| *** dklyle has joined #openstack-placement | 00:16 | |
| *** dklyle has quit IRC | 00:31 | |
| *** e0ne has quit IRC | 00:32 | |
| *** efried has quit IRC | 06:05 | |
| *** efried has joined #openstack-placement | 06:44 | |
| *** helenafm has joined #openstack-placement | 08:15 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs master: Amend the detach-boot-volume design https://review.openstack.org/619161 | 08:26 |
|---|---|---|
| *** tssurya has joined #openstack-placement | 08:36 | |
| *** takashin has left #openstack-placement | 09:32 | |
| *** e0ne has joined #openstack-placement | 09:38 | |
| *** e0ne has quit IRC | 09:58 | |
| *** e0ne has joined #openstack-placement | 10:02 | |
| *** e0ne has quit IRC | 11:03 | |
| *** e0ne has joined #openstack-placement | 11:10 | |
| *** cdent has joined #openstack-placement | 11:19 | |
| *** tetsuro has joined #openstack-placement | 12:18 | |
| *** e0ne has quit IRC | 12:20 | |
| *** e0ne has joined #openstack-placement | 12:35 | |
| *** cdent has quit IRC | 12:36 | |
| *** cdent has joined #openstack-placement | 13:02 | |
| *** e0ne has quit IRC | 13:17 | |
| *** takashin has joined #openstack-placement | 13:18 | |
| *** e0ne has joined #openstack-placement | 13:20 | |
| cdent | thanks for responding on that email sean-k-mooney, was mostly an effort to draw people out on why not to do what I suggest | 13:41 |
| *** mriedem has joined #openstack-placement | 13:43 | |
| sean-k-mooney | cdent: did the responce make sense? | 13:45 |
| sean-k-mooney | i did not have a chance to read the full thread so it may have been out of context | 13:45 |
| cdent | yes, but it doesn't address the "already pretty confusing" aspect of things | 13:45 |
| cdent | no, it's fairly relevant | 13:45 |
| cdent | a worthwhile addition | 13:45 |
| sean-k-mooney | from my past experince working on kolla a year or two ago im not sure if nova for example maintining its own blessed image would be a good thing or not | 13:46 |
| cdent | presumably it would be several images | 13:48 |
| sean-k-mooney | well it coulbd be one image with multiple contianes useing the same image with different entry points | 13:49 |
| sean-k-mooney | or it could be multiple images each with a hardcode entrypoint and a common base | 13:49 |
| cdent | that seems to violate much of the point (to me) | 13:49 |
| cdent | they should be as small as possible | 13:49 |
| cdent | or course nova is not oriented towards small or independent from other stuff | 13:50 |
| sean-k-mooney | true but multiple images does not make nova smaller | 13:50 |
| cdent | what I would want is access to a container which is solely the nova-api and nova-api's config and the python code that it (and only it) requires. same for the other processes | 13:53 |
| cdent | but that ship is gone | 13:53 |
| sean-k-mooney | cdent: well what distro would be provideing the python/uwsgi/apache that is running it | 13:55 |
| sean-k-mooney | cdent: loci kind of does that | 13:55 |
| cdent | uwsgi from pip, no apache | 13:55 |
| sean-k-mooney | so that would work for nova but cider i bevilve or swift sitll cant run under uwsgi due to its perodic tasks | 13:56 |
| cdent | right, but one project not being able to uwsgi doesn't mean another shouldn't | 13:56 |
| sean-k-mooney | cdent: what you want can certenly be dont but can i ask what is your intinet with the image. test? dev? production? | 13:56 |
| sean-k-mooney | cdent: true but if they were offical images i assume we would want them to be consitent in there usage? | 13:57 |
| cdent | my hope is that we could produce tiny images for use in production. I don't think we are there yet because of issues with observability but I think that trying to get there would help us improve our systems | 13:57 |
| cdent | why would they need to be consistent if they worked? (i'm not disputing that, I'm asking why) | 13:58 |
| sean-k-mooney | i can certenly agree with that goal | 13:58 |
| sean-k-mooney | they would need to be consitent to be consumable downstram and prodctisable | 13:58 |
| efried | n-sch meeting in 2 minutes in #openstack-meeting-alt | 13:59 |
| sean-k-mooney | if that is not a goal then it would not need to be a requirement | 13:59 |
| sean-k-mooney | that said kolla defiend an abstract api and kolla-ansbile can work with anything that implement that api | 13:59 |
| cdent | if an api-service container present an http interface, why does it matter which http server is within? | 13:59 |
| sean-k-mooney | so if you did the same for does then your fine | 13:59 |
| sean-k-mooney | cdent: it only matter when it comes to configuring the service | 14:00 |
| cdent | not if it as http port that is being exposed | 14:00 |
| cdent | if it is a unix socket of something, then yes, it gets messy | 14:00 |
| sean-k-mooney | from a prodcut perspective the versions and distor also matter but for generic usecase it does not | 14:01 |
| * cdent likes to live the lie that products and distros are dead :) | 14:01 | |
| sean-k-mooney | cdent: i deploy kolla container form source git repos personally | 14:02 |
| sean-k-mooney | cdent: so a tiny image with python and pip installing everyting form git is/would be my go to | 14:03 |
| *** cdent has quit IRC | 14:27 | |
| *** tetsuro has quit IRC | 14:29 | |
| *** cdent has joined #openstack-placement | 14:52 | |
| *** cdent has quit IRC | 14:59 | |
| *** takashin has left #openstack-placement | 15:02 | |
| *** cdent has joined #openstack-placement | 15:03 | |
| *** leakypipes is now known as jaypipes | 15:49 | |
| *** openstackgerrit has quit IRC | 15:51 | |
| *** dklyle has joined #openstack-placement | 16:06 | |
| *** helenafm has quit IRC | 16:25 | |
| *** efried has quit IRC | 16:28 | |
| *** ttsiouts has joined #openstack-placement | 16:53 | |
| *** ttsiouts has quit IRC | 17:00 | |
| *** ttsiouts has joined #openstack-placement | 17:01 | |
| *** ttsiouts has quit IRC | 17:05 | |
| *** dims has quit IRC | 17:08 | |
| *** openstackgerrit has joined #openstack-placement | 17:21 | |
| openstackgerrit | Merged openstack/placement master: Use local config for placement-status CLI https://review.openstack.org/632600 | 17:21 |
| openstackgerrit | Merged openstack/placement master: Add upgrade status check for missing root ids https://review.openstack.org/632599 | 17:21 |
| cdent | yay | 17:22 |
| *** e0ne has quit IRC | 17:42 | |
| *** jaypipes has quit IRC | 18:57 | |
| *** jaypipes has joined #openstack-placement | 18:57 | |
| *** e0ne has joined #openstack-placement | 18:59 | |
| *** dims has joined #openstack-placement | 19:15 | |
| openstackgerrit | Chris Dent proposed openstack/placement master: Use one ConfigOpts in placement-status https://review.openstack.org/633595 | 19:57 |
| *** efried has joined #openstack-placement | 20:24 | |
| *** tssurya has quit IRC | 20:40 | |
| *** takashin has joined #openstack-placement | 21:49 | |
| *** tssurya has joined #openstack-placement | 22:26 | |
| *** efried1 has joined #openstack-placement | 22:32 | |
| *** efried has quit IRC | 22:34 | |
| *** efried1 is now known as efried | 22:34 | |
| *** e0ne has quit IRC | 22:36 | |
| *** cdent has quit IRC | 22:41 | |
| *** tssurya has quit IRC | 22:53 | |
| *** mriedem has quit IRC | 23:15 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!