Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-01-13
22:56:25 mriedem ok, RP ?resources= patch update coming
22:56:31 mriedem with a bunch more edge case testing
22:57:39 openstackgerrit Matt Riedemann proposed openstack/nova: Expose a REST API for a specific list of RPs https://review.openstack.org/392569
22:57:40 mriedem OH F YEAH ^
22:57:40 openstackgerrit Dan Smith proposed openstack/nova: Add some cellsv2 setup docs https://review.openstack.org/420198
22:57:49 openstackgerrit Davanum Srinivas (dims) proposed openstack/nova: [WIP] Ignore me - ry py35 functional with a clean slate https://review.openstack.org/419968
22:57:54 mriedem bauzas: i addressed my concerns in https://review.openstack.org/392569
22:57:57 mriedem and added a bunch more testing
22:58:53 mriedem plus more docs about min/max unit and step size consideratoins
22:58:56 mriedem *considerations
23:00:03 mwhahaha dansmith: it should noted that these command in the case of puppet may get logged which is why I asked for using the config and not cmd line params. we have things to hide secrets when they get put into the config files but not so much for command line executions
23:00:39 dansmith mwhahaha: I understand your plight
23:01:02 mriedem does it matter that we leak tokens in the logs all over the place?
23:01:05 dansmith mwhahaha: hopefully you understand my point about illustrative examples with big fat caveats
23:01:10 mriedem and people post the logs to bug reports and kibana
23:01:23 mwhahaha we're not supposed to be leaking tokens anywhere
23:01:27 mriedem if your token expires fast enough it's not an issue
23:01:33 mriedem but...
23:01:40 mwhahaha but yea, i get it but passwords don't change like tokens :D
23:01:55 melwitt mriedem: yeah, it matters. we had some patches to redact tokens in the logs
23:02:04 mriedem i know, but it's whack a mole
23:02:10 melwitt yes. that's true
23:02:12 mwhahaha always is
23:02:51 mwhahaha database passwords are worse because many times the permissions people are setting with the service users are wide open
23:02:51 dansmith these credentials are super easy for you to programmatically rotate too, if you're concerned about it
23:03:37 mwhahaha well they should be, but i've done my fair share of db password rotations and it never goes well
23:03:55 mwhahaha cause someone ultimately uses it for something aren't supposed to and the world falls apart :D
23:04:36 melwitt that depends on your infra whether it's easy to rotate them, but yeah
23:04:38 dansmith well, I see we're far outside the realm of realism here
23:04:52 dansmith melwitt: have you got anything brewing for that create_cell patch?
23:04:55 mriedem sdague: alex_xu: i'm +2 on this now https://review.openstack.org/#/c/392569/
23:05:05 melwitt it was not a picnic when the passwords got compromised and had to be changed, in my recollection
23:05:22 dansmith oh,
23:05:22 melwitt dansmith: yeah, it's failing py35 test only and I'm trying to fix it
23:05:32 dansmith I see you already did
23:06:14 mriedem alright i'm heading out
23:06:19 mriedem will be back on this week i'm sure
23:06:24 mriedem *weekend
23:06:33 melwitt *sad trombone*
23:06:41 mriedem but getting that scheduler RPs query thing +2ed unblocks the scheduler using placement
23:06:49 mriedem so is a high priority to get that patch in
23:07:04 melwitt is that the one you mentioned to me this morning?
23:07:13 mriedem i don't think so?
23:07:34 melwitt okay. nvm then
23:07:38 mriedem if you're looking, https://review.openstack.org/#/c/398472/ and https://review.openstack.org/#/c/417962/ are +2 and need to get in
23:08:19 melwitt okay yeah it was the first one you mentioned
23:08:48 mriedem ok now i'm out - i'm told there is casserole at home dontchaknow
23:11:15 openstackgerrit Anusha Unnam proposed openstack/nova: Add keypairs field to InstancePayload https://review.openstack.org/419730
23:11:44 slaweq mriedem: hello, can You maybe have a look on https://review.openstack.org/#/c/420176/ ? thx in advance
23:24:43 openstackgerrit Davanum Srinivas (dims) proposed openstack/nova: [WIP] Ignore me - Try py35 functional with a clean slate https://review.openstack.org/419968
23:27:51 openstackgerrit Ken'ichi Ohmichi proposed openstack/nova: Add hacking rule to avoid unnecessary space in method https://review.openstack.org/417582
23:28:12 openstackgerrit melanie witt proposed openstack/nova: Add nova-manage cell_v2 create_cell command https://review.openstack.org/332713
23:49:01 dims down to 1 test in py35 functional black list. yay!
