Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-09-21
13:56:18 mriedem haven't gone through all of it, i just know it's a special snowflake in the existing code
13:56:32 openstackgerrit Merged openstack/nova master: Remove 400 as expected error https://review.openstack.org/505485
13:56:35 mriedem i think, however, we have a functional regression test that i added for filtering on ip
13:57:00 mriedem because i remember once last year accusing steve kaufer of regressing that
13:57:02 dansmith oh, filtering not sorting?
13:57:06 mriedem right
13:57:10 mriedem filter by ip regex
13:57:12 dansmith yeah, that shouldn't be any different
13:57:40 dansmith you can only sort on access_ip_vX, which are strings and fine
13:57:46 openstackgerrit sahid proposed openstack/nova master: libvirt: bandwidth param should be set in guest migrate https://review.openstack.org/497455
13:57:47 openstackgerrit sahid proposed openstack/nova master: libvirt: add method to configure migration speed https://review.openstack.org/497456
13:57:48 openstackgerrit sahid proposed openstack/nova master: libvirt: slow live-migration to ensure network is ready https://review.openstack.org/497457
13:57:53 dansmith filters basically go straight through like they did before
13:58:00 openstackgerrit Merged openstack/nova master: pci: Validate behavior of empty devname https://review.openstack.org/396130
13:58:20 mriedem straight through to the db api?
13:58:25 openstackgerrit Jesse Pretorius (odyssey4me) proposed openstack/nova master: Add default configuration files to data_files https://review.openstack.org/506188
13:58:39 mriedem if yes, that's my concern, since we don't filter ips in the db
13:59:04 mriedem but i'll look for this test before declaring that the sky has fallen
13:59:07 dansmith if they were passed to the db api before then they still are, and if they weren't then they're not
14:00:10 dansmith mriedem: nova/tests/functional/test_list_servers_ip_filter.py
14:00:39 mriedem dansmith: that's the one
14:00:40 mriedem cool
14:03:05 sahid mriedem: can i have at some point your eyes on https://review.openstack.org/#/c/485522/, the code https://review.openstack.org/#/c/458513/
14:03:11 sahid you were asking for a spec
14:04:52 mriedem sahid: sure, i'll star it for later today
14:05:32 sahid mriedem: ok thanks
14:07:08 openstackgerrit Merged openstack/nova master: Adopt new pypowervm power_off APIs https://review.openstack.org/476945
14:22:18 dansmith tssurya: this is the place :)
14:23:16 tssurya dansmith: yep I will be here then :) thank you
14:36:29 gibi dansmith: I'm looking at the scatter_gather code and I don't think I spotted the part that prevents us to return a generator as a result.
14:36:51 gibi dansmith: I even replaced the list() call with a generator expression in your first patch and that seems to still pass the tests
14:39:41 dansmith gibi: hmm, I initially was returning a generator from it and was hitting something in there that was trying to count the result
14:40:17 dansmith gibi: but I guess maybe I was generating an error in one of the threads and that threw me off
14:40:27 dansmith let me try to change it again locally now that things are all working and see
14:41:06 gibi the only length calulation is here https://github.com/openstack/nova/blob/193729b93a11ff54da99386e881270209797f020/nova/context.py#L439 but that is only checking the number of results not the lenght of each result
14:42:26 dansmith yeah
14:42:35 dansmith hmm, yeah, that seems to work
14:43:04 dansmith I dunno what I was hitting before, but I initially had this so it was generators all the way down, convinced myself that wouldn't work across that thread boundary and then wrote that docstring about it
14:43:21 dansmith so... I dunno, but that was super early, so maybe something else was going on
14:43:26 dansmith I shall change it
14:45:27 stephenfin sdague: Any tips on how to find changes that have not been reviewed by anyone? 'NOT label:Code-Review<=2 age:5d' no longer seems to work
14:46:01 stephenfin This is in relation to Gerrit and your 'dashboard query changes since upgrade' mail (which I've yet to receive :()
14:46:03 johnthetubaguy I think sdague refreshed the openstack dashboard url for the new gerrit version
14:46:03 mriedem dansmith: you won me over on the sort keys/dirs defaults thing in https://review.openstack.org/#/c/504983/
14:46:10 mriedem i'd be cool with dropping that handling
14:46:24 stephenfin johnthetubaguy: Aye, but I don't think that change is in there.
14:46:27 gibi dansmith: cool
14:46:47 johnthetubaguy stephenfin: ah, fair enough
14:47:04 dansmith mriedem: why not just leave what I have?
14:47:16 dansmith mriedem: you can't pass dirs without keys, and can't pass dirs that don't match keys in terms of length
14:47:19 mriedem you can, it's just kind of dead code as you said yesterday
14:47:29 dansmith yeah, but now I have tests for it
14:47:34 mriedem :/
14:47:53 mriedem up to you, i'm +2 either way
14:48:01 mriedem are you going to redo this series to make do_query return a generator?
