | 12:33 |
| odyssey4me | oanson is that on ubuntu xenial? if so, that may just be a default and have nothing to do with what we've deployed | 12:33 |
| odyssey4me | unless you've added the configuration of etcd into the applicable role/play? | 12:33 |
| oanson | odyssey4me, yes, ubuntu xenial | 12:33 |
| oanson | but i see etcd stuff in dragonflow_inventory: | 12:34 |
| odyssey4me | if it's the default, it's very possibly socket activated rather than listening via a network interface | 12:34 |
| oanson | etcd_all, etcd:children | 12:34 |
| odyssey4me | ok, those are just groups if I understand you correctly | 12:34 |
| odyssey4me | is there an entry in a role or playbook that consumes that group | 12:34 |
| oanson | I understood from logan- that there is a role deploying etcd. Available at 127.0.0.1:4001 or 2379. | 12:35 |
| oanson | I looked again, and I found it in the infra1 container. | 12:35 |
| odyssey4me | sure, there's a role - but what's activating that role in your deployment? | 12:35 |
| oanson | I thought having it in the inventory, and assigning it a host/container activates it? | 12:36 |
| odyssey4me | ah, you've assigned it to a host - that'll do it | 12:36 |
| oanson | I am using tox to deploy/test. I understood that downloads and applies all the other necessary roles | 12:36 |
| oanson | Is there something that needs etcd in infra1? Or can I move it to localhost? | 12:36 |
| oanson | Or do I just add localhost to [etcd_all] and then it will deploy on both? | 12:37 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Speed up V-38496 https://review.openstack.org/392966 | 12:37 |
| odyssey4me | that depends on what you're wanting to do - it all depends on your deployment architecture | 12:37 |
| oanson | Right now I'm using tox -e dragonflow from os_neutron. I am using the version in the patch: https://review.openstack.org/#/c/391524/ | 12:38 |
| odyssey4me | mhayden I just had a thought. You could, very possibly, do something funky like put all packages to be deployed into a var that builds up over time (maybe through a notifier), then execute the install at the end through a subsequent notifier | 12:38 |
| *** thorst is now known as thorst_afk | 12:39 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [Docs] Adjust docs for Ocata https://review.openstack.org/391964 | 12:40 |
| *** hw_wutianwei has quit IRC | 12:41 | |
| mhayden | odyssey4me: possibly | 12:41 |
| mhayden | odyssey4me: i thought about gathering all of the packages i need to install since i know what vars are set, and install them in one big swoop | 12:41 |
| mhayden | something like i did here https://review.openstack.org/#/c/391972/7/tasks/rhel7stig/packages.yml | 12:41 |
| *** jheroux has joined #openstack-ansible | 12:43 | |
| *** gouthamr has joined #openstack-ansible | 12:49 | |
| mhayden | odyssey4me: if you have a moment to review this gigantic diff, i'd be much obliged https://review.openstack.org/391903 | 12:49 |
| *** smatzek has quit IRC | 12:51 | |
| *** thorst_afk is now known as thorst | 12:51 | |
| *** schwicht has joined #openstack-ansible | 12:52 | |
| logan- | oanson: I'll take a look at your patch once I get to the office in about 30 mins | 12:54 |
| *** david-lyle has quit IRC | 12:55 | |
| odyssey4me | mhayden yep, something like that patch - although I wonder if you couldn't do some sort of setting-package mapping to simplify that into a loop of some sort | 12:55 |
| *** aheczko has joined #openstack-ansible | 12:55 | |
| oanson | logan-, thanks. | 12:55 |
| mhayden | odyssey4me: yeah, i thought about doing a with_items but i couldn't figure out how to deal with the conditionals | 12:56 |
| mhayden | not sure i can stick those conditionals into with_items | 12:56 |
| odyssey4me | argh, not with_items -rather a loop in jinja | 12:56 |
| mhayden | i guess that would be possible, but there are only 5-6 to remove there | 12:57 |
| odyssey4me | something more like ... lemme work up a sketch | 12:57 |
| openstackgerrit | Merged openstack/openstack-ansible-security: [Docs] Add 'only' to clarify status https://review.openstack.org/391903 | 12:59 |
| *** v1k0d3n has joined #openstack-ansible | 12:59 | |
| *** smatzek has joined #openstack-ansible | 13:01 | |
| evrardjp | oanson: logan- would be probably the one that can help you the most with the etcd part... According to your deployment graph, I think etcd should be part of infra1 in the tests, and all the compute nodes (here "localhost", the bare instance/metal) should have access to the etcd cluster | 13:02 |
| evrardjp | I think logan- told indeed there was a proxy installed on the compute nodes, so it's maybe worth looking with him how we would do that | 13:02 |
| oanson | evrardjp, that sounds great. | 13:03 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Speed up V-38496 https://review.openstack.org/392966 | 13:04 |
| evrardjp | this way you keep the infrastructure (what you call NB/SB) separate in infra, and the rest is just connecting to it | 13:04 |
| evrardjp | for the moment, the rest would be just one node | 13:04 |
| evrardjp | We probably have to look at how to bring ovs for proper bridging there, I didn't got the chance to look at it yet | 13:04 |
| oanson | evrardjp, this would also allow easier use in deployments where the etcd is in its own container as well | 13:04 |
| oanson | evrardjp, ovs is bought in. It's already installed, up, and running. | 13:05 |
| *** kjw3 has joined #openstack-ansible | 13:05 | |
| evrardjp | cool, I didn't check so it's good news | 13:05 |
| oanson | the br-int is missing, but I can look at how ml2.ovs configures it and do the same | 13:05 |
| *** cathrichardson has joined #openstack-ansible | 13:07 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 13:09 |
| odyssey4me | mhayden it's a sketch, and you may want to do it quite differently, but this is an idea: http://pastebin.com/78ywLNiB | 13:10 |
| *** klamath has joined #openstack-ansible | 13:10 | |
| *** klamath has quit IRC | 13:10 | |
| *** jmhale has joined #openstack-ansible | 13:10 | |
| odyssey4me | the settings var could be in the vars file, whereas the stuff to turn them on or off would be in the defaults | 13:11 |
| *** klamath has joined #openstack-ansible | 13:11 | |
| *** v1k0d3n has quit IRC | 13:12 | |
| *** haad1 has joined #openstack-ansible | 13:12 | |
| *** haasn has quit IRC | 13:13 | |
| mhayden | odyssey4me: oh, i think i see where you're going with that | 13:13 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Speed up V-38496 https://review.openstack.org/392966 | 13:13 |
| evrardjp | oanson: yes if we want to integrate dragonflow in openstack-ansible it would be good to properly adjust inventory and its attached vars | 13:13 |
| mhayden | odyssey4me: not a bad idea -- let me see what i can do there | 13:13 |
| automagically | Morning y’all | 13:13 |
| odyssey4me | logan- it looks like there may be a legit failure in https://review.openstack.org/392360 | 13:14 |
| *** v1k0d3n has joined #openstack-ansible | 13:14 | |
| *** haasn` has joined #openstack-ansible | 13:15 | |
| *** jperry has joined #openstack-ansible | 13:16 | |
| *** chhavi has quit IRC | 13:17 | |
| mhayden | odyssey4me: i'm wondering if variable scope would work correctly if i put it in vars/main.yml | 13:17 |
| mhayden | i don't think it could | 13:17 |
| mhayden | s/could/would/ | 13:17 |
| openstackgerrit | Merged openstack/openstack-ansible-security: [Docs] Adjust docs for Ocata https://review.openstack.org/391964 | 13:17 |
| openstackgerrit | Merged openstack/openstack-ansible-security: [Docs] Fix indentation for bullets https://review.openstack.org/391940 | 13:19 |
| odyssey4me | mhayden I think you'd have to duplicate a bit unfortunately - but you're creative enough to figure out something that works for you :) | 13:19 |
| mhayden | haha | 13:19 |
| mhayden | i'm glad you have faith in me | 13:19 |
| *** haasn` is now known as haasn | 13:19 | |
| *** victorhe has joined #openstack-ansible | 13:23 | |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Better functional tests for monasca-api. https://review.openstack.org/392515 | 13:23 |
| *** haad1 has quit IRC | 13:25 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Replace github with git.o.o https://review.openstack.org/392347 | 13:27 |
| openstackgerrit | Donovan Francesco proposed openstack/openstack-ansible-os_monasca: Better functional tests for monasca-api. https://review.openstack.org/392515 | 13:28 |
| *** aheczko has quit IRC | 13:29 | |
| *** jmhale is now known as hale | 13:29 | |
| *** haad1 has joined #openstack-ansible | 13:29 | |
| *** smatzek has quit IRC | 13:30 | |
| *** fguillot has joined #openstack-ansible | 13:31 | |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client: Replace github.com with git.o.o https://review.openstack.org/392340 | 13:32 |
| mgariepy | anyone here would have some times to help me figure out where the issue is in my centos test ? | 13:32 |
| drifterza | automagically, hi there :) | 13:33 |
| *** pegmanm_ has quit IRC | 13:33 | |
| automagically | o/ drifterza | 13:33 |
| *** Jack_Iv has quit IRC | 13:33 | |
| mgariepy | I cannot ping the router ips (10.1.3.x from the hosts or from infra1) but inside the netns it's working correctly, i can ssh to the instance) | 13:33 |
| mgariepy | router ips or floating ip. | 13:34 |
| *** schwicht has quit IRC | 13:35 | |
| mgariepy | evrardjp ^^ :D ? | 13:36 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-security: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392360 | 13:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Updated from global requirements https://review.openstack.org/392780 | 13:39 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-security: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392360 | 13:39 |
