Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-07-28
14:46:15 lyarwood superdan: not helped by my `yeah sure I can help with that` follow up ;)
14:46:23 kashyap lyarwood: When you craw out of your cave, your comment isn't clear - does that mean, as a stable maint, do you find it acceptable to backport to Newton, or not?
14:46:28 superdan lyarwood: yeah :/
14:46:34 mriedem lyarwood: you're the new tonyb
14:46:49 mriedem kashyap: bug fixes to newton are ok
14:46:56 kashyap mriedem: Okido, thanks
14:47:10 lyarwood mriedem: what did you do to the old tonyb?
14:47:39 mriedem lyarwood: tonyb signs up to help with everything
14:48:00 sdague fried_rice: I'm about to lose power here as they move our power to another pole
14:48:14 sdague so... poke me later
14:48:18 fried_rice sdague Sweet. I think I may have figured it out. Thanks.
14:55:28 mriedem superdan: looking over my old todo list, i had a note about online data migrations for multiple cells,
14:55:47 mriedem e.g. object code looks in the api db first and if not found, looks in the 'cell' db for looking things up
14:55:57 mriedem but with multiple cells, you wouldn't know which cell db to lookup from
14:56:16 mriedem so that probably goes in our list of things you have to do before you can do multi cell
14:56:48 superdan depends on the migration I guess
14:56:57 superdan mriedem: but you could use multiple configs to handle that today
14:57:28 superdan mriedem: i.e. use the config for one of the cell conductors you have if it doesn't hit the api db, and if it does, just add creds for the api db and the one cell you're looking at
14:57:39 superdan but yeah I'd have to think about the best way to test those things
14:59:47 mriedem ok i was just thinking about like migrating quotas, we look in api, then look in cell, but really if we did it like the api code, we'd iterate the cells to find the quota limit - although something like quota limits in multiple cells would be wonky - it could be default in a new cell but custom in an old cell
15:00:03 mriedem probably just best to say, make sure your online data migrations are done before attempting adding a new cell
15:00:30 superdan yeah
15:00:40 mriedem i should also get this cleaned up https://review.openstack.org/#/c/460377/
15:02:56 superdan ugh yeah
15:03:28 openstackgerrit Ed Leafe proposed openstack/nova master: Handle hash ring rebalancing in ironic flavor migration https://review.openstack.org/487954
15:12:54 mriedem bauwser: going to update the reno in https://review.openstack.org/#/c/484828/ per cdent's comment?
15:13:46 bauwser mriedem: sure, can do
15:14:26 bauwser mriedem: I'm just going to end up my work week prematurely because of corporate network requirements that require major upgrades for me
15:15:26 mriedem bauwser: you know how when i asked if you were talking to your dog when in boston and i said the dog said 'le woof' and you looked like that doesn't make sense/
15:15:27 mriedem ?
15:15:50 mriedem i said the same thing to my wife, who knows a little french, and she did the same, and said, "that doesn't make sense, it would be la woof"
15:16:00 mriedem i can't win
15:16:04 bauwser mriedem: heh
15:16:35 bauwser mriedem: I was litterally about to show my dog yesterday on the hangout as it was bugging me
15:17:28 bauwser anyway, my internal network connection is about to blow out in a few, so \o
15:20:18 bauwser cdent: one tuple (openvpn 2.4, fedora 23)
15:27:57 openstackgerrit Matt Riedemann proposed openstack/nova master: DNM: simulate rpc delay when cleaning up allocations from source node https://review.openstack.org/488500
15:30:26 mriedem superdan: ^ like this?
15:30:56 superdan mriedem: yeah exactly
15:32:06 mriedem leakypipes: i'm going to close out the placement-claims bp
15:32:16 mriedem the main bits are done for pike, rest can be something else in queens
15:32:18 leakypipes k
15:37:32 leakypipes superdan: what was that bug number again? I remember it was a palindrome but my scrollback is too little to grab it..
15:37:41 leakypipes 1707071?
