| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-02-08 | |||
| 15:25:48 | mdbooth | raj_singh: Yeah, that's cool. I still wonder if this needs a spec, though. Doesn't seem very 'architectural', just a bunch of implementation detail. That's not what mriedem is saying, though, so I suggest you don't listen to me :) | |
| 15:25:58 | EmilienM | dansmith: but not 430571 | |
| 15:26:05 | EmilienM | dansmith: and 430571 depends-on 429008 | |
| 15:26:07 | dansmith | oh right I see, I was confusing the two | |
| 15:26:08 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Consider startup scenario in _get_compute_nodes_in_db https://review.openstack.org/430994 | |
| 15:26:10 | mriedem | johnthetubaguy: this is for you (and me) ^ | |
| 15:26:22 | EmilienM | dansmith: I wanted to see more Ci jobs | |
| 15:26:25 | mdbooth | raj_singh: However, I really would prefer to see lyarwood's approach to rescue instead. It's both simpler and more functional. | |
| 15:26:37 | dansmith | EmilienM: sounds like a likely making for a setup race or something | |
| 15:26:56 | raj_singh | mdbooth: I think its good to have discussion on the spec and sort out details as this does not seems to be trivial. | |
| 15:27:11 | raj_singh | mdbooth: sure will ping him as well | |
| 15:27:14 | mdbooth | raj_singh: True. | |
| 15:27:23 | EmilienM | dansmith: ok I'm doing recheck and in 40m if I still see multinode failing, I'll unpin Nova in RDO and see how it goes | |
| 15:28:08 | raj_singh | mdbooth: siva_krishnan pinged you because you were thinking to sort out how to persist domain xml and we thought your input will be valuable. | |
| 15:28:25 | mdbooth | raj_singh siva_krishnan: I'm glad you did, thanks! | |
| 15:28:35 | cdent | EmilienM: I'm around pretty much all evening, so check in with me a things proceed. At the moment I'm pretty much bewildered and don't really understand how the pieces are fitting together ... :( | |
| 15:29:42 | mriedem | mdbooth: raj_singh: and i've -1ed https://review.openstack.org/#/c/429913/1 | |
| 15:29:58 | mriedem | i'd prefer to continue the review on the original change for ocata, | |
| 15:30:18 | mdbooth | mriedem: Are there ownership issues? | |
| 15:30:25 | raj_singh | mriedem: agree that is the direction I gave to siva_krishnan. | |
| 15:31:17 | EmilienM | cdent: no worries, I'll let you know | |
| 15:32:32 | EmilienM | owalsh: during your irc away, I decided to recheck the CI job to see if it's consistly failing and unpin Nova in RDO in a few time. So we'll be able to see how things work in our promotion job because I have some doubts about CI results. If you look 429008 - it passed tripleo some ovb & multinode jobs | |
| 15:32:33 | siva_krishnan | mriedem I saw your comments on previous patch. you wanted a rpc call to check compute capabilities to live migrate instance. As we are already doing it pre live migrate check, would that be really necessary at API level? | |
| 15:32:47 | raj_singh | mdbooth: I doubt it, LM is one of the priority item that OSIC is pushing on and we talked with Paul last cycle and he was OK for us to take it over. | |
| 15:33:19 | raj_singh | mdbooth: and we are happy to work with anyone who is interested in this work. | |
| 15:34:04 | mriedem1 | siva_krishnan: when we do the pre live migration check haven't we already casted off from the API request? | |
| 15:34:32 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Improve assertJsonEqual error reporting https://review.openstack.org/428199 | |
| 15:36:58 | raj_singh | mriedem1: yes | |
| 15:37:08 | siva_krishnan | mriedem1: yes we do. Currently most of the live migration operation happens like that. | |
| 15:38:14 | raj_singh | mriedem1: I talked to johnthetubaguy and he said adding a query before that cast could be expensive. | |
| 15:38:51 | mriedem | raj_singh: siva_krishnan: ok, most of that context is out of my head now, and i'm in meetings, so can't really dive into it again atm | |
| 15:39:00 | mriedem | i'd suggest putting it on the PTG topic list | |
| 15:39:02 | johnthetubaguy | I don't like APIs that wait on a RPC call to return, unless we really need to do that | |
| 15:39:09 | mriedem | https://etherpad.openstack.org/p/nova-ptg-pike | |
| 15:39:14 | johnthetubaguy | yeah | |
| 15:39:23 | mriedem | mdbooth: lyarwood: are you guys going to be at the ptg? | |
| 15:39:25 | raj_singh | mriedem: sure, lets discuss it on a spec and PTG | |
| 15:39:59 | mdbooth | mriedem: Unfortunately not, as I'd already booked a skiing trip. | |
| 15:40:25 | mdbooth | mriedem: And Atlanta just didn't seem quite as much fun as the Als. | |
| 15:40:28 | mdbooth | Alps. | |
| 15:40:33 | mriedem | grenoble?! | |
| 15:40:39 | mriedem | bauzas is giving lessons | |
| 15:40:42 | mdbooth | Not far from bauzas | |
| 15:40:46 | siva_krishnan | mriedem: will add it | |
| 15:40:47 | mdbooth | Val Thorens | |
| 15:41:31 | mdbooth | mriedem: In all seriousness, though, I'm annoyed to be missing it. | |
| 15:42:25 | raj_singh | johnthetubaguy: Can you please use your core power and restore https://review.openstack.org/#/c/347161/. | |
| 15:43:09 | johnthetubaguy | raj_singh: sure | |
| 15:43:16 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'ipv6_backend' https://review.openstack.org/430716 | |
| 15:43:17 | openstackgerrit | Stephen Finucane proposed openstack/nova master: netutils: Ignore 'use_ipv6' for network templates https://review.openstack.org/430910 | |
