| *** annegentle has joined #openstack-nova | 00:09 | |
| *** alexpilotti has quit IRC | 00:10 | |
| *** dims__ has joined #openstack-nova | 00:11 | |
| *** VW has quit IRC | 00:11 | |
| *** dims_ has quit IRC | 00:12 | |
| *** vilobhmm has joined #openstack-nova | 00:12 | |
| *** markvoelker has joined #openstack-nova | 00:13 | |
| *** annegentle has quit IRC | 00:14 | |
| *** otter768 has joined #openstack-nova | 00:15 | |
| *** iamjarvo has quit IRC | 00:16 | |
| *** dims__ has quit IRC | 00:17 | |
| *** SourabhP has quit IRC | 00:17 | |
| *** nickchase has quit IRC | 00:17 | |
| *** dims_ has joined #openstack-nova | 00:17 | |
| *** markvoelker has quit IRC | 00:18 | |
| *** otter768 has quit IRC | 00:20 | |
| *** annegentle has joined #openstack-nova | 00:22 | |
| *** vilobhmm has quit IRC | 00:27 | |
| *** Sukhdev has quit IRC | 00:29 | |
| *** yingjun has joined #openstack-nova | 00:30 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 test https://review.openstack.org/188243 | 00:32 |
|---|---|---|
| eliqiao1 | dims: hi | 00:32 |
| eliqiao1 | dims: do we need to take py34 testing now ? I see py34 is no voting now. | 00:33 |
| jaypipes | my chromium top bar once again looks like a saw-tooth comb... | 00:35 |
| *** jogo is now known as flashgordon | 00:37 | |
| *** nelsnelson has joined #openstack-nova | 00:40 | |
| *** gholler__ has quit IRC | 00:41 | |
| *** marun has quit IRC | 00:42 | |
| *** tsekiyam_ has joined #openstack-nova | 00:42 | |
| jaypipes | melwitt: around still? | 00:42 |
| *** tsekiyam_ has quit IRC | 00:42 | |
| jaypipes | I have a question about https://review.openstack.org/#/c/188261/ | 00:42 |
| jaypipes | melwitt: how is it that an Instance record cannot be created before a request to schedule that instance (or delete it) has been received? | 00:43 |
| jaypipes | dansmith: ^^ too | 00:43 |
| *** mtanino has quit IRC | 00:43 | |
| dansmith | jaypipes: cells | 00:44 |
| jaypipes | dansmith: yeah, but why? | 00:44 |
| jaypipes | or rather, how... | 00:44 |
| jaypipes | is it possible? | 00:44 |
| melwitt | jaypipes: yeah. it's when you request delete of an instance that hasn't yet been scheduled. since cells doesn't know where it could be, it broadcasts "delete" to all cells and then local deletes at the top cell | 00:45 |
| dansmith | jaypipes: I think that it happens when you boot an instance, we're in the middle of syncing the pre-scheduler decision when delete flies in underneath us | 00:45 |
| jaypipes | melwitt: how can anything request to delete something if it doesn't have the UUID of the instance yet? | 00:45 |
| *** tsekiyama has quit IRC | 00:45 | |
| melwitt | jaypipes: any cells that don't have the instance row (which will be at least N - 1) will have this situation where they're asked to delete an instance that doesn't exist in their db | 00:45 |
| jaypipes | melwitt: I see... so this is a condition where the child cell database hasn't had data replicated to it yet, and the delete request comes through? | 00:46 |
| melwitt | jaypipes: before the rest api call returns, there is row created in the top cell db, and the caller has the uuid from there. then if they delete really immediately after that, the broadcast goes out with that uuid | 00:47 |
| jaypipes | gotcha. | 00:47 |
| jaypipes | melwitt: had another querstion about why the tags field is left off the test? | 00:48 |
| melwitt | jaypipes: yeah, when host is None it will send to delete to every cell. so it's not that it's not replicated necessarily, it's that all but one cell will never get that instance anyway | 00:48 |
| jaypipes | k | 00:48 |
| melwitt | jaypipes: I found that tags doesn't have a _save_<field> method like all the other ObjectField fields, so it doesn't get saved during instance.save(). I don't really know what the tags is | 00:50 |
| *** promulo__ has joined #openstack-nova | 00:50 | |
| *** unicell has quit IRC | 00:50 | |
| jaypipes | melwitt: k. tags is the server simple string tagging one-to-many relationship listfield | 00:50 |
| melwitt | jaypipes: I think I could robustify the test by doing a hasattr check instead of just excluding tags, if you want me to do that | 00:52 |
| dansmith | right, there's no _save_tags() handler so this isn't a thing we can hit/test with that | 00:52 |
| jaypipes | melwitt: sure, that would be nice please. :) | 00:52 |
| *** ssurana has quit IRC | 00:52 | |
| *** ssurana has joined #openstack-nova | 00:53 | |
| jaypipes | melwitt: left a comment about adding to the comment in the instance.save() method. otherwise looks good to me. | 00:53 |
| *** amotoki has joined #openstack-nova | 00:53 | |
| *** salv-orlando has joined #openstack-nova | 00:54 | |
| *** promulo has quit IRC | 00:54 | |
| melwitt | jaypipes: thanks for looking at it | 00:54 |
| jaypipes | melwitt: sure thing. push a quick revision and I'll +W it. | 00:55 |
| dansmith | so, | 00:56 |
| dansmith | if we were to pre-generate a uuid and call instance.create(), | 00:56 |
| *** mohitsharma has joined #openstack-nova | 00:56 | |
| dansmith | we could end up with things partially in the database such that we could hit this during cleanup | 00:56 |
| jaypipes | dansmith: that's what is currently happening, no? | 00:56 |
| dansmith | jaypipes: I mean without cells and the content of the comment you asked for | 00:56 |
| dansmith | certainly in our tests, but I think there was a place in the ec2 api code where they generate the uuid of the instance before creation, | 00:57 |
| *** ssurana has quit IRC | 00:57 | |
| dansmith | in which case we could potentially hit this without cells, although that code is going away sometimeish | 00:57 |
| jaypipes | dansmith: instance.create() calls save(), though, no? | 00:57 |
| *** sajeesh has joined #openstack-nova | 00:58 | |
| dansmith | jaypipes: no | 00:58 |
| dansmith | create is a separate thing | 00:58 |
| jaypipes | hmmm? | 00:58 |
| dansmith | instance_update() and instance_create() are the primitives underneath, | 00:58 |
| melwitt | I was thinking over that too. I wrote the comment more generally because if someone does something like create an Instance() dirty a field that's in another table, this would get hit. and I was thinking maybe we don't ever want to have DBReferenceError raised in nova anyway | 00:58 |
| dansmith | and we do both of those after we do things like updating/creating sysmeta, flavors, etc | 00:58 |
| jaypipes | dansmith: no, sorry, I mean that instance.create() creates a record in the DB, regardless. | 00:58 |
| dansmith | melwitt: right, I think your comment is good | 00:58 |
| dansmith | jaypipes: I think you could fail with half of it created | 00:59 |
| * dansmith looks | 00:59 | |
| jaypipes | melwitt: I don't see how you could ever create a child record (instance_extra) without first have created the parent (instances) record when not using cells. | 00:59 |
| dansmith | jaypipes: https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L1655-1655 | 01:00 |
| dansmith | if you fail there, you end up with an ec2 mapping referencing your instance that doesn't exist | 01:00 |
| dansmith | although maybe there's not a FK there? | 01:00 |
| jaypipes | dansmith: line 1652 has already committed the instance to the DB though.\ | 01:01 |
| melwitt | jaypipes: you "could" but if you did it would be a bug, I guess. this will happen if you ever do like inst = Instance(context=context) and then inst.save() if inst.create() has never been called. so I guess the question is, can it happen in practice in things that would make it past CI, I'm not sure | 01:02 |
| jaypipes | melwitt: sure, that's true I suppose... | 01:02 |
| jaypipes | is the main problem here that we have a create() and a save() method instead of just a save() method? | 01:02 |
| jaypipes | or that we don't check for the existence of the parent row in the Instance.save() method before we go adding child records? | 01:03 |
| dansmith | jaypipes: er, yeah, oops | 01:03 |
| dansmith | jaypipes: the real problem is because we don't do this all atomically in save I think | 01:03 |
| *** baoli has joined #openstack-nova | 01:04 | |
| dansmith | well, that's not the problem she's trying to solve, but it's kinda related | 01:04 |
| jaypipes | well, regardless of how we can improve all that, I think melwitt's patch fixes this problem. melwitt, if you could add a comment saying that the cells thing is one case where this might happen, but not the only case, I'd be happy | 01:04 |
| jaypipes | just to keep some documentation for future readers of the code... | 01:05 |
| dansmith | yes, this is definitely a thing we want, I just don't want it to say "this is because cells" and we remove it without thinking when we yank cellsv1 some day | 01:05 |
| jroll | I don't want to interrupt, but when someone has a second, question: if a virt driver artifically reports all resources as used (but has no instances), will the resource tracker try to correct that by looking at what instances are on the compute host/node? | 01:05 |
| *** alanf-mc has quit IRC | 01:05 | |
| jaypipes | jroll: yes. | 01:06 |
| *** annegentle has quit IRC | 01:06 | |
| jroll | jaypipes: I assume that's _update_usage_from_instances (still trying to parse that)? is there a way to configure that to not behave that way? :) | 01:07 |
| jaypipes | jroll: this: https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L393 | 01:07 |
| jaypipes | is called in a periodic task running in the nova-compute daemon. | 01:08 |
| jroll | jaypipes: yeah, tracing through these bits now | 01:08 |
| jaypipes | jroll: here is the piece that reconciles stuff the DB has but the hypervisor has no record of: | 01:08 |
| jaypipes | https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L475-L478 | 01:08 |
| jaypipes | jroll: sorry, I meant what the hypervisor thinks exist but the DB doesn't... | 01:09 |
| jroll | jaypipes: so the ironic driver reports unavailable nodes as having all resources consumed, but does not report instances attached to that compute node | 01:09 |
| *** yingjun has quit IRC | 01:10 | |
| *** salv-orlando has quit IRC | 01:10 | |
| jaypipes | jroll: wait, say what? | 01:10 |
| jroll | not sure if that would count as orphaned | 01:10 |
| jroll | unavailable being in a state that cannot be provisioned from | 01:10 |
| jaypipes | jroll: I actually cannot parse your above statement... | 01:10 |
| *** yingjun has joined #openstack-nova | 01:10 | |
| jroll | jaypipes: https://github.com/openstack/nova/blob/master/nova/virt/ironic/driver.py#L280-286 | 01:10 |
| jroll | jaypipes: so e.g. when an instance is deleted, the node goes through a cleaning process *after* the instance is marked deleted in nova. ironic continues to report the resources as consumed. | 01:11 |
| jaypipes | right... | 01:12 |
| jroll | jaypipes: and I have a feeling we are (once again) breaking world views here | 01:12 |
| jaypipes | hmm. | 01:12 |
| jroll | because it seems nova sees that there's no instance on that compute node and thus exposes resources in the resource tracker | 01:12 |
| jroll | I'm not 100% sure, but a theory I have | 01:12 |
| jaypipes | gime a few | 01:13 |
| jroll | sure thing | 01:13 |
| jaypipes | jroll: I think the Ironic virt driver needs to implement the base driver per_instance_usage() method: https://github.com/openstack/nova/blob/master/nova/virt/driver.py#L1208 | 01:15 |
| morgabra | jroll: https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L780 | 01:15 |
| jaypipes | jroll: so that it can report to nova-compute exactly what resource usages it knowes about. | 01:15 |
| morgabra | right? | 01:15 |
| morgabra | oh, I wasn't scrolled down | 01:15 |
| * morgabra goes away | 01:15 | |
| jaypipes | morgabra: :) | 01:16 |
| *** ijw_ has quit IRC | 01:16 | |
| *** jyuso1 has joined #openstack-nova | 01:16 | |
| jroll | jaypipes: hmm | 01:16 |
| *** ijw has joined #openstack-nova | 01:16 | |
| * jroll pokes aroudn | 01:16 | |
| jroll | jaypipes: so we would report a fake instance consuming that resource? | 01:18 |
| jaypipes | jroll: that way, if Ironic returned "hey, this instance (still in cleanup) is still consuming resources", then nova-compute would remove that set of resources from its deleted instances and write a log message saying there was an orphaned instance (which is correct-ish) during the time that Iironic is cleaning up the deleted node. | 01:18 |
| jroll | seems like it would just create an orphan | 01:18 |
| jaypipes | yeah, you *want* to create an orphan in this case... | 01:18 |
| jaypipes | jroll: because you want to tell nova-compute not to consider that instance UUID as deleted... | 01:19 |
| jaypipes | jroll: for the purposes of resource tracking. | 01:19 |
| jroll | jaypipes: well, there's other states that aren't necessarily related to an instance... | 01:19 |
| jroll | that wouldn't ressurect the instance or whatever, right? | 01:19 |
| jaypipes | nope. | 01:19 |
| jaypipes | it would just keep the amount of resources "used" to a correct amount. | 01:19 |
| jroll | right. interesting. | 01:20 |
| jaypipes | and allow for Ironic to have "instances" in intermediary states that Nova doesn't really know about. | 01:20 |
| jaypipes | make sense? | 01:20 |
| jroll | yes | 01:20 |
| jroll | however | 01:20 |
| jroll | it doesn't even need to be a real instance, right? | 01:20 |
| morgabra | it apparently needs to have a uuid not in the set of uuids the resource tracker knows about | 01:21 |
| jroll | just needs to report resources consumed for some instance ID that isn't in the DB | 01:21 |
| jroll | AIUI | 01:21 |
| morgabra | https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L841 | 01:21 |
| morgabra | yea | 01:21 |
| morgabra | like, we don't have the uuid anymore either | 01:22 |
| morgabra | but could return anything | 01:22 |
| jroll | the interesting part is, this has nothing to do with the compute node, only the compute host | 01:22 |
| *** ijw has quit IRC | 01:22 | |
| jaypipes | jroll: correct. as long as it returns a dict of things keyed by the instance UUIDs that the Ironic driver believes are "active", that is fine. | 01:22 |
| jroll | I presume the node is in the resources dict that method returns | 01:22 |
| jaypipes | jroll: there is a separate instance of the resource tracker class for each node in the compute host. | 01:23 |
| jaypipes | jroll: wonky, I know. | 01:23 |
| jroll | o.o | 01:23 |
| jroll | woah. | 01:23 |
| jaypipes | jroll: all that "worldview" you talked about :) | 01:23 |
| jroll | yeah :) | 01:23 |
| jroll | but get_per_instance_usage is called without any context as to which node | 01:24 |
| jroll | afaict | 01:24 |
| jaypipes | jroll: right. it's the worldview that Nova has without the node stuff :) | 01:24 |
| jroll | so there's no way for our driver to decide which node it's looking at for that | 01:24 |
| jroll | hrm | 01:24 |
| jaypipes | jroll: that get_per_instance_usage() is going to return the exact same (possibly huge) list of instance UUIDs regardless of the compute host and node the resource tracker class instance is for. | 01:24 |
| jroll | jaypipes: right. good times. | 01:25 |
| jaypipes | indeed :) | 01:25 |
| *** Jeffrey4l has joined #openstack-nova | 01:26 | |
| jaypipes | jroll: incidentally, this is why I recommended at the summit that Ironic have its own ironic-compute daemon that understands all of the node stuff and can report things in aggregate to nova-conductor and other things that don't think in terms of a host/node tuple identifier | 01:26 |
| jroll | these are probably instance objects, too | 01:26 |
| jroll | so I'll need to build fake instances for anything with artificially consumed resources | 01:27 |
| jroll | jaypipes: yeah, I understand | 01:27 |
| jaypipes | jroll: what do you mean by "artificially-consumed resources"? | 01:27 |
| jroll | jaypipes: nodes without instances that we want to show as in use | 01:27 |
| jaypipes | jroll: why do you want to show them as "in use" (other than this "cleaning up after a delete instance" thing?) | 01:28 |
| jroll | jaypipes: by "in use" I mean "resources consumed" | 01:29 |
| jroll | so they aren't scheduled to | 01:29 |
| jroll | and there's other states besides "cleaning" | 01:29 |
| jaypipes | yeah, this all goes back IMHO to needing an Ironic-compute daemon that understands its own definition of a "compute node" | 01:30 |