14:48:24 dansmith yeah, I'm doing it now
14:48:33 dansmith I wish I hadjust done it at the end because of the conflict it makes,
14:48:37 dansmith but I'm in the middle now anyway
14:49:32 openstackgerrit Dan Smith proposed openstack/nova master: Add base implementation for efficient cross-cell instance listing https://review.openstack.org/504983
14:49:33 openstackgerrit Dan Smith proposed openstack/nova master: Make instance_list honor global query limit https://review.openstack.org/504984
14:49:33 openstackgerrit Dan Smith proposed openstack/nova master: Add db.instance_get_by_sort_filters() https://review.openstack.org/504985
14:49:34 openstackgerrit Dan Smith proposed openstack/nova master: Support pagination in instance_list https://review.openstack.org/504986
14:49:34 openstackgerrit Dan Smith proposed openstack/nova master: Add fault-filling into instance_get_all_by_filters_sort() https://review.openstack.org/505391
14:49:35 openstackgerrit Dan Smith proposed openstack/nova master: Add tests to validate instance_list handles faults correctly https://review.openstack.org/505392
14:49:35 openstackgerrit Dan Smith proposed openstack/nova master: Add get_instance_objects_sorted() https://review.openstack.org/505417
14:49:36 openstackgerrit Dan Smith proposed openstack/nova master: Copy some tests to a cellsv1 mixin https://review.openstack.org/505442
14:49:36 openstackgerrit Dan Smith proposed openstack/nova master: Fix a pagination logic bug in test_bug_1689692 https://review.openstack.org/505661
14:49:37 openstackgerrit Dan Smith proposed openstack/nova master: Use improved instance_list module in compute API https://review.openstack.org/505418
14:49:37 openstackgerrit Dan Smith proposed openstack/nova master: Remove legacy fault-loading routines https://review.openstack.org/505456
14:49:47 dansmith mriedem: gibi: ^
14:49:58 gibi dansmith: looking...
14:50:55 dansmith gibi: I'm still trying to figure out why that weird raise is .. weird.. but I can stack whatever the fix is on top
14:53:45 gibi dansmith: OK. thanks for the explanation. As you guessed I haven't tried to move the raise in the except block myself
14:55:27 gibi dansmith: It seems you forget to update the code comments when you change the list to the generator expression in https://review.openstack.org/#/c/504983
14:55:32 mriedem andreykurilin: hey i assume you like to do some scale and performance testing since you work on rally, right?
14:56:54 dansmith gibi: I updated the docstring... did I miss a reference?
14:57:38 gibi dansmith: https://review.openstack.org/#/c/504983/6/nova/compute/instance_list.py
14:57:43 dansmith what in the
14:57:48 gibi dansmith: L69 and L93
14:57:58 dansmith craaap
14:58:06 dansmith I must have dumped it during a rebase
14:58:09 dansmith urgh
14:58:12 dansmith I _did_ update ;)
14:58:20 gibi I believe you :)
14:58:23 openstackgerrit Elod Illes proposed openstack/nova master: Add instance.interface_attach notification https://review.openstack.org/503089
14:58:50 dansmith gibi: from my history: https://pastebin.com/u7nUYi38
14:58:51 dansmith :P
14:59:20 openstackgerrit Eric Fried proposed openstack/nova-specs master: Spec: Use keystoneauth1 Adapter for endpoints https://review.openstack.org/500190
14:59:31 efried mriedem edmondsw ^
14:59:52 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix hyperlinks in document https://review.openstack.org/506101
14:59:54 edmondsw efried ack
15:01:06 openstackgerrit Takashi NATSUME proposed openstack/nova master: Fix hyperlinks in document https://review.openstack.org/506101
15:01:53 openstackgerrit Dan Smith proposed openstack/nova master: Add base implementation for efficient cross-cell instance listing https://review.openstack.org/504983
15:01:54 openstackgerrit Dan Smith proposed openstack/nova master: Make instance_list honor global query limit https://review.openstack.org/504984
15:01:54 openstackgerrit Dan Smith proposed openstack/nova master: Add db.instance_get_by_sort_filters() https://review.openstack.org/504985
15:01:55 openstackgerrit Dan Smith proposed openstack/nova master: Support pagination in instance_list https://review.openstack.org/504986
15:01:55 openstackgerrit Dan Smith proposed openstack/nova master: Add fault-filling into instance_get_all_by_filters_sort() https://review.openstack.org/505391
15:01:56 openstackgerrit Dan Smith proposed openstack/nova master: Add tests to validate instance_list handles faults correctly https://review.openstack.org/505392
15:01:56 openstackgerrit Dan Smith proposed openstack/nova master: Add get_instance_objects_sorted() https://review.openstack.org/505417
15:01:57 openstackgerrit Dan Smith proposed openstack/nova master: Copy some tests to a cellsv1 mixin https://review.openstack.org/505442

Earlier   Later