| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-06-28 | |||
| 13:57:01 | andrewbogott | mnaser: you'd expect that to be keyed using the uuid? Or the 'id' from the instances table? | |
| 13:57:18 | mnaser | both | |
| 13:57:18 | mnaser | :) | |
| 13:57:44 | mnaser | nova is complex and in my experience you really dont want to muck around the db manually | |
| 13:57:55 | jangutter | andrewbogott: the scary part of this is not accidentally re-inventing all the instance creation code... | |
| 13:58:05 | melwitt | gmann: I put it on the agenda for the nova meeting today | |
| 13:58:09 | mnaser | it might work today but break down in terrible ways in your next upgrade | |
| 13:58:34 | andrewbogott | ok, I'll see if I can do use the instance creation code instead, then. | |
| 13:58:40 | mnaser | melwitt: i know this is pretty last minute but if we have time to discuss https://review.openstack.org/#/c/577933/ that'd be nice | |
| 13:58:42 | andrewbogott | thank you for your thoughts, all | |
| 13:58:45 | mnaser | though i am not attached or that is not a priority | |
| 13:58:50 | yikun | mriedem: "if we check db_policy._policies anywhere", why we check this? | |
| 13:59:17 | mriedem | yikun: replied | |
| 13:59:22 | mriedem | yikun: yeah i don't get it, | |
| 13:59:38 | melwitt | mnaser: you mean during open discussion in the nova meeting? that's fine, I'm sure we'll have time | |
| 13:59:53 | mriedem | yikun: i'd say leave that part as you have it now, address the other comments and then we can come back to it when dan is back | |
| 13:59:55 | mnaser | melwitt: ok cools | |
| 14:00:22 | mriedem | yikun: i can maybe get jaypipes to take a look later too | |
| 14:00:36 | mriedem | he knows sqla db objects and versioned object mysteries | |
| 14:00:58 | Guest88823 | mriedem: nothing to do with ovo, just sqla | |
| 14:01:08 | mriedem | TheLurker | |
| 14:01:17 | yikun | hah | |
| 14:01:19 | dansmith | mriedem: yikun: You always need to check the old field until we're fully migrated yeah? | |
| 14:01:45 | mriedem | but it doesn't exist in the model https://review.openstack.org/#/c/563375/17/nova/db/sqlalchemy/api_models.py | |
| 14:01:55 | mriedem | _policies was changed to policy in the model | |
| 14:01:57 | mriedem | for the backref | |
| 14:02:12 | mriedem | crap meeting time | |
| 14:02:29 | dansmith | ohh | |
| 14:02:33 | dansmith | I never got that far | |
| 14:02:36 | dansmith | why is that changing? | |
| 14:02:48 | mriedem | i asked that as well, | |
| 14:03:00 | mriedem | i assume it's for model sanity since we don't have many policies per group, we have one | |
| 14:03:10 | mriedem | it's not a column, just a backref attribute | |
| 14:03:16 | dansmith | I thought we were adding a new backref | |
| 14:03:23 | dansmith | not a column obviously, but a new backreg | |
| 14:03:29 | dansmith | alright, well, nevermind me then | |
| 14:03:37 | mriedem | heh ok :) | |
| 14:03:39 | mriedem | well that was easy | |
| 14:04:17 | mriedem | yikun: so leave it like you have it for that joinedload | |
| 14:04:38 | dansmith | can we break the backref change out to something separate? | |
| 14:04:43 | dansmith | because that would be less confusing | |
| 14:05:00 | dansmith | 1. refactor the name, 2. make the changes to actually support the new object field | |
| 14:05:00 | mriedem | yeah that would be nice | |
| 14:05:55 | mriedem | left a comment | |
| 14:06:07 | jaypipes | mriedem: whatup? | |
| 14:06:08 | mriedem | yikun: ^ let me know if that all makes sense | |
| 14:06:11 | mriedem | jaypipes: nvm | |
| 14:07:56 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Merge server create for security group extension https://review.openstack.org/578714 | |
| 14:09:49 | yikun | yeah, got it, :), put refactor _policies stuff in a separate patch~ | |
| 14:10:04 | yikun | mriedem, dansmith ^ | |
| 14:11:21 | efried | gmann: Having the word "merge" in your commit title, that's great psychology. | |
| 14:11:36 | efried | Like, we had a guy running for mayor in Austin whose name was Will Wynn. | |
| 14:11:56 | mriedem | but he lost right? | |
| 14:12:10 | efried | Heh. I actually don't think so. I don't keep up on politics much though. | |
| 14:12:33 | mriedem | i'll be damned, mayor from 2003 to 2009 | |
| 14:12:34 | efried | mriedem: No, apparently he served as mayor from 2003 to 2009. | |
| 14:12:39 | efried | jynx | |
| 14:12:41 | mriedem | https://en.wikipedia.org/wiki/Will_Wynn#/media/File:Will_Wynn_2005.jpg | |
| 14:12:42 | mriedem | real hair? | |
| 14:12:50 | yikun | mriedem: yeah, maybe go to the vocation. : ) | |
| 14:13:30 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Merge server create for user data extension https://review.openstack.org/578709 | |