| jroll | jaypipes: sure, and maybe that's the long-term thing to do. but today I have a real bug that I'd like to get fixed quickly :) | 01:31 |
| jaypipes | jroll: understood. | 01:31 |
| *** nastooh1 has joined #openstack-nova | 01:37 | |
| jroll | jaypipes: looking through this code, it seems get_per_instance_usage should only return instances for that node | 01:37 |
| jroll | everything else done in that context is in the context of a single node | 01:37 |
| jroll | not the entire host | 01:37 |
| *** tobe has joined #openstack-nova | 01:38 | |
| jaypipes | jroll: but there is no "node" parameter to get_per_instance_usage(), so how do you know which node? | 01:38 |
| jroll | jaypipes: right, I think we need to add that to do this right | 01:38 |
| jaypipes | jroll: perhaps the first thing is to add a node param to get_per_instance_usage()? | 01:39 |
| jroll | otherwise everything will look like it's consuming a bazillion RAMs | 01:39 |
| jroll | yeah | 01:39 |
| jroll | jaypipes: I (poorly) assumed folks would be against that while y'all are trying to untangle this mess | 01:39 |
| openstackgerrit | melanie witt proposed openstack/nova: Raise InstanceNotFound when save FK constraint fails https://review.openstack.org/188261 | 01:41 |
| jaypipes | jroll: meh, I wouldn't be opposed to adding a new node=None kwarg to that method of the virt driver. | 01:42 |
| jaypipes | jroll: well, hold up :) | 01:42 |
| jroll | jaypipes: at any rate, thanks for your help. things are much clearer now :) | 01:42 |
| * jroll holds up | 01:42 | |
| jaypipes | jroll: maybe I'm not so keen on that... | 01:43 |
| jroll | for philosophical reasons or technical reasons? :) | 01:43 |
| jaypipes | heh, a bit of both :) | 01:43 |
| jaypipes | jroll: can I think on it more tonight? | 01:44 |
| jroll | jaypipes: sure thing, I'm probably going to hack on adding that argument tonight or first thing tomorrow | 01:44 |
| jroll | but I'm willing to throw it out if we come up with something better | 01:44 |
| jaypipes | k, let's chat about this tomorrow. | 01:45 |
| jroll | cool. thanks again | 01:45 |
| jaypipes | any time | 01:45 |
| *** vilobhmm has joined #openstack-nova | 01:46 | |
| *** baoli has quit IRC | 01:48 | |
| *** patrickeast has quit IRC | 01:50 | |
| *** pmoosh has joined #openstack-nova | 01:51 | |
| *** dboik has joined #openstack-nova | 01:52 | |
| *** dboik_ has joined #openstack-nova | 01:54 | |
| *** dboik has quit IRC | 01:57 | |
| *** Jeffrey4l has quit IRC | 02:00 | |
| *** vilobhmm has quit IRC | 02:01 | |
| *** markvoelker has joined #openstack-nova | 02:02 | |
| openstackgerrit | Jens Rosenboom proposed openstack/nova-specs: Allow ip6 server search for non-admin https://review.openstack.org/187835 | 02:03 |
| *** yamahata has quit IRC | 02:04 | |
| *** radek_ has joined #openstack-nova | 02:05 | |
| *** yingjun has quit IRC | 02:05 | |
| *** markvoelker has quit IRC | 02:07 | |
| *** salv-orlando has joined #openstack-nova | 02:09 | |
| *** oomichi has joined #openstack-nova | 02:11 | |
| *** Jeffrey4l has joined #openstack-nova | 02:12 | |
| oomichi | alex_xu: hi | 02:12 |
| *** pixelbeat_ has quit IRC | 02:12 | |
| alex_xu | oomichi: hi | 02:13 |
| oomichi | alex_xu: can you check Kevin's comment on https://review.openstack.org/#/c/160202/12/nova/tests/unit/api/openstack/compute/contrib/test_quotas.py | 02:13 |
| oomichi | alex_xu: ? | 02:13 |
| *** sks has joined #openstack-nova | 02:13 | |
| oomichi | alex_xu: I also have same quiestion | 02:13 |
| alex_xu | oomichi: let me check | 02:14 |
| oomichi | alex_xu: thanks :) | 02:14 |
| *** unicell has joined #openstack-nova | 02:14 | |
| *** iamjarvo has joined #openstack-nova | 02:15 | |
| *** yingjun has joined #openstack-nova | 02:15 | |
| *** iamjarvo has quit IRC | 02:15 | |
| *** gholler has joined #openstack-nova | 02:15 | |
| *** iamjarvo has joined #openstack-nova | 02:15 | |
| *** otter768 has joined #openstack-nova | 02:16 | |
| *** radek_ has quit IRC | 02:17 | |
| *** salv-orlando has quit IRC | 02:17 | |
| *** angdraug has quit IRC | 02:20 | |
| *** otter768 has quit IRC | 02:21 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: Don't report resources for nodes in cleaning https://review.openstack.org/188652 | 02:24 |
| alex_xu | oomichi: np, I forget why...that patch stay at here for a long time, will try to figure out this afternoon, thanks for the review | 02:24 |
| jroll | jaypipes: I think the fix is much simpler :) https://review.openstack.org/#/c/188652/ | 02:25 |
| jroll | I'll file a bug and such tomorrow | 02:26 |
| *** gholler_ has joined #openstack-nova | 02:26 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests https://review.openstack.org/188243 | 02:26 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] getting vmware api tests to work under python34 https://review.openstack.org/188653 | 02:26 |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: Don't report resources for nodes in cleaning https://review.openstack.org/188652 | 02:28 |
| *** oomichi has quit IRC | 02:28 | |
| *** gholler has quit IRC | 02:29 | |
| *** baoli has joined #openstack-nova | 02:32 | |
| *** achanda has joined #openstack-nova | 02:46 | |
| *** dpaterson has joined #openstack-nova | 02:50 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests https://review.openstack.org/188243 | 02:51 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] getting vmware api tests to work under python34 https://review.openstack.org/188653 | 02:51 |
| *** gholler__ has joined #openstack-nova | 02:53 | |
| dpaterson | Stumped on following error: No valid host was found. Exceeded max scheduling attempts 3 for instance e5003edc-2b51-4b93-8287-b3385ed7963f | 02:53 |
| dpaterson | This has started happening on a cluster that was working fine and all services and resources seem to be fine so I am stumped as to what could be causing this, logs have not been useful thus far. Any assistance much appreciated. | 02:54 |
| *** gholler_ has quit IRC | 02:57 | |
| *** gholler__ has quit IRC | 02:57 | |
| *** boris-42 has quit IRC | 02:58 | |
| *** shaohe_robot1 has joined #openstack-nova | 02:58 | |
| *** achanda has quit IRC | 03:00 | |
| *** ssurana has joined #openstack-nova | 03:00 | |
| *** yingjun has quit IRC | 03:03 | |
| flashgordon | johnthetubaguy: want to resping https://review.openstack.org/#/c/187272/3 ? | 03:04 |
| *** dims_ has quit IRC | 03:07 | |
| *** zul has quit IRC | 03:08 | |
| *** yingjun has joined #openstack-nova | 03:09 | |
| *** yingjun has quit IRC | 03:10 | |
| *** sajeesh has quit IRC | 03:14 | |
| openstackgerrit | Zhenzan Zhou proposed openstack/nova: Bypass ironic server not available issue https://review.openstack.org/173681 | 03:14 |
| *** xyang1 has quit IRC | 03:15 | |
| *** lpetrut has joined #openstack-nova | 03:18 | |
| *** baoli has quit IRC | 03:18 | |
| *** liusheng has joined #openstack-nova | 03:19 | |
| *** zul has joined #openstack-nova | 03:20 | |
| *** amotoki has quit IRC | 03:22 | |
| *** dave-mccowan has joined #openstack-nova | 03:30 | |
| *** mohitsharma has quit IRC | 03:41 | |
| *** mohitsharma has joined #openstack-nova | 03:42 | |
| *** nelsnelson has quit IRC | 03:43 | |
| *** markvoelker has joined #openstack-nova | 03:50 | |
| *** salv-orlando has joined #openstack-nova | 03:51 | |
| *** coolsvap|afk is now known as coolsvap | 03:53 | |
| *** markvoelker has quit IRC | 03:55 | |
| *** exploreshaifali has joined #openstack-nova | 03:56 | |
| *** matrohon has joined #openstack-nova | 03:56 | |
| *** oomichi has joined #openstack-nova | 04:01 | |
| *** salv-orlando has quit IRC | 04:01 | |
| *** nihilifer has quit IRC | 04:04 | |
| *** dims_ has joined #openstack-nova | 04:05 | |
| *** dave-mccowan has quit IRC | 04:06 | |
| *** dims_ has quit IRC | 04:10 | |
| *** dpaterson has quit IRC | 04:15 | |
| *** otter768 has joined #openstack-nova | 04:17 | |
| *** lpetrut has quit IRC | 04:21 | |
| *** oomichi has quit IRC | 04:21 | |
| *** pradipta has joined #openstack-nova | 04:22 | |
| *** otter768 has quit IRC | 04:22 | |
| *** rushiagr_away is now known as rushiagr | 04:22 | |
| *** rushiagr is now known as rushiagr_away | 04:27 | |
| *** Nisha has joined #openstack-nova | 04:28 | |
| *** Longgeek has joined #openstack-nova | 04:30 | |
| *** Longgeek has quit IRC | 04:30 | |
| *** amotoki has joined #openstack-nova | 04:31 | |
| *** iamjarvo has quit IRC | 04:31 | |
| *** jamielennox is now known as jamielennox|away | 04:32 | |
| *** Longgeek has joined #openstack-nova | 04:32 | |
| *** matrohon has quit IRC | 04:36 | |
| *** iamjarvo has joined #openstack-nova | 04:36 | |
| *** prashant-upadhya has joined #openstack-nova | 04:37 | |
| *** destiny has joined #openstack-nova | 04:37 | |
| *** oomichi has joined #openstack-nova | 04:40 | |
| *** rushiagr_away is now known as rushiagr | 04:40 | |
| *** oomichi has quit IRC | 04:41 | |
| *** oomichi has joined #openstack-nova | 04:41 | |
| *** oro_ has joined #openstack-nova | 04:42 | |
| *** kaisers1 has quit IRC | 04:44 | |
| *** kaisers1 has joined #openstack-nova | 04:45 | |
| *** nastooh1 has left #openstack-nova | 04:46 | |
| *** ajayaa has joined #openstack-nova | 04:49 | |
| *** Longgeek_ has joined #openstack-nova | 04:50 | |
| *** deepthi has joined #openstack-nova | 04:51 | |
| *** Longgeek has quit IRC | 04:54 | |
| *** _gryf|afk is now known as _gryf | 04:59 | |
| flashgordon | respin* | 05:00 |
| *** cfriesen has quit IRC | 05:03 | |
| *** unicell has quit IRC | 05:05 | |
| *** unicell has joined #openstack-nova | 05:07 | |
| *** lpetrut has joined #openstack-nova | 05:08 | |
| *** ssurana has quit IRC | 05:14 | |
| *** prashant-upadhya has quit IRC | 05:21 | |
| *** achanda has joined #openstack-nova | 05:25 | |
| *** exploreshaifali_ has joined #openstack-nova | 05:29 | |
| *** exploreshaifali has quit IRC | 05:29 | |
| *** oomichi has quit IRC | 05:31 | |
| *** oomichi has joined #openstack-nova | 05:32 | |
| *** iamjarvo has quit IRC | 05:32 | |
| openstackgerrit | Roman Dobosz proposed openstack/nova-specs: New nova API call to mark nova-compute down https://review.openstack.org/169836 | 05:33 |
| *** prashant-upadhya has joined #openstack-nova | 05:34 | |
| *** oomichi has quit IRC | 05:36 | |
| *** armax has quit IRC | 05:38 | |
| *** markvoelker has joined #openstack-nova | 05:39 | |
| *** mitz has quit IRC | 05:41 | |
| *** tobe has quit IRC | 05:41 | |
| *** tobe has joined #openstack-nova | 05:42 | |
| *** achanda has quit IRC | 05:43 | |
| openstackgerrit | Masaki Matsushita proposed openstack/python-novaclient: Adds support to set admin password from the cli https://review.openstack.org/188670 | 05:43 |
| *** tobe has quit IRC | 05:44 | |
| *** markvoelker has quit IRC | 05:44 | |
| *** sudipto has joined #openstack-nova | 05:47 | |
| *** mdbooth has quit IRC | 05:48 | |
| *** nelsnelson has joined #openstack-nova | 05:49 | |
| *** nihilifer has joined #openstack-nova | 05:49 | |
| *** tojuvone has joined #openstack-nova | 05:50 | |
| *** josecastroleon has joined #openstack-nova | 05:50 | |
| *** nihilifer has quit IRC | 05:51 | |
| *** nihilifer has joined #openstack-nova | 05:51 | |
| *** mdbooth has joined #openstack-nova | 05:53 | |
| *** belmoreira has joined #openstack-nova | 05:55 | |
| *** vishwana_ has quit IRC | 05:56 | |
| *** tojuvone has quit IRC | 06:02 | |
| *** nelsnelson has quit IRC | 06:03 | |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova-specs: Add support for shared volumes between guests https://review.openstack.org/179104 | 06:06 |
| *** achanda has joined #openstack-nova | 06:09 | |
| *** romainh has joined #openstack-nova | 06:11 | |
| *** vilobhmm has joined #openstack-nova | 06:12 | |
| *** amotoki has quit IRC | 06:16 | |
| *** otter768 has joined #openstack-nova | 06:18 | |
| *** lpetrut has quit IRC | 06:19 | |
| *** tobe has joined #openstack-nova | 06:19 | |
| *** romainh has quit IRC | 06:20 | |
| *** otter768 has quit IRC | 06:23 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: Add config drive support for PCS containers https://review.openstack.org/149253 | 06:24 |
| *** sahid has joined #openstack-nova | 06:24 | |
| *** amotoki has joined #openstack-nova | 06:25 | |
| *** ajayaa has quit IRC | 06:28 | |
| *** slong has quit IRC | 06:28 | |
| *** amotoki has quit IRC | 06:36 | |
| jaypipes | melwitt: feel link +2'ing https://review.openstack.org/#/c/187250/? | 06:36 |
| *** amotoki has joined #openstack-nova | 06:37 | |
| *** ajayaa has joined #openstack-nova | 06:40 | |
| *** tojuvone has joined #openstack-nova | 06:40 | |
| *** amotoki has quit IRC | 06:41 | |
| *** abhishekk has joined #openstack-nova | 06:42 | |
| *** exploreshaifali_ has quit IRC | 06:45 | |
| *** vilobhmm has quit IRC | 06:46 | |
| *** Nisha has quit IRC | 06:49 | |
| *** pbandzi has joined #openstack-nova | 06:50 | |
| *** ociuhandu has joined #openstack-nova | 06:50 | |
| *** dims_ has joined #openstack-nova | 06:54 | |
| *** takedakn has joined #openstack-nova | 06:56 | |
| *** oomichi has joined #openstack-nova | 06:56 | |
| *** jistr has joined #openstack-nova | 06:57 | |
| *** abhishekk has quit IRC | 06:58 | |
| *** dims_ has quit IRC | 06:59 | |
| *** takedakn has quit IRC | 07:00 | |
| *** takedakn has joined #openstack-nova | 07:01 | |
| *** ajayaa has quit IRC | 07:03 | |
| *** takedakn has quit IRC | 07:04 | |
| *** bzhao has joined #openstack-nova | 07:05 | |
| Kevin_Zheng | Hi, could anyone kindly review this tiny fix https://review.openstack.org/#/c/184662/ :-) | 07:05 |
| Kevin_Zheng | Already alot +1 | 07:05 |
| *** markus_z has joined #openstack-nova | 07:15 | |
| *** matrohon has joined #openstack-nova | 07:15 | |
| *** mohitsha_ has joined #openstack-nova | 07:16 | |
| *** takedakn has joined #openstack-nova | 07:16 | |
| *** pixelbeat_ has joined #openstack-nova | 07:17 | |
| *** takedakn has quit IRC | 07:17 | |
| *** takedakn has joined #openstack-nova | 07:18 | |
| *** takedakn has quit IRC | 07:18 | |
| *** mohitsharma has quit IRC | 07:18 | |
| *** takedakn has joined #openstack-nova | 07:19 | |
| rgerganov_ | alex_xu: hi | 07:20 |
| alex_xu | rgerganov_: ni | 07:20 |
| rgerganov_ | alex_xu: I don't think your last comment on https://review.openstack.org/#/c/148509/ is correct | 07:20 |
| rgerganov_ | the liberty spec is using "console" not "consoles" | 07:21 |
| rgerganov_ | and this is what I have implemented in my patch | 07:21 |
| alex_xu | rgerganov_: ah...just found that | 07:21 |
| *** takedakn has quit IRC | 07:22 | |
| rgerganov_ | alex_xu: could you please remove the -1? I will update the patch if your proposal is accepted | 07:22 |
| alex_xu | rgerganov_: I can remove my -1, but really think we have two resource one is consoles another is console is confuse | 07:22 |
| *** ajayaa has joined #openstack-nova | 07:23 | |
| alex_xu | rgerganov_: and after you patch merge, that mean this api already available for user, if want to change it again, it mean another microversion bump | 07:23 |
| rgerganov_ | alex_xu: I am not really sure if "consoles" is useful any more | 07:24 |
| rgerganov_ | maybe we should consider to remove that? | 07:24 |
| alex_xu | rgerganov_: but it still existed in our api... | 07:24 |
| rgerganov_ | and have only "console" | 07:24 |
| *** Jeffrey4l has quit IRC | 07:24 | |
| alex_xu | rgerganov_: that sounds like another propose we need get generic discussion | 07:24 |
| alex_xu | rgerganov_: maybe we should push the review on https://review.openstack.org/#/c/185844/3, it is pretty easy to review | 07:25 |
| *** oro_ has quit IRC | 07:26 | |
| rgerganov_ | alex_xu: I am fine for this, I just want to move this effort forward | 07:26 |
| alex_xu | rgerganov_: yea, I understand that. you already work this patch servral month ago. So that why I help you send out a patch correct the spec. Let me try if i can help you to push that review | 07:27 |
| rgerganov_ | alex_xu: thanks | 07:28 |
| *** ociuhandu has quit IRC | 07:28 | |
| alex_xu | rgerganov_: np, thanks for your patient | 07:28 |
| *** markvoelker has joined #openstack-nova | 07:28 | |
| *** achanda has quit IRC | 07:28 | |
| *** lpetrut has joined #openstack-nova | 07:29 | |
| *** rgerganov_ is now known as rgerganov | 07:30 | |
| *** markvoelker has quit IRC | 07:33 | |
| *** exploreshaifali has joined #openstack-nova | 07:36 | |
| *** Jeffrey4l has joined #openstack-nova | 07:37 | |