15:37:51 superdan yep
15:38:05 leakypipes got it, thanks
15:56:55 s-dean curl -g -i --cacert "/etc/keystone/tls/ca-cert.pem" -X GET https://10.30.0.2:8774/v2.1/servers/detail -H "User-Agent: python-novaclient" -H "Accept: application/json" -H "X-Auth-Token: {SHA1}3ac08c79ab47c87d8050040c8a18d0774c326933"
15:56:55 s-dean Hi could somebody tell me what, this command queries
15:57:13 s-dean obviously the nova-api
15:57:27 mriedem list servers for that tenant with details
15:57:34 s-dean but what does the servers detail part look at
15:57:48 mriedem https://developer.openstack.org/api-ref/compute/#list-servers-detailed
15:57:51 s-dean because its not returning anything
15:57:57 s-dean i never get a response back
15:58:13 mriedem it hangs?
15:58:17 s-dean yeah
15:58:19 mriedem or you get an empty body?
15:58:24 s-dean no response
15:58:28 mriedem the response should timeout either way
15:58:36 s-dean mhh
15:58:56 mriedem there should be a request id you can trace through the nova-api logs
15:59:11 mriedem look for the GET /servers/detail request in the nova-api debug logs
15:59:16 mriedem there is a req-<uuid> thing
15:59:19 s-dean nova.osapi_compute.wsgi.server [req-b884b02e-e9c0-4845-9dd8-1f4cd1a6215a 2456405c5d3547e5aeece36828578f99 be5ea0b1031d4949ab94107522f305b4 - default default] 10.30.0.2 "GET /v2.1/servers/detail HTTP/1.1" status: 500 len: 566 time: 600.7115021
15:59:20 mriedem you can follow that through the logs
15:59:26 mriedem req-b884b02e-e9c0-4845-9dd8-1f4cd1a6215a
15:59:27 mriedem yeah
15:59:32 mriedem so something exploded
15:59:34 mriedem 500 error
15:59:38 mriedem look for errors
15:59:52 s-dean DBError: Can't reconnect until invalid transaction is rolled back
16:00:18 mriedem huh, don't know what that is
16:00:28 s-dean its asking me to report a bug
16:00:36 mriedem s-dean: are you hand-rolling all this stuff or using some deployment tooling?
16:00:45 s-dean hand-rolling baby
16:00:47 s-dean haha
16:00:48 mriedem misconfiguration of the db wouldn't be a nova bug
16:00:57 s-dean probably a config error again
16:00:59 mriedem may i suggest using a deployment tool
16:01:17 mriedem like openstack-ansible or kolla
16:01:17 s-dean yeah but which oone
16:01:24 mriedem ^ are the ones i'd look at
16:01:25 s-dean i was looking into kolla
16:01:39 s-dean deploys it onto docker right
16:02:21 mriedem yeah
16:02:29 mriedem openstack-ansible deploys into lxc containers
16:02:39 leakypipes builds openstack as docker images is what you mean, but yeah.
16:03:01 s-dean ive just spent so much time learning how to deploy to bare metal
16:03:10 s-dean 3 node setup
16:03:26 s-dean i dont even now how all that stuff ties into my backend network
16:04:10 s-dean i wanted to deploy a self service network with overlay network provider and management hence why ive been doing it by hand
16:05:06 s-dean been one heck of a learning experience i'm glad its the weekend though hahaha
16:10:00 openstackgerrit Jay Pipes proposed openstack/nova master: remove source provider allocs in confirm_resize() https://review.openstack.org/488510
16:11:51 leakypipes mriedem, superdan, bauwser, cdent, figleaf: ^^
16:12:25 leakypipes that didn't mock everything out..
16:12:26 mriedem leakypipes: we do-ish
16:12:39 leakypipes mriedem: see my qualifying statement above :)
16:12:53 mriedem like what?
16:13:08 mriedem we mock out cinder/glance/neutron yeah
16:13:14 mriedem but we don't really need those for testing this

Earlier   Later