Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-04-06
16:52:15 arvindn05 mriedem: is the placement api called anytime during the migrations?
16:52:15 mriedem but, when migrating to a new host, the scheduler will validate the image used to create the instance as well
16:52:20 mriedem yes
16:52:25 mriedem as part of scheduling to find a new dest host
16:52:36 mriedem rebuild is weird in that the instance doesn't move
16:52:50 mriedem but the image can change so we have to re-validate that the new image can work on the same host
16:53:19 mriedem the original image used to create the instance is stored in the request_spec for that instance
16:53:31 mriedem so that's what the scheduler uses during a move operation
16:54:23 arvindn05 so when you say the scheduler will validate the image, the placement API will be requested for a matching compute host?
16:54:57 arvindn05 i see there is a "Manual selection of the destination host" per https://docs.openstack.org/nova/pike/admin/live-migration-usage.html
16:55:03 mriedem yes. scheduling asks placement for compute node resource providers that can fit the flavor constraints, and optionally qualitative traits defined in the flavor extra specs
16:55:37 mriedem arvindn05: yes, but that still runs through the scheduler but with the requested host
16:55:54 mriedem if you specify the force parameter to live migration with a host, then we bypass the scheduler and, well, you get what you ask for
16:56:17 arvindn05 ok...and we would get no valid host if when run through the scheduler, the host does not match
16:56:24 mriedem correct
16:57:22 arvindn05 can you point me to the code where the live migration calls the scheduler?(assuming it calls select destination method)
16:58:14 mriedem arvindn05: https://github.com/openstack/nova/blob/0de6ebd086656066c4dcc5c093b5cc7e92e035ac/nova/conductor/tasks/live_migrate.py#L319
16:59:29 arvindn05 mriedem: thank you very much :)
17:01:57 arvindn05 mriedem: btw https://review.openstack.org/#/c/557795/ -Update ImageMetaProp object to expose traits is ready for review as well
17:06:35 mriedem arvindn05: yeah i wanted to post this https://review.openstack.org/#/c/557795/14/nova/objects/image_meta.py@560 and get the ack from superdan and leakypipes first so we're on the same page
17:13:56 mtreinish superdan: is that ordering documented anywhere?
17:14:09 superdan mtreinish: it's documented in the right order
17:14:12 superdan mtreinish: this matches the docs
17:14:13 mriedem mtreinish: it is now https://review.openstack.org/#/c/559357/
17:14:18 superdan mriedem: dude, shsssshh
17:14:21 mriedem ha
17:14:51 mtreinish heh, ok well as long as it's written down now
17:15:17 mriedem mtreinish is like the substitute teacher on a friday that lets things slide
17:15:19 mtreinish +A
17:15:23 mriedem see!?
17:15:35 superdan hah
17:17:12 mtreinish lol
17:17:12 sean-k-mooney by the way when ye refer to cells in that doc does the ordering of api first apply equally to v1 and v2
17:20:06 sean-k-mooney i generally try to not know how cells v1 works in case anyone will ask me to help fix somthing with cellsv1 in the furutre but just taught i would ask.
17:21:36 mriedem don't ask
17:21:47 mriedem the rule with "helping people with cells v1" is, don't help them
17:22:01 sean-k-mooney :)
17:22:02 mriedem failing that, goto superdan or jgwentworth
17:23:59 sean-k-mooney i approch cellsv1 the opisitway i approch most of the rest of the codebase. i want to know how most of the codebase works so i can help others but with cellsv1 i tend to follow the above advice
17:25:10 mriedem finucannot: wtf does this mean? http://logs.openstack.org/70/559370/1/check/build-sphinx-docs/bda526f/job-output.txt.gz#_2018-04-06_16_47_02_376553
17:25:13 mriedem and how do i fix it?
17:27:04 ildikov mriedem: it looks a little like when the 'todo' extension was missing from the config
17:27:09 openstackgerrit Matt Riedemann proposed openstack/nova master: Cleanup _get_request_spec_for_select_destinations for live migrate https://review.openstack.org/559381
17:28:07 mriedem ildikov: it's something with this new PTI docs change i think, but i don't have the heart to read that thread right now
17:36:18 sean-k-mooney mriedem: do you need the "-" in query: >- also gr say sphinx>=1.6.2,!=1.6.6,!=1.6.7 # BSD but the test requirements in elastic-rechec is sphinx>=1.1.2,<1.2
17:37:03 sean-k-mooney mriedem: so maybe just use a non achient version of sphinx?
17:38:46 sean-k-mooney mriedem: it does not look like the openstack requirement proposal bot keeps the repo in sync so when they bumped to new pbr on march 1st perhaps that has issues since pbr is trying to move to sphinx >1.6
17:41:24 sean-k-mooney lower constratings also reqiures 1.6.5 https://github.com/openstack/requirements/blob/master/lower-constraints.txt#L359 so the 1.2 max in openstack-infra/elastic-recheck is proable the issue
17:43:44 arvindn05 mriedem: thanks. jaypipes said he would approve as well...will work on my next patch for the BP and also start thinking about the resize issue for rocky
17:55:21 sean-k-mooney mriedem: https://review.openstack.org/559390 fixes your issue. do you want me to rebase your patch on top.