| *** jlanoux has joined #openstack-nova | 07:40 | |
| gilliard | Morning! | 07:41 |
| alex_xu | gilliard: morning | 07:42 |
| gilliard | Hi There :) | 07:42 |
| alex_xu | gilliard: hi, looks like ironic guys didn't very agree remove headers for min/max version | 07:45 |
| gilliard | oh well :( | 07:45 |
| alex_xu | gilliard: but think about back-compatible, remove something they already have...sound bad | 07:45 |
| alex_xu | gilliard: I think I didn't the strong point we should have body | 07:46 |
| gilliard | Yes that is a difficult point. | 07:46 |
| *** jistr is now known as jistr|biab | 07:47 | |
| *** romainh has joined #openstack-nova | 07:47 | |
| gilliard | We can't expect them to make a change like that to their API if they've released it already wit headers - that kind of change is not possible by increasing the microversion. | 07:47 |
| gilliard | So, honestly I think we should consider adding them to nova, even if we don't really like it. | 07:47 |
| gilliard | Because being compatible with other projects is important. | 07:47 |
| *** rwsu has joined #openstack-nova | 07:48 | |
| alex_xu | gilliard: yes | 07:48 |
| alex_xu | gilliard: actually we already have body to return error message | 07:50 |
| alex_xu | gilliard: {"computeFault": {"message": "Version 2.0 is not supported by the API. Minimum is 2.1 and maximum is 2.3.", "code": 406}} | 07:50 |
| gilliard | That's in already? | 07:51 |
| alex_xu | gilliard: yea | 07:51 |
| * gilliard goes looking | 07:51 | |
| *** coolsvap is now known as coolsvap|afk | 07:52 | |
| gilliard | I'd have thought we'd want the min and max to be individual fields, otherwise a client reading that has to parse an english sentence to know what it can do next. | 07:52 |
| *** achanda has joined #openstack-nova | 07:52 | |
| alex_xu | gilliard: yes, but if we decide to use header, why we still need put that in body | 07:53 |
| *** zxen has joined #openstack-nova | 07:53 | |
| *** rotbart has joined #openstack-nova | 07:53 | |
| gilliard | OK. Well I suppose the thing to do is put a patch up which adds the headers, and another one which updates the spec. | 07:54 |
| *** gszasz has joined #openstack-nova | 07:54 | |
| gilliard | Slightly sad news - but it's Friday so... :D | 07:54 |
| *** gilliard is now known as gillliard | 07:54 | |
| gillliard | alex_xu: does either of those patches exist yet? | 07:55 |
| alex_xu | gillliard: no, didn't have patch yet | 07:55 |
| gillliard | OK I can do those if you like? | 07:56 |
| alex_xu | gillliard: sure | 07:56 |
| gillliard | I expect sdague to be -1 all over them :) | 07:56 |
| alex_xu | gillliard: hah | 07:57 |
| alex_xu | gillliard: or waiting for sdague wakeup before send patch out :) | 07:57 |
| *** lucasagomes has joined #openstack-nova | 07:57 | |
| gillliard | By the way did you see sdague's draft blog post about nova api? | 07:59 |
| alex_xu | gillliard: yea, I read last night, it a great post | 07:59 |
| gillliard | Agree - I think there was going to be an etherpad to talk about any changes before he publishes it. Do you know about that? | 08:00 |
| alex_xu | gillliard: I didn't saw the etherpad, I remember edleafe talk about that etherpad, but I didn't saw a link | 08:01 |
| *** oomichi has quit IRC | 08:01 | |
| *** achanda_ has joined #openstack-nova | 08:01 | |
| *** ajayaa has quit IRC | 08:04 | |
| *** achanda has quit IRC | 08:04 | |
| *** ajo_ has joined #openstack-nova | 08:05 | |
| *** jyuso1 has quit IRC | 08:05 | |
| *** Jeffrey4l has quit IRC | 08:06 | |
| openstackgerrit | ZhaoBo proposed openstack/nova: Fixed raise SecurityGroupCannotBeApplied as network port_security off https://review.openstack.org/188260 | 08:08 |
| *** jlanoux has quit IRC | 08:09 | |
| johnthetubaguy | alex_xu: whats all this about v2.0 not being supported? | 08:10 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to erase instance config https://review.openstack.org/188347 | 08:10 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to get vcpus info https://review.openstack.org/188346 | 08:10 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make _create_domain return a Guest object https://review.openstack.org/182185 | 08:10 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method poweroff to Guest object https://review.openstack.org/182194 | 08:10 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method resume to Guest object https://review.openstack.org/182195 | 08:10 |
| alex_xu | johnthetubaguy: sorry, I didn't quite understand you | 08:11 |
| alex_xu | johnthetubaguy: 'all' means what? | 08:11 |
| johnthetubaguy | alex_xu: its just me looking at the error message in the IRC scrollback | 08:11 |
| johnthetubaguy | alex_xu: I think ironic's headers get around the client having to parse english, its a good idea really, although its extra headers for no reason for most of the time | 08:12 |
| alex_xu | johnthetubaguy: yea | 08:13 |
| gillliard | Do they give min/max headers on _all_ responses? | 08:14 |
| johnthetubaguy | gillliard: apparently | 08:14 |
| johnthetubaguy | I haven't checked | 08:14 |
| johnthetubaguy | feels like overkill | 08:14 |
| johnthetubaguy | but handy | 08:14 |
| alex_xu | gillliard: Devananda said yes in the email | 08:15 |
| gillliard | It makes more sense than only sending them on microversion-mismatch error responses. | 08:15 |
| * gillliard not caught up with ML yet this morning. | 08:15 | |
| gillliard | I'll go read | 08:15 |
| alex_xu | gillliard: agree | 08:15 |
| *** ajayaa has joined #openstack-nova | 08:16 | |
| johnthetubaguy | gillliard: its consistent at least | 08:17 |
| johnthetubaguy | maybe I would me happier if it was a single header, but that seems ridiculous... | 08:18 |
| *** otter768 has joined #openstack-nova | 08:19 | |
| gillliard | so do we still need a /versions endpoint? | 08:20 |
| alex_xu | gillliard: I think so | 08:21 |
| *** danpb has joined #openstack-nova | 08:21 | |
| *** yassine_ has joined #openstack-nova | 08:21 | |
| alex_xu | johnthetubaguy: btw, could you help on this patch https://review.openstack.org/185844 , the original spec use wrong resource name, I submit another propose. The patch auther already work out patch for a long time, hope we can quick merge this. | 08:23 |
| *** ajayaa has quit IRC | 08:23 | |
| gillliard | Yea I feel sorry for that patch | 08:23 |
| *** otter768 has quit IRC | 08:24 | |
| alex_xu | rgerganov: ^ :) | 08:24 |
| johnthetubaguy | alex_xu: why do we have both specs updated? | 08:24 |
| alex_xu | johnthetubaguy: emm...I'm not sure the policy, we should keep the old one as before? | 08:25 |
| johnthetubaguy | alex_xu: well its just a record of what we approved before, so I think so | 08:25 |
| alex_xu | johnthetubaguy: ok, got it, let me update | 08:25 |
| johnthetubaguy | alex_xu: although I guess that could be confusing | 08:25 |
| johnthetubaguy | alex_xu: gillliard: what is the old consoles extension actually doing? | 08:26 |
| gillliard | somethign xen-specific I think? | 08:26 |
| alex_xu | johnthetubaguy: yea I never used it before, it looks like xen-specific | 08:27 |
| johnthetubaguy | alex_xu: so... I think I wanted this to be same extension on purpose, if I remember correctly (which I probably don't) | 08:28 |
| sahid | danpb: hello, we have the gate quiet empty. can i ask you to re+W some patches | 08:28 |
| sahid | https://review.openstack.org/#/c/182185/ | 08:29 |
| johnthetubaguy | alex_xu: it looks like v3 is broken here... possibly: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/contrib/consoles.py#L40 | 08:29 |
| sahid | i had to rebase locally the first one, but no conflict actually | 08:29 |
| *** mohitsha_ has quit IRC | 08:29 | |
| alex_xu | johnthetubaguy: the v2 is this one https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/consoles.py | 08:30 |
| johnthetubaguy | alex_xu: oh.. consoles vs os-consoles? | 08:30 |
| *** ZZelle has quit IRC | 08:31 | |
| alex_xu | johnthetubaguy: when we move contrib/consoles.py into v3, we rename it to remote-consoles | 08:31 |
| johnthetubaguy | alex_xu: which isn't allowed, I presume? | 08:31 |
| johnthetubaguy | well I guess its just the py file name | 08:31 |
| *** ociuhandu has joined #openstack-nova | 08:32 | |
| johnthetubaguy | alex_xu: gillliard: oh, I think I get you now, because we have /console you don't want /servers/<uuid>/console | 08:33 |
| alex_xu | johnthetubaguy: exactly say, we have /consoles now, don't want another one called /console | 08:33 |
| *** sdake has quit IRC | 08:34 | |
| alex_xu | /servers/<uuid>/consoles vs /servers/<uuid>/console | 08:34 |
| * gillliard => Breakfast :D | 08:35 | |
| *** foexle has joined #openstack-nova | 08:35 | |
| *** rwsu has quit IRC | 08:36 | |
| alex_xu | johnthetubaguy: and the orignal spec didn't say we will remove action added by remote-consoles plugin | 08:37 |
| * johnthetubaguy is totally confused about what consoles api is meant to do | 08:37 | |
| *** ndipanov has joined #openstack-nova | 08:37 | |
| *** oro_ has joined #openstack-nova | 08:41 | |
| *** bzhao has quit IRC | 08:41 | |
| bauzas | morning Nova - what a heat here | 08:42 |
| * bauzas complains about not having AC in his home office: ) | 08:42 | |
| *** ankit_ag has joined #openstack-nova | 08:45 | |
| ekarlso | bauzas: where is it you live ? | 08:46 |
| bauzas | ekarlso: Grenoble, FR surroudings | 08:46 |
| *** achanda_ has quit IRC | 08:46 | |
| *** ociuhandu has quit IRC | 08:47 | |
| *** jistr|biab is now known as jistr | 08:50 | |
| *** lennyb has quit IRC | 08:52 | |
| *** jyuso1 has joined #openstack-nova | 08:53 | |
| *** coolsvap|afk is now known as coolsvap | 08:54 | |
| *** gfa has joined #openstack-nova | 08:56 | |
| gfa | i always thought that nova don't honor aggregates/az when doing live migration/resize/migration. but i saw this bug | 08:57 |
| gfa | https://bugs.launchpad.net/nova/+bug/1444841 | 08:57 |
| openstack | Launchpad bug 1444841 in OpenStack Compute (nova) "Resize instance fails after creating host aggregate" [Medium,In progress] - Assigned to Qin Zhao (zhaoqin) | 08:57 |
| gfa | what is the current status? will nova honor aggregates and az when doing live migration/resize? if it does. since when? | 08:58 |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Removed explicit return from __init__ method https://review.openstack.org/188708 | 08:58 |
| gfa | i read release notes but no mention to it | 08:58 |
| *** jistr has quit IRC | 08:59 | |
| openstackgerrit | Marian Horban proposed openstack/nova-specs: Add ability to run OpenStack nova APIs by Apache/Nginx https://review.openstack.org/186503 | 09:00 |
| bauzas | gfa: that's a little bit more complicated | 09:01 |
| bauzas | gfa: when doing migrations, you have 2 options | 09:01 |
| bauzas | gfa: 1/ pass a destination host | 09:01 |
| bauzas | gfa: 2/ only provide the instance to migrate, not the dest host | 09:02 |
| bauzas | gfa: if 1/, the conductor will directly call the destination compute node to ask for the migration, without checking if the scheduler is fine | 09:02 |
| bauzas | gfa: if 2/, the scheduler is called to find a destination which matches the user request, but since the user request is not persisted, it sometimes violates the previous request | 09:03 |
| bauzas | gfa: so, back to your problem | 09:03 |
| bauzas | gfa: if you pass a destination host to the migration, Nova will put your instance there (provided some checks are OK), even if your destination host is not in the same aggregate, or in another aggregate having a different metadata | 09:04 |
| bauzas | gfa: in addition to that, resize is calling the scheduler, but it checks the instance AZ which is buggy | 09:05 |
| *** alex_klimov has joined #openstack-nova | 09:06 | |
| *** ajayaa has joined #openstack-nova | 09:07 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: Add config drive support for PCS containers https://review.openstack.org/149253 | 09:08 |
| openstackgerrit | heha proposed openstack/nova: fff https://review.openstack.org/188709 | 09:11 |
| openstackgerrit | heha proposed openstack/nova: Fix the little spell mistake of the annotation. In the annotation of class Service,the last "it" should be "its". Fixes Bug1462266 https://review.openstack.org/188710 | 09:11 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: Hyper-V: restart serial console workers after instance power change https://review.openstack.org/163080 | 09:13 |
| *** prashant-upadhya has quit IRC | 09:13 | |
| gfa | bauzas: i'm referring when doing 2. let nova select the destination | 09:13 |
| bauzas | gfa: yeah since a resize needs a schedule | 09:14 |
| *** ildikov_ has quit IRC | 09:14 | |
| bauzas | gfa: was mentioning the live migration as it was also part of the confusion | 09:14 |
| gfa | so resize will honor aggregates but live migration no? | 09:14 |
| gfa | i understand that live migration, migration and resize has more or less the same codepath | 09:15 |
| bauzas | gfa: "honoring aggregates" doesn't mean nothing unless you have filters in place :) | 09:15 |
| bauzas | doesn't mean anything | 09:15 |
| bauzas | (sorry) | 09:15 |
| gfa | of course i have them, they work when i boot the vm but no when i live migrate some time latter | 09:15 |
| gfa | that's icehouse. i'm searching if that behavior was improved latter. i think not | 09:16 |
| gfa | but is not clear to me | 09:16 |
| bauzas | hem | 09:16 |
| bauzas | we had quite some bugfixes since icehouse, I would suggest you to try to reproduce your problems with devstack | 09:17 |
| *** markvoelker has joined #openstack-nova | 09:17 | |
| bauzas | gfa: but there is still a problem with the instance AZ and I would like to kick it off | 09:17 |
| bauzas | sorry, to drop it I mean | 09:17 |
| gfa | i know new releases are much better :) | 09:18 |
| gfa | but i didn't found anything specific to that problem | 09:18 |
| gfa | and some blueprint half merged so i wasn't really sure | 09:19 |
| *** jistr has joined #openstack-nova | 09:19 | |
| gfa | bauzas: please tell me about that problem. if i go up to kilo i will be able to test the fix for you | 09:19 |
| *** marzif_ has joined #openstack-nova | 09:21 | |
| *** markvoelker has quit IRC | 09:22 | |
| *** exploreshaifali has quit IRC | 09:23 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to delete domain config https://review.openstack.org/188347 | 09:32 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to get vcpus info https://review.openstack.org/188346 | 09:32 |
| *** ajo_ has quit IRC | 09:33 | |
| *** afazekas_mtg has joined #openstack-nova | 09:33 | |
| *** prashant-upadhya has joined #openstack-nova | 09:34 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to delete domain config https://review.openstack.org/188347 | 09:34 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to get vcpus info https://review.openstack.org/188346 | 09:34 |
| *** aix has joined #openstack-nova | 09:34 | |
| *** dims_ has joined #openstack-nova | 09:43 | |
| *** alexpilotti has joined #openstack-nova | 09:45 | |
| *** dims__ has joined #openstack-nova | 09:47 | |
| *** jlanoux has joined #openstack-nova | 09:47 | |
| *** dims_ has quit IRC | 09:48 | |
| mnestratov|3 | Hi there | 09:51 |
| *** romainh has quit IRC | 09:52 | |
| mnestratov|3 | I discovered that yesterday on nova meeting was a discussion about PCS config drive support | 09:52 |
| mnestratov|3 | unfortunately i could't participate | 09:53 |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova: Consolidate the APIs for getting consoles https://review.openstack.org/148509 | 09:54 |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: Fix rebuild of an instance with a volume attached https://review.openstack.org/176891 | 09:54 |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: rebuild: make sure server is shut down before volumes are detached https://review.openstack.org/176892 | 09:54 |
| rgerganov | johnthetubaguy: alex_xu gillliard guys, I have updated https://review.openstack.org/#/c/148509 to use 'remote-consoles' instead of 'console' | 09:55 |
| rgerganov | any feedback is appreciated | 09:55 |
| gillliard | thanks rgerganov | 09:55 |
| johnthetubaguy | alex_xu: gillliard: do you remember who was going to do the big v3->v2.1 code rename? | 09:56 |
| gillliard | no, sorry. | 09:56 |
| mnestratov|3 | I wanted to mention it here that I updated https://review.openstack.org/#/c/149253/ where I got gid of explicit filesystem defining | 09:56 |