| oanson | logan-, hi, are you in? | 13:40 |
| logan- | hi oanson, yes i'm here now | 13:41 |
| oanson | logan-, I was wondering if you have any ideas about the etcd proxy? | 13:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client: Replace github.com with git.o.o https://review.openstack.org/393319 | 13:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Replace github with git.o.o https://review.openstack.org/393320 | 13:42 |
| logan- | one moment im going to spin up an instance and see how the patch looks once deployed | 13:42 |
| oanson | all right. Thanks | 13:43 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 13:44 |
| *** Jack_Iv has joined #openstack-ansible | 13:44 | |
| logan- | just to verify, you are using the neutron role tests to try to deploy this correct? | 13:44 |
| drifterza | anyone keen on reviewing : https://review.openstack.org/#/c/392515/ ? Please :D | 13:45 |
| logan- | ie the tox env defined here https://review.openstack.org/#/c/391524/13/tox.ini | 13:45 |
| oanson | yes | 13:45 |
| logan- | ok thanks | 13:45 |
| oanson | I use tox -e dragonflow | 13:45 |
| *** oanson has quit IRC | 13:45 | |
| *** oanson has joined #openstack-ansible | 13:46 | |
| mattt | can anyone remind me how to search infra's elasticsearch for gate logs? | 13:47 |
| odyssey4me | matt logstash.openstack.org | 13:47 |
| *** schwicht has joined #openstack-ansible | 13:48 | |
| mattt | there it is, thanks odyssey4me :) | 13:48 |
| *** michaelgugino has joined #openstack-ansible | 13:49 | |
| *** smatzek has joined #openstack-ansible | 13:50 | |
| *** chhavi has joined #openstack-ansible | 13:50 | |
| *** aheczko has joined #openstack-ansible | 13:52 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 13:53 |
| *** pcaruana has joined #openstack-ansible | 13:53 | |
| *** drifterza has quit IRC | 13:55 | |
| *** Mudpuppy has joined #openstack-ansible | 13:56 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 13:57 |
| mhayden | odyssey4me: here's my take on that -> https://review.openstack.org/#/c/393307/4/vars/redhat.yml | 13:57 |
| *** agrebennikov has joined #openstack-ansible | 13:58 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest: Allow user defined nova flavor resources https://review.openstack.org/393004 | 13:59 |
| oanson | logan-, hi, I have to go. | 13:59 |
| oanson | Could we continue this on Monday? | 13:59 |
| logan- | yes oanson. i will drop comments in the review | 14:00 |
| oanson | logan-, Great. Thanks a lot for your help! | 14:00 |
| logan- | np! | 14:00 |
| *** cathrichardson has quit IRC | 14:02 | |
| odyssey4me | mhayden commented | 14:04 |
| *** oanson has quit IRC | 14:05 | |
| *** TxGirlGeek has joined #openstack-ansible | 14:07 | |
| mhayden | odyssey4me: ah, i see what you're saying | 14:10 |
| odyssey4me | mhayden either the with_items needs the complete list, or the packages needs to be a complete list | 14:11 |
| mhayden | let me see what i can work up | 14:11 |
| mhayden | odyssey4me: it seems like i'd need 2 loops -- one for installation and one for removal | 14:13 |
| odyssey4me | yeah, that would be the case | 14:13 |
| *** mheremans has joined #openstack-ansible | 14:14 | |
| *** dxiri has joined #openstack-ansible | 14:17 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 14:17 |
| *** hale is now known as jmhale | 14:18 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 14:21 |
| *** chris_hultin|AWA is now known as chris_hultin | 14:21 | |
| *** yolanda has quit IRC | 14:22 | |
| *** yolanda has joined #openstack-ansible | 14:23 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_keystone: Implement zero downtime upgrades https://review.openstack.org/384269 | 14:24 |
| cloudnull | mornings | 14:26 |
| *** spotz_zzz is now known as spotz | 14:27 | |
| admin0 | \o | 14:27 |
| *** johnmilton has quit IRC | 14:28 | |
| cloudnull | o/ admin0 | 14:28 |
| *** johnmilton has joined #openstack-ansible | 14:28 | |
| cloudnull | dxiri: did you ever get the console bits going ? | 14:33 |
| *** yolanda has quit IRC | 14:33 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_magnum: Create magnum flavor using tempest role https://review.openstack.org/393345 | 14:34 |
| dxiri | cloudnull: yes! | 14:34 |
| dxiri | memcached service was down | 14:34 |
| *** Mudpuppy has quit IRC | 14:34 | |
| dxiri | after starting it, it started working | 14:34 |
| dxiri | can you explain what memcached has to do with the console and enlight a noob? | 14:35 |
| *** Mudpuppy has joined #openstack-ansible | 14:36 | |
| *** adrian_otto has joined #openstack-ansible | 14:36 | |
| *** adrian_otto has quit IRC | 14:36 | |
| cloudnull | odyssey4me: re: https://review.openstack.org/#/c/393145/ -- we could do vcpus however we still want the number of forks to be something sane. IE if i have a dual proc 16 core machine with hyperthreading I don't believe we want 64 builds happening all at the same time. | 14:37 |
| odyssey4me | cloudnull sure, that's sensible - then maybe a lower and upper bound? | 14:37 |
| cloudnull | yea. i get on that. | 14:38 |
| odyssey4me | maybe 2-8 based on the vcpu value | 14:38 |
| *** yolanda has joined #openstack-ansible | 14:38 | |
| spotz | morning all | 14:38 |
| cloudnull | also I think we should implement something like https://review.openstack.org/#/c/392969/ in our other places where we do address discovery | 14:38 |
| cloudnull | it seems to be quite a bit faster. | 14:39 |
| odyssey4me | 2 would cater for the horrible slowness when using vagrant on a laptop | 14:39 |
| cloudnull | dxiri: nova-consoleauth has a hard dependency on memcached when on a distributed system . | 14:39 |
| odyssey4me | and yes, I absolutely agree that we should do much more of that wherever we find opporunities to squash tasks | 14:39 |
| cloudnull | it auth's against keyston and then stores a token | 14:39 |
| cloudnull | dxiri: http://docs.openstack.org/admin-guide/compute-remote-console-access.html | 14:40 |
| odyssey4me | dxiri memcache does all the token caching across all of openstack, the cache not only speeds up repeated validation requests, but also allows failover between the same services without losing the existing sessions | 14:40 |
| cloudnull | that has a good diagram within it, and while it's not super apparent the "cache connection information and token" part is all memcached. | 14:41 |
| openstackgerrit | Merged openstack/openstack-ansible: Replace github with git.o.o https://review.openstack.org/392339 | 14:43 |
| openstackgerrit | Merged openstack/openstack-ansible: Remove outdated aio heat template https://review.openstack.org/393015 | 14:43 |
| dxiri | thanks cloudnull, odyssey4me. I was on that page cloudnull posted and it doesn't mention anything about memcached or keystone. Much appreciated for the explanation :) | 14:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Replace github with git.o.o https://review.openstack.org/393350 | 14:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Remove outdated aio heat template https://review.openstack.org/393351 | 14:44 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393145 | 14:45 |
| cloudnull | ^ odyssey4me i think that'll do it | 14:45 |
| stevelle | cloudnull: odyssey4me on https://review.openstack.org/#/c/392969 wondering if that could be rolled nicely into a network discovery module instead, as I find the proposed change to be a bit impenetrable using that block of jinja in playbook. if it's going to be opaque anyway, maybe move to a module? | 14:47 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Update branch names for Ocata https://review.openstack.org/392517 | 14:47 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393145 | 14:47 |
| cloudnull | sorry needed another set of parens | 14:48 |
| *** haad1 has quit IRC | 14:48 | |
| cloudnull | stevelle: that's not a bad idea | 14:48 |
| odyssey4me | stevelle using modules for wherever we use scripts and that sort of thing would make sense as a target, using jinja instead of multiple tasks is an improvement for speed | 14:48 |
| odyssey4me | and honestly I find the jinja more penetrable than the previous set of tasks! | 14:49 |
| *** drifterza has joined #openstack-ansible | 14:49 | |
| cloudnull | maybe we can just roll it into a single common task | 14:49 |
| stevelle | module should be even faster :D | 14:49 |
| cloudnull | while we really only use this for cinder in the playbooks | 14:50 |
| odyssey4me | realistically if we really want speed, we need to use more modules, less tasks and less scripts | 14:50 |
| cloudnull | the swift role has a very similar method for the repl and storage networks. | 14:50 |
| odyssey4me | isn't there one for neutron and nova too? | 14:50 |
| drifterza | helllo | 14:50 |
| cloudnull | I think nova has something similar for the management addr. | 14:50 |
| cloudnull | i'm not sure about neutron | 14:50 |
| * cloudnull looking | 14:50 | |
| cloudnull | ohai drifterza | 14:51 |
| drifterza | cloudnull, +2 on your blog post dude. | 14:51 |
| drifterza | nice work | 14:51 |
| odyssey4me | cloudnull processor count is cores, right? why not vcpu count in https://review.openstack.org/#/c/393145/3/defaults/main.yml ? | 14:52 |
| *** aheczko has quit IRC | 14:53 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393145 | 14:53 |
| cloudnull | done. | 14:53 |
| cloudnull | count on an intel die is cores, on amd is procs, im not sure what it is on ppc | 14:54 |
