| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-07-26 | |||
| 17:16:02 | cburgess | dansmith This is a nova.conf option or API thing... or CLI thing? Sorry a bit slow this morning it seems. | |
| 17:16:03 | dansmith | cfriesen: I'm happy with your patch if you make it only do that thing on --strict | |
| 17:16:17 | dansmith | cfriesen: or if you want to go down the path of requiring either --strict or --loose and do the major version bump dance | |
| 17:16:25 | dansmith | cburgess: yeah, nova-compute conf | |
| 17:16:27 | dansmith | cburgess: hold on | |
| 17:16:55 | dansmith | cburgess: https://github.com/openstack/nova/blob/master/nova/conf/compute.py#L588-L601 | |
| 17:16:58 | dansmith | cburgess: you're welcome :) | |
| 17:17:25 | dansmith | cburgess: we have a build limit too, in case you hadn't seen it | |
| 17:17:35 | dansmith | cburgess: to avoid building 20 instances in parallel on a single compute node | |
| 17:17:39 | dansmith | s/building/failing to build/ :) | |
| 17:18:34 | cburgess | dansmith Interesting... thanks. I will have to read up on this. I vaguely recall some conversation about this several summits ago. | |
| 17:21:14 | dansmith | I feel like this is a 3-coffee day | |
| 17:30:07 | cburgess | dansmith So if I'm reading the code right... the max_concurrent_live_migrations option protects the source hypervisor from having more then 1 migrations at a time but the destination isn't protected at all. Does this jive with your understanding? | |
| 17:30:38 | dansmith | cburgess: yep, it's outbound | |
| 17:30:55 | dansmith | cburgess: could do the same for inbound, or obey the build counter for inbounds | |
| 17:31:17 | dansmith | the reason was, | |
| 17:31:40 | dansmith | on host-evac-live, you're necessarily hitting one compute node for outbound, but scheduler could spread the targets out | |
| 17:31:52 | dansmith | obviously if you're packing and have an empty host, they'll all want to go to the same place | |
| 17:32:32 | cburgess | dansmith inbound seems like it would be harder... you would have to modify the flow so that the source makes a call to the dest to try and acquire a migration slot or something like that and then keep retrying to get it before it does the migration. Seems like it would be tough to do right an not end up with deadlocks. | |
| 17:33:04 | openstackgerrit | Merged openstack/python-novaclient master: Microversion 2.53 - services and hypervisors using UUIDs https://review.openstack.org/485435 | |
| 17:33:09 | dansmith | cburgess: ah, yeah I guess because we make a blocking call for live migrate but not for build, fair point | |
| 17:34:20 | cburgess | dansmith Yeah... it would be very tricky.. which is a bummer since protecting the dest is something we would want to do as well (we actually try very hard to ensure we never have more then 1 in-coming migration at time in liberty due to some pretty nasty os-brick race conditions). | |
| 17:35:40 | cburgess | dansmith But its good to know about these 2 options. I think we will tune the max_concurrent_builds down some to help with performance. | |
| 17:36:06 | dansmith | cburgess: donuts appreciated :P | |
| 17:36:53 | cburgess | dansmith hehe... yeah I didn't do donuts in Boston because mriedem said we wouldn't really have the right setup. I'll look into the donuts making a return in Denver. | |
| 17:37:24 | dansmith | cburgess: I'm only joking.. you're way over your donuts requirement quota | |
| 17:38:28 | vdrok | dansmith: ouch http://logs.openstack.org/58/487458/2/check/gate-tempest-dsvm-ironic-pxe_ipmitool-postgres-ubuntu-xenial-nv/40fb4fb/logs/devstacklog.txt.gz waiting for hypervisors fail | |
| 17:39:05 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/487473 | |
| 17:40:27 | vdrok | the variable picked up by nova http://logs.openstack.org/58/487458/2/check/gate-tempest-dsvm-ironic-pxe_ipmitool-postgres-ubuntu-xenial-nv/40fb4fb/logs/devstacklog.txt.gz#_2017-07-26_16_06_41_216 | |
| 17:40:46 | dansmith | vdrok: gdi, nothing but bad news from you ever | |
| 17:40:56 | vdrok | :D | |
| 17:40:56 | dansmith | vdrok: you never just stop by to say hi, always to say someting is broken! :) | |
| 17:41:20 | vdrok | I promise to start the next day with 'good morning' to nova channel :) | |
| 17:41:25 | cburgess | dansmith You have sent me down a rabbit whole here reading all the options with that link to the code. Its nice to have all the options in one place but man its a bit scary reading some of this stuff. | |
| 17:42:05 | mriedem | is there anything i need to care about in the scrollback? | |
| 17:42:11 | mriedem | besides chet letting us down on donuts in boston? | |
| 17:42:14 | cburgess | mriedem donuts | |
| 17:42:18 | cburgess | lol | |
| 17:42:21 | dansmith | vdrok: http://logs.openstack.org/58/487458/2/check/gate-tempest-dsvm-ironic-pxe_ipmitool-postgres-ubuntu-xenial-nv/40fb4fb/logs/screen-n-cpu.txt.gz#_Jul_26_16_08_05_910044 | |
| 17:42:28 | dansmith | vdrok: looks like it created the compute node... | |
| 17:43:22 | vdrok | yup, but for some reason it's not picked up by hypervisor-stats | |
| 17:44:47 | dansmith | vdrok: and never by discover either | |