| bauzas | johnthetubaguy: re: https://etherpad.openstack.org/p/YVR-nova-liberty-summit-action-items and your email, I was already putting my name somewhere :) | 09:57 |
| johnthetubaguy | bauzas: line 11? | 09:57 |
| bauzas | johnthetubaguy: I also think we should identify some quickwins that we could possibly give to new contributors | 09:57 |
| bauzas | johnthetubaguy: and L91 | 09:57 |
| johnthetubaguy | bauzas: yeah, I was thinking I should write down my crazy plan as a backlog spec, for line 11 | 09:58 |
| bauzas | johnthetubaguy: L11 is a bit tricky, since you have ideas that I don't | 09:58 |
| bauzas | johnthetubaguy: I was more seeing L11 as a joint effort | 09:58 |
| johnthetubaguy | bauzas: me too, thats cool | 09:58 |
| johnthetubaguy | L91 could work, thanks | 09:58 |
| bauzas | np | 09:59 |
| johnthetubaguy | bauzas: although, you have loads to do already, it might be worth waiting to see if someone else steps up for that while you get the ones you are already doing, but happy to leave you name there, and race you to see who gets their first! | 10:00 |
| bauzas | johnthetubaguy: I'm primarly involved with ReqSpec and cells | 10:00 |
| bauzas | johnthetubaguy: since ReqSpec is on its way, that's why I'm caring about helping | 10:01 |
| bauzas | johnthetubaguy: that's a natural behaviour I try to put | 10:01 |
| johnthetubaguy | yup, there are both super high priorities, so I don't want to derail those :) | 10:01 |
| bauzas | johnthetubaguy: ie. the tasks I'm committed to are mostly regarding how to help new contributors to join | 10:01 |
| johnthetubaguy | cools | 10:02 |
| bauzas | johnthetubaguy: so it's about offloading my tasks | 10:02 |
| bauzas | johnthetubaguy: I mean scaling out | 10:02 |
| johnthetubaguy | yeah, although scaling out will mean it takes longer as folks ramp up, so its not "free", but you know all that, just saying | 10:02 |
| bauzas | johnthetubaguy: if you're seeing my duties, I'm just extra-committed to design discussions and drafts | 10:03 |
| johnthetubaguy | cools | 10:03 |
| johnthetubaguy | that will be a big help | 10:03 |
| flashgordon | johnthetubaguy: random question, have you seen nova-compute hang | 10:03 |
| johnthetubaguy | flashgordon: in the past yes, but it got fixed | 10:03 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: libvirt: Don't try to confine a non-NUMA instance https://review.openstack.org/188468 | 10:03 |
| bauzas | johnthetubaguy: agreed, I'm just investing some time for helping people | 10:03 |
| flashgordon | johnthetubaguy: because http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/screen-n-sch.txt.gz#_2015-05-29_23_46_46_736 | 10:04 |
| flashgordon | johnthetubaguy: only happening in multinode | 10:04 |
| johnthetubaguy | flashgordon: the hang was mostly a pretend one though, the greenlet doing periodic stuff got stuck waiting on a lock that was held by a create instance task, and the create instance was downloading a big windows image, so was waiting ages | 10:04 |
| bauzas | flashgordon: mmm, that's something I was discussing downstream | 10:05 |
| bauzas | flashgordon: I still need to figure out the problem | 10:05 |
| johnthetubaguy | oh, I wonder about a deadlock inside conductor, maybe? | 10:05 |
| johnthetubaguy | is conductor only singular in the setup, or something? | 10:06 |
| flashgordon | I am thinking sinding a SIGUSR1 to nova-compute before we stop it in devstack | 10:06 |
| johnthetubaguy | yeah | 10:06 |
| bauzas | dumb question, how can we check 2 n-cpu log in a multinode dsvm ? :) | 10:06 |
| * johnthetubaguy just heard thunder, quickly presses save on his work... oh wait its a laptop, phew | 10:06 | |
| flashgordon | bauzas: http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/subnode-2/ | 10:07 |
| johnthetubaguy | flashgordon: are you sure its not clock skew / drift? | 10:07 |
| bauzas | flashgordon: thanks, missed it | 10:07 |
| flashgordon | johnthetubaguy: positive, look at the size of the log for n-cpu http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/ | 10:08 |
| flashgordon | plus scheduler and the bad n-cpu are on the same node | 10:08 |
| johnthetubaguy | flashgordon: when I do two node devstack in VMs, its almost always clock skew that kills me | 10:08 |
| johnthetubaguy | flashgordon: what about conductor, it might be conductor time thats used for the DB write | 10:09 |
| johnthetubaguy | flashgordon: oh, I see, the logs just stop from nova-compute | 10:10 |
| bauzas | johnthetubaguy: yup | 10:11 |
| flashgordon | johnthetubaguy: yeah its very odd | 10:11 |
| sdague | morning folks | 10:12 |
| gillliard | Hi sdague! up early? | 10:13 |
| sdague | gillliard / alex_xu - https://etherpad.openstack.org/p/sdague_api_post is the etherpad that edleafe made to copy edit things, I'm starting to apply those now | 10:13 |
| gillliard | thanks | 10:13 |
| sdague | gillliard: this is normal time for me | 10:13 |
| flashgordon | sdague: just the person I was looking for | 10:13 |
| sdague | flashgordon: oh dear | 10:13 |
| johnthetubaguy | flashgordon: do you have the log from the working compute node? | 10:13 |
| johnthetubaguy | whats it called? | 10:13 |
| flashgordon | johnthetubaguy: http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/subnode-2/ | 10:14 |
| flashgordon | sdague: we are seeing nova-compute just hang in the multinode test | 10:14 |
| flashgordon | so I was thinking in devstack-gate it would be nice to have a step where we try to cleanly shutdown all the services (and we can send a SIGUSR1 to n-cpu for guru meditation) | 10:14 |
| flashgordon | so make devstack-gate do clean shutdowns | 10:15 |
| johnthetubaguy | flashgordon: so I will put money on an lock decorator deadlock, just wondering where | 10:15 |
| sdague | well, there is an updown devstack job that does such a thing, you could do it from there. That being said, the services running long have been useful to debug | 10:15 |
| sdague | because neutron reporting sending an event after the tests were over with let us know it was just timing issues | 10:16 |
| sdague | did it trigger a fail? | 10:16 |
| flashgordon | sdague: yes, 1 compute goes down so the multi node specific tests start failing | 10:17 |
| flashgordon | strange thing is this only happens in multinode | 10:17 |
| flashgordon | message:"has not been heard from in a while" AND tags:"screen-n-sch.txt" | 10:17 |
| sdague | so, are you sure the ncpu didn't crash? | 10:17 |
| *** markvoelker has joined #openstack-nova | 10:17 | |
| flashgordon | sdague: good question, I can actually answer that | 10:18 |
| sdague | honestly, I wouldn't focus on clean shutdown, I'd focus on doing a dump at the end of a failed run | 10:18 |
| flashgordon | sdague: that works for me | 10:19 |
| sdague | I wrote the worlddump script in devstack for such purposes, putting guru meditation into that as well seems fine | 10:19 |
| sdague | and call that on nodes if it fails | 10:19 |
| flashgordon | nova-compute is still running is still running at the end | 10:19 |
| *** otter768 has joined #openstack-nova | 10:20 | |
| *** achanda has joined #openstack-nova | 10:20 | |
| flashgordon | sdague: so it fails after tempest, so we would need to run worlddump if tempest fails? | 10:20 |
| *** boris-42 has joined #openstack-nova | 10:20 | |
| flashgordon | err causes tempest to fail* | 10:20 |
| sdague | yeh, https://github.com/openstack-infra/devstack-gate/blob/master/devstack-vm-gate-wrap.sh#L510 here ish | 10:22 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py https://review.openstack.org/187785 | 10:22 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests https://review.openstack.org/188243 | 10:22 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/scheduler/test*.py https://review.openstack.org/187968 | 10:22 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] getting vmware api tests to work under python34 https://review.openstack.org/188653 | 10:22 |
| *** markvoelker has quit IRC | 10:23 | |
| johnthetubaguy | flashgordon: it seems like the compute node locks up just after the migration starts | 10:24 |
| johnthetubaguy | flashgordon: http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/subnode-2/screen-n-cpu.txt.gz#_2015-05-29_23_27_48_079 | 10:24 |
| *** otter768 has quit IRC | 10:24 | |
| johnthetubaguy | http://logs.openstack.org/67/175067/2/check/check-tempest-dsvm-multinode-full/7a95fb0/logs/subnode-2/screen-n-cpu.txt.gz#_2015-05-29_23_27_48_079 | 10:24 |
| *** ildikov_ has joined #openstack-nova | 10:26 | |
| flashgordon | johnthetubaguy: oh good catch | 10:30 |
| flashgordon | johnthetubaguy: and I have to head out, dinner time here in Tokyo | 10:30 |
| johnthetubaguy | flashgordon: I would say lock ordering in the iptables updates when you deletes and migrates at the same time, are worth a look | 10:30 |
| johnthetubaguy | flashgordon: cool, I should do lunch soon, and start on some spec reviews! | 10:30 |
| flashgordon | johnthetubaguy: can you file a bug for that so I can track this later | 10:31 |
| flashgordon | johnthetubaguy: looks like multinode testing has caught yet another multinode nova bug :) | 10:31 |
| johnthetubaguy | flashgordon: yeah, I will try write something down | 10:31 |
| *** ajayaa has quit IRC | 10:31 | |
| dims__ | garyk: quick one - https://review.openstack.org/#/c/181987/ (oslo.vmware) | 10:32 |
| flashgordon | dims__: looks like we found a fun nova migration bug if you wan to take a look at it | 10:33 |
| flashgordon | o/ | 10:33 |
| dims__ | flashgordon: sure, got a bug? | 10:34 |
| *** ajayaa has joined #openstack-nova | 10:34 | |
| mnestratov|3 | Guys, regarding yesterday PCS config drive disscussion, i've updated https://review.openstack.org/#/c/149253/ where I hid fs choosing for PCS exe instances | 10:36 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds RemoteFX support to the Hyper-V driver https://review.openstack.org/42529 | 10:36 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds hostutilsv2 to HyperV https://review.openstack.org/188734 | 10:36 |
| gillliard | sdague: I'm done scribbling on the etherpad. Thanks for writing all that up. | 10:39 |
| *** aignatov has quit IRC | 10:39 | |
| *** DinaBelova has quit IRC | 10:39 | |
| johnthetubaguy | flashgordon: https://bugs.launchpad.net/nova/+bug/1462305 | 10:41 |
| openstack | Launchpad bug 1462305 in OpenStack Compute (nova) "multi-node test causes nova-compute to lockup" [Undecided,Incomplete] - Assigned to Joe Gordon (jogo) | 10:41 |
| *** pradipta has quit IRC | 10:43 | |
| *** ajayaa has quit IRC | 10:47 | |
| *** DinaBelova has joined #openstack-nova | 10:49 | |
| *** aignatov has joined #openstack-nova | 10:49 | |
| *** wasmum has quit IRC | 10:55 | |
| sdague | gillliard: ok, I think I just got everything applied - https://dague.net/?p=4461&shareadraft=baba4461_55718053e3086 | 10:56 |
| sdague | if anyone still has brain power to read words in english for a final look, let me know | 10:57 |
| openstackgerrit | garyk proposed openstack/nova: Neutron: add in API method for updating VNIC index https://review.openstack.org/147126 | 10:59 |
| *** coolsvap is now known as coolsvap|afk | 10:59 | |
| *** prashant-upadhya has quit IRC | 11:03 | |
| *** ajayaa has joined #openstack-nova | 11:04 | |
| sdague | ok, I just added a Want to learn more? bit at the end. | 11:08 |
| sdague | johnthetubaguy: anything else you want in there before I let this 2700 word essay loose on the world? | 11:09 |
| *** prashant-upadhya has joined #openstack-nova | 11:09 | |
| sdague | also, is the todo to make the extensions deprecated formal policy in your hands somewhere? So we can be more explicit about that than just my blog post :) | 11:11 |
| *** marzif_ has quit IRC | 11:11 | |
| *** marzif_ has joined #openstack-nova | 11:12 | |
| *** romainh has joined #openstack-nova | 11:12 | |
| *** marzif_ has quit IRC | 11:13 | |
| *** marzif_ has joined #openstack-nova | 11:13 | |
| johnthetubaguy | sdague: I think that one is in the etherpad of doom | 11:15 |
| johnthetubaguy | sdague: curd its not listed | 11:16 |
| johnthetubaguy | sdague: once the config goes away, I think we can rename the files appropriatly | 11:17 |
| johnthetubaguy | sdague: that blog looks good to me, its a massive step forward in our communication about v2.1 | 11:18 |
| *** markvoelker has joined #openstack-nova | 11:18 | |
| flip214 | johnthetubaguy: Hi. any news regarding the DRBD driver for Nova? | 11:19 |
| *** wasmum has joined #openstack-nova | 11:19 | |
| johnthetubaguy | flip214: none that I can remember, what news are you wanting? | 11:20 |
| flip214 | Is it decided whether the (existing) spec is really needed, and whether/when/how it gets accepted? | 11:20 |
| sdague | johnthetubaguy: ok, cool, well it would be nice to land a policy spec or dev ref or something soon so it's on record with a pointer when people come with pitchforks | 11:20 |
| flip214 | https://review.openstack.org/#/c/134153/ is the spec | 11:20 |
| sdague | and we can say, nope, nova team agreed, this is where we are headed | 11:20 |
| johnthetubaguy | sdague: +1 I added it as a todo | 11:20 |
| sdague | I expect there will be a few pitchforks | 11:21 |
| flip214 | https://review.openstack.org/#/c/149244/ the code (that needs to get rebased) | 11:21 |
| johnthetubaguy | sdague: yeah, I hope that will engage them in helping make policy easier to configure | 11:21 |
| johnthetubaguy | flip214: so I hope to get to your spec today, turns out I have got 28 days behind in where I want to be with spec reviews, oops | 11:22 |
| flip214 | okay, thanks. | 11:22 |
| *** markvoelker has quit IRC | 11:22 | |
| flip214 | looking forward to your +2 ;] | 11:23 |
| johnthetubaguy | flip214: sorry this is so long and drawn out, not getting the review velocity we need there right now | 11:23 |
| * johnthetubaguy runs away to get some food | 11:23 | |
| sdague | ok, the button is pushed | 11:24 |
| *** mnestratov|3 is now known as mnestratov | 11:25 | |
| *** claudiub has joined #openstack-nova | 11:29 | |
| *** aysyanne has joined #openstack-nova | 11:33 | |
| gillliard | Nova API meeting in 25 minutes | 11:34 |
| *** achanda has quit IRC | 11:35 | |
| aloga | bauzas: ping | 11:36 |
| alex_xu | sdague: johnthetubaguy gillliard, add something if you want to talk about at meeting https://wiki.openstack.org/wiki/Meetings/NovaAPI#Agenda | 11:36 |
| sdague | alex_xu: ok, cool. It would be good to talk about - https://review.openstack.org/#/c/173243/12/specs/liberty/approved/api-relax-validation.rst,cm in details, because I think the direction for that changed a little bit. | 11:37 |
| sdague | during summit discussions. | 11:38 |
| kashyap | johnthetubaguy: "when I do two node devstack in VMs, its almost always clock skew that kills me" -- NTP to the rescue? | 11:38 |
| sdague | kashyap: the gate forces ntp update at node boot, so we should be pretty good | 11:38 |
| *** mwagner_afk has joined #openstack-nova | 11:38 | |
| alex_xu | sdague: ok, no problem | 11:38 |
| kashyap | sdague: Aha, thanks! | 11:38 |
| sdague | we did that because one rax region was once misconfigured so that all the clocks were 4 hours back | 11:39 |
| sdague | and our timeout code went wonky | 11:39 |
| kashyap | sdague: Can imagine it causing needless debugging exercise | 11:39 |
| kashyap | sdague: Do you also run it locally when you do multi-node DevStack testing? | 11:39 |
| sdague | I don't, but my multinode local testing has been in vms on the same box, so I'm relying on the clock sync from the host | 11:40 |
| kashyap | Yeah, that's as far as multi-node testing goes too, for migration testing. | 11:40 |
| kashyap | sdague: On a different topic, any reason why not let your nice writeup on API be in devref? | 11:41 |
| *** prashant-upadhya has quit IRC | 11:42 | |
| *** lucasagomes is now known as lucas-hungry | 11:42 | |
| *** tobe has quit IRC | 11:42 | |
| *** rwsu has joined #openstack-nova | 11:43 | |
| *** nelsnelson has joined #openstack-nova | 11:45 | |
| sdague | kashyap: honestly, for right now, exposure. It's written for a wider audience. The goal is to merge it back into devref (with some extended history bits) later in the cycle. Especially after we field any questions from people about stuff that wasn't clear. | 11:47 |