#openstack-nova - 2017-01-14
00:04:47 mriedem i'm going to push an infra change to run the placement job on newton only
00:04:57 mriedem since everything unique about the placement job in newton is now on by default in ocata
00:07:32 melwitt I put up the simple_cell_setup backport to newton at https://review.openstack.org/#/c/420226/ assuming the simple_cell_setup command stays relevant
00:19:42 mriedem cool
01:11:01 mriedem melwitt: reviewed the create_cell patch, i'm not crazy about how it creates a new cell mapping each time, even if the input params point to an existing cell mapping
01:11:11 mriedem it seems our unique constraint on a cell mapping is very loose - it's only the uuid
01:11:44 melwitt mriedem: oh. hm
01:13:11 mriedem i also feel secret shame a bit about how all of this is done via nova-manage for CRUD operations on these resources, but i guess we can skirt that a bit because these are technically things that you setup before starting up nova-api
01:13:39 mriedem thought about that when list and delete operations were mentioned
01:13:55 melwitt yeah
01:14:27 melwitt I thought the same thing and noticed we have the same for the cells v1 nova-manage commands, for probably the same reasons
01:16:21 melwitt to do the checks you mentioned, I think we need some new stuff in the CellMappingList object to query with some filtering. otherwise we have to do a manual scan of everything
01:23:36 mriedem melwitt: yeah we'd have to call CellMappingList.get_all() and manually filter
01:23:46 mriedem unless we add a new method as you said,
01:23:56 mriedem but for this one-off it's easier to just do it in the CLI itself
01:24:36 mriedem i could be wrong about wanting to guard against duplicates, but considering tooling using this i can see someone at some point saying this is a bug because they have 500 cell mappings with different uuids but all the other values are the same
01:24:51 melwitt mriedem: you mean you are for manually filtering for this one-off?
01:25:00 mriedem yeah
01:28:50 mriedem btw, this is funny for 2 reasons: https://github.com/openstack/nova/blob/master/nova/cmd/manage.py#L1338
01:28:57 mriedem 1. --quiet says don't print anything, oops
01:29:09 mriedem 2. uuid is required by argparse already and will validate that it's provided or fail, so we don't need to manually
01:31:18 melwitt heh, yeah
01:31:55 melwitt I hope I didn't review that one :P
01:33:35 mriedem will have a patch directly
01:34:33 mriedem btw, remember the leaked token thing? https://review.openstack.org/#/q/I4d996a2631f61a2c9bbbc7f959e97c7279be023d,n,z
01:35:54 melwitt woo, secure_headers
01:36:09 openstackgerrit Dan Smith proposed openstack/nova: Make metadata server know about cell mappings https://review.openstack.org/417962
01:43:36 openstackgerrit Matt Riedemann proposed openstack/nova: Remove redundant arg check in nova-manage cell_v2 verify_instance https://review.openstack.org/420261
01:47:31 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571
02:44:23 openstackgerrit Matt Riedemann proposed openstack/nova: placement: validate member_of values are uuids https://review.openstack.org/420272
02:47:53 openstackgerrit Matt Riedemann proposed openstack/nova: placement: validate member_of values are uuids https://review.openstack.org/420272
05:46:22 openstackgerrit Tiger Li proposed openstack/nova: modified: nova/virt/libvirt/driver.py https://review.openstack.org/420298
06:28:06 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142
07:17:58 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142
11:40:05 openstackgerrit Francesco Santoro proposed openstack/nova: os-vif: convert libvirt driver to use os-vif for fast path vhostuser https://review.openstack.org/385061
11:55:23 openstackgerrit Francesco Santoro proposed openstack/nova: os-vif-util: set vif_name for vhostuser ovs os-vif port https://review.openstack.org/410737
11:55:23 openstackgerrit Francesco Santoro proposed openstack/nova: os-vif: convert libvirt driver to use os-vif for fast path vhostuser https://review.openstack.org/385061
14:47:05 openstackgerrit OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/420345
15:24:45 weshay_mod anyone around who can review https://review.openstack.org/#/c/399735/
15:55:34 openstackgerrit Matt Riedemann proposed openstack/nova: [placement] Add an "Adding a Handler" section to placement_dev https://review.openstack.org/416304
15:55:35 openstackgerrit Matt Riedemann proposed openstack/nova: [placement] placement_dev info for microversion handling https://review.openstack.org/411946
15:55:35 openstackgerrit Matt Riedemann proposed openstack/nova: [placement] placement_dev info for testing and gabbi https://review.openstack.org/412600
15:55:36 openstackgerrit Matt Riedemann proposed openstack/nova: [placement] Add a bit about extraction plans to placement_dev https://review.openstack.org/416676
16:10:13 openstackgerrit Dirk Mueller proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/409077
18:09:48 openstackgerrit Davanum Srinivas (dims) proposed openstack/nova: Fixes to get all functional tests working on py35 https://review.openstack.org/419968
19:12:44 openstackgerrit Dan Smith proposed openstack/nova: Add some cellsv2 setup docs https://review.openstack.org/420198
19:20:44 dansmith mriedem: I kinda think we should fix that nova_api_cell0 silliness, and I'm happy to work the two-step dance with devstack, but I left it the way it was until you comment

Earlier   Later