Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-02-28
14:49:42 johnthetubaguy mdbooth: but why not guard the count with a regular lock? with eventlet I can't see there being much of a gain between the two, and one is more obviously "OK ish"
14:49:54 johnthetubaguy I am probably missing something big
14:50:35 mdbooth johnthetubaguy: Yeah, so I want to wait until the in-use count reaches zero.
14:50:46 mdbooth I take the lock
14:50:51 mdbooth It's not zero
14:50:55 mdbooth What do I do?
14:51:12 mdbooth This is exactly what the Condition api is for
14:51:30 johnthetubaguy mdbooth: ah, so its because you want to wait rather than just sleep or something dumb like that, thats what I was missing
14:51:49 mdbooth Yeah, otherwise you have to drop the lock and poll
14:52:09 mdbooth That uses more resources, and it's more complicated
14:52:22 mdbooth And there's an existing API which designed for exactly this
14:52:39 johnthetubaguy mdbooth: we could just do callbacks on triggers I guess?
14:52:51 mdbooth That's not sounding simpler :)
14:53:04 mdbooth Condition gives us: wait until X is true
14:53:10 openstackgerrit Jay Pipes proposed openstack/nova master: placement: ironic inventory reporting https://review.openstack.org/437602
14:54:51 mdbooth johnthetubaguy: I also wrote a slightly funky test library for this, btw
14:54:56 sdague mriedem: thank you for sharing
14:55:08 mdbooth It's in that patch, but it's entirely generic
14:55:30 mdbooth It allows you to insert arbitrary wait points in test code so you can force race conditions
15:03:28 openstackgerrit Sylvain Bauza proposed openstack/nova master: Fix default_availability_zone docs https://review.openstack.org/438887
15:03:45 sfinucan mdbooth: The fact that your Red Hat email isn't also mdbooth annoys me
15:04:00 sfinucan mriedem: That run_tests patch you requested is here https://review.openstack.org/#/c/437947
15:05:13 johnthetubaguy mdbooth: maybe the bit I am not getting is why you have to reset the state when you loose connection to libvirt?
15:14:28 mdbooth johnthetubaguy: That was originally danpb's suggestion. It's a robustness measure.
15:14:49 mdbooth johnthetubaguy: We need to tie initialisation to host_up anyway, because initialisation uses the libvirt connection.
15:15:01 johnthetubaguy yeah, host up makes sense
15:15:07 mdbooth So re-initialising it after host_down is no great complication.
15:15:08 johnthetubaguy just seems odd to drop it
15:15:28 mdbooth johnthetubaguy: Well, if libvirt went down we don't really know what happened.
15:16:06 mdbooth That doesn't really complicate the code all that much, anyway, as it's separate in the manager
15:17:24 bauzas jaypipes: around ?
15:17:43 bauzas jaypipes: I have a question for you before +Wign https://review.openstack.org/#/c/376198/11
15:18:24 mdbooth The code is especially paranoid as it's reference counting. We want to do our best to be correct, even if operators are doing unexpected things underneath us, or things are crashing.
15:18:41 bauzas jaypipes: ie. we set the collation to latin1 encoding, I guess it's for memory reasons, right?
15:19:19 bauzas I just wonder if that's really worth it in case we would need to add another column needing utf8 encoding, like a description TEXT field
15:20:34 jaypipes bauzas: gimme a bit.. back to back meetings
15:20:43 bauzas heh, no worries
15:21:10 mriedem bauzas: before you go approving that code,
15:21:11 mriedem https://blueprints.launchpad.net/openstack/?searchtext=resource-provider-traits
15:21:15 mriedem can we get the bp actually approved?
15:21:34 bauzas oh very good point
15:21:49 bauzas was there a spec reproposed ?
15:22:09 mriedem https://review.openstack.org/#/c/345138/
15:22:12 bauzas there it is https://review.openstack.org/#/c/345138/
15:22:14 bauzas yeah
15:23:00 bauzas here is the thing, given it would take more time for that as we haven't previously approved the spec
15:23:17 bauzas mriedem: I would then wonder if it would be worth decoupling that spec into 2 pieces
15:23:30 mriedem i haven't read the spec,
15:23:31 spotz Hey nova, anyone about who could review the doc we're working on for the placement API and cells and give feedback? We can't build instances so we're missing something https://review.openstack.org/#/c/438328/
15:23:33 bauzas mriedem: one would be the the basement piece we agreed as the top prio
15:23:38 mriedem if it's the full hog of traits, then yeah probably
15:23:42 bauzas mriedem: the other would be exposing it thru the API
15:23:48 mriedem since we just need the piece to fix can_host
15:24:14 mriedem spotz: have you seen the docs in the nova devref?