| cloudnull | thanks ! :) more soon. | 14:54 |
| odyssey4me | blog post? link? | 14:54 |
| cloudnull | https://cloudnull.io/2016/11/multi-hypervisor-openstack-cloud/ | 14:55 |
| cloudnull | here's the general collection https://cloudnull.io/tag/multi-hypervisor-cloud/ | 14:56 |
| cloudnull | i have a post for lxd which i'm not done with | 14:56 |
| *** hw_wutianwei has joined #openstack-ansible | 14:56 | |
| smatzek | adreznec, you dug into Ansible for ppc64le with regards to vcpus, core counts, etc, can you and answer the question above for ppc? | 14:56 |
| cloudnull | drifterza: I'm working on IOMMU and SR-IOV as well. so more on that soon. :) | 14:57 |
| drifterza | nice | 14:57 |
| odyssey4me | smatzek adreznec well, IIRC the reason there was a default there was due to ppc64le not responding with anything useful on the cpu count... we retain a default there so there's no real risk | 14:57 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Fix bind mount of glance images https://review.openstack.org/393358 | 14:58 |
| odyssey4me | cloudnull oh neat - you got some kit for it? | 14:58 |
| cloudnull | I did | 14:58 |
| cloudnull | https://cloudnull.io/2016/11/home-server-episode-1/ | 14:58 |
| adreznec | Reading backscroll quick | 14:58 |
| cloudnull | not crazy expensive either. | 14:58 |
| *** karimb has quit IRC | 14:59 | |
| adreznec | But yeah odyssey4me that's the net of the issue. I proposed a fix for that to ansible proper and it was accepted | 14:59 |
| adreznec | But I'm not sure what version it's in | 14:59 |
| *** oanson has joined #openstack-ansible | 14:59 | |
| *** karimb has joined #openstack-ansible | 15:00 | |
| adreznec | With that ansible_processor_count on ppc64le essentially the count of procs that come out of /proc/cpuinfo | 15:02 |
| adreznec | Without that fix it just gets returned as zero | 15:02 |
| adreznec | hence the need for the default | 15:03 |
| *** TxGirlGeek has quit IRC | 15:03 | |
| openstackgerrit | Merged openstack/openstack-ansible-security: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392360 | 15:03 |
| *** TxGirlGeek has joined #openstack-ansible | 15:04 | |
| drifterza | cloudnull, you keen to review this quick ? : https://review.openstack.org/#/c/392515/ | 15:04 |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible-os_neutron: Fix OVS kernel module vars https://review.openstack.org/393360 | 15:05 |
| cloudnull | drifterza: done :) | 15:06 |
| adreznec | FWIW the fix is in 2.2, so when we make it there we wouldn't need the workaround for ppc64le anymore | 15:07 |
| cloudnull | nice | 15:08 |
| *** TxGirlGeek has quit IRC | 15:10 | |
| cloudnull | anyone want to punt this through https://review.openstack.org/#/c/388864/ | 15:10 |
| *** TxGirlGeek has joined #openstack-ansible | 15:10 | |
| mgariepy | odyssey4me, http://paste.openstack.org/show/587792/ ;) | 15:10 |
| odyssey4me | mgariepy is that the last hurdle? | 15:11 |
| odyssey4me | oh hang on - that passed? | 15:11 |
| mgariepy | I dont know. | 15:12 |
| mgariepy | lol | 15:12 |
| odyssey4me | it failed once, then retried and passed | 15:12 |
| mgariepy | yeah it passed but it's really unstable. | 15:12 |
| mgariepy | haha | 15:12 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392355 | 15:13 |
| mgariepy | neutron linuxbridge_agent.ini.j2 physical_interface_mappings isn't defined for some reason on openstack1 container. | 15:13 |
| mgariepy | I don't know why | 15:13 |
| mgariepy | nova-compute is working-ish but sometimes it seems to require restart to complete stuff.. | 15:14 |
| mgariepy | i had to restart nova-compute to make the tempest test pass | 15:15 |
| *** chhavi has quit IRC | 15:15 | |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-haproxy_server: Added ACL module in service.conf.j2 https://review.openstack.org/392328 | 15:15 |
| *** haad1 has joined #openstack-ansible | 15:16 | |
| *** pegmanm_ has joined #openstack-ansible | 15:17 | |
| *** weezS has joined #openstack-ansible | 15:18 | |
| *** TxGirlGeek has quit IRC | 15:19 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Removing extras/ https://review.openstack.org/384676 | 15:19 |
| *** TxGirlGeek has joined #openstack-ansible | 15:19 | |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-haproxy_server: Added ACL module in service.conf.j2 https://review.openstack.org/392328 | 15:21 |
| openstackgerrit | Nish Patwa(nishpatwa_) proposed openstack/openstack-ansible-haproxy_server: Added ACL module in service.conf.j2 https://review.openstack.org/392328 | 15:22 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392648 | 15:24 |
| *** TxGirlGeek has quit IRC | 15:25 | |
| *** aheczko has joined #openstack-ansible | 15:25 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_nova: Add CentOS support for os_nova https://review.openstack.org/378963 | 15:27 |
| *** pegmanm_ has quit IRC | 15:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic: Simplify pip options/constraints mechanism https://review.openstack.org/392654 | 15:27 |
| mgariepy | the neutron missing mappings is because of my ansible run since i ran the commands in manual-test.rc which doesn't include overrides. | 15:28 |
| drifterza | mhayden, odyssey4me could one of you guys do a +W on my PR please ?: https://review.openstack.org/#/c/392515/ | 15:32 |
| * mhayden ganders | 15:32 | |
| drifterza | thanks dude. | 15:32 |
| nishpatwa007 | Hi Can anyone review this patch https://review.openstack.org/#/c/383822/ | 15:33 |
| *** TxGirlGeek has joined #openstack-ansible | 15:34 | |
| *** hw_wutianwei has quit IRC | 15:37 | |
| *** hw_wutianwei has joined #openstack-ansible | 15:38 | |
| *** TxGirlGeek has quit IRC | 15:39 | |
| openstackgerrit | Samuel Matzek proposed openstack/openstack-ansible-os_swift: Add object-reconstructor for swift EC https://review.openstack.org/393374 | 15:39 |
| *** TxGirlGe_ has joined #openstack-ansible | 15:39 | |
| agrebennikov | hey drifterza qq | 15:40 |
| agrebennikov | https://github.com/drifterza/openstack-ansible-os_elasticsearch/blob/master/templates/elasticsearch_hosts_file.j2#L4 | 15:40 |
| agrebennikov | I'm working on refactoring the elastic role to fit into openstack-ansible | 15:40 |
| agrebennikov | jsut wanted to ask why it was done this way in your repo | 15:40 |
| drifterza | umn | 15:41 |
| *** hw_wutianwei has quit IRC | 15:41 | |
| *** hw_wutianwei has joined #openstack-ansible | 15:41 | |
| drifterza | oh right, thats because of clustering | 15:42 |
| drifterza | https://github.com/drifterza/openstack-ansible-os_elasticsearch/blob/master/templates/elasticsearch.yml.j2#L42-L50 | 15:42 |
| drifterza | elasticsearch uses hostnames | 15:42 |
| drifterza | and that was the only way I could get it to work without dns entries | 15:42 |
| drifterza | make sense ? | 15:43 |
| agrebennikov | drifterza, not really, it should be {{ hostvars[host]['container_address'] }} {{ host }} | 15:44 |
| drifterza | thats true, that could also work. | 15:44 |
| *** TxGirlGe_ has quit IRC | 15:45 | |
| *** adrian_otto has joined #openstack-ansible | 15:45 | |
| agrebennikov | also this should be completely refactored anyway since it destroys existing hosts file | 15:45 |
| agrebennikov | the problem with "eth1" is it is not necessarily eth1)) | 15:45 |
| drifterza | you are more than welcome to do a pull request | 15:45 |
| agrebennikov | I'm doping right now)) | 15:46 |
| agrebennikov | *doing | 15:46 |
| drifterza | okay awesome ;) | 15:46 |
| agrebennikov | just wanted to confirm | 15:46 |
| agrebennikov | also did you get success with working out kibana together with elastic? | 15:46 |
| *** TxGirlGeek has joined #openstack-ansible | 15:47 | |
| drifterza | yes i did | 15:47 |
| *** aheczko has quit IRC | 15:47 | |
| drifterza | Umn, I can upload my kibana role | 15:47 |
| drifterza | if you want | 15:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Fix linting issues for ansible-lint 3.4.1 https://review.openstack.org/392605 | 15:49 |
| openstackgerrit | Merged openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393145 | 15:50 |
| agrebennikov | drifterza, sure)) | 15:50 |
| *** m4rx has quit IRC | 15:51 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [docs] Update branch names for Ocata https://review.openstack.org/392517 | 15:51 |
| drifterza | you can fork this : https://github.com/drifterza/openstack-ansible-os_kibana | 15:52 |
| drifterza | let me galaxy it aswell | 15:52 |
| agrebennikov | cool, tnx | 15:53 |
| agrebennikov | does it set up apache auth for it? | 15:53 |
| openstackgerrit | Merged openstack/openstack-ansible-os_monasca: Better functional tests for monasca-api. https://review.openstack.org/392515 | 15:53 |
| drifterza | umn, no I don't think I did auth yet | 15:53 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Replace github with git.o.o https://review.openstack.org/393350 | 15:55 |
| mhayden | OSA weekly meeting in #openstack-meeting-4 in a few minutes! cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, erikmwilson, mancdaz, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung, prometheanfire, evrardjp, arbrandes, mhayden, scarlisle, luckyinva, ntt, javeriak, automagically, spotz, vdo, jmccrory, alextricity25, | 15:55 |
| mhayden | jasondotstar, admin0, michaelgugino, ametts, v1k0d3n, severion, bgmccollum, darrenc, JRobinson__, asettle, colinmcnamara, thorst, adreznec, eil397, qwang,nishpatwa_ | 15:55 |
| drifterza | https://galaxy.ansible.com/drifterza/openstack-ansible-os_kibana | 15:55 |
| drifterza | agrebennikov, got nuts :) | 15:55 |
| drifterza | I mean go nuts * | 15:56 |