| 17:47:46 | dansmith | vdrok: okay I think this might be because this hack was for grenade and it's not complete enough for a fresh install | |
| 17:48:06 | dansmith | vdrok: I think we're pointing at super-conductor (which we want) but it's configured to point at cell0 (which we do not) | |
| 17:48:14 | dansmith | vdrok: so give me a few to mull this over | |
| 17:49:31 | openstackgerrit | Jackie Truong proposed openstack/nova master: Add trusted certificates to InstanceExtras https://review.openstack.org/457711 | |
| 17:56:29 | dansmith | vdrok: updated the devstack change | |
| 17:57:51 | vdrok | dansmith: thanks! will come back if it fails again :D | |
| 17:57:55 | vdrok | good night | |
| 17:58:07 | dansmith | vdrok: o/ | |
| 18:30:22 | mriedem | jangutter: really only nits in the release note now https://review.openstack.org/#/c/483459/ | |
| 18:30:28 | mriedem | but see what you think | |
| 18:32:51 | jangutter | mriedem: I'm working on getting our docs cleared for general release on our support site. Currently customers should have subscriptions, but they should be generally available by the time Pike gets released. | |
| 18:33:17 | jangutter | mriedem: should I respin and remove the soon? | |
| 18:47:58 | sdague | dansmith: your cell1 db change didn't work | |
| 18:47:58 | jaypipes | bauzas: answered. sorry, went to lunch | |
| 18:48:08 | sdague | https://review.openstack.org/#/c/487478 | |
| 18:49:19 | sdague | dansmith: is there a reason you changed that? | |
| 19:01:24 | dansmith | sdague: didn't work how? but yes, we need to make that change | |
| 19:01:55 | dansmith | sdague: remember this was just a grenade hack, which meant conductor and other services were still configured to point at the cell db, not cell0 | |
| 19:02:07 | dansmith | on fresh with this they're configured wrong now | |
| 19:02:09 | bauzas | jaypipes: just reading your comment | |
| 19:02:14 | dansmith | sdague: didn't want to ask before pushing over top eh? | |
| 19:02:29 | bauzas | jaypipes: I'm a bit in and out till' 9pm my time but will think of your comment | |
| 19:02:32 | sdague | dansmith: I did, all the tests failed | |
| 19:02:36 | bauzas | oops 9.30pm | |
| 19:02:43 | sdague | cell1 db does not exist | |
| 19:03:09 | sdague | dansmith: http://logs.openstack.org/78/487478/4/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/69e2d2e/ | |
| 19:03:24 | dansmith | sdague: nova_cell1 should be created on L693 | |
| 19:03:36 | sdague | so given that jenkins -1ed that version of the patch I figured it was fine to run it the old way | |
| 19:03:49 | sdague | dansmith: well, it failed pretty hard | |
| 19:03:52 | openstackgerrit | Jan Gutter proposed openstack/nova master: Netronome SmartNIC Enablement https://review.openstack.org/483459 | |
| 19:04:27 | dansmith | it won't actually work without it, and the ironic people found it before jenkins | |
| 19:04:48 | sdague | dansmith: ok, but it doesn't pass with it | |
| 19:05:13 | dansmith | I get it, but we need to make the change | |
| 19:05:19 | sdague | http://logs.openstack.org/78/487478/4/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/69e2d2e/logs/devstacklog.txt.gz#_2017-07-26_18_12_02_728 | |
| 19:05:36 | sdague | dansmith: that's fine, you need to help me figure out what other thing is also needed for that change to work | |
| 19:05:45 | dansmith | I'm doing it now | |
| 19:07:14 | dansmith | oh it's failing on the sync before we create it.. I wonder why that's different now | |
| 19:09:33 | sdague | because this function is creating the top level nova.conf | |
| 19:10:07 | sdague | https://github.com/openstack-dev/devstack/blob/62edb2f0f64bf4ac2c75e9bbfffbad5aac5ad41c/lib/nova#L671-L680 | |
| 19:10:20 | sdague | the cell1 db is not created before the sync | |
| 19:10:24 | sdague | for the top level | |
| 19:11:01 | dansmith | right, because now that points to cell0 always in the toplevel and it's created already | |
| 19:11:16 | sdague | yes | |
| 19:12:22 | dansmith | so I switched the order of those things which I think will be fine for both cases | |
| 19:14:31 | sdague | ok, if that works, good enough | |
| 19:21:33 | sdague | dansmith: thanks for the grenade patch fix, I should put a safety wrapper around that for things that only need to set variables | |
| 19:22:03 | dansmith | sdague: np | |
| 19:33:01 | openstackgerrit | Sean Dague proposed openstack/nova master: Remove the useless fake ExtensionManager from API unittests https://review.openstack.org/486416 | |
| 19:33:11 | openstackgerrit | Sean Dague proposed openstack/nova master: Move the note about '/os-volume_boot' to the correct place https://review.openstack.org/486071 | |
| 19:37:28 | mriedem | jangutter: +2, just need jaypipes to +W now | |
| 19:38:28 | jangutter | mriedem: I sincerely owe you guys. Thanks! | |
| 19:52:19 | bauzas | jaypipes: so I'm back | |
| 19:52:28 | jaypipes | bauzas: from outer space? | |
| 19:53:22 | bauzas | jaypipes: heh, just from my backyard :p | |
| 19:54:04 | bauzas | anyway | |
| 19:54:08 | bauzas | so I saw your comment | |
| 19:54:40 | bauzas | jaypipes: (and others) https://review.openstack.org/#/c/483566/14/nova/scheduler/filter_scheduler.py@170 | |
| 19:55:34 | bauzas | so, yeah, we have a self-heal by the RT.update_available_resource(), sure, agreed | |