15:24:28 mriedem https://docs.openstack.org/developer/nova/cells.html
15:24:33 mriedem https://docs.openstack.org/developer/nova/placement.html
15:24:59 dansmith spotz: also, kinda need to know the error you're hitting to know what you're missing
15:27:05 spotz dansmith: I'm getting the very unhelpful No valid hosts, I'm not sure what Brian got last night though
15:27:24 dansmith spotz: got host_mappings in the apidb?
15:28:41 spotz dansmith. Let me bring up the VM and double check
15:30:08 spotz dansmith: Nice call, table is empty
15:30:20 dansmith spotz: nova-manage cell_v2 discover_hosts
15:30:21 dansmith -v
15:32:19 spotz dansmith: Nada, not even with debug nova-manage --debug cell_v2 discover_hosts
15:32:40 dansmith spotz: got hosts actually registering with their cell db?
15:32:51 dansmith spotz: compute_nodes and services tables in the cell1 db
15:35:41 spotz dansmith: I thought so, but let me go back to the DB vs output from list_cells
15:36:02 dansmith spotz: -v should show you something at least I thought
15:36:15 dansmith spotz: like "checking cell foo"
15:36:57 dansmith yeah, confirmed
15:37:01 spotz dansmith: I have a nova_cell0 DB no cell1 DB
15:37:09 dansmith spotz: well that's a real problem :P
15:37:31 spotz dansmith: So our create_cell didn't work?
15:38:04 dansmith spotz: create_cell just tells nova where the db is for a given cell
15:38:16 dansmith spotz: you should have an apidb, a cell0 db, and a cell1 db
15:38:26 spotz dansmith: Ok so we're definitely missing steps, I do get this for list_cells though cell1 | 143df967-eb22-4d2c-baff-bfc2538816fd |
15:38:45 dansmith yeah, but that doesn't mean anything if you don't have a db to back it up
15:38:53 dansmith spotz: have you looked through our devref docs mriedem pointed to?
15:39:44 spotz dansmith: yeah that's where the steps came from
15:40:31 dansmith spotz: okay I see, you're doing both of those mappings with the same config, which means they probably point to the same place?
15:40:47 dansmith spotz: an unfortunate side-effect of the config file method for getting to this point :/
15:41:00 macsz sfinucan: lol what about no :D
15:41:03 dansmith spotz: can you pastebin your cel_mappings table?
15:41:20 spotz dansmith: sure hang on
15:41:22 macsz sfinucan: and also from nose's docs: "nose itself supports python 3, but many 3rd-party plugins do not!"
15:41:43 sfinucan macsz: Who says we need plugins, eh? ;)
15:42:48 spotz dansmith: https://gist.github.com/spotz/42b9d15f016073458742268735644578
15:43:12 dansmith spotz: oh, no you do have a cell1 db
15:43:24 dansmith spotz: in that db (nova) do you have services and compute_nodes?
15:45:03 macsz sfinucan: I have had enough of beers last week, so i can invest my friday's evenings to convert tox to nose, if you want it :P
15:45:29 sfinucan macsz: *testrepository to nose
15:45:33 sfinucan tox would stay
15:45:49 spotz dansmith: added as a comment on the gist
15:45:58 macsz sfinucan: ah, sorry, was focused on typing "beer"
15:46:18 sfinucan Ima take a look and see if it would even be possible. The testrepository library annoys me in many ways
15:46:24 sfinucan macsz: :D
15:46:38 dansmith spotz: cool, so can you show me the exact result of "nova-manage cell_v2 discover_hosts -v" ?
15:46:59 spotz dansmith: it didn't like -v as a switch when I tried earlier
15:47:09 spotz --debug gave no results to screen
15:47:39 spotz moved it but no help

Earlier   Later