| sdague | Consider this the first draft of eventual devref | 11:47 |
| kashyap | sdague: Yep, noted. Thanks. | 11:48 |
| sdague | alex_xu: so... I actually am not clear I'll be able to make the API meeting today, because I need to get some stuff done around the house before we drive to Mass this afternoon for a family thing. However, I left some notes in https://etherpad.openstack.org/p/liberty-api-working-list | 11:48 |
| sdague | also, I think we should do an exercise about trying to chunk up the API work into milestones | 11:49 |
| sdague | I moved a couple of really near term things that are needed up into Liberty 1 | 11:49 |
| *** tojuvone has quit IRC | 11:49 | |
| alex_xu | sdague: ok, no problem | 11:50 |
| sdague | alex_xu: early next week, lets get some folks together and start trying to sort priorities and be able to divide and conquer on things. I think finding a volunteer for that middleware to relax the validation is the next hot item. | 11:50 |
| sdague | and update the spec that goes with it | 11:51 |
| alex_xu | sdague: ok | 11:52 |
| *** jecarey has joined #openstack-nova | 11:52 | |
| alex_xu | sdague: what plan for api doc, as this is priority item. do you know the plan, or we should talk with Anne directly. | 11:54 |
| *** iamjarvo has joined #openstack-nova | 11:54 | |
| *** nelsnelson has quit IRC | 11:55 | |
| *** sks has quit IRC | 11:56 | |
| openstackgerrit | p-draigbrady proposed openstack/nova: rbd: use rbd snapshots and cloning for ephemeral disk snapshots https://review.openstack.org/125963 | 11:58 |
| bauzas | aloga: pong | 12:01 |
| *** eliqiao has joined #openstack-nova | 12:01 | |
| bauzas | aloga: could you try to avoid naked pings ? | 12:01 |
| *** rotbart has quit IRC | 12:01 | |
| bauzas | aloga: FYI https://blogs.gnome.org/markmc/2014/02/20/naked-pings/ | 12:01 |
| *** dave-mccowan has joined #openstack-nova | 12:03 | |
| aloga | bauzas: sorry | 12:03 |
| bauzas | aloga: np, just for helping you with IRC | 12:04 |
| *** ildikov_ has quit IRC | 12:04 | |
| aloga | bauzas: thanks :) | 12:04 |
| aloga | bauzas: it is regarding https://bugs.launchpad.net/nova/+bug/1227925 | 12:04 |
| openstack | Launchpad bug 1227925 in OpenStack Compute (nova) "Resources tracker doesn't free data on instance terminate" [Medium,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 12:04 |
| aloga | bauzas: are you still looking at it? | 12:05 |
| bauzas | aloga: checking | 12:05 |
| bauzas | aaaaah this one | 12:05 |
| bauzas | aloga: well it's not a trivial ojne | 12:05 |
| aloga | indeed | 12:05 |
| bauzas | aloga: because the RT is made like this | 12:06 |
| *** dprince has joined #openstack-nova | 12:06 | |
| bauzas | aloga: so I owned it because I would like to discuss on how to provide resources to the Scheduèler | 12:07 |
| aloga | bauzas: apart from that, I've a doubt | 12:07 |
| aloga | from an operator point of view | 12:07 |
| bauzas | yup ? | 12:07 |
| aloga | shouldn't deleted but not claimed instances free their resources? | 12:08 |
| bauzas | aloga: you mean deleted instances ? yes they do | 12:08 |
| aloga | deleted but not reclaimed | 12:09 |
| bauzas | aloga: technically, that's soft-deleted even | 12:09 |
| *** sks has joined #openstack-nova | 12:09 | |
| *** dave-mccowan has quit IRC | 12:09 | |
| openstackgerrit | p-draigbrady proposed openstack/nova: rbd: use rbd snapshots and cloning for ephemeral disk snapshots https://review.openstack.org/125963 | 12:09 |
| aloga | bauzas: the point is that we found that intances do not free their resources until they're reclaimed | 12:09 |
| aloga | so, I assumed that a soft-deleted instance could free them so that they could be used by another instance | 12:10 |
| bauzas | aloga: mmm | 12:10 |
| aloga | if then the restore fails because of a lack of resources... that's another story | 12:10 |
| bauzas | aloga: you mean that if you terminate an instance, the resources are not freed ? | 12:10 |
| aloga | bauzas yes | 12:11 |
| bauzas | aloga: weirod | 12:11 |
| bauzas | weirdo even | 12:11 |
| *** ferest has joined #openstack-nova | 12:11 | |
| bauzas | aloga: because I'm pretty sure that the RT is making sure that the resources are freed, at least when update_resource_stats() is called every 60 secs | 12:12 |
| *** sudipto has quit IRC | 12:13 | |
| aloga | bauzas: sorry, didn't meant that! | 12:13 |
| aloga | bauzas: I mean they are not freed when the machine is deleted | 12:13 |
| aloga | only when the update_resource_stats() is called | 12:13 |
| aloga | i.e. the periodic task | 12:14 |
| *** dave-mccowan has joined #openstack-nova | 12:15 | |
| *** zul has quit IRC | 12:16 | |
| aloga | let me check again... | 12:16 |
| *** zul has joined #openstack-nova | 12:16 | |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds hostutilsv2 to HyperV https://review.openstack.org/188734 | 12:16 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Adds RemoteFX support to the Hyper-V driver https://review.openstack.org/42529 | 12:16 |
| *** ferest has quit IRC | 12:17 | |
| bauzas | aloga: oh that, yes that's how it's done unfortunatelty | 12:17 |
| aloga | ah, ok, ok | 12:18 |
| *** ildikov_ has joined #openstack-nova | 12:19 | |
| aloga | bauzas: so, in summary: a request to soft-delete an instance will not free its resources, but they'll get freed when the resources are updated, right? | 12:19 |
| bauzas | aloga: correct | 12:20 |
| aloga | bauzas: great | 12:20 |
| aloga | finally, I got it :) | 12:20 |
| *** otter768 has joined #openstack-nova | 12:21 | |
| aloga | due to the usage of our cloud for us this is a problem sometimes :( | 12:24 |
| bauzas | aloga: ;) | 12:24 |
| bauzas | aloga: yeah, the whole design is subject to debate I truly admin | 12:24 |
| bauzas | admit | 12:24 |
| bauzas | oh TGIF ! | 12:25 |
| *** bauzas is now known as bauwser | 12:25 | |
| *** otter768 has quit IRC | 12:25 | |
| *** MVenesio has joined #openstack-nova | 12:25 | |
| *** MVenesio has quit IRC | 12:25 | |
| aloga | bauwser: thanks for your time | 12:26 |
| aloga | bauwser: enjoy the weekend ;-) | 12:26 |
| bauwser | aloga: np, here to help :) | 12:26 |
| *** edleafe is now known as figleaf | 12:30 | |
| *** sks has quit IRC | 12:31 | |
| openstackgerrit | Alexis Lee proposed openstack/nova-specs: Amend resource objects proposal https://review.openstack.org/188759 | 12:32 |
| *** rfolco has joined #openstack-nova | 12:32 | |
| *** annegentle has joined #openstack-nova | 12:32 | |
| *** nihilifer has quit IRC | 12:33 | |
| *** markvoelker has joined #openstack-nova | 12:34 | |
| *** deepthi has quit IRC | 12:35 | |
| *** ndipanov is now known as ndipanoff | 12:36 | |
| *** VW has joined #openstack-nova | 12:36 | |
| *** zhenguo has quit IRC | 12:37 | |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: Allow adjust ephemeral disks during resize https://review.openstack.org/145736 | 12:38 |
| *** markvoelker has quit IRC | 12:38 | |
| *** ociuhandu has joined #openstack-nova | 12:39 | |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Remove unit tests for _hard_reboot https://review.openstack.org/177398 | 12:39 |
| *** ivasev has joined #openstack-nova | 12:42 | |
| *** sks has joined #openstack-nova | 12:44 | |
| *** lucas-hungry is now known as lucasagomes | 12:52 | |
| *** ajayaa has quit IRC | 12:54 | |
| *** gszasz has quit IRC | 12:57 | |
| *** exploreshaifali has joined #openstack-nova | 12:57 | |
| *** _gryf has quit IRC | 12:58 | |
| *** romainh has left #openstack-nova | 12:58 | |
| *** dave-mccowan has quit IRC | 12:59 | |
| *** dpaterson has joined #openstack-nova | 13:02 | |
| *** alaski is now known as lascii | 13:02 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to attach device https://review.openstack.org/188763 | 13:02 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: remove _get_disk_xml to use get_disk from Guest https://review.openstack.org/188764 | 13:02 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method detach_device to Guest object https://review.openstack.org/188765 | 13:02 |
| gillliard | So johnthetubaguy, we need to make a decision about the schema-relaxation patch, right? | 13:04 |
| johnthetubaguy | gillliard: yeah | 13:04 |
| *** baoli has joined #openstack-nova | 13:04 | |
| *** gholler has joined #openstack-nova | 13:04 | |
| johnthetubaguy | so the plan was, leave /v2.1 alone | 13:04 |
| sdague | alex_xu: I don't know the API doc plan, I think that needs to be figured out still | 13:04 |
| alex_xu | sdague: ok | 13:05 |
| sdague | gillliard: I tried to write the thing we agreed in the etherpad | 13:05 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to attach device https://review.openstack.org/188763 | 13:05 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method detach_device to Guest object https://review.openstack.org/188765 | 13:05 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: remove _get_disk_xml to use get_disk from Guest https://review.openstack.org/188764 | 13:05 |
| sdague | build it as middleware, if enabled the schema is relaxed, and we block microversions | 13:05 |
| johnthetubaguy | for /v2/ we said fails when it gets given version headers for v2.1 (whether its using v2 or v2.1 code) | 13:05 |
| gillliard | do we just need a POC in code to chew over? | 13:06 |
| johnthetubaguy | sdague: there was questions about it being middleware, but wanted to get the requirements straight | 13:06 |
| *** ajayaa has joined #openstack-nova | 13:06 | |
| sdague | so... I think the reason we thought middleware would be good is that then the /v2.0 paste could just be /v2.1 + that middleware | 13:07 |
| johnthetubaguy | gillliard: we could, just seems like we don't agree on what it should do yet | 13:07 |
| johnthetubaguy | sdague: +1 | 13:07 |
| johnthetubaguy | and that /v2.1 is left with full validtion | 13:07 |
| sdague | right, exactly | 13:07 |
| *** iamjarvo has quit IRC | 13:08 | |
| johnthetubaguy | sdague: attempted a draft of all that here: https://review.openstack.org/#/c/173243/ | 13:08 |
| openstackgerrit | Paul Murray proposed openstack/nova: Convert RT compute_node to be a ComputeNode object https://review.openstack.org/160942 | 13:09 |
| *** gholler has quit IRC | 13:09 | |
| *** dims__ has quit IRC | 13:09 | |
| *** dims_ has joined #openstack-nova | 13:10 | |
| sdague | johnthetubaguy: ok, cool, I won't be able to really look today, got to get back to house choirs so we can get back there. But will look first thing on monday | 13:10 |
| johnthetubaguy | sdague: thats cool | 13:10 |
| alex_xu | sdague: so that means people can remove that middleware. | 13:11 |
| lpetrut | Hi guys. We're having issues w/ the Hyper-V driver caused by the fact that the snapshot operation is not synchronized in the manager. Would it make sense to use a lock in the manager, or should we do this at the Hyper-V driver level? | 13:12 |
| *** ildikov_ has quit IRC | 13:13 | |
| gillliard | alex_xu: what do you mean "remove that middleware"? | 13:13 |
| *** raph_ has left #openstack-nova | 13:13 | |
| *** baoli has quit IRC | 13:14 | |
| alex_xu | gillliard: api-paste.ini is configuable | 13:14 |
| *** baoli has joined #openstack-nova | 13:14 | |
| sdague | alex_xu: yes, they could | 13:15 |
| sdague | but they could also set up the v20 controller on some other endpoint as well | 13:15 |
| sdague | honestly, the api-paste bits are another place where we allow too many things to be crazy configurable in a way that will probably just make things nutty at the end of the day | 13:16 |
| alex_xu | sdague: we support that? if yes, version api looks bug, it always return '/v2' and '/v2.1' | 13:16 |
| sdague | things would break in unexpected ways I think | 13:16 |
| *** mestery is now known as mestery_afk | 13:16 | |
| sdague | anyway... the point is, if we want to use the same code on v2 and v2.1, and have v2 not do enforcement and not do microversions, this seemed reasonable | 13:17 |
| alex_xu | sdague: ++ | 13:17 |
| sdague | because, if you put the v2.1 code on v2.0, v2.0 will do microversions, which was really not intended | 13:17 |
| lpetrut | sdague: sorry for interrupting for a second, could you please give me some hints on the above stated issue, related to synchronizing the snapshot operation in the manager? | 13:18 |
| *** boris-42 has quit IRC | 13:18 | |
| johnthetubaguy | lpetrut: why do you want it synchronized? | 13:20 |
| *** gholler has joined #openstack-nova | 13:21 | |
| lpetrut | well, it causes us a lot of trouble in case you attempt to destroy the instance while it's being snapshoted or if you attempt to change the instance power state | 13:21 |
| johnthetubaguy | lpetrut: we use the oslo.concurrency lock decorator to do that sort of thing, see how the resource tracker does some basic locking | 13:21 |
| johnthetubaguy | lpetrut: ah, so you want to lock on the instance_uuid I suppose? | 13:22 |
| lpetrut | I know, that's easy to add. I just wanted to get some feedback, to see if people would agree w/ synchronizing that | 13:22 |
| lpetrut | johnthetubaguy: exactly | 13:22 |
| johnthetubaguy | lpetrut: generally we like to let people delete as quickly as possible though, because it has billing impact | 13:22 |
| *** pbandzi has quit IRC | 13:22 | |
| johnthetubaguy | lpetrut: you can create some horrid deadlock issues, if you are not careful, but it could work... | 13:23 |
| lpetrut | johnthetubaguy: in our case, the delete fails so you end up having leaked instances. The other option would be to use this lock in the driver | 13:23 |
| johnthetubaguy | lpetrut: it should be the other way around | 13:23 |
| *** cknight has joined #openstack-nova | 13:24 | |
| johnthetubaguy | lpetrut: I mean a lock kinda fixes it up, but it doesn't really solve the issue, but its probably a half reasonable quick fix | 13:24 |
| johnthetubaguy | others may hate me for saying that, but it seems a pragmatic step forward from totally broken | 13:25 |
| jaypipes | danpb, sahid: question... why do we not pass VIR_DOMAIN_AFFECT_CONFIG if we are attaching a PCI or SR-IOV device? | 13:25 |
| lpetrut | johnthetubaguy: so just to be sure, you suggest using the lock in the hyper-v driver for now, right? | 13:25 |
| lpetrut | johnthetubaguy: because I was not sure which option you were reffering to by "the other way arround" | 13:27 |
| *** mriedem has joined #openstack-nova | 13:27 | |
| sahid | jaypipes: yes i thought about that too but since the code was like that i did not update it | 13:28 |
| sahid | jaypipes: but honestly i do not no why | 13:28 |
| *** ildikov_ has joined #openstack-nova | 13:28 | |
| jaypipes | sahid: oh, I'm not complaining about that :) I was just curious. | 13:28 |
| *** vladikr has joined #openstack-nova | 13:28 | |
| sahid | jaypipes: sure you question make a lot of sense and unfortunatly i do not know the answer | 13:28 |
| *** rushiagr is now known as rushiagr_away | 13:28 | |
| jaypipes | sahid: I'm diggin this refactor-libvirt patch series from you. | 13:29 |
| jaypipes | very nice work. | 13:29 |
| sahid | jaypipes: thank you i'm happy to see you following that work | 13:29 |
| jaypipes | sahid: +2 from me on the series. | 13:30 |
| sahid | thanks if i have to respin i will address you comments | 13:31 |
| sahid | i will try to find the answer about sriov device, perhpas we have a bug behind this | 13:31 |
| openstackgerrit | Lan Qi Song proposed openstack/nova: Remove useless volume when boot from volume failed https://review.openstack.org/177084 | 13:32 |
| *** iamjarvo has joined #openstack-nova | 13:32 | |
| openstackgerrit | Alex Xu proposed openstack/nova-specs: Correct resource name for consolidate console API https://review.openstack.org/185844 | 13:33 |
| *** artom has joined #openstack-nova | 13:33 | |
| *** VW has quit IRC | 13:34 | |
| *** thangp has joined #openstack-nova | 13:35 | |
| *** burt has joined #openstack-nova | 13:36 | |
| *** nihilifer has joined #openstack-nova | 13:39 | |
| *** afazekas_mtg has quit IRC | 13:40 | |
| *** jecarey has quit IRC | 13:40 | |