| drifterza | lol | 15:56 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_build: Statically define the repo build concurrency https://review.openstack.org/393383 | 15:56 |
| mhayden | d34dh0r53: ^^ | 15:56 |
| mhayden | drifterza: ol' d34dh0r53 loves some kibana | 15:56 |
| d34dh0r53 | :) | 15:57 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_ironic: Simplify pip options/constraints mechanism https://review.openstack.org/393385 | 15:57 |
| d34dh0r53 | I was thinking about moving our ELK stuff to openstack-ansible-ops | 15:57 |
| drifterza | I didn't know we were considering ELK as an integrated role | 15:58 |
| drifterza | it would be cool though | 15:59 |
| admin0 | if you ask me, i say its a must :) | 15:59 |
| admin0 | we want osa to be in production .. production does not end with just openstack .. monitoring, backup, alerting, capacity planning :) | 15:59 |
| drifterza | I did those roles a few months ago | 16:00 |
| drifterza | but got distracted with monasca | 16:00 |
| d34dh0r53 | yeah, there is some really sweet monitoring/graphing stuff in osa-ops which cloudnull did, so I was thinking that the ELK stuff that I work on would be a good fit | 16:00 |
| admin0 | anyone needs multi node clusters to test stuff — i got 3 of them now :D | 16:00 |
| *** cathrichardson has joined #openstack-ansible | 16:01 | |
| agrebennikov | drifterza, we should merge our activity :/ | 16:02 |
| drifterza | Agreed, d34dh0r53 ? | 16:02 |
| logan- | d34dh0r53: would be great to see those roles broken out into separate repos at least | 16:02 |
| agrebennikov | spending 90% of time on monasca rol | 16:03 |
| agrebennikov | role | 16:03 |
| logan- | to make them more consumable | 16:03 |
| drifterza | monaca is a very large beast | 16:03 |
| d34dh0r53 | agrebennikov: agreed | 16:03 |
| d34dh0r53 | logan-: yeah I totally agree | 16:03 |
| d34dh0r53 | I'll get to work on it early next week | 16:04 |
| mhayden | #startmeeting openstack_ansible_meeting | 16:04 |
| openstack | Meeting started Thu Nov 3 16:04:15 2016 UTC and is due to finish in 60 minutes. The chair is mhayden. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:04 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:04 |
| openstack | The meeting name has been set to 'openstack_ansible_meeting' | 16:04 |
| mhayden | #topic Roll call | 16:04 |
| *** stevelle is now known as qwebirc78377 | 16:04 | |
| *** qwebirc78377 is now known as stevelle | 16:04 | |
| mhayden | #endmeeting | 16:04 |
| openstack | Meeting ended Thu Nov 3 16:04:48 2016 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 16:04 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2016/openstack_ansible_meeting.2016-11-03-16.04.html | 16:04 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2016/openstack_ansible_meeting.2016-11-03-16.04.txt | 16:04 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/openstack_ansible_meeting/2016/openstack_ansible_meeting.2016-11-03-16.04.log.html | 16:04 |
| admin0 | :D | 16:05 |
| admin0 | awesome meeting notes and action items | 16:05 |
| admin0 | sorry .. found it funny | 16:05 |
| mhayden | we're over in #openstack-meeting-4 | 16:06 |
| mhayden | a previous meeting ran over | 16:06 |
| admin0 | now a days meeting always run between my driving hours :( | 16:06 |
| *** h5t4 has quit IRC | 16:07 | |
| *** david-lyle has joined #openstack-ansible | 16:10 | |
| *** hughmFLEXin has joined #openstack-ansible | 16:12 | |
| *** haad1 has quit IRC | 16:14 | |
| *** maximov_ has quit IRC | 16:15 | |
| *** maximov_ has joined #openstack-ansible | 16:15 | |
| *** haad1 has joined #openstack-ansible | 16:20 | |
| *** pcaruana has quit IRC | 16:20 | |
| admin0 | have to run :( driving time .. .. always miss these meeting .. will check from the notes .. | 16:22 |
| *** McMurlock1 has quit IRC | 16:28 | |
| *** jlockwood has joined #openstack-ansible | 16:30 | |
| *** karimb has quit IRC | 16:31 | |
| *** david-lyle has quit IRC | 16:32 | |
| *** david-lyle has joined #openstack-ansible | 16:33 | |
| *** f13o_ has quit IRC | 16:36 | |
| *** f13o_ has joined #openstack-ansible | 16:36 | |
| *** jmhale- has joined #openstack-ansible | 16:40 | |
| *** chhavi has joined #openstack-ansible | 16:40 | |
| *** jmhale- has quit IRC | 16:46 | |
| *** dogsec has joined #openstack-ansible | 16:46 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: [WIP] Speed up package install https://review.openstack.org/393307 | 16:47 |
| dxiri | guys quick question, can you use ceph with openstack ansible? | 16:50 |
| dxiri | as in, can it configure it and use it? | 16:50 |
| palendae | Rackspace uses https://github.com/ceph/ceph-ansible, as does logan- with https://logan.protiumit.com/2016/08/02/openstack-ansible-ceph.html | 16:51 |
| palendae | But I myself haven't touched anything with ceph so I'm not the best to answer questions | 16:52 |
| palendae | There are not, however, any OSA roles for Ceph | 16:52 |
| stevelle | logan-: I'm in agreement that improving the role testing would be helpful there, so that we can ensure we can deploy the keystone role in a non-degenerate state | 16:53 |
| logan- | yep! | 16:54 |
| logan- | dxiri: as palendae said, no ceph deployment is not integrated into OSA at this point. OSA can consume an existing ceph cluster only. but there are some examples integrating the two projects, and I use the methods outlined in that blog article to deploy both ceph-ansible and OSA in the same environment/inventory. | 16:56 |
| *** pester has quit IRC | 16:57 | |
| dxiri | logan-: great thanks :) | 16:58 |
| logan- | if you have any questions on how the pieces fit together feel free to ping me | 16:58 |
| *** karimb has joined #openstack-ansible | 17:00 | |
| *** hw_wutianwei has quit IRC | 17:00 | |
| odyssey4me | palendae logan- yep, we can improve the tests to ensure that memcache is deployed (if it isn't already) | 17:00 |
| *** cathrichardson has quit IRC | 17:02 | |
| *** h5t4 has joined #openstack-ansible | 17:03 | |
| *** karimb has quit IRC | 17:03 | |
| *** gouthamr has quit IRC | 17:05 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add supported Operating System check https://review.openstack.org/392089 | 17:05 |
| *** schwicht has quit IRC | 17:05 | |
| *** schwicht has joined #openstack-ansible | 17:06 | |
| odyssey4me | palendae lbragstad dolphm evrardjp please see feedback in https://review.openstack.org/#/c/326748/ :) it'd be good to understand the recommended change here, because our integrated build already implements the memcache back-end by default... the role doesn't do so on purpose because that should be a decision taken by the deployer consuming the role | 17:06 |
| *** haad1 has quit IRC | 17:06 | |
| odyssey4me | (when the role is used independently) | 17:06 |
| odyssey4me | also, the default tests do deploy memcache: https://github.com/openstack/openstack-ansible-tests/blob/master/test-install-infra.yml#L16 | 17:07 |
| *** mhereman1 has joined #openstack-ansible | 17:08 | |
| *** gouthamr has joined #openstack-ansible | 17:08 | |
| odyssey4me | ah, but the tests don't wire it up | 17:08 |
| odyssey4me | right, that we can fix | 17:09 |
| *** h5t4 has quit IRC | 17:13 | |
| *** h5t4 has joined #openstack-ansible | 17:14 | |
| evrardjp | odyssey4me: on this issue it's also my fault, I said I wanted to help, I didn't got the chance to look at group_vars and everything, so it's kinda a not finished work :p | 17:15 |
| evrardjp | I will take all the blame! | 17:15 |
| evrardjp | odyssey4me: the question is , I guess, should we enable this by default or not in os_keystone role testing | 17:17 |
| evrardjp | maybe we should not have this dependency on memcache in the meta/ at all | 17:18 |
| odyssey4me | It looks to me like the meta dep is there specifically for *not* fernet... (ie UUID) | 17:19 |
| odyssey4me | but perhaps the UUID token caching should be normalised to work in the same way | 17:19 |
| odyssey4me | ie remove the meta-dep, change https://github.com/openstack/openstack-ansible-os_keystone/blob/master/templates/keystone.conf.j2#L38 to use the generally deployed memcache servers, not the list of keystone servers | 17:20 |
| odyssey4me | but I don't know what the best practise is here | 17:20 |
| odyssey4me | to make memcache caching enabled for our role tests, we just need https://github.com/openstack/openstack-ansible/blob/master/playbooks/inventory/group_vars/keystone_all.yml#L18 either set in openstack-ansible-tests/test-vars.yml, or as a specific keystone role test override | 17:21 |
| odyssey4me | it probably makes the most sense to do it for all roles, so the central test-vars likely makes the most sense | 17:22 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Add supported Operating System check https://review.openstack.org/392478 | 17:22 |
| evrardjp | odyssey4me: 100% agreed this time | 17:24 |
| *** hughmFLEXin has quit IRC | 17:24 | |
| evrardjp | it's the best for our community and for our gates | 17:24 |
| evrardjp | it's also conform with my review :D | 17:24 |
| *** weezS has quit IRC | 17:25 | |
| *** hughmFLEXin has joined #openstack-ansible | 17:25 | |
| evrardjp | so you would not enable this, keep the conditional in the template, and have this overriden in the test repo? | 17:30 |
| *** javeriak has joined #openstack-ansible | 17:33 | |
| odyssey4me | evrardjp personally, yes - I think there's nothing wrong and all that needs to happen is that the role is configured right via the vars given it | 17:33 |
| odyssey4me | when uuid goes, we can ditch the dep | 17:34 |
| *** dxiri_ has joined #openstack-ansible | 17:34 | |
| *** dxiri has quit IRC | 17:34 | |
| evrardjp | well I don't see why we shouldn't ditch the dep already | 17:35 |
