| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-08-21 | |||
| 13:39:37 | mriedem | despite that one person saying it was never even used | |
| 13:39:45 | mriedem | tssurya: so just report a bug and deprecate --instance-name | |
| 13:39:47 | mriedem | i'll +2 that | |
| 13:39:54 | tssurya | mriedem: cool | |
| 13:40:02 | mriedem | it predates gerrit so i'm not surprised it's a mess | |
| 13:40:41 | stephenfin | mriedem: Yeah, just oslo.config, I think | |
| 13:40:58 | stephenfin | though I had it in my head oslo.policy wasn't broken in the first place. Obviously not | |
| 13:41:14 | efried | mriedem: | |
| 13:41:14 | efried | - nova patch to twiddle a couple of options to prove it works | |
| 13:41:14 | efried | - oslo.config patch to address complaint that using the rst role in help text shows up ugly in the sample: https://review.openstack.org/#/c/583064/ | |
| 13:41:25 | efried | https://review.openstack.org/#/c/583025/ shoulda been that first link, sorry | |
| 13:41:42 | stephenfin | efried: I think that's a different issue | |
| 13:42:08 | stephenfin | efried: mriedem's asking why newlines and the likes in policy.help aren't being parsed | |
| 13:42:24 | stephenfin | ...in the HTML output. Your patch affects the ini output, right? | |
| 13:43:18 | tssurya | okay, so instance.display_name is name and instance.hostname is OS-EXT-SRV-ATTR:hostname and we don't care about OS-EXT-SRV-ATTR:instance_name. | |
| 13:43:22 | efried | stephenfin: the oslo.config patch, yes. | |
| 13:44:01 | efried | oh, reread what mriedem was actually saying. Yeah, I don't know about that, sorry. | |
| 13:44:09 | efried | I would have asked stephenfin :) | |
| 13:44:28 | mriedem | ha | |
| 13:44:58 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Filter out instances without a host when populating AZ https://review.openstack.org/594050 | |
| 13:45:17 | mriedem | ^ is likely an RC3 issue | |
| 13:47:51 | mriedem | regarding petr's email about install guide testing, | |
| 13:47:55 | stephenfin | mriedem: Agreed | |
| 13:48:20 | mriedem | i wonder how valid it is, or time would be saved, by starting up devstack but not enabling nova, so that you can do that manually after keystone/glance/cinder/neutron are already setup | |
| 13:48:58 | mriedem | i think the only major thing in the install guide in rocky was the placement db | |
| 13:49:29 | sean-k-mooney | mriedem: i think you will hit depency issues | |
| 13:50:00 | mriedem | on other openstack services? | |
| 13:50:06 | mriedem | or things like setting up libvirt? | |
| 13:50:34 | sean-k-mooney | well neutron would expect to be able to talks to placement for things like routed networks | |
| 13:50:48 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/rocky: Filter out instances without a host when populating AZ https://review.openstack.org/594178 | |
| 13:50:50 | mriedem | routed networks are optional and devstack doesn't set those up anyway | |
| 13:51:01 | mriedem | we definitely *should* have a ci job that uses routed networks | |
| 13:51:13 | mriedem | across a 2-node deploy where each host is in a separate aggregate | |
| 13:51:39 | mriedem | but that would require time and people that care to make sure it continues to work | |
| 13:52:12 | sean-k-mooney | in theroy devstack should be able to help i guess | |
| 13:52:45 | sean-k-mooney | mriedem: is placement installation considered part of the nova install guide | |
| 13:55:40 | dansmith | tssurya: looks like the down-cell stack needs rebasing again | |
| 13:56:00 | dansmith | presumably its review-able regardless? | |
| 13:56:01 | efried | mriedem: https://review.openstack.org/594179 <== alternative uuidsentinel impl | |
| 13:56:20 | tssurya | dansmith: yea, its ready for a first time review | |
| 13:56:26 | tssurya | still working on filtering part | |
| 13:56:32 | dansmith | okay | |
| 13:56:36 | tssurya | but would be nice to get opinions | |
| 13:56:58 | tssurya | I have them as seperate patches for now, will squash them with the version BUMP | |
| 13:57:09 | tssurya | once we review the approach | |
| 13:57:59 | tssurya | and, mriedem: sorry about missing the instance.host None case earlier on and the backport headaches. | |
| 13:58:03 | openstackgerrit | Jiri Suchomel proposed openstack/nova stable/pike: Filter out instances without a host when populating AZ https://review.openstack.org/594184 | |
| 14:00:20 | openstackgerrit | Surya Seetharaman proposed openstack/nova stable/queens: Filter out instances without a host when populating AZ https://review.openstack.org/594185 | |
| 14:02:37 | stephenfin | mriedem: https://bugs.launchpad.net/oslo.policy/+bug/1788183 | |
| 14:02:37 | openstack | Launchpad bug 1788183 in oslo.policy "Rule description not rendered as rST" [Undecided,New] | |
| 14:04:14 | mriedem | tssurya: not your fault, we have reviewers for a reason | |
| 14:04:21 | mriedem | and i obviously missed it as well | |
| 14:04:33 | mriedem | sean-k-mooney: i think so yes | |