| 14:13:56 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Merge server create for security group extension https://review.openstack.org/578714 | |
| 14:14:43 | openstackgerrit | Merged openstack/nova master: Remove compatibility code for instance groups https://review.openstack.org/578192 | |
| 14:14:53 | openstackgerrit | Merged openstack/nova master: Fix the duplicated config options of api_database and placement_database https://review.openstack.org/577023 | |
| 14:19:26 | openstackgerrit | xulei proposed openstack/nova master: Disable limits if force_hosts or force_nodes is set https://review.openstack.org/576693 | |
| 14:26:37 | mriedem | can i get a stable review on this? https://review.openstack.org/#/c/563700/ | |
| 14:26:43 | mriedem | johnthetubaguy: ^? | |
| 14:27:30 | johnthetubaguy | mriedem: oh that is a tasty one, will take a look | |
| 14:36:26 | openstackgerrit | Matthew Booth proposed openstack/nova master: Remove irrelevant comment https://review.openstack.org/578821 | |
| 14:36:27 | openstackgerrit | Matthew Booth proposed openstack/nova master: Move static _get_power_off_values to compute_utils https://review.openstack.org/578822 | |
| 14:44:23 | openstackgerrit | Stephen Finucane proposed openstack/nova master: trivial: Remove 'tools/releasenotes_tox.sh' https://review.openstack.org/534383 | |
| 14:44:24 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tox: Reuse envdirs https://review.openstack.org/534382 | |
| 14:44:25 | openstackgerrit | Stephen Finucane proposed openstack/nova master: tox: Document and dedupe mostly everything https://review.openstack.org/578825 | |
| 14:44:26 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Convert 'placement_api_docs' into a Sphinx extension https://review.openstack.org/578826 | |
| 14:45:30 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Convert 'placement_api_docs' into a Sphinx extension https://review.openstack.org/578826 | |
| 14:51:46 | openstackgerrit | Merged openstack/python-novaclient master: Add release note link in README https://review.openstack.org/578664 | |
| 14:57:22 | efried | stephenfin: You split https://review.openstack.org/#/c/534382/ into https://review.openstack.org/#/c/534382/9 and https://review.openstack.org/#/c/578825/1 right? | |
| 14:57:47 | mnaser | i asked this a bit earlier but | |
| 14:57:54 | mnaser | https://docs.openstack.org/nova/latest/user/cells.html#adding-a-new-cell-to-an-existing-deployment | |
| 14:57:57 | stephenfin | efried: Yes. https://review.openstack.org/#/c/534382/ is the one I really want but I need to sent a note to the mailing lsit first | |
| 14:57:59 | mnaser | "To expand your deployment with a new cell, first follow the usual steps for standing up a new Cells V1 cell." | |
| 14:58:00 | stephenfin | *list | |
| 14:58:00 | efried | stephenfin: but you moved the larger part of the change to the new change-id, making it harder to diff. | |
| 14:58:18 | stephenfin | efried: Sec. I can fix that, if you'd like? | |
| 14:58:20 | mnaser | is that a typo or do we really have to setup cells v1 and upgrade to cells v2... | |
| 14:58:39 | efried | stephenfin: Meh, it'll probably do me good to re-review from scratch anyway. | |
| 14:58:44 | mnaser | i'll be very happy to fix the docs if that's incorrect | |
| 14:58:46 | efried | was just giving you grief | |
| 14:59:13 | mriedem | mnaser: i think that's a typo | |
| 14:59:16 | mriedem | or out of context | |
| 14:59:22 | melwitt | mnaser: uh no, what I meant there was create a new db, mq, etc whatever you would have normally done to set up a new cell | |
| 14:59:56 | mnaser | okay, i'll be working on adding a new cells v2 over the next few days so there might be a bunch of doc patches that hopefully help | |
| 14:59:58 | melwitt | so it should be reworded if that's how it sounds to other people | |
| 14:59:59 | melwitt | (if it sounds like it's saying set up cells v1) | |
| 15:00:05 | melwitt | thanks mnaser | |
| 15:00:21 | mnaser | and hopefully adding cells v2 native support in openstack ansible (not just 1 cell) | |
| 15:01:04 | mriedem | yeah i'd just generalize https://docs.openstack.org/nova/latest/user/cells.html#adding-a-new-cell-to-an-existing-deployment into saying, create new db and mq, sync db, create cell with new db and mq, discover hosts, etc | |
| 15:01:26 | mriedem | don't need to run map_instances if it's a new deploy | |
| 15:01:36 | melwitt | right | |
| 15:02:37 | mriedem | if you're adding existing child cells to cells v2 mappings, then you skip the "create db and mq" part, and just create cell mapping with existing db and mq for child cell, discover hosts and map_instances | |