| evrardjp | it's a terrible idea to force collocation | 17:35 |
| evrardjp | more of it if it's not used | 17:35 |
| evrardjp | :p | 17:36 |
| *** adogrant has quit IRC | 17:36 | |
| evrardjp | ditch the dependency, and let our playbooks + inventory vars do the automatic thing | 17:36 |
| *** chhavi_ has joined #openstack-ansible | 17:36 | |
| *** chhavi has quit IRC | 17:37 | |
| odyssey4me | if you can re-jig the uuid token config to make that work, and add a release note to highlight the design change, then I would support that | 17:38 |
| *** pjm6 has quit IRC | 17:38 | |
| odyssey4me | deployers can still opt to colocate the two services if they want to, but they will have to use joint groups for the services or execute the memcache role against the keystone hosts and override the vars | 17:39 |
| odyssey4me | so we're not precluding it | 17:39 |
| *** f13o_ has quit IRC | 17:39 | |
| *** hughmFLE_ has joined #openstack-ansible | 17:40 | |
| *** hughmFLEXin has quit IRC | 17:42 | |
| *** hughmFLE_ has quit IRC | 17:46 | |
| *** hughmFLEXin has joined #openstack-ansible | 17:48 | |
| *** weezS has joined #openstack-ansible | 17:49 | |
| *** sdake has joined #openstack-ansible | 18:05 | |
| *** chhavi_ has quit IRC | 18:07 | |
| *** cathrichardson has joined #openstack-ansible | 18:08 | |
| *** retreved has quit IRC | 18:08 | |
| *** retreved has joined #openstack-ansible | 18:10 | |
| *** hughmFLE_ has joined #openstack-ansible | 18:10 | |
| *** chris_hultin is now known as chris_hultin|AWA | 18:12 | |
| *** hughmFLEXin has quit IRC | 18:12 | |
| mhayden | odyssey4me: i finally got that darned jinja2 loopdeloop working https://review.openstack.org/393307 | 18:15 |
| *** chris_hultin|AWA is now known as chris_hultin | 18:17 | |
| odyssey4me | mhayden so that does two loops - one for present and one for absent? | 18:18 |
| odyssey4me | dang, it does | 18:19 |
| odyssey4me | that's some neat work there :) | 18:19 |
| odyssey4me | it halves the code | 18:20 |
| mhayden | indeedydo | 18:21 |
| mhayden | took a lot of monkeying around to get that | 18:22 |
| mhayden | groupby is funky | 18:22 |
| *** sdake has quit IRC | 18:22 | |
| odyssey4me | I think that's pretty neat, and saves a bunch of maintenance | 18:22 |
| odyssey4me | I've added a comment just to make things easier to grok. | 18:23 |
| odyssey4me | it seems like you're shaving 10-30 % off the time there? | 18:23 |
| admin0 | shaving :D | 18:25 |
| *** mhereman1 has quit IRC | 18:26 | |
| mhayden | possibly | 18:27 |
| *** oanson has quit IRC | 18:27 | |
| * odyssey4me is waiting for automagically and jmccrory to do another review run :) | 18:29 | |
| *** f13o has joined #openstack-ansible | 18:29 | |
| openstackgerrit | Ravi Gummadi proposed openstack/openstack-ansible-galera_server: Fix galera packages installation failure https://review.openstack.org/393440 | 18:31 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Speed up package install/removal https://review.openstack.org/393307 | 18:32 |
| mhayden | odyssey4me: fixed it | 18:33 |
| *** adrian_otto has quit IRC | 18:34 | |
| *** RaviGummadi has joined #openstack-ansible | 18:35 | |
| mgariepy | odyssey4me, https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tasks/neutron_init_systemd.yml#L35 | 18:36 |
| *** hughmFLEXin has joined #openstack-ansible | 18:36 | |
| odyssey4me | mgariepy ? | 18:36 |
| mgariepy | only the last service gets the entry in the file. | 18:36 |
| openstackgerrit | Merged openstack/openstack-ansible-ops: Replace github with git.o.o https://review.openstack.org/393134 | 18:36 |
| mgariepy | https://github.com/openstack/openstack-ansible-os_neutron/blob/master/tasks/neutron_init.yml | 18:37 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server: Added ACL module in service.conf.j2 https://review.openstack.org/392328 | 18:37 |
| odyssey4me | haha, oh dear - bad bug | 18:37 |
| mgariepy | haha | 18:37 |
| openstackgerrit | Merged openstack/openstack-ansible-plugins: Replace github with git.o.o https://review.openstack.org/393136 | 18:37 |
| mgariepy | doesn't make debugging centos easy haha | 18:37 |
| odyssey4me | hmm, although - doesn't the template handle it properly? | 18:37 |
| *** GheRivero has joined #openstack-ansible | 18:37 | |
| mgariepy | depend if all the neutron sevices runs in the same container | 18:37 |
| odyssey4me | heh, nope | 18:38 |
| odyssey4me | ok, that's a bug | 18:38 |
| mgariepy | like in the gate check | 18:38 |
| *** oanson has joined #openstack-ansible | 18:38 | |
| mgariepy | renaming the file to neutron-sub-service.conf ? | 18:38 |
| *** hughmFLE_ has quit IRC | 18:38 | |
| mgariepy | i can do the patch if you want. | 18:38 |
| odyssey4me | uh, not sure - push up a patch and let's see what people who know better say | 18:39 |
| odyssey4me | perhaps just {{ program_name }}.conf | 18:39 |
| odyssey4me | yeah, that seems to make the most sense | 18:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_heat: Simplify pip options/constraints mechanism https://review.openstack.org/393211 | 18:40 |
| mhayden | if i could get one more core review on https://review.openstack.org/391910, i'd be much obliged | 18:40 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Implement LXC cache timeout https://review.openstack.org/393138 | 18:41 |
| openstackgerrit | Merged openstack/openstack-ansible-haproxy_server: Remove dynamic includes from package installations https://review.openstack.org/393270 | 18:41 |
| *** asettle has joined #openstack-ansible | 18:41 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_neutron: Fix tempfile when multiple service are running. https://review.openstack.org/393442 | 18:41 |
| odyssey4me | mgariepy I suspect that issue applies to other roles too, so we should probably replicate and backport that one. | 18:41 |
| odyssey4me | but let's get a sample merged first | 18:42 |
| mgariepy | ok | 18:42 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client: Ensure upper constraints are respected https://review.openstack.org/393204 | 18:42 |
| odyssey4me | mhayden do you know if a config change in /etc/tmpfiles.d/*.conf should require some sort of systemd reload? | 18:43 |
| mhayden | hmm, that's a good question -- i haven't mucked in there much | 18:46 |
| mhayden | odyssey4me: from the looks of the docs, a restart isn't req'd | 18:46 |
| mgariepy | odyssey4me, what about rename from tempfile.d to tmpfiles.d at the same time ? | 18:47 |
| *** cathrichardson has quit IRC | 18:47 | |
| odyssey4me | mhayden would the service need to be restarted though? | 18:47 |
| mhayden | doesn't look like it | 18:47 |
| mgariepy | the 2 tasks above create the dirs described in the conf file. | 18:48 |
| odyssey4me | mgariepy commented - yeah, may as well fix up the name | 18:48 |
| *** b3rn-n00dl3s has quit IRC | 18:49 | |
| *** b3rn-n00dl3s has joined #openstack-ansible | 18:50 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_neutron: Fix tempfile when multiple service are running. https://review.openstack.org/393442 | 18:50 |
| *** pwnall1337 has joined #openstack-ansible | 18:51 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_neutron: Fix tmpfiles.d when multiple service are running. https://review.openstack.org/393442 | 18:51 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Simplify pip options/constraints mechanism https://review.openstack.org/393216 | 18:52 |
| mgariepy | I'll wait for this one to merge and submit fixes for all the other roles. | 18:53 |
| odyssey4me | thanks mgariepy | 18:53 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Simplify pip options/constraints mechanism https://review.openstack.org/393214 | 18:53 |
| odyssey4me | the structure is slightly different for many of the other roles - this just reminded me to update the os_neutron role to be in line with those | 18:53 |
| odyssey4me | we flattened the structure a bit and implemented smarter handlers | 18:54 |
| mgariepy | unless you think i should push them right now ? | 18:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi: Replace github with git.o.o https://review.openstack.org/393137 | 18:54 |
| odyssey4me | mhayden your thoughts on https://review.openstack.org/393442 ? | 18:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh: Use updated get_url module checksum functionality https://review.openstack.org/392955 | 18:56 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh: Simplify pip options/constraints mechanism https://review.openstack.org/393219 | 18:56 |
| *** javeriak has quit IRC | 18:56 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance: Simplify pip options/constraints mechanism https://review.openstack.org/393217 | 18:57 |
| *** spotz is now known as spotz_zzz | 18:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder: Add cinder_profiler_hmac_key to cinder.conf https://review.openstack.org/392979 | 18:57 |
| *** hughmFLEXin has quit IRC | 18:58 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally: Simplify pip options/constraints mechanism https://review.openstack.org/392910 | 18:58 |
| *** schwicht has quit IRC | 18:58 | |
| mgariepy | hmm maybe align all the templates to templates/systemd-tmpfiles.conf.j2 ? | 18:59 |
| *** adrian_otto has joined #openstack-ansible | 18:59 | |
| *** asettle has quit IRC | 18:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder: Simplify pip options/constraints mechanism https://review.openstack.org/393220 | 18:59 |
| odyssey4me | mgariepy if that patch merges, go ahead with the same type of change elsewhere | 18:59 |
| *** dxiri has joined #openstack-ansible | 18:59 | |