18:01:18 mriedem sean-k-mooney: i can, was looking at the pti email and doing some stuff, but if this works, sure
18:02:40 sean-k-mooney mriedem: locally at least this fixes it for me. i cherry picked you patch locallly and that passed the docs job too after i bumped the version
18:03:09 cdent leakypipes, fried_rolls, figleaf : this might be of interest, a sort of placement test env: https://github.com/cdent/placecat (not fully fleshed out yet, just playing so far)
18:04:05 sean-k-mooney mriedem: by the way pti do you mean the "[openstack-dev] Following the new PTI for document build, broken local builds" thread from finucannot
18:04:18 leakypipes cdent: nice :)
18:06:31 sean-k-mooney mriedem: hum the chages for pti seam simple ill try making those instead
18:06:45 sean-k-mooney mriedem: then i need to go home because its 7 on a friday
18:35:28 figleaf cdent: sweet. Wish I had time to play with it. :)
18:43:37 openstackgerrit Merged openstack/python-novaclient master: Fix validation for command arguments https://review.openstack.org/558324
18:51:16 openstackgerrit Merged openstack/python-novaclient master: Fix AttributeError in getting a resource ID https://review.openstack.org/558328
18:51:17 openstackgerrit Merged openstack/python-novaclient master: Replace GB with GiB https://review.openstack.org/558364
18:54:26 openstackgerrit sean mooney proposed openstack/os-vif master: convert os-vif docs to follow PTI https://review.openstack.org/559398
18:54:36 openstackgerrit Matt Riedemann proposed openstack/nova master: doc: add a link in the install guides about configuring neutron https://review.openstack.org/559115
18:54:36 openstackgerrit Matt Riedemann proposed openstack/nova master: Log a more useful error when cinder auth isn't configured https://review.openstack.org/559399
19:00:46 fried_rice cdent: Neat!
19:01:30 fried_rice cdent: It's basically a fully functional but not-for-production placement service that you can set up to have something to test/play against?
19:01:49 cdent fried_rice: since you're not one with the twitters, you may not be aware of the placedock (or the blog posts associated with it) stuff, if you go to the placedock repo there's loads of links to context
19:01:57 cdent fried_rice: pretty much
19:02:15 cdent depending on the variables you pass into the container
19:02:26 cdent (and I recently found a bug with that)
19:03:07 cdent and in terms of production, you could, except that the container builds from non master code
19:07:09 openstackgerrit sean mooney proposed openstack/os-vif master: make legacy-tempest-dsvm-nova-os-vif voteing https://review.openstack.org/559401
19:31:25 sean-k-mooney cdent: i have been following your tweets on it on and off its a cool idea
19:33:48 sean-k-mooney cdent: have you considerd useing any of this work as part of functional testing for placemnet
19:34:24 cdent sean-k-mooney: not realy: because of the way the existing gabbi tests work with wsgi-intercept, it's not really needed
19:34:32 cdent using it for some kind of integration tests might be useful thought
19:34:39 sean-k-mooney cdent: when placement has its own repo this could be an interesting way for folks to do placement dev/testing
19:34:46 cdent yes
19:34:56 sean-k-mooney cdent: well i was thinking it would be kind of like neutron fullstack tests
19:35:13 sean-k-mooney they are more then fuctional tests but not quite intergration tests
19:35:30 cdent sean-k-mooney: what do you see as the difference between using the container and what the current wsgi-intercept driven tests do?
19:36:51 sean-k-mooney cdent: noting really other then the fact that you are using sqlite right so i dont need to setup a db first so it would mean a smaller testing requiremtes set
19:38:19 cdent the container can use any database, depending on environment variables
19:38:31 cdent the functional tests use sqlite, yeah?
19:39:00 sean-k-mooney am im not sure what they use but were you not tweeting about using sqlight
19:39:43 sean-k-mooney i could be getting confused between what can work and what that suff is using today
19:41:01 sean-k-mooney anyway this time im actully going to leave since its 8:40 on a friday. enjoy your weekends o/
19:43:40 cdent o/
20:45:19 mriedem for the life of me i can't figure out why this isn't a problem anymore https://bugs.launchpad.net/nova/+bug/1718512
20:45:19 openstack Launchpad bug 1718512 in OpenStack Compute (nova) "migration fails if instance build failed on destination host" [Medium,In progress] - Assigned to Matt Riedemann (mriedem)
20:46:01 mriedem or, how it was a problem to begin with - since i don't know where we save 'retry' changes to a request spec during the initial build
20:49:44 mriedem superdan: i think i see a fun little weirdness, we update the requestspec with a new flavor during resize before we've actually confirmed that the resize happened https://github.com/openstack/nova/blob/master/nova/conductor/manager.py#L349
20:49:54 mriedem b/c at that point we've rpc cast to the dest compute to start the resize
20:50:13 mriedem if it fails on the compute, the request spec now says that the instance is using a new flavor
20:51:02 superdan mriedem: and that matters because why? if you later evacuate maybe?
20:51:12 mriedem it might not matter right now,
20:51:16 mriedem i was thinking,
20:51:33 mriedem if resize does a reschedule to the cell conductor, we could try saving changes to the request spec, because of the retry stuff
20:51:36 superdan actually, I think evac is probably fine since it likely uses the instance
20:51:39 mriedem which would be an upcall to the api db

Earlier   Later