| 15:43:17 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Fix indentation in conf/netconf https://review.openstack.org/430717 | |
| 15:43:18 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Ignore 'use_ipv6' for port filters https://review.openstack.org/430909 | |
| 15:43:18 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'use_ipv6' https://review.openstack.org/430911 | |
| 15:43:19 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Ignore 'allow_same_net_traffic' for port filters https://review.openstack.org/430912 | |
| 15:43:19 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'allow_same_net_traffic' https://review.openstack.org/430913 | |
| 15:43:20 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'metadata_host', 'metadata_port' https://review.openstack.org/430914 | |
| 15:43:20 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate various 'iptables_' options https://review.openstack.org/430915 | |
| 15:43:21 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate various 'dns_' options https://review.openstack.org/431017 | |
| 15:43:21 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'firewall_driver' https://review.openstack.org/412390 | |
| 15:43:22 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'ebtables_' options https://review.openstack.org/431019 | |
| 15:43:22 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate various 'dns_', 'dhcp_' options https://review.openstack.org/431018 | |
| 15:43:23 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'send_arp_for_ha', 'send_arp_for_ha_count' https://review.openstack.org/431021 | |
| 15:43:23 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'fake_network' https://review.openstack.org/431020 | |
| 15:43:24 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Final cleanups in conf/network https://review.openstack.org/431023 | |
| 15:43:24 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate remaining options in conf/network https://review.openstack.org/431022 | |
| 15:43:51 | mdbooth | sfinucan: Um... I guess the gate can take one for the team? | |
| 15:44:29 | sfinucan | mdbooth: Damn it. Only four/five of those should have gone up | |
| 15:44:34 | sfinucan | *the last four/five | |
| 15:44:50 | mdbooth | Accidental rebase? | |
| 15:45:22 | sfinucan | Looks like it :( | |
| 15:45:30 | sfinucan | Soz | |
| 15:46:07 | sfinucan | Also, me getting carried away deprecating all the things | |
| 15:46:23 | mdbooth | By all means carry on :) | |
| 15:51:23 | mdbooth | mriedem johnthetubaguy: NFS patch thoroughly re-picked over, btw. lyarwood and I went over the CI logs carefully. | |
| 15:55:07 | jaypipes | mdbooth: link pls? | |
| 15:55:37 | mdbooth | jaypipes: https://review.openstack.org/#/c/383859/ | |
| 15:56:04 | jaypipes | mdbooth: on it. | |
| 15:56:11 | mdbooth | jaypipes: Thanks! | |
| 15:56:14 | jaypipes | np | |
| 15:56:40 | jaypipes | mdbooth: need to have something productive to do while on status calls. | |
| 15:56:47 | mdbooth | jaypipes: Hehe | |
| 15:57:34 | mriedem | mdbooth: ok maybe after calls | |
| 15:58:29 | mdbooth | jaypipes: If you're reviewing it for the first time, I'd split it in 2. First convince yourself that mount and umount in _HostMountState do the right thing. Then entirely separately convince yourself that the locking in mount.py means that we only ever let 1 thing operate on 1 mountpoint, and that we make everything wait during initialisation. | |
| 15:59:03 | jaypipes | mdbooth: will do, thanks for the heads up. | |
| 16:02:57 | openstackgerrit | Gábor Antal proposed openstack/nova master: Enable global hacking checks and removed local checks https://review.openstack.org/431030 | |
| 16:09:04 | macsz | sfinucan: hmm good point :) | |
| 16:09:55 | lyarwood | mriedem: sorry wasn't watching irc, yeah I'll be at the PTG, happy to talk about stable rescue and the LM spec in more detail then | |
| 16:23:16 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'default_floating_pool' https://review.openstack.org/431038 | |
| 16:23:16 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Add neutron.floating_pool https://review.openstack.org/431037 | |
| 16:23:17 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Move 'floating_ips' opts into 'network' https://review.openstack.org/431039 | |
| 16:23:47 | sfinucan | mdbooth: I think I've got it out of my system for now :) | |
| 16:29:28 | cfriesen | Got a compute node question...in commit 2c0a306 we changed things to spawn a helper thread. Doesn't this mean that any errors in the helper thread won't be caught by the function decorators? (@wrap_instance_fault, @wrap_instance_event, etc.) | |
| 16:29:40 | cfriesen | I'm wondering if those decorators shouldn't also be on dispatch_live_migration. | |
| 16:32:45 | cfriesen | come to think of it, won't having @wrap_instance_event on live_migration() give odd results because the function will spawn a new helper and immediately return? | |
| 16:45:17 | openstackgerrit | Chris Friesen proposed openstack/nova-specs master: Support Instance Rebuild With Userdata https://review.openstack.org/431052 | |
| 16:56:46 | oomichi | alex_xu: thanks for review | |
| 16:57:05 | oomichi | mriedem: could you take a look at https://review.openstack.org/#/c/410006/ ? | |
| 16:58:12 | mriedem | oomichi: i'm trying to sort through https://review.openstack.org/#/c/430497/ first | |
| 17:05:07 | johnthetubaguy | pumaranikar: did you say you had a question for me? | |
| 17:11:29 | mdbooth | lyarwood: Still around? | |