| 14:05:09 | mriedem | efried: why not in oslotest? because of the circular dep? | |
| 14:05:54 | efried | mriedem: And because it's... a UUID util. And because just because I can't think of a reason for it to be used outside of test, doesn't mean it can't be. See commit message. | |
| 14:06:33 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/queens: Filter out instances without a host when populating AZ https://review.openstack.org/594185 | |
| 14:07:03 | sean-k-mooney | mriedem: then in that case if you wanted to test the nova install guide, and use devstack to help, you would just have devstack install keystone,mysql, and rabbitmq. and perhapse memcached | |
| 14:07:39 | mriedem | but nova also needs neutron | |
| 14:07:46 | mriedem | and i don't want to go through the neutron install guide to test nova's install | |
| 14:07:49 | mriedem | same for glance | |
| 14:08:15 | mriedem | cue kevin fox to say it should all be one monolithic install | |
| 14:09:06 | sean-k-mooney | well i would not expect the install guide for nova to cover the glance or neutron parts | |
| 14:09:23 | sean-k-mooney | i also would not assmue you could boot a vm after finishing it | |
| 14:09:44 | sean-k-mooney | i would just assumne i had the nova compontes deployed and fuctioning | |
| 14:10:06 | sean-k-mooney | e.g. nova hypervior list should show all the resouces but openstack server create would fail | |
| 14:10:28 | mriedem | well, if i'm installing nova, i would like to be able to create a vm by the end of it | |
| 14:10:35 | mriedem | otherwise i don't know if i f'ed up the install somewhere | |
| 14:11:08 | sean-k-mooney | in that case it does have to be a multi service install guide | |
| 14:11:12 | mriedem | also, https://docs.openstack.org/nova/latest/install/controller-install-ubuntu.html#install-and-configure-components refers to the neutron install guide | |
| 14:13:19 | sean-k-mooney | i guess refering to the other guide also works. that said untill nova networks if fully dead neutron is technical not a nova depency | |
| 14:13:45 | sean-k-mooney | but i could see adding neutron to the devstack install. | |
| 14:14:24 | openstackgerrit | Matt Riedemann proposed openstack/nova stable/pike: Filter out instances without a host when populating AZ https://review.openstack.org/594184 | |
| 14:14:46 | sean-k-mooney | glace i guess would also be required because there is no way to boot a vm otherwise. unless you used the fake drivers | |
| 14:27:42 | mriedem | jroll: pretty sure this has always been true yeah? https://bugs.launchpad.net/nova/+bug/1787509 | |
| 14:27:42 | openstack | Launchpad bug 1787509 in OpenStack Compute (nova) "Baremetal filters and default filters cannot be used simultaneously in the same nova" [Undecided,New] | |
| 14:27:47 | mriedem | until pike anyway | |
| 14:34:52 | dansmith | mriedem: so reviewing tssurya's series just now made me (re-)realize | |
| 14:35:07 | dansmith | we're still iterating all of the instances from a cell before returning them in order to do the fault stuff | |
| 14:35:29 | mriedem | dansmith: i think when you were adding instance lister, | |
| 14:35:40 | mriedem | you pre-joined faults in the db api and it didn't seem to make a difference in perf | |
| 14:35:48 | mriedem | and it might have caused some other issue | |
| 14:36:05 | dansmith | so I'm surprised we gained as much as we did by my batching, and so I wonder if we push the faults into the batches if it would help | |
| 14:36:10 | dansmith | mriedem: yeah, I remember that now | |
| 14:36:13 | mriedem | we also only show fault if the vm state is ERROR or DELETED | |
| 14:36:16 | dansmith | right | |
| 14:37:12 | mriedem | well this bug says nova list is too slow https://bugs.launchpad.net/nova/+bug/1788149 | |
| 14:37:12 | openstack | Launchpad bug 1788149 in OpenStack Compute (nova) "nova list too slow" [Undecided,New] | |
| 14:37:15 | mriedem | so there is that | |
| 14:37:57 | dansmith | heh | |
| 14:38:41 | tssurya | nice | |
| 14:41:28 | mriedem | alex_xu: did you say gmann was on vacation? https://review.openstack.org/#/c/584223/ | |
| 14:41:45 | mriedem | (8:09:57 AM) alex_xu: gmann: enjoy your vacation! | |
| 14:41:50 | tssurya | mriedem: yes untill 31st | |
| 14:42:17 | tssurya | until* | |
| 14:45:21 | sean-k-mooney | mriedem: regarding the migration issue i dont see "Binding ports to destination host" in either the source or dest compute logs | |
| 14:46:08 | sean-k-mooney | the dest does have "Plugging VIFs using destination host port bindings before live migration." and "Deleted binding for port 3218fd70-ea82-4ee1-9a5b-2d3c9d8b9fa0 and host devstack2." | |
| 14:46:47 | mriedem | the former is when we do pre_live_migration on the dest host, | |
| 14:46:54 | mriedem | at that point port bindings are still active for the source host | |
| 14:47:03 | mriedem | the latter is when we're rolling back after the failed migration | |
| 14:47:21 | mriedem | so i'm not sure that my patch would fix your issue if we never deactivated the source host bindings | |