| *** boris-42 has joined #openstack-nova | 13:41 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add to_dict() to ImageMeta and ImageMetaProps objects https://review.openstack.org/186474 | 13:41 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add RequestSpec methods for primitiving into dicts https://review.openstack.org/187197 | 13:41 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Remove unnecessary method in FilterScheduler https://review.openstack.org/188485 | 13:41 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Consider that all scheduler calls are IO Ops https://review.openstack.org/188486 | 13:42 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Create RequestSpec object https://review.openstack.org/145528 | 13:42 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler https://review.openstack.org/188487 | 13:42 |
| bauwser | jaypipes: ^ above is for your pleasure | 13:42 |
| bauwser | jaypipes: still missing to update the filters and the RPC API, but that's a first usage | 13:42 |
| *** exploreshaifali has quit IRC | 13:43 | |
| jaypipes | sweet. wil review today, bauwser | 13:43 |
| *** jaypipes is now known as leakypipes | 13:43 | |
| bauwser | leakypipes: np, will be stuck in Disney world by the next 4 days | 13:43 |
| bauwser | leakypipes: ie. on PTO Mon-Tues | 13:43 |
| danpb | leakypipes: no idea really | 13:44 |
| leakypipes | bauwser: lol, sounds like hell on Earth to me :( | 13:45 |
| *** gholler has quit IRC | 13:46 | |
| bauwser | leakypipes: and add a 5yo daughter wanting to see Cinderella and you can see what looks like Hell | 13:46 |
| mriedem | sc68cal: got some more info on that vif_type=binding_failed issue in nova from yesterday - turns out the neutron ovs agent was dead due to some rpc misconfiguration | 13:48 |
| mriedem | i'm not sure what to do on the nova side when we detect a busted port in the nw info model for an instance | 13:49 |
| mriedem | i guess we just fail to spawn the instance and the user had to find out that way | 13:50 |
| mriedem | the 'user' wouldn't know though, the admin would have to find out the ovs agent is down | 13:50 |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: Don't report resources for nodes in unavailable states https://review.openstack.org/188652 | 13:50 |
| beagles | mriedem, just a related FYI: https://review.openstack.org/#/c/156112/ | 13:51 |
| jroll | leakypipes: ^ the fix was way easier than we thought :P | 13:51 |
| mriedem | beagles: cool, i'll take a look - seems exactly the issue i'm seeing | 13:51 |
| beagles | sc68cal, you too.. if you weren't aware of that patch ^^^ | 13:51 |
| *** jecarey has joined #openstack-nova | 13:52 | |
| *** eliqiao has quit IRC | 13:53 | |
| lxsli | leakypipes: I've had a go at updating your resource objects spec for you | 13:53 |
| lxsli | leakypipes: it turned into a bit of a major rewrite so I put it up as a dependent patch rather than a new patchset: https://review.openstack.org/#/c/188759 | 13:54 |
| lxsli | I haven't done the implementation or later sections yet but if you get a chance could you give an initial yay/nay look please? | 13:54 |
| lxsli | bauwser n0ano: would also welcome any feedback you have | 13:55 |
| lxsli | edl | 13:55 |
| bauwser | lxsli: CC'd yup | 13:55 |
| *** eharney has joined #openstack-nova | 13:55 | |
| openstackgerrit | Ritesh proposed openstack/nova: Nova flavor-list swap size to show 0 when no value is specified https://review.openstack.org/188781 | 13:55 |
| bauwser | lxsli: btw. I won't be able to attend next scheduler weekly meeting | 13:56 |
| lxsli | bauwser: okey dokey | 13:56 |
| lxsli | anything fun? | 13:57 |
| *** READ10 has joined #openstack-nova | 13:58 | |
| *** VW has joined #openstack-nova | 14:00 | |
| *** exploreshaifali has joined #openstack-nova | 14:00 | |
| *** tonytan4ever has joined #openstack-nova | 14:01 | |
| *** zzzeek has joined #openstack-nova | 14:01 | |
| *** tonytan4ever has quit IRC | 14:02 | |
| *** yamahata has joined #openstack-nova | 14:03 | |
| openstackgerrit | Baodong (Robert) Li proposed openstack/nova: Detach obsolete ports during periodic info cache refresh https://review.openstack.org/188783 | 14:03 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests - take 2 https://review.openstack.org/188785 | 14:04 |
| openstackgerrit | Auktavian Garrett proposed openstack/nova: Use unique values for hypervisor_hostnames https://review.openstack.org/187844 | 14:05 |
| *** markvoelker has joined #openstack-nova | 14:05 | |
| openstackgerrit | Daniel Berrange proposed openstack/nova: libvirt: convert driver to use nova.objects.ImageMeta https://review.openstack.org/188119 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: libvirt: make instance compulsory in blockinfo APIs https://review.openstack.org/188118 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: compute: only use non_inheritable_image_properties if snapshotting https://review.openstack.org/187250 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: compute: remove get_image_metadata method https://review.openstack.org/187251 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: objects: add os_require_quiesce image property https://review.openstack.org/188787 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: objects: fix parsing of NUMA cpu/mem properties https://review.openstack.org/188788 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: virt: convert hardware module to use nova.objects.ImageMeta https://review.openstack.org/188789 | 14:06 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: xenapi: convert driver to use nova.objects.ImageMeta https://review.openstack.org/188790 | 14:06 |
| *** ajayaa has quit IRC | 14:06 | |
| *** exploreshaifali has quit IRC | 14:06 | |
| bauwser | lxsli: just Disneyland Paris... | 14:07 |
| danpb | wow, that's pretty impressive - XenProject CI posted a comment "Build successful" on https://review.openstack.org/#/c/188790/ a mere 3 seconds after i first uploaded it | 14:07 |
| danpb | either xen has some crazy fast hardware, or the test system is reporting lies | 14:08 |
| bauwser | danpb: while I had a merge issue for all my series with Xen... | 14:08 |
| lxsli | bauwser: pretty nice! I love rollercoasters, though I guess you won't get to go on the big ones | 14:08 |
| *** iamjarvo has joined #openstack-nova | 14:09 | |
| *** iamjarvo has quit IRC | 14:09 | |
| *** markvoelker has quit IRC | 14:09 | |
| bauwser | lxsli: it will be less sexy, ie. meeting princesses | 14:09 |
| *** iamjarvo has joined #openstack-nova | 14:10 | |
| *** achanda has joined #openstack-nova | 14:10 | |
| mriedem | eliqiao1: beagles: comments in https://review.openstack.org/#/c/156112/ - needs some work yet | 14:11 |
| mriedem | pretty easy changes though | 14:11 |
| lxsli | bauwser: I'm sure a confident man like you will be able to enjoy meeting some princesses | 14:12 |
| lxsli | maybe even try on a tiara! | 14:12 |
| beagles | mriedem, cool. thanks for looking at it. Its been iterating a *lot*, it'd be good to bring it "home". | 14:14 |
| mriedem | yeah i see since february | 14:14 |
| bauwser | lxsli: em, will maybe be less verbose once I wait 1 hour in a queue for meeting a princess :) | 14:15 |
| lxsli | bauwser: hehe | 14:15 |
| openstackgerrit | Baodong (Robert) Li proposed openstack/nova: Handle port delete initiated by neutron https://review.openstack.org/179390 | 14:15 |
| *** rushiagr_away is now known as rushiagr | 14:18 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add to_dict() to ImageMeta and ImageMetaProps objects https://review.openstack.org/186474 | 14:19 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add RequestSpec methods for primitiving into dicts https://review.openstack.org/187197 | 14:19 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Remove unnecessary method in FilterScheduler https://review.openstack.org/188485 | 14:19 |
| *** cfriesen has joined #openstack-nova | 14:19 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Consider that all scheduler calls are IO Ops https://review.openstack.org/188486 | 14:19 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler https://review.openstack.org/188487 | 14:19 |
| *** otter768 has joined #openstack-nova | 14:21 | |
| *** tonytan4ever has joined #openstack-nova | 14:22 | |
| *** cfriesen has quit IRC | 14:24 | |
| *** otter768 has quit IRC | 14:26 | |
| *** foexle has quit IRC | 14:27 | |
| *** sks has quit IRC | 14:28 | |
| *** mtanino has joined #openstack-nova | 14:29 | |
| leakypipes | jroll: cool, I'll review it in a bit :) | 14:30 |
| leakypipes | lxsli: looking... | 14:30 |
| *** salv-orlando has joined #openstack-nova | 14:31 | |
| dansmith | oh my.. manners! | 14:32 |
| *** dansmith is now known as superdan | 14:32 | |
| * gillliard didn't want to be the first to say... | 14:32 | |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: api: relax validation for unversioned requests https://review.openstack.org/173243 | 14:34 |
| openstackgerrit | John Garbutt proposed openstack/nova-specs: api: relax validation for unversioned requests https://review.openstack.org/173243 | 14:34 |
| *** ankit_ag has quit IRC | 14:34 | |
| *** garyk has quit IRC | 14:35 | |
| johnthetubaguy | danpb: bauwser: there is a zuul bug there, when it choses there are no tests to run, it sends an empty +1, sigh | 14:36 |
| johnthetubaguy | the XenCI that is | 14:36 |
| bauwser | johnthetubaguy: uh | 14:36 |
| *** tjones1 has joined #openstack-nova | 14:36 | |
| *** tjones1 has left #openstack-nova | 14:36 | |
| bauwser | johnthetubaguy: I thought it was fixed on the regular CI ? | 14:37 |
| bauwser | johnthetubaguy: if so, maybe Zuul would need to be updated on the Xen 3rd-party CI ? | 14:37 |
| johnthetubaguy | bauwser: that never goes down to zero tests, as I understand it, but I suspect they just need to update to a new zuul | 14:37 |
| *** packet has joined #openstack-nova | 14:37 | |
| johnthetubaguy | yeah, +1 | 14:37 |
| johnthetubaguy | BobBall: our zuul might be out of date and causing some confusion ^ | 14:37 |
| johnthetubaguy | BobBall: by our I mean Xen | 14:37 |
| bauwser | okay, that's actually just a bit painful because Xen is voting :) | 14:38 |
| BobBall | humz? | 14:38 |
| johnthetubaguy | bauwser: yeah, there was another one as well | 14:38 |
| bauwser | BobBall: sec, showing | 14:38 |
| BobBall | danpb: This is an infra bug | 14:38 |
| bauwser | XenProject CI 3:56 PM Patch Set 2: Verified-1This change depends on a change that failed to merge. | 14:38 |
| bauwser | BobBall: ^ | 14:38 |
| danpb | BobBall: ah ok | 14:39 |
| bauwser | BobBall: so I guess it's a zuul bug, right ? | 14:39 |
| BobBall | danpb: Zuul allows you to selectively run on parts of nova (e.g. run only on doc changes) - but it does not allow you to selectively post messages. In this case we exclude doc changes and all non-XenServer hypervisor | 14:39 |
| BobBall | bauwser: That one is less certain to be a zuul bug; What's the changeset? | 14:39 |
| BobBall | A verified -1 shouldn't be sent in the case of the zuul bug I was talking about | 14:39 |
| johnthetubaguy | BobBall: there was another CI with the same issue the other day, possibly mellonox I think | 14:40 |
| johnthetubaguy | BobBall: we don't want a +1 either right? | 14:40 |
| bauwser | BobBall: I just pushed again the series, so you won't find it now but it was beginning with https://review.openstack.org/#/c/145528/15 | 14:40 |
| johnthetubaguy | should be +0 | 14:40 |
| mnestratov | BobBall: JFYI there is a fix pending for this https://review.openstack.org/#/c/188383/ | 14:40 |
| BobBall | Correct johnthetubaguy - Just saw the ML - but I don't think you can configure that | 14:40 |
| *** sks has joined #openstack-nova | 14:40 | |
| BobBall | Joyous mnestratov! | 14:40 |
| *** IanGovett has joined #openstack-nova | 14:40 | |
| johnthetubaguy | oh I haven't see the ML, but thats cool | 14:40 |
| *** moshele has joined #openstack-nova | 14:41 | |
| openstackgerrit | Daniel Berrange proposed openstack/nova: libvirt: convert driver to use nova.objects.ImageMeta https://review.openstack.org/188119 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: libvirt: make instance compulsory in blockinfo APIs https://review.openstack.org/188118 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: objects: add os_require_quiesce image property https://review.openstack.org/188787 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: virt: convert hardware module to use nova.objects.ImageMeta https://review.openstack.org/188789 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: compute: only use non_inheritable_image_properties if snapshotting https://review.openstack.org/187250 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: objects: fix parsing of NUMA cpu/mem properties https://review.openstack.org/188788 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: compute: remove get_image_metadata method https://review.openstack.org/187251 | 14:41 |
| openstackgerrit | Daniel Berrange proposed openstack/nova: xenapi: convert driver to use nova.objects.ImageMeta https://review.openstack.org/188790 | 14:41 |
| *** belmoreira has quit IRC | 14:42 | |
| *** yamahata has quit IRC | 14:43 | |
| *** electrocucaracha has joined #openstack-nova | 14:43 | |
| BobBall | bauwser: I don't see that failure on that patch; The only CI that failed to merge patch set 2 was the IBM PowerKVM CI? | 14:43 |
| bauwser | BobBall: my bad, that one was the dependency, the one which failed was the one which needed it | 14:44 |
| bauwser | BobBall: ie. https://review.openstack.org/#/c/186474/4 | 14:44 |
| *** zzzeek has quit IRC | 14:45 | |
| BobBall | bauwser: Digging. | 14:45 |
| bauwser | BobBall: thanks | 14:45 |
| *** zzzeek has joined #openstack-nova | 14:46 | |
| *** zzzeek has quit IRC | 14:46 | |
| *** armax has joined #openstack-nova | 14:46 | |
| *** _gryf has joined #openstack-nova | 14:47 | |
| mmedvede | johnthetubaguy: do you know if the zuul bug you're referring to is filed? I did manually patch our zuul (IBM PowerKVM CI) not to post empty comments, I wondered if it was intended design to post empties | 14:47 |
| johnthetubaguy | mmedvede: no idea I am afraid | 14:48 |
| johnthetubaguy | it could just be a config bug, to my knowledge | 14:48 |
| BobBall | bauwser: Curious - Seems that zuul was quite happy with the merge. I have no idea why it posted that it wasn't... | 14:49 |
| BobBall | mmedvede: See mnestratov's posted fix https://review.openstack.org/#/c/188383/ | 14:49 |
| bauwser | BobBall: indeed :) | 14:49 |
| *** xgsa has quit IRC | 14:50 | |
| BobBall | I can't even see what the dependencies were at the time of patchset 4 to re-test that. | 14:50 |
| BobBall | If you see it again can you let me know preferably before uploading a new change? :) | 14:50 |
| BobBall | Or just try a xen: recheck to see if it gets through then | 14:50 |
| *** xgsa has joined #openstack-nova | 14:50 | |
| BobBall | because I'm confused from the logs. | 14:50 |
| mmedvede | BobBall: thanks for the link, that is the proper fix | 14:52 |
| beagles | danpb: regarding the vif-plug-script spec. Outside of mleroy's comments on vif_script_dir, what do you feel remains on that to resolve? | 14:54 |
| bauwser | BobBall: nevermind, I'll put another change because I need it :) | 14:55 |
| BobBall | Fair enough - just ping me if you see another issue. | 14:56 |
| *** achanda has quit IRC | 14:56 | |
| *** hemnafk is now known as hemna | 14:58 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add IOUtils class for asynchronous IO operations https://review.openstack.org/184035 | 15:00 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add IOQueue class used for named pipe IO operations https://review.openstack.org/184036 | 15:00 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add named pipe handler https://review.openstack.org/184037 | 15:00 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add serial console proxy https://review.openstack.org/184038 | 15:00 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add serial console handler class https://review.openstack.org/184039 | 15:01 |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Add SerialConsoleOps class https://review.openstack.org/184040 | 15:01 |
| *** agireud has joined #openstack-nova | 15:04 | |
| *** dmsimard has joined #openstack-nova | 15:04 | |
| *** unicell1 has joined #openstack-nova | 15:04 | |
| dmsimard | melwitt: Re: https://bugs.launchpad.net/python-novaclient/+bug/1420192 have you ended up finding what was the cause ? Also seem to be bumping into this with Cells. | 15:05 |