| *** dxiri_ has quit IRC | 18:59 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer: Replace github with git.o.o https://review.openstack.org/393273 | 19:00 |
| *** schwicht has joined #openstack-ansible | 19:00 | |
| mgariepy | odyssey4me, ok perfect. | 19:00 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_rally: Simplify pip options/constraints mechanism https://review.openstack.org/393446 | 19:01 |
| *** Mudpuppy_ has joined #openstack-ansible | 19:02 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_gnocchi: Simplify pip options/constraints mechanism https://review.openstack.org/393218 | 19:02 |
| *** adrian_otto1 has joined #openstack-ansible | 19:03 | |
| mgariepy | odyssey4me, https://review.openstack.org/#/c/393442/ please take a look at my comment | 19:04 |
| odyssey4me | mgariepy what sort of clean up are you thinking is needed? | 19:05 |
| mgariepy | the file neutron.conf will be present | 19:05 |
| odyssey4me | yes, but for master that's OK - nothing is deployed using master | 19:05 |
| *** adrian_otto has quit IRC | 19:05 | |
| mgariepy | when appling to the service neutron-metering-agent it will create the new file. | 19:05 |
| odyssey4me | for the stable branches that's not OK | 19:05 |
| odyssey4me | for your test environment you can manually remove the old file | 19:05 |
| mgariepy | ok | 19:05 |
| *** Mudpuppy has quit IRC | 19:05 | |
| odyssey4me | ok, time for me to go home for the night | 19:06 |
| odyssey4me | have a great day everyone :) | 19:07 |
| mgariepy | almost time for me too :P | 19:07 |
| mgariepy | cya | 19:07 |
| *** TxGirlGeek has quit IRC | 19:10 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:11 | |
| *** javeriak has joined #openstack-ansible | 19:11 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova: Change protocol of novalink url. https://review.openstack.org/393133 | 19:12 |
| *** fandi__ has joined #openstack-ansible | 19:15 | |
| *** fandi has quit IRC | 19:16 | |
| *** TxGirlGeek has quit IRC | 19:16 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:16 | |
| *** jlockwood has quit IRC | 19:18 | |
| castulo | I'm trying to find where Tempest gets configured in an openstack-ansible deployment, can someone point me in the right direction please? | 19:19 |
| *** TxGirlGeek has quit IRC | 19:20 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:20 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Use centralised Ansible test scripts https://review.openstack.org/393224 | 19:22 |
| *** retreved has quit IRC | 19:22 | |
| *** retreved_ has joined #openstack-ansible | 19:22 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum: Simplify pip options/constraints mechanism https://review.openstack.org/393222 | 19:22 |
| *** david-lyle has quit IRC | 19:23 | |
| *** johnmilton has quit IRC | 19:24 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add supported Operating System check https://review.openstack.org/392478 | 19:24 |
| *** david-lyle has joined #openstack-ansible | 19:25 | |
| *** aheczko has joined #openstack-ansible | 19:27 | |
| *** dxiri_ has joined #openstack-ansible | 19:27 | |
| *** hybridpollo has joined #openstack-ansible | 19:27 | |
| *** retreved has joined #openstack-ansible | 19:27 | |
| *** dxiri has quit IRC | 19:28 | |
| cloudnull | castulo: https://github.com/openstack/openstack-ansible-tests/blob/master/test-install-tempest.yml | 19:28 |
| cloudnull | which pulls in the test-vars | 19:28 |
| dxiri_ | is there anything special I need to do in order to get my flat network working? | 19:28 |
| cloudnull | if you're looking into the roles. | 19:28 |
| dxiri_ | If I set an IP statically via the console of the VM, connectivity works | 19:29 |
| *** adrian_otto1 has quit IRC | 19:29 | |
| *** retreved_ has quit IRC | 19:29 | |
| cloudnull | castulo: most of the resource config can be found here https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_resources.yml | 19:29 |
| dxiri_ | but if I want to use dhcp on a small range, the vm never gets the ip | 19:29 |
| cloudnull | dxiri_: that sounds like dhcp isn't running in the agent contianer. | 19:30 |
| cloudnull | or metadata is making it's way to the vm | 19:30 |
| dxiri_ | I even set a dhcp server on the host itself using isc-dhcp-server | 19:30 |
| dxiri_ | and I see the replies and everything...but the VM never gets the lease | 19:30 |
| *** electrofelix has quit IRC | 19:31 | |
| cloudnull | maybe something within the image ? | 19:31 |
| *** adrian_otto has joined #openstack-ansible | 19:31 | |
| cloudnull | if you login and set the ip statically can you also login and run dhclient | 19:31 |
| cloudnull | to see if you can force the lease? | 19:31 |
| *** hybridpollo has quit IRC | 19:32 | |
| *** TxGirlGeek has quit IRC | 19:32 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:32 | |
| castulo | cloudnull: ok, thanks, I noticed the tempest.conf generated when deploying osa is very outdated, I wanted to see if we could update it a little bit | 19:33 |
| cloudnull | please do | 19:33 |
| dxiri_ | I am testing with cirros, and if I do that, I see "sending discover" and on the isc-dhcp-server I see it does send out an ip | 19:33 |
| automagically | castulo - That would be great | 19:33 |
| dxiri_ | but the vm never gets it | 19:33 |
| dxiri_ | cirros-dhcpc up eth0 | 19:33 |
| dxiri_ | that's what I am running inside the VM | 19:33 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon: Workaround upstream issue with apache2_module https://review.openstack.org/392649 | 19:35 |
| *** TxGirlGeek has quit IRC | 19:42 | |
| openstackgerrit | Marc Gariépy proposed openstack/openstack-ansible-os_neutron: Fix tmpfiles.d when multiple service are running. https://review.openstack.org/393442 | 19:43 |
| *** hybridpollo has joined #openstack-ansible | 19:44 | |
| *** aheczko has quit IRC | 19:45 | |
| ioni | https://bugs.launchpad.net/openstack-ansible/+bug/1639026 | 19:46 |
| openstack | Launchpad bug 1639026 in openstack-ansible "ceilometer init is not create/updated on compute nodes on head stable/newton" [Undecided,New] | 19:46 |
| ioni | anyone can confirm the bug? | 19:46 |
| ioni | i can always replicate it | 19:46 |
| *** aheczko has joined #openstack-ansible | 19:47 | |
| *** aheczko has joined #openstack-ansible | 19:48 | |
| *** aheczko has joined #openstack-ansible | 19:49 | |
| *** aheczko has quit IRC | 19:49 | |
| *** aheczko has joined #openstack-ansible | 19:50 | |
| *** aheczko has quit IRC | 19:50 | |
| *** aheczko has joined #openstack-ansible | 19:50 | |
| *** aheczko has quit IRC | 19:50 | |
| *** TxGirlGeek has joined #openstack-ansible | 19:51 | |
| *** javeriak has quit IRC | 19:53 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible: [TEST] Update to Ansible v2.2.0.0-1 https://review.openstack.org/389292 | 19:54 |
| *** adrian_otto has quit IRC | 19:55 | |
| *** fandi__ has quit IRC | 19:56 | |
| *** drifterza_ has joined #openstack-ansible | 19:56 | |
| *** fandi__ has joined #openstack-ansible | 19:56 | |
| *** javeriak has joined #openstack-ansible | 19:57 | |
| *** drifterza has quit IRC | 19:59 | |
| jmccrory | ioni : looks like a definite bug. ceilometer-agent-compute is missing from newton | 20:01 |
| ioni | jmccrory, on 14.0.0 it was with; | 20:03 |
| ioni | ceilome+ 5629 0.9 0.0 102552 42748 ? Ss Oct28 81:08 ceilometer-polling: master process [/openstack/venvs/ceilometer-14.0.1/bin/ceilometer-polling --polling-namespaces compute -- | 20:03 |
| ioni | ceilome+ 5653 0.4 0.0 851420 65072 ? Sl Oct28 39:39 \_ ceilometer-polling: AgentManager worker(0) | 20:03 |
| ioni | i did modify myself the init to start with 14.0.1 | 20:03 |
| jmccrory | ioni : yep, looks like this accidentally removed it https://review.openstack.org/#/c/380164/3/defaults/main.yml | 20:04 |
| ioni | indeed | 20:06 |
| ioni | odyssey4me, your bug :D | 20:14 |
| openstackgerrit | Merged openstack/openstack-ansible-rabbitmq_server: RabbitMQ Upgrade Testing https://review.openstack.org/391899 | 20:15 |
| *** RaviGummadi has quit IRC | 20:16 | |
| *** oanson has quit IRC | 20:17 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_ceilometer: Re-add ceilometer-agent-compute service https://review.openstack.org/393470 | 20:19 |
| *** smatzek has quit IRC | 20:19 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_ceilometer: Re-add ceilometer-agent-compute service https://review.openstack.org/393470 | 20:20 |
| *** askb has joined #openstack-ansible | 20:24 | |
| *** khamtamtun has joined #openstack-ansible | 20:24 | |
| *** jlockwood has joined #openstack-ansible | 20:25 | |
| *** c-mart has quit IRC | 20:28 | |
| *** johnmilton has joined #openstack-ansible | 20:35 | |
| *** hughmFLEXin has joined #openstack-ansible | 20:36 | |
| *** c-mart has joined #openstack-ansible | 20:46 | |
| *** h5t4 has quit IRC | 20:46 | |
| *** weezS has quit IRC | 20:49 | |
| ioni | jmccrory, i tested your patch | 20:50 |
| ioni | jmccrory, it does create now the init file but it doesn't start/restart service on compute node | 20:51 |
| *** johnmilton has quit IRC | 20:51 | |
| *** hughmFLEXin has quit IRC | 20:52 | |
| jmccrory | hmm would you be able to share the playbook output? | 20:52 |
| ioni | sure, let me run it again | 20:53 |
| ioni | i did modify the handler to match yours | 20:53 |
| ioni | i had in stable shell | | 20:53 |
| ioni | at command reload | 20:53 |
| *** armaan has quit IRC | 20:53 | |
| *** hughmFLEXin has joined #openstack-ansible | 20:53 | |
| *** armaan has joined #openstack-ansible | 20:54 | |
| *** askb has quit IRC | 20:54 | |
| mrda | Morning OSA | 20:55 |
| *** askb has joined #openstack-ansible | 20:55 | |
