| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-03-29 | |||
| 14:07:54 | dansmith | I haven't gotten to the two specific tests yet | |
| 14:07:58 | bauzas | ah ok | |
| 14:12:59 | sdague | dansmith: the only thing I can imagine is virtualenv on your base computer is really old? | |
| 14:13:28 | dansmith | sdague: just uninstalling tox and re-installing with python3 seems to have fixed it | |
| 14:13:33 | sdague | dansmith: ah | |
| 14:13:38 | sdague | any idea how old old tox was? | |
| 14:13:45 | sdague | is that in your uninstall logs | |
| 14:14:08 | dansmith | it was from pip before, probably fairly old, not sure I'd have logs of that kind of uninstall | |
| 14:14:13 | dansmith | and it's long since gone off my screen | |
| 14:14:52 | sdague | dansmith: ok, because if there is a new min we need it would be good to know, we can do that in tox.ini, then it will at least freak out if things aren't new enough with an understood message | |
| 14:15:18 | dansmith | sdague: yeah, wish I had thought of it | |
| 14:15:24 | dansmith | I don't think I have pip logs of that anywhere | |
| 14:17:16 | sdague | I honestly expect we probably need 2.3.0 as a minimum | |
| 14:17:21 | sdague | there is a 2.2ish fix around env | |
| 14:17:49 | dansmith | bauzas: and yes, those tests all pass locally :) | |
| 14:18:03 | bauzas | logstash-ing it | |
| 14:18:13 | bauzas | I'm a bit rusty | |
| 14:18:42 | dansmith | sdague: well, the weirdness was that it was importing things from py27 and py35 on the system | |
| 14:18:44 | dansmith | but yeah | |
| 14:19:05 | sdague | yeh, but there was something about basepython and env clobbering each other | |
| 14:21:04 | rpodolyaka | frickler: it's in instance detailed info - http://paste.openstack.org/show/604665/ - i.e. GET /servers/ |
|
| 14:21:05 | dansmith | okay | |
| 14:24:53 | openstackgerrit | Stephen Finucane proposed openstack/os-vif master: Convert all 'ip-link set' commands to pyroute2 https://review.openstack.org/451433 | |
| 14:27:02 | frickler | rpodolyaka: ah, great, the v2.41 is the trick, osc uses 2.1 by default and then it is missing. if I explicitly call for 2.41, I see the correct information. thx alot | |
| 14:27:39 | rpodolyaka | frickler: oops, that may be a red herring - that just what I get for `nova --debug show <uuid>` | |
| 14:27:59 | rpodolyaka | frickler: so just test it on 2.1, I | |
| 14:28:29 | rpodolyaka | *I'm not aware of any changes related to the info we return on attached volumes | |
| 14:28:34 | frickler | rpodolyaka: no, I confirmed that "openstack --os-compute-api-version=2.41 server show test1" shows the attribute, without setting the api version it doesn't | |
| 14:28:42 | frickler | rpodolyaka: was added in 2.16 it seems | |
| 14:28:49 | rpodolyaka | ah, ok | |
| 14:29:15 | frickler | makes me wonder why OSC doesn't use latest API by default | |
| 14:30:35 | rpodolyaka | frickler: neither does python-novaclient. Every new microversion is potentially backward-incompatible - you have to explicitly ask for it and make changes to the code (if necessary) | |
| 14:31:54 | openstackgerrit | Dan Smith proposed openstack/nova master: Make conductor ask scheduler to limit migrates to same cell https://review.openstack.org/438025 | |
| 14:32:07 | sdague | rpodolyaka: that's not true, novaclient CLI uses latest by default | |
| 14:33:31 | rpodolyaka | sdague: hmm, won't it break novaclient if we remove field X from the output of some endpoint and novaclient expects it to be present? | |
| 14:33:54 | rpodolyaka | I thought we defaulted to the minimum | |
| 14:34:06 | rpodolyaka | and only ask for newer version if that's needed | |
| 14:34:26 | rpodolyaka | or if the command, let's say nova show now supports it (and so does the nova-api we talk to) | |
| 14:35:51 | efried | Did nova recently change what kind of threads the compute process uses? | |
| 14:36:12 | rpodolyaka | sdague: or by latest you mean latest supported by novaclient? not the latest returned by nova-api? | |
| 14:36:43 | sdague | rpodolyaka: we default to latest novaclient understands that the server supports | |
| 14:36:55 | rpodolyaka | ah, I see, that makes sense | |
| 14:36:59 | rpodolyaka | frickler: ^ | |
| 14:37:08 | rpodolyaka | frickler: sorry for misleading you :P | |
| 14:37:25 | sdague | efried: nothing that I know of, what kind of things are you running into? | |
| 14:37:55 | efried | sdague Got an open() hanging in a subthread - and it's hanging the entire compute process. | |
| 14:38:55 | sdague | efried: there was an eventlet bump not that long ago | |
| 14:38:56 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Rename '[neutron] ovs_bridge' https://review.openstack.org/450266 | |
| 14:38:56 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Move 'floating_ips' opts into 'network' https://review.openstack.org/431039 | |
| 14:38:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'default_floating_pool' https://review.openstack.org/431038 | |