| openstack | Launchpad bug 1420192 in python-novaclient "Nova interface-attach command has optional arguments to add network details. It should be positional arguments otherwise command fails." [Low,Confirmed] - Assigned to jui chandwaskar (jui-136) | 15:05 |
| *** unicell has quit IRC | 15:06 | |
| dims_ | mriedem: seen the python34 breaks because of jsonschema release? | 15:06 |
| mriedem | nope | 15:07 |
| mriedem | i'm sadly debugging trove gate issues in stable/juno :( | 15:07 |
| *** Longgeek_ has quit IRC | 15:07 | |
| *** Longgeek has joined #openstack-nova | 15:07 | |
| dims_ | mriedem: ack. i am digging in | 15:08 |
| dims_ | http://logs.openstack.org/81/155581/32/check/gate-python-glanceclient-python26/620ca24/console.html | 15:08 |
| mriedem | that's a py26 job | 15:08 |
| dims_ | oops wrong url https://jenkins03.openstack.org/job/gate-nova-python34/82/console | 15:09 |
| *** beagles is now known as beagles|lunch | 15:09 | |
| *** lsmola has quit IRC | 15:12 | |
| mriedem | dims_: i'd open an issue against the jsonschema project | 15:12 |
| mriedem | the last 2 changes here are suspect https://github.com/Julian/jsonschema/commits/master/setup.py | 15:13 |
| mriedem | dims_: the problem isn't jsonschema, it's functools32 | 15:14 |
| johnthetubaguy | superdan: where you going to document data migrations and the generic plan, I just saw this spec come up, it looks like it could be quite messy to do online: https://review.openstack.org/#/c/172412 | 15:14 |
| mriedem | https://github.com/MiCHiLU/python-functools32/blob/master/setup.py#L8 | 15:14 |
| mriedem | dims_: ^ | 15:14 |
| dims_ | mriedem: functools32 is optional https://github.com/Julian/jsonschema/blob/master/setup.py#L29 | 15:14 |
| superdan | johnthetubaguy: devref I guess | 15:14 |
| openstackgerrit | Eugeniya Kudryashova proposed openstack/python-novaclient: Raise exception when API does't support microversions https://review.openstack.org/188816 | 15:15 |
| johnthetubaguy | superdan: yeah | 15:15 |
| mriedem | dims_: yeah, we shouldn't be pullling that in | 15:15 |
| danpb | someone want to +W this docs fix for NUMA .... https://review.openstack.org/#/c/178773/ | 15:15 |
| *** READ10 has quit IRC | 15:15 | |
| superdan | johnthetubaguy: that probably just has to be a two-step thing.. we shouldn't allow it to be an "offline migration" IMHO | 15:15 |
| *** electrocucaracha has quit IRC | 15:15 | |
| johnthetubaguy | superdan: +1 | 15:15 |
| *** cknight has quit IRC | 15:15 | |
| *** electrocucaracha has joined #openstack-nova | 15:15 | |
| superdan | johnthetubaguy: by the way, just yesterday, I tested a running juno deployment with kilo schema applied at runtime | 15:16 |
| superdan | johnthetubaguy: and it worked fine, which was our goal.. I don't think we should let that behavior regress.. ever. | 15:16 |
| *** dprince has quit IRC | 15:16 | |
| *** ildikov_ has quit IRC | 15:16 | |
| johnthetubaguy | superdan: +100 | 15:17 |
| johnthetubaguy | superdan: should we change grenade? | 15:17 |
| *** tjones1 has joined #openstack-nova | 15:17 | |
| superdan | johnthetubaguy: we could maybe add another mode to try this, but grenade is really not intended to test this sort of arrangement | 15:18 |
| superdan | so doing it there would require just not restarting any nova services, which might be okay, but it also might be easier to test this in a different environment | 15:18 |
| johnthetubaguy | superdan: I was just thinking run the migrate before restarting the services | 15:18 |
| mriedem | dims_: i see the problem | 15:19 |
| mriedem | opening an issue | 15:19 |
| mriedem | dims_: already done i guess https://github.com/Julian/jsonschema/issues/228 | 15:19 |
| superdan | johnthetubaguy: well, we have to run the schema upgrade and then test the old services before restart to make sure they work | 15:19 |
| *** electrocucaracha has quit IRC | 15:19 | |
| johnthetubaguy | superdan: oh, true, oops | 15:20 |
| dims_ | mriedem: ack thanks | 15:20 |
| superdan | johnthetubaguy: but would like to figure out how to prove that in a job, yeah | 15:20 |
| johnthetubaguy | superdan: anyways, its so awesome to see the effort starting to pay off! | 15:20 |
| *** electrocucaracha has joined #openstack-nova | 15:20 | |
| superdan | johnthetubaguy: I'm on a call now, but will add comments to this review when I'm done | 15:20 |
| superdan | johnthetubaguy: totes! | 15:20 |
| *** electrocucaracha has quit IRC | 15:20 | |
| johnthetubaguy | superdan: cools | 15:20 |
| *** dane-fichter has joined #openstack-nova | 15:21 | |
| *** electrocucaracha has joined #openstack-nova | 15:22 | |
| *** agireud has quit IRC | 15:24 | |
| dims_ | mriedem: so it breaks both py26 and 34 jobs that need jsonschema | 15:24 |
| dims_ | oh well. time for a break :) | 15:24 |
| *** agireud has joined #openstack-nova | 15:28 | |
| *** markvoelker has joined #openstack-nova | 15:30 | |
| mriedem | dims_: ack, i'll get an LP bug opened to track the issue in e-r | 15:31 |
| openstackgerrit | Qin Zhao proposed openstack/nova: Refactor show_port() in neutron api https://review.openstack.org/135020 | 15:31 |
| *** patrickeast has joined #openstack-nova | 15:31 | |
| *** _gryf is now known as _gryf|afk | 15:31 | |
| *** arnaud____ has joined #openstack-nova | 15:32 | |
| dims_ | thanks mriedem | 15:32 |
| *** marun has joined #openstack-nova | 15:33 | |
| *** oro_ has quit IRC | 15:33 | |
| *** achanda has joined #openstack-nova | 15:34 | |
| *** garyk has joined #openstack-nova | 15:34 | |
| *** markvoelker has quit IRC | 15:34 | |
| *** thedodd has joined #openstack-nova | 15:36 | |
| *** tsekiyama has joined #openstack-nova | 15:36 | |
| lxsli | superdan: is there anything written about passing objects over REST please? | 15:37 |
| superdan | lxsli: not that I know of, no | 15:37 |
| *** mohitsharma has joined #openstack-nova | 15:37 | |
| lxsli | superdan: also are there any plans for passing object definitions over the wire? | 15:37 |
| superdan | lxsli: definitions? | 15:38 |
| lxsli | like "class MyObj(base.NovaObject)" | 15:38 |
| bauwser | lxsli: well, I don't see your point of passing objects in a request body ?. | 15:38 |
| superdan | lxsli: one of the design tenets is that we don't pass the definition (or code) over the wire | 15:39 |
| lxsli | so given my amendment to the resource objects spec, as it stands if we break out the scheduler Nova will have to provide a plugin to the scheduler giving EG RAMAmount | 15:39 |
| superdan | lxsli: for security and stability | 15:39 |
| lxsli | superdan: sensible - just checking, thanks | 15:39 |
| bauwser | lxsli: don't get your point, still | 15:40 |
| bauwser | lxsli: you have JSON schemas | 15:40 |
| bauwser | and you have input validation | 15:40 |
| *** marun has quit IRC | 15:40 | |
| lxsli | bauwser: so you're saying although an object can't be passed directly over REST, it can be serialized and deserialized with schema validation which is just as good? | 15:41 |
| superdan | lxsli: the object can be passed over REST easily | 15:41 |
| superdan | lxsli: the definition cannot | 15:41 |
| bauwser | lxsli: I'm not saying that | 15:41 |
| bauwser | lxsli: I'm saying that you can expose a model | 15:41 |
| lxsli | superdan: ah! OK thanks | 15:41 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: Adding user_id handling to keypair index, show and create api calls https://review.openstack.org/70485 | 15:42 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: functional tests for keypairs create in API microversion 2.4 https://review.openstack.org/183645 | 15:42 |
| danpb | beagles|lunch: from my pov it looks detailed enough for the spec | 15:42 |
| lxsli | bauwser: still not getting it | 15:42 |
| *** agireud has quit IRC | 15:44 | |
| bauwser | lxsli: probably because I'm not getting your problem too :) | 15:44 |
| *** david-lyle has quit IRC | 15:46 | |
| *** tjones1 has left #openstack-nova | 15:46 | |
| *** david-lyle has joined #openstack-nova | 15:46 | |
| lxsli | bauwser: so the resource objects spec (with my amendment) introduces abstract classes like ResourceType with Nova-specific subclasses like NUMAResourceType | 15:48 |
| bauwser | yup | 15:48 |
| lxsli | bauwser: I expect Nova to receive the ResourceType using a scheduler-client pip, which will also contain EG the SchedulerReportClient | 15:48 |
| bauwser | and you would like to find some way to say that a REST API resource would allow to pass them in | 15:48 |
| bauwser | ? | 15:48 |
| bauwser | bauwser: okay, then, that's not a problem, rich objects can be declared | 15:49 |
| leakypipes | jroll: you see JayF's comment on your patch? | 15:49 |
| lxsli | bauwser: but we also need to pass the NUMAResourceType back to the scheduler, as we can't pass definitions over the wire, that has to be a plugin | 15:49 |
| bauwser | oops | 15:49 |
| bauwser | lxsli: | 15:49 |
| lxsli | bauwser: and each service that wants to use the scheduler must register its own plugin to it, explaining its specific resources | 15:49 |
| jroll | leakypipes: yes, I did, and responded | 15:49 |
| bauwser | lxsli: Nova (or the scheduler) API has to expect certain sets of resources | 15:49 |
| lxsli | bauwser: I'd like the scheduler API to only refer to abstract classes from the scheduler-client | 15:50 |
| bauwser | lxsli: okay, I see your question and I think you're overthinking for now :) | 15:50 |
| lxsli | bauwser: however the specific ResourceTypes have custom comparison logic in them, so they need to be made available to the scheduler | 15:50 |
| *** markus_z has quit IRC | 15:50 | |
| bauwser | lxsli: just because the REST API question needs to be addressed only once the scheduler will expose its API | 15:50 |
| *** beagles|lunch is now known as beagles | 15:51 | |
| lxsli | bauwser: I'm not actually worried about that, passing objects around should be easy like dan said | 15:51 |
| lxsli | bauwser: it's getting the definitions in place | 15:51 |
| lxsli | bauwser: I'm not super happy that Nova and the Scheduler will have to exchange plugins | 15:51 |
| bauwser | lxsli: because I'm seriously -1 about passing definitions to the server | 15:51 |
| bauwser | lxsli: I mean, the API server exposes the resources it knows | 15:52 |
| bauwser | lxsli: any client can provide those resources | 15:52 |
| *** matrohon has quit IRC | 15:52 | |
| bauwser | lxsli: but the client shouldn't define its own resource to the API server | 15:52 |
| bauwser | ie. the scheduler accepts a set of resources it knows | 15:52 |
| bauwser | the client provides them | 15:53 |
| lxsli | Well, we could put Nova, Cinder etc resource definitions in the Scheduler tree just like Nova has virt drivers | 15:53 |
| lxsli | It means you'd need a new scheduler release to change your resources rather than just a plugin upgrade but ^^ | 15:54 |
| bauwser | lxsli: I wouldn't use the virt drivers as an analogy to your point | 15:54 |
| lxsli | bauwser: because? | 15:55 |
| bauwser | lxsli: because it's not something API-related :) | 15:55 |
| bauwser | lxsli: coming from the WSME world, it's quite simple to me that your API definition proposes some resources | 15:56 |
| lxsli | Do you mean scheduler resources or REST resources? | 15:56 |
| bauwser | REST resources having types | 15:57 |
| bauwser | anyway, we're far beyond the current situation, and I don't yet think the scheduler would need a REST API | 15:58 |
| bauwser | lxsli: what I know is that other services may expose some resource usage to Nova/the scheduler | 15:58 |
| *** vishwanathj has joined #openstack-nova | 15:58 | |
| bauwser | how we expose that is a matter of discussion :) | 15:58 |
| bauwser | lxsli: but frankly not a Friday afternoon :p | 15:59 |
| *** achanda_ has joined #openstack-nova | 15:59 | |
| *** electrocucaracha has quit IRC | 16:00 | |
| *** achanda has quit IRC | 16:00 | |
| lxsli | Nova-Glance comms are over REST, right? For scheduler to be independent like those, I thought a REST interface was a given | 16:00 |
| danpb | yes, its over REST | 16:02 |
| johnthetubaguy | lxsli: for the scheduler, I am thinking a lib that was backwards compatible would be enough/better | 16:02 |
| johnthetubaguy | lxsli: it could be REST, but that might not make sense, dunno | 16:02 |
| lxsli | johnthetubaguy: so EG oslo_scheduler? | 16:03 |
| *** jistr has quit IRC | 16:03 | |
| *** electrocucaracha has joined #openstack-nova | 16:03 | |
| johnthetubaguy | lxsli: kinda, although its not really oslo, its a client lib for the scheduler (hence the scheduler "lib" we have in tree) | 16:04 |
| lxsli | yeah we have the Scheduler{Report,Query}Client already | 16:05 |
| *** danpb has quit IRC | 16:05 | |
| lxsli | I thought the aim was for the scheduler to become a completely separate project, which other projects could use for whatever scheduling they needed - maybe I'm off base there | 16:06 |
| superdan | long-term that's a goal | 16:07 |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: Don't report resources for nodes without instances https://review.openstack.org/188652 | 16:08 |
| *** IanGovett has quit IRC | 16:08 | |
| bauwser | lxsli: you're surely off-base for now :) | 16:08 |
| *** jlanoux has quit IRC | 16:08 | |
| jroll | leakypipes: I think this is much more sane now | 16:08 |
| johnthetubaguy | lxsli: the thing is, I don't want us spending time on the REST API until we have the lines clearer, and we don't need to create a REST API to do that | 16:08 |
| bauwser | ++ | 16:08 |
| lxsli | oh sure I'm not imagining it'll happen tomorrow | 16:09 |
| bauwser | some people would :) | 16:09 |
| bauwser | I know it's Friday, so we can open the cans of worms right ? :) | 16:09 |
| figleaf | lxsli: it's always good to keep that in mind as a design goal, but not as a requirement any time soon | 16:10 |
| lxsli | sure I just don't want to put additional obstacles in the way | 16:10 |
| *** claudiub has quit IRC | 16:10 | |
| figleaf | lxsli: We've talked about how scheduler would "understand" various resource types in the past | 16:11 |
| bauwser | lxsli: having given and clean APIs doesn't mean we're going to the wrong direction :) | 16:11 |
| figleaf | lxsli: we never came up with a satisfactory solution | 16:11 |
| lxsli | figleaf: !!! ... v.v | 16:12 |
| *** tjones1 has joined #openstack-nova | 16:12 | |
| *** agireud has joined #openstack-nova | 16:12 | |
| lxsli | well, good to know. I'll try to avoid getting bogged down in that then | 16:12 |
| figleaf | lxsli: yeah, you're falling into the same sticky pit that we've been in before | 16:13 |
| *** xgsa has quit IRC | 16:13 | |
| figleaf | lxsli: the scheduler lib approach seems to be the only workable one. A separate service will have the problems you've outlined: how does it understand different resources? | 16:14 |
| *** exploreshaifali has joined #openstack-nova | 16:14 | |
| *** xgsa has joined #openstack-nova | 16:15 | |
| *** alex_klimov has quit IRC | 16:15 | |
| lxsli | Well there's food for thought over the weekend | 16:16 |
| *** tjones1 has quit IRC | 16:16 | |
| lxsli | So we'd have EG oslo_scheduler that would bundle an API to a scheduling algorithm with a bunch of abstract classes to pass over that API. Then Nova would wrap that up into Nova-Scheduler by providing subclasses and an RPC API? | 16:20 |
| lxsli | figleaf: ^ | 16:20 |
| figleaf | lxsli: the scheduler lib would understand resources, requests, and claims. The service needing scheduling (e.g., Nova) would define classes of each | 16:21 |
| figleaf | lxsli: the scheduler provides the comparison, filtering, and weighting logic | 16:22 |
| lxsli | figleaf: the service could provide comparison logic, given appropriate hooks | 16:22 |
| *** matrohon has joined #openstack-nova | 16:22 | |
| *** otter768 has joined #openstack-nova | 16:22 | |
| figleaf | lxsli: it would also define the abstract classes and how they relate | 16:23 |
| openstackgerrit | Qin Zhao proposed openstack/nova: Refactor show_port() in neutron api https://review.openstack.org/135020 | 16:23 |
| figleaf | lxsli: Nova would define subclasses of each | 16:23 |
| lxsli | figleaf: got it, thank you | 16:23 |
| figleaf | lxsli: cool | 16:23 |
| figleaf | lxsli: there might be a better way, but we haven't really come up with one | 16:24 |
| bauwser | my internal program (called sbtables) is filtering out any design discussion led on Friday | 16:24 |
| lxsli | johnthetubaguy: does that match what you were thinking please? | 16:24 |