| *** khamtamtun has quit IRC | 20:55 | |
| *** javeriak has quit IRC | 20:56 | |
| *** javeriak has joined #openstack-ansible | 20:56 | |
| ioni | jmccrory, https://paste.xinu.at/cg7E/ | 20:56 |
| dxiri_ | is there anything here you guys see that can prevent a dhcp packet to enter a VM? | 20:57 |
| dxiri_ | as an example...one of the VMs is tap91b7c922-82 | 20:57 |
| *** gouthamr has quit IRC | 20:58 | |
| ioni | jmccrory, is it me or it doesn't even use the handler | 20:58 |
| *** schwicht has quit IRC | 20:59 | |
| jmccrory | yep, looks like it wasn't triggered and only would be if the init file changes | 21:00 |
| ioni | well, the init file was created but the service wasn't started | 21:01 |
| ioni | if you want, I can delete the init file and rerun it | 21:01 |
| jmccrory | yeah, try that and see if the handler goes off | 21:01 |
| ioni | jmccrory, https://paste.xinu.at/3BeQ94/ | 21:05 |
| ioni | jmccrory, not triggered | 21:06 |
| ioni | jmccrory, in fact, i tried to start manuall the agent and it fails | 21:07 |
| *** thorst_ has joined #openstack-ansible | 21:07 | |
| *** jheroux has quit IRC | 21:08 | |
| *** thorst has quit IRC | 21:09 | |
| *** david-lyle has quit IRC | 21:09 | |
| ioni | -bash: /openstack/venvs/ceilometer-14.0.1/bin/ceilometer-agent-compute: No such file or directory | 21:09 |
| ioni | ah lol | 21:09 |
| ioni | i need to rebuild repo | 21:09 |
| ioni | :D | 21:09 |
| ioni | hmm | 21:10 |
| ioni | jmccrory, it should be ceilometer-polling or ceilometer-agent-compute ? | 21:11 |
| ioni | on 14.0.0 i had ceilometer-polling as agent on compute | 21:11 |
| ioni | so the init script is generated incorrectly? | 21:11 |
| *** schwicht has joined #openstack-ansible | 21:14 | |
| *** retreved has quit IRC | 21:15 | |
| *** karimb has joined #openstack-ansible | 21:16 | |
| *** askb has quit IRC | 21:21 | |
| *** schwicht has quit IRC | 21:23 | |
| *** adrian_otto has joined #openstack-ansible | 21:23 | |
| *** khamtamtun has joined #openstack-ansible | 21:27 | |
| jmccrory | ioni: well that's confusing, maybe agent-compute isn't a thing anymore. don't see a binary file for it, and the ceilometer docs are inconsistent about mentioning it | 21:27 |
| jmccrory | stevelle : do you know what ceilometer processes are supposed to be running on compute nodes? | 21:28 |
| *** javeriak has quit IRC | 21:28 | |
| ioni | i'm not sure what item.value.service_name was before | 21:29 |
| *** adrian_otto has quit IRC | 21:30 | |
| *** chris_hultin is now known as chris_hultin|AWA | 21:30 | |
| ioni | on mitaka was ceilometer-polling | 21:30 |
| ioni | newton until 14.0.1 was ceilometer-polling | 21:31 |
| *** schwicht has joined #openstack-ansible | 21:31 | |
| ioni | i've been running newton 14.0.0 | 21:31 |
| stevelle | jmccrory: I think at this point (newton) ceilometer should rely on polling + events | 21:32 |
| stevelle | but I'm not really sure | 21:32 |
| *** khamtamtun has quit IRC | 21:34 | |
| *** askb has joined #openstack-ansible | 21:34 | |
| *** khamtamtun has joined #openstack-ansible | 21:34 | |
| *** Jeffrey4l has quit IRC | 21:34 | |
| *** Jeffrey4l has joined #openstack-ansible | 21:35 | |
| *** thorst_ has quit IRC | 21:36 | |
| *** javeriak has joined #openstack-ansible | 21:37 | |
| alextricity25 | This is probably a dumb questions, but what happened to all of the env.d files? | 21:40 |
| *** weezS has joined #openstack-ansible | 21:40 | |
| alextricity25 | I must have missed that change | 21:40 |
| logan- | alextricity25: the env.d is loaded from osa/playbooks/inventory/env.d now. the files in /etc/openstack_deploy/env.d are treated as overrides to the base configuration and loaded on top of the ones in the inventory directory | 21:42 |
| alextricity25 | logan-: Ah! There they are :) Thanks! | 21:43 |
| *** aheczko has joined #openstack-ansible | 21:43 | |
| *** khamtamtun has quit IRC | 21:45 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Move storage_address discovery into a single task https://review.openstack.org/392969 | 21:46 |
| *** hughmFLEXin has quit IRC | 21:48 | |
| *** hughmFLEXin has joined #openstack-ansible | 21:48 | |
| *** askb has quit IRC | 21:50 | |
| *** askb has joined #openstack-ansible | 21:52 | |
| *** hughmFLEXin has quit IRC | 21:53 | |
| dxiri_ | ok after about 5 days looking at wtf with this problem (dhcp not working for flat working) I found out the reason, but I want you guys input on the best way to solve this here | 21:54 |
| dxiri_ | turn out the iptables setup is dropping the dhcp response when it tries to come back | 21:54 |
| dxiri_ | *turns out | 21:54 |
| dxiri_ | 9130 933K DROP all -- * * 0.0.0.0/0 0.0.0.0/0 /* Default drop rule for unmatched traffic. */ | 21:54 |
| dxiri_ | specifically on that rule | 21:55 |
| dxiri_ | since this is all ansible/openstack managed, adding a custom iptables rule doesn't look right | 21:55 |
| *** victorhe has quit IRC | 21:55 | |
| dxiri_ | on chain neutron-linuxbri-sg-fallback | 21:56 |
| *** jperry has quit IRC | 22:00 | |
| *** david-lyle_ has joined #openstack-ansible | 22:01 | |
| *** schwicht has quit IRC | 22:01 | |
| *** david-lyle_ has quit IRC | 22:02 | |
| *** david-lyle has joined #openstack-ansible | 22:03 | |
| *** c-mart has quit IRC | 22:06 | |
| *** aheczko has quit IRC | 22:07 | |
| *** f13o has quit IRC | 22:10 | |
| *** kjw3 has quit IRC | 22:17 | |
| *** gouthamr has joined #openstack-ansible | 22:21 | |
| *** Jack_Iv has quit IRC | 22:21 | |
| cloudnull | dxiri_: does adding the rule make it go ? | 22:21 |
| dxiri_ | can anyone mind to explain how does the neutron_agents_container ends up receiving the dhcp request? | 22:21 |
| *** javeriak has quit IRC | 22:22 | |
| dxiri_ | cloudnull: not really, cause I don't really now how to add it correctly, I put a regular "ACCEPT" and it matches but nothing happens, I guess I have to do some king of those PHYSDEV rules like the others but not sure how to do that | 22:22 |
| dxiri_ | *s/king/kind | 22:23 |
| *** c-mart has joined #openstack-ansible | 22:23 | |
| cloudnull | you really shouldn't need anything there. | 22:23 |
| cloudnull | dxiri_: from more on how neutron is wired up http://docs.openstack.org/security-guide/networking/architecture.html is fairly acurate. | 22:25 |
| dxiri_ | exactly, I don't know why it isn't matching any of the rules it is supposed to match: http://paste.ubuntu.com/23422450/ | 22:25 |
| dxiri_ | tap91b7c922-82 is the interface in question | 22:25 |
| dxiri_ | you can see there are rules already for input and output, for some reason that isn't matching the dhcp stuff so it ends up being droped | 22:25 |
| cloudnull | and when you do a console-log on the booted vm you're seeing it loop trying to get DHCP for 240 seconds before giving up ? | 22:26 |
| *** schwicht has joined #openstack-ansible | 22:29 | |
| dxiri_ | correct | 22:29 |
| dxiri_ | and this ONLY happens with my external/flat network | 22:30 |
| dxiri_ | internal ones work perfectly | 22:30 |
| dxiri_ | here is another question....if I go inside the dhcp namespace by doing ip netns exec [namespace-id] bash | 22:31 |
| dxiri_ | should I be able to reach any other host on the external physical network? | 22:32 |
| dxiri_ | I'm guessing no | 22:32 |
| cloudnull | no. | 22:32 |
| *** javeriak has joined #openstack-ansible | 22:33 | |
| dxiri_ | k, so how is the dhcp request going to end up in this namespace? | 22:33 |
| dxiri_ | for this dhcp-agent, its the only one out of 3 that I see no dhcp traffic coming into the namespace | 22:34 |
| cloudnull | this is the qdhcp NS right? | 22:34 |
| *** TxGirlGeek has quit IRC | 22:36 | |
| dxiri_ | correct | 22:36 |
| cloudnull | some of this is a bit out of my wheelhosuse but I believe on flat network, the instances will need to have access to the agnet container on the same L2. | 22:36 |
| cloudnull | so if you IP the device on the compute node you should be able to hit agent container. | 22:36 |
| cloudnull | ping, etc. | 22:37 |
| dxiri_ | ip the device? which device are you refering to here? | 22:37 |
| c-mart | Do I really need to join the OpenStack Foundation (https://www.openstack.org/join/) in order to submit a patch to OSA? | 22:38 |
| cloudnull | c-mart: yes. | 22:39 |
| cloudnull | sorry. | 22:39 |
| cloudnull | if that is a no go for you (some folks can't due to conflicts at work), you can email or upload a patch to a launchpad bug and someone else can pick it up for you. | 22:40 |
| c-mart | I can do it cloudnull, just checking. At the moment I have no ambition to research candidates and vote in elections, which looks to be required for members. (But maybe I'll want to do that someday.) | 22:42 |
| cloudnull | you are not required to participate in elections and such | 22:43 |
| dxiri_ | cloudnull: can you expand a little bit on what you said earlier | 22:43 |
| cloudnull | but you can if you so choose. | 22:43 |
| *** baz has joined #openstack-ansible | 22:44 | |
| cloudnull | dxiri_: on the compute node you will have an ethernet device which will be on the same l2 as the host running the neutron agent container which should be plugged into a bridge. . | 22:44 |
| cloudnull | so if you add an IP address to the ethernet device on the compute node you should be able to ping your agent container. | 22:45 |
| * cloudnull assuming you dont already have an address on that device. | 22:45 | |
| dxiri_ | and that ethernet device you are mentioning, is it a tap interface? tap91b7c922-82 >> this is the one mapped to the VM | 22:48 |
| dxiri_ | from the compute node I can already ping the agents container over the br-mgmt network | 22:49 |