| 14:38:57 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Add neutron.default_floating_pool https://review.openstack.org/431037 | |
| 14:39:45 | efried | sdague Ah, that could do it. Got a pointer on hand, save me digging? | |
| 14:39:54 | efried | thorst ^^ | |
| 14:40:49 | efried | It's the open() in the GlanceImageServiceV2.download method. I know why it's hanging - it's opening a FIFO and the reader side hasn't been opened yet - but the problem is that it's blocking all the other threads in the process - one of which is the reader side of that fifo :( | |
| 14:41:11 | frickler | sdague: rpodolyaka: so do you happen to know why OSC behaves different? or is it just lagging behind? | |
| 14:41:11 | sdague | efried: check out the openstack dev list on some issues around poll that were found | |
| 14:41:21 | sdague | frickler: ask the osc team? | |
| 14:41:26 | efried | sdague rgr that. | |
| 14:41:37 | efried | frickler I actually ran into this same thing yesterday. | |
| 14:42:10 | frickler | sdague: efried: o.k., moving over to -sdks, thx | |
| 14:42:22 | efried | Talked about it on #openstack-sdks - let me get you an eavesdrop reference... | |
| 14:43:06 | efried | http://eavesdrop.openstack.org/irclogs/%23openstack-sdks/%23openstack-sdks.2017-03-28.log.html#t2017-03-28T17:33:58 | |
| 14:43:19 | efried | frickler ^^ Read down to the 20:21 mark. | |
| 14:44:34 | efried | frickler Adding your voice to the "use latest version" argument might help to get things moving in that direction \o/ | |
| 14:44:50 | frickler | efried: will do, thx for the pointer | |
| 14:45:44 | openstackgerrit | Matthew Booth proposed openstack/nova master: Add InstanceMappingList.get_by_instance_uuids https://review.openstack.org/445139 | |
| 14:45:44 | openstackgerrit | Matthew Booth proposed openstack/nova master: Make server_external_events use an optimal number of db queries https://review.openstack.org/445142 | |
| 14:45:45 | openstackgerrit | Matthew Booth proposed openstack/nova master: Add RequestContext.for_cell() https://review.openstack.org/445140 | |
| 14:45:45 | openstackgerrit | Matthew Booth proposed openstack/nova master: Add InstanceList.get_by_instance_mapping_list() https://review.openstack.org/445141 | |
| 14:52:18 | EmilienM | sdague or mriedem: can you please click on https://bugs.launchpad.net/nova/+configure-bugtracker and paste the text you use ? thanks | |
| 14:52:47 | mriedem | paste the text we use? | |
| 14:52:52 | mriedem | i don't understand the request | |
| 14:53:03 | mdbooth | mriedem: If you like my meta-test, could we re-merge https://review.openstack.org/#/c/450383/ ? | |
| 14:53:29 | EmilienM | mriedem: yes, the text that appears when someone fils a bug | |
| 14:53:34 | mriedem | mdbooth: i'm leaving in about 10 minutes so won't have time for that right now. | |
| 14:53:37 | EmilienM | mriedem: I want to re-use it in tripleo somehow | |
| 14:53:50 | mdbooth | mriedem: Ok. | |
| 14:54:01 | mriedem | EmilienM: like http://paste.openstack.org/show/604670/ ? | |
| 14:54:16 | sdague | EmilienM: is that not visible to you if you just try to open a new bug? | |
| 14:54:32 | EmilienM | mriedem: perfect | |
| 14:54:44 | EmilienM | sdague: yes but I didn't want to spam your bug tracker | |
| 14:54:54 | sdague | EmilienM: you don't have to save the bug :) | |
| 14:55:19 | EmilienM | sdague: in fact, I'm interested by the popup that appears when the bug is saved | |
| 14:55:35 | sdague | EmilienM: ah, yeh, that's another field, let me grab | |
| 14:55:42 | EmilienM | thanks a lot | |
| 14:55:51 | sdague | http://paste.openstack.org/show/604672/ | |
| 14:55:56 | EmilienM | awesome, thx again | |
| 14:56:00 | sdague | no prob | |
| 15:27:33 | mdbooth | dansmith: Did you see this change, btw: https://review.openstack.org/#/c/445142/ | |
| 15:28:10 | mdbooth | It's not incredibly important itself, although I think it's useful, but it illustrates the idea of using object contexts | |
| 15:28:18 | dansmith | mdbooth: yes, but I've been completely swamped with other stuff | |
| 15:28:36 | mdbooth | dansmith: Ok, cool. Just wondered if it had hit your radar. | |
| 15:28:45 | openstackgerrit | Sujitha proposed openstack/nova master: Adding tags field to InstancePayload https://review.openstack.org/407228 | |
| 15:39:36 | jaypipes | sfinucan: hmm, weird. for some reason I was under the impression that emulator threads would all be pinned to the same physical CPU core (when emulator thread policy was ISOLATE) | |
| 15:40:09 | jaypipes | sfinucan: since, my impression was that the emulator threads were actually pretty lightly loaded in a RT environment | |
| 15:40:53 | sfinucan | jaypipes: Not per my understanding, no - emulator cores aren't really any different to instance cores | |
| 15:41:28 | sfinucan | the only difference is that emulator cores are always on node 0 | |
| 15:41:31 | jaypipes | sfinucan: ok, no worries. just a bad recollection of mine, I guess :) | |
| 15:41:42 | sfinucan | plus how they're tracked is totally different :) | |
| 15:41:45 | sfinucan | cool cool | |