| evrardjp | cloudnull: do you have an idea on how to fetch from a playbook/role the current ansible version in a friendly manner? (so not using python import) | 22:51 |
| evrardjp | I mean no usage of from ansible import __version__ | 22:53 |
| *** adrian_otto has joined #openstack-ansible | 22:54 | |
| evrardjp | I will try if ansible_version exists | 22:54 |
| *** weezS has quit IRC | 22:55 | |
| cloudnull | evrardjp: you can use ansible_version | 22:55 |
| evrardjp | yup just tried, it works | 22:55 |
| *** klamath has quit IRC | 22:55 | |
| cloudnull | if you're trying to limit a task to a specific version of ansible there used to be a magic conditional "ansible_required: INSERT_VERSION_HERE" | 22:56 |
| cloudnull | IDK if that's still around. | 22:56 |
| evrardjp | ohoh | 22:57 |
| evrardjp | that's what I was trying to look at indeed | 22:57 |
| cloudnull | or you can do a version compare like -- when: ansible_version.full | version_compare('2.1.0', '>') -- or similar | 22:58 |
| evrardjp | ansible 2.2 breaks a conditional that is terribly written but worked | 22:58 |
| cloudnull | ah. | 22:58 |
| evrardjp | Maybe I was drunk when writing it.... Anyway, now it properly breaks, but I don't need the conditional depending on the version | 22:59 |
| evrardjp | (it's about the apt module that you fixed) | 22:59 |
| evrardjp | now I will move the tasks to cleaner version | 22:59 |
| evrardjp | I'm just reluctant to fix something that will disappear soon | 23:00 |
| evrardjp | jmccrory: you plan to move to 2.2 soon ? I see you're testing it | 23:00 |
| evrardjp | with odyssey4me | 23:00 |
| jmccrory | evrardjp : hope we can and that it brings some speed improvements | 23:01 |
| evrardjp | well we'll never know until properly tested but according to what I read it's supposed to be better | 23:02 |
| evrardjp | :p | 23:02 |
| evrardjp | ok I will remove the apt hack and consider master as 2.2, we can then decide to merge it whenever appropriate | 23:03 |
| *** adrian_otto has quit IRC | 23:04 | |
| evrardjp | cloudnull I'll check on the ansible_required | 23:04 |
| evrardjp | seems interesting | 23:04 |
| dxiri_ | cloudnull: if I move the agents container to the compute node, will that help? | 23:06 |
| baz | cloudnull: https://github.com/openstack/openstack-ansible-os_horizon/commit/c0abd22c49373326b637f8b88112b25a4af77cc3#diff-7eeda618087b49ae876084ab6c73fdbbR86 causes image uploads to fail if | 23:14 |
| baz | https://github.com/openstack/openstack-ansible-os_horizon/blob/master/defaults/main.yml#L73 is set to internalURL | 23:14 |
| baz | a different issue shows up if it's set to publicURL. Only clean combination I've found if Horizon is on the internal network is for "horizon_images_upload_mode" to be set to "legacy" | 23:16 |
| cloudnull | baz: does setting that to legacy fix it ? | 23:16 |
| cloudnull | ok. | 23:16 |
| cloudnull | do you mind changing that in the os_horizon role? | 23:16 |
| *** askb has quit IRC | 23:16 | |
| baz | with it set to direct, the browser tries to talk to the endpoint selected by OPENSTACK_ENDPOINT_TYPE which is default (in OSA) to internalURL, which of course will fail. | 23:17 |
| baz | with legacy Horizon does the proxy between the browser and glance. | 23:18 |
| *** javeriak has quit IRC | 23:19 | |
| *** javeriak has joined #openstack-ansible | 23:19 | |
| baz | I think this probably should be an issue raised with Horizon itself, the interaction between HORIZON_IMAGES_UPLOAD_MODE and OPENSTACK_ENDPOINT_TYPE | 23:20 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Fix BaseException.message deprecation messages https://review.openstack.org/392279 | 23:24 |
| *** askb has joined #openstack-ansible | 23:29 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-ceph_client: Remove ansible<2.2 apt cache hack https://review.openstack.org/393501 | 23:30 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_client: Remove ansible<2.2 apt cache hack https://review.openstack.org/393502 | 23:30 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-galera_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393503 | 23:30 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-haproxy_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393504 | 23:31 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_hosts: Remove ansible<2.2 apt cache hack https://review.openstack.org/393505 | 23:31 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-memcached_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393506 | 23:31 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-openstack_hosts: Remove ansible<2.2 apt cache hack https://review.openstack.org/393507 | 23:32 |
| *** armaan has quit IRC | 23:32 | |
| *** dxiri_ has quit IRC | 23:32 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_aodh: Remove ansible<2.2 apt cache hack https://review.openstack.org/393508 | 23:33 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_ceilometer: Remove ansible<2.2 apt cache hack https://review.openstack.org/393509 | 23:33 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_cinder: Remove ansible<2.2 apt cache hack https://review.openstack.org/393511 | 23:34 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_cloudkitty: Remove ansible<2.2 apt cache hack https://review.openstack.org/393512 | 23:34 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_designate: Remove ansible<2.2 apt cache hack https://review.openstack.org/393513 | 23:34 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_glance: Remove ansible<2.2 apt cache hack https://review.openstack.org/393514 | 23:35 |
| openstackgerrit | Merged openstack/openstack-ansible: Remove outdated aio heat template https://review.openstack.org/393351 | 23:35 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_gnocchi: Remove ansible<2.2 apt cache hack https://review.openstack.org/393515 | 23:35 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_heat: Remove ansible<2.2 apt cache hack https://review.openstack.org/393516 | 23:35 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_horizon: Remove ansible<2.2 apt cache hack https://review.openstack.org/393517 | 23:36 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_magnum: Remove ansible<2.2 apt cache hack https://review.openstack.org/393518 | 23:36 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_neutron: Remove ansible<2.2 apt cache hack https://review.openstack.org/393519 | 23:37 |
| *** Mudpuppy_ has quit IRC | 23:37 | |
| *** Mudpuppy has joined #openstack-ansible | 23:37 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_nova: Remove ansible<2.2 apt cache hack https://review.openstack.org/393520 | 23:37 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_rally: Remove ansible<2.2 apt cache hack https://review.openstack.org/393521 | 23:38 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_sahara: Remove ansible<2.2 apt cache hack https://review.openstack.org/393522 | 23:38 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_swift: Remove ansible<2.2 apt cache hack https://review.openstack.org/393523 | 23:38 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest: Remove ansible<2.2 apt cache hack https://review.openstack.org/393524 | 23:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_trove: Remove ansible<2.2 apt cache hack https://review.openstack.org/393525 | 23:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_watcher: Remove ansible<2.2 apt cache hack https://review.openstack.org/393526 | 23:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_zaqar: Remove ansible<2.2 apt cache hack https://review.openstack.org/393527 | 23:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-pip_install: Remove ansible<2.2 apt cache hack https://review.openstack.org/393528 | 23:40 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rabbitmq_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393529 | 23:41 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_build: Remove ansible<2.2 apt cache hack https://review.openstack.org/393530 | 23:41 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-repo_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393531 | 23:42 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_client: Remove ansible<2.2 apt cache hack https://review.openstack.org/393532 | 23:42 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-rsyslog_server: Remove ansible<2.2 apt cache hack https://review.openstack.org/393533 | 23:42 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-security: Remove ansible<2.2 apt cache hack https://review.openstack.org/393534 | 23:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove: Simplify pip options/constraints mechanism https://review.openstack.org/392921 | 23:46 |
| openstackgerrit | Merged openstack/openstack-ansible-galera_client: Replace github.com with git.o.o https://review.openstack.org/393319 | 23:48 |
| openstackgerrit | Chris Martin proposed openstack/openstack-ansible-ceph_client: Support Ceph monitors on non-default SSH port https://review.openstack.org/393536 | 23:52 |
| openstackgerrit | Merged openstack/openstack-ansible-security: Replace github with git.o.o https://review.openstack.org/393320 | 23:53 |
| *** hw_wutianwei has joined #openstack-ansible | 23:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_zaqar: Simplify pip options/constraints mechanism https://review.openstack.org/392926 | 23:57 |
| openstackgerrit | Merged openstack/openstack-ansible-os_watcher: Simplify pip options/constraints mechanism https://review.openstack.org/392924 | 23:57 |
| *** dxiri has joined #openstack-ansible | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