| figleaf | bauwser: shouldn't that be 'bstables'? :) | 16:24 |
| lxsli | bauwser: as I said to my colleague this morning, I didn't get a lot done Mon-Thu so this is last chance saloon :) | 16:24 |
| bauwser | figleaf: you don't figure how many people are putting "s" at the end of my last name... | 16:25 |
| bauwser | like Sylvain Bauzas | 16:25 |
| figleaf | bauwser: Heh, I had 30 minutes before I gotta pick the kid up from school | 16:25 |
| figleaf | bauwser: this was fun | 16:25 |
| johnthetubaguy | lxsli: honestly my brain is a bit fried, not sure I have enough context to tell, sorry | 16:25 |
| lxsli | johnthetubaguy: ok, thanks all, see you Monday :) | 16:26 |
| *** achanda_ has quit IRC | 16:26 | |
| figleaf | bauwser: yeah, but I meant 'bstables' as 'BS tables' (bullshit) | 16:26 |
| figleaf | bauwser: for Friday | 16:26 |
| bauwser | figleaf: oooooh | 16:26 |
| *** alex_klimov has joined #openstack-nova | 16:26 | |
| lxsli | our JIRA tag was BS for months... | 16:26 |
| bauwser | here sbtables was just referring to my name, sorry for the total miss of ingenuity | 16:27 |
| *** otter768 has quit IRC | 16:27 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova: HyperV: Nova serial console access support https://review.openstack.org/145004 | 16:28 |
| *** kiwik-chenrui has joined #openstack-nova | 16:28 | |
| *** unicell has joined #openstack-nova | 16:28 | |
| figleaf | bauwser: :) | 16:28 |
| *** unicell1 has quit IRC | 16:30 | |
| *** RuiChen has quit IRC | 16:31 | |
| *** alex_klimov has quit IRC | 16:32 | |
| *** nic has joined #openstack-nova | 16:32 | |
| bauwser | folks, going out for 4 days | 16:33 |
| bauwser | call the police if you don't see me by Wed | 16:33 |
| *** whenry has joined #openstack-nova | 16:34 | |
| *** dprince has joined #openstack-nova | 16:34 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/objects/test*.py https://review.openstack.org/187785 | 16:34 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests https://review.openstack.org/188243 | 16:34 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable python34 tests for nova/tests/unit/scheduler/test*.py https://review.openstack.org/187968 | 16:34 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Adding more py34 tests - take 2 https://review.openstack.org/188785 | 16:34 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] getting vmware api tests to work under python34 https://review.openstack.org/188653 | 16:34 |
| *** arnaud____ has quit IRC | 16:35 | |
| dims_ | bauwser: safe travels | 16:35 |
| *** tjones1 has joined #openstack-nova | 16:36 | |
| *** bauwser is now known as bauzas | 16:36 | |
| *** sahid has quit IRC | 16:37 | |
| *** claudiub has joined #openstack-nova | 16:39 | |
| *** sdake has joined #openstack-nova | 16:39 | |
| *** ZZelle has joined #openstack-nova | 16:42 | |
| *** mohitsharma has quit IRC | 16:43 | |
| *** achanda has joined #openstack-nova | 16:43 | |
| *** IanGovett has joined #openstack-nova | 16:44 | |
| *** sdake_ has joined #openstack-nova | 16:44 | |
| *** yassine_ has quit IRC | 16:45 | |
| *** tonytan4ever has quit IRC | 16:46 | |
| *** tonytan4ever has joined #openstack-nova | 16:46 | |
| *** annegentle has quit IRC | 16:46 | |
| *** sdake has quit IRC | 16:48 | |
| *** tonytan4ever has quit IRC | 16:51 | |
| *** moshele has quit IRC | 16:52 | |
| *** oro_ has joined #openstack-nova | 16:52 | |
| *** alanf-mc has joined #openstack-nova | 16:52 | |
| *** annegentle has joined #openstack-nova | 16:54 | |
| *** cknight has joined #openstack-nova | 16:55 | |
| *** achanda has quit IRC | 16:55 | |
| *** cknight has quit IRC | 16:56 | |
| *** cknight has joined #openstack-nova | 16:56 | |
| *** cknight has left #openstack-nova | 16:57 | |
| *** VW has quit IRC | 16:57 | |
| *** alanf-mc_ has joined #openstack-nova | 17:00 | |
| *** alanf-mc has quit IRC | 17:02 | |
| *** tjones1 has quit IRC | 17:02 | |
| openstackgerrit | Petrut Lucian proposed openstack/nova: Hyper-V: Lock snapshot operation using instance uuid https://review.openstack.org/188859 | 17:03 |
| *** tjones1 has joined #openstack-nova | 17:03 | |
| *** annegentle has quit IRC | 17:08 | |
| *** IanGovett has quit IRC | 17:10 | |
| *** vilobhmm has joined #openstack-nova | 17:10 | |
| *** alanf-mc_ has quit IRC | 17:11 | |
| *** alanf-mc has joined #openstack-nova | 17:11 | |
| openstackgerrit | Maxim Nestratov proposed openstack/nova: libvirt: rename parallels driver to virtuozzo https://review.openstack.org/184311 | 17:11 |
| *** agireud has quit IRC | 17:11 | |
| *** lucasagomes is now known as lucas-beer | 17:13 | |
| *** artom has quit IRC | 17:15 | |
| openstackgerrit | Jens Rosenboom proposed openstack/nova: Add microversion to allow server search option ip6 for non-admin https://review.openstack.org/179569 | 17:15 |
| *** achanda has joined #openstack-nova | 17:16 | |
| *** dane-fichter has quit IRC | 17:17 | |
| *** electrocucaracha has quit IRC | 17:18 | |
| *** markvoelker has joined #openstack-nova | 17:19 | |
| *** electrocucaracha has joined #openstack-nova | 17:21 | |
| openstackgerrit | Dmitry Guryanov proposed openstack/nova: libvirt: resize ploop disks with prl_disk_tool https://review.openstack.org/182390 | 17:23 |
| *** markvoelker has quit IRC | 17:24 | |
| *** electrocucaracha has quit IRC | 17:25 | |
| *** ssurana has joined #openstack-nova | 17:26 | |
| openstackgerrit | Dmitry Guryanov proposed openstack/nova: libvirt: support snapshots with parallels virt_type https://review.openstack.org/184153 | 17:27 |
| *** harlowja has quit IRC | 17:27 | |
| *** electrocucaracha has joined #openstack-nova | 17:28 | |
| *** harlowja has joined #openstack-nova | 17:32 | |
| *** achanda has quit IRC | 17:34 | |
| openstackgerrit | dane-fichter proposed openstack/nova-specs: Nova Support of Glance Image Signing & Encryption https://review.openstack.org/188874 | 17:37 |
| *** oro_ has quit IRC | 17:38 | |
| *** rushiagr is now known as rushiagr_away | 17:40 | |
| *** achanda has joined #openstack-nova | 17:40 | |
| *** dane-fichter has joined #openstack-nova | 17:40 | |
| *** lpetrut has quit IRC | 17:41 | |
| *** mwagner_afk has quit IRC | 17:42 | |
| mriedem | looking for a +W on https://review.openstack.org/#/c/135020/ - it's a nice cleanup that's been in the works for awhile | 17:45 |
| *** marzif_ has quit IRC | 17:45 | |
| *** krtaylor has quit IRC | 17:46 | |
| *** sdake has joined #openstack-nova | 17:50 | |
| *** annegentle has joined #openstack-nova | 17:50 | |
| *** marzif_ has joined #openstack-nova | 17:51 | |
| *** zz_jgrimm is now known as jgrimm | 17:53 | |
| *** sdake_ has quit IRC | 17:54 | |
| *** sdake_ has joined #openstack-nova | 17:56 | |
| *** penick has joined #openstack-nova | 17:57 | |
| *** achanda_ has joined #openstack-nova | 17:59 | |
| *** thangp has quit IRC | 17:59 | |
| *** sdake has quit IRC | 17:59 | |
| *** agireud has joined #openstack-nova | 18:01 | |
| *** achanda has quit IRC | 18:01 | |
| *** artom has joined #openstack-nova | 18:01 | |
| *** mingy has joined #openstack-nova | 18:03 | |
| *** rwsu has quit IRC | 18:04 | |
| *** penick has quit IRC | 18:06 | |
| *** ildikov_ has joined #openstack-nova | 18:09 | |
| *** achanda has joined #openstack-nova | 18:10 | |
| *** achanda_ has quit IRC | 18:10 | |
| *** sdake_ is now known as sdake | 18:10 | |
| *** Nisha has joined #openstack-nova | 18:18 | |
| *** achanda_ has joined #openstack-nova | 18:18 | |
| *** achanda has quit IRC | 18:20 | |
| *** krtaylor has joined #openstack-nova | 18:20 | |
| *** otter768 has joined #openstack-nova | 18:23 | |
| *** agireud has quit IRC | 18:23 | |
| *** mwagner_afk has joined #openstack-nova | 18:23 | |
| *** tonytan4ever has joined #openstack-nova | 18:26 | |
| *** agireud has joined #openstack-nova | 18:26 | |
| *** otter768 has quit IRC | 18:28 | |
| *** marun has joined #openstack-nova | 18:28 | |
| *** artom has quit IRC | 18:30 | |
| *** VW has joined #openstack-nova | 18:32 | |
| *** sks has quit IRC | 18:33 | |
| *** agireud has quit IRC | 18:33 | |
| *** VW has quit IRC | 18:33 | |
| *** VW has joined #openstack-nova | 18:34 | |
| *** VW has quit IRC | 18:34 | |
| *** kaisers1 has quit IRC | 18:35 | |
| *** VW has joined #openstack-nova | 18:35 | |
| *** artom has joined #openstack-nova | 18:35 | |
| *** mriedem is now known as mriedem_madmax | 18:35 | |
| *** mriedem_madmax has quit IRC | 18:35 | |
| *** rohitk has joined #openstack-nova | 18:36 | |
| *** kaisers1 has joined #openstack-nova | 18:37 | |
| *** kaisers has quit IRC | 18:37 | |
| *** marun has quit IRC | 18:38 | |
| *** achanda has joined #openstack-nova | 18:38 | |
| *** VW has quit IRC | 18:39 | |
| *** kaisers has joined #openstack-nova | 18:39 | |
| *** sabari has quit IRC | 18:41 | |
| *** achanda_ has quit IRC | 18:41 | |
| *** xyang1 has joined #openstack-nova | 18:41 | |
| *** tjones1 has quit IRC | 18:43 | |
| *** iamjarvo has quit IRC | 18:44 | |
| *** VW has joined #openstack-nova | 18:44 | |
| *** penick has joined #openstack-nova | 18:45 | |
| *** marun has joined #openstack-nova | 18:46 | |
| *** penick has quit IRC | 18:46 | |
| *** harlowja has quit IRC | 18:46 | |
| *** achanda_ has joined #openstack-nova | 18:48 | |
| *** achanda has quit IRC | 18:49 | |
| *** iamjarvo has joined #openstack-nova | 18:50 | |
| *** vladikr has quit IRC | 18:50 | |
| *** htruta has quit IRC | 18:52 | |
| *** harlowja has joined #openstack-nova | 18:53 | |
| *** ociuhandu has quit IRC | 18:53 | |
| *** wanghao has quit IRC | 18:54 | |
| *** ameade has quit IRC | 18:54 | |
| *** briancurtin has quit IRC | 18:54 | |
| *** zhiyan has quit IRC | 18:54 | |
| *** destiny has quit IRC | 18:54 | |
| *** sabari has joined #openstack-nova | 18:55 | |
| *** penick has joined #openstack-nova | 18:55 | |
| *** lpetrut has joined #openstack-nova | 18:57 | |
| *** exploreshaifali has quit IRC | 18:58 | |
| *** marun has quit IRC | 18:59 | |
| *** achanda has joined #openstack-nova | 19:00 | |
| *** achanda_ has quit IRC | 19:01 | |
| *** dane-fichter has quit IRC | 19:03 | |
| *** achanda has quit IRC | 19:04 | |
| *** iamjarvo has quit IRC | 19:06 | |
| *** sdake has quit IRC | 19:08 | |
| *** markvoelker has joined #openstack-nova | 19:08 | |
| *** Sukhdev has joined #openstack-nova | 19:11 | |
| *** markvoelker has quit IRC | 19:12 | |
| *** annegentle has quit IRC | 19:14 | |
| *** agireud has joined #openstack-nova | 19:15 | |
| *** Longgeek has quit IRC | 19:16 | |
| *** Nisha has quit IRC | 19:18 | |
| *** pradipta has joined #openstack-nova | 19:19 | |
| *** iamjarvo has joined #openstack-nova | 19:19 | |
| *** iamjarvo has quit IRC | 19:31 | |
| *** electrocucaracha has quit IRC | 19:33 | |
| *** electrocucaracha has joined #openstack-nova | 19:33 | |
| *** leakypipes has quit IRC | 19:34 | |
| *** rohitk has quit IRC | 19:35 | |
| *** _gryf|afk is now known as _gryf | 19:35 | |
| *** iamjarvo has joined #openstack-nova | 19:37 | |
| *** iamjarvo has joined #openstack-nova | 19:38 | |
| *** zzzeek has joined #openstack-nova | 19:41 | |
| *** angdraug has joined #openstack-nova | 19:45 | |
| *** Mike_D_laptop has joined #openstack-nova | 19:46 | |
| *** claudiub has quit IRC | 19:49 | |
| *** _gryf has quit IRC | 19:54 | |
| *** annegentle has joined #openstack-nova | 19:54 | |
| *** Mike_D_laptop1 has joined #openstack-nova | 19:54 | |
| *** amotoki has joined #openstack-nova | 19:55 | |
| *** Mike_D_laptop has quit IRC | 19:56 | |
| *** penick has quit IRC | 19:56 | |
| *** marun has joined #openstack-nova | 19:57 | |
| *** dims__ has joined #openstack-nova | 19:57 | |
| *** dims_ has quit IRC | 20:00 | |
| *** penick has joined #openstack-nova | 20:01 | |
| *** boris-42 has quit IRC | 20:02 | |
| *** erw has quit IRC | 20:02 | |
| *** rex_lee has quit IRC | 20:02 | |
| *** dougwig has quit IRC | 20:02 | |
| *** penick has quit IRC | 20:03 | |
| *** annegentle has quit IRC | 20:04 | |
| *** annegentle has joined #openstack-nova | 20:04 | |
| *** penick has joined #openstack-nova | 20:05 | |
| *** penick has quit IRC | 20:07 | |
| *** electrocucaracha has quit IRC | 20:10 | |
| *** thedodd has quit IRC | 20:11 | |
| *** thedodd has joined #openstack-nova | 20:11 | |
| *** dims__ has quit IRC | 20:11 | |
| *** electroc_ has joined #openstack-nova | 20:12 | |
| *** electroc_ has quit IRC | 20:13 | |
| *** oro_ has joined #openstack-nova | 20:13 | |
| *** tjones1 has joined #openstack-nova | 20:14 | |
| *** tellesnobrega_ has joined #openstack-nova | 20:17 | |
| *** ociuhandu has joined #openstack-nova | 20:19 | |
| *** VW has quit IRC | 20:19 | |
| *** ijw has joined #openstack-nova | 20:22 | |
| *** markvoelker has joined #openstack-nova | 20:24 | |
| *** otter768 has joined #openstack-nova | 20:24 | |
| *** VW has joined #openstack-nova | 20:25 | |
| *** electrocucaracha has joined #openstack-nova | 20:25 | |
| *** ijw has quit IRC | 20:28 | |
| *** markvoelker has quit IRC | 20:28 | |
| *** otter768 has quit IRC | 20:29 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: API: make sure a blank volume with no size is requested https://review.openstack.org/188913 | 20:31 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: Make Nova better at keeping track of volume sizes in BDM https://review.openstack.org/188914 | 20:31 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: compute: Make swap_volume with resize updates BDM size https://review.openstack.org/188915 | 20:31 |
| *** annashen has joined #openstack-nova | 20:34 | |
| *** READ10 has joined #openstack-nova | 20:35 | |
| *** tellesnobrega_ has quit IRC | 20:38 | |
| *** ivasev has quit IRC | 20:38 | |
| *** tellesnobrega_ has joined #openstack-nova | 20:39 | |
| *** alex_klimov has joined #openstack-nova | 20:39 | |
| *** iamjarvo has quit IRC | 20:39 | |
| *** alex_klimov has quit IRC | 20:45 | |
| vilobhmm | superdan : ping | 20:47 |
| superdan | vilobhmm: pong | 20:47 |
| vilobhmm | like the friday names….meanwhile thought will check with you on https://review.openstack.org/#/c/138607/ | 20:48 |
| vilobhmm | superdan : ^^ | 20:49 |
| *** pradipta has quit IRC | 20:49 | |
| superdan | vilobhmm: like I said before, it's on my list, I'll get to it, you don't need to remind me | 20:49 |
| *** jecarey has quit IRC | 20:49 | |
| *** lpetrut has quit IRC | 20:49 | |
| superdan | vilobhmm: have you reviewed any of my patches? | 20:49 |
| *** dougwig has joined #openstack-nova | 20:50 | |
| vilobhmm | superdan : i will try to help….please let me know which ones need review | 20:50 |
| vilobhmm | will be happy to help | 20:51 |
| *** ameade has joined #openstack-nova | 20:51 | |
| *** dprince has quit IRC | 20:51 | |
| superdan | vilobhmm: https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 20:52 |
| *** erw has joined #openstack-nova | 20:52 | |
| vilobhmm | superdan : sure will have a look | 20:52 |
| *** zhiyan has joined #openstack-nova | 20:54 | |
| *** dave-mccowan has joined #openstack-nova | 20:56 | |
| *** alex_klimov has joined #openstack-nova | 20:57 | |
| *** amotoki has quit IRC | 20:57 | |
| *** tonytan4ever has quit IRC | 20:59 | |
| *** tonytan4ever has joined #openstack-nova | 20:59 | |
| *** penick has joined #openstack-nova | 20:59 | |
| *** boris-42 has joined #openstack-nova | 21:00 | |
| *** tonytan4ever has quit IRC | 21:04 | |
| *** alex_klimov has quit IRC | 21:04 | |
| *** penick has quit IRC | 21:04 | |
| *** briancurtin has joined #openstack-nova | 21:05 | |
| *** rfolco has quit IRC | 21:05 | |
| *** dpaterson has quit IRC | 21:08 | |
| *** oro_ has quit IRC | 21:08 | |
| *** krtaylor has quit IRC | 21:08 | |
| *** ociuhandu has quit IRC | 21:09 | |
| *** Sukhdev has quit IRC | 21:10 | |
| *** tellesnobrega_ has quit IRC | 21:11 | |
| *** artom has quit IRC | 21:12 | |
| *** iamjarvo has joined #openstack-nova | 21:12 | |
| *** iamjarvo has quit IRC | 21:12 | |
| *** aysyanne has quit IRC | 21:12 | |
| *** destiny has joined #openstack-nova | 21:12 | |
| *** iamjarvo has joined #openstack-nova | 21:13 | |
| *** Mike_D_laptop1 has quit IRC | 21:15 | |
| *** ijw has joined #openstack-nova | 21:16 | |
| *** wanghao has joined #openstack-nova | 21:17 | |
| *** VW has quit IRC | 21:18 | |
| *** ijw has quit IRC | 21:21 | |
| *** dave-mccowan has quit IRC | 21:21 | |
| *** ijw has joined #openstack-nova | 21:22 | |
| vilobhmm | superdan : reviewed https://review.openstack.org/#/c/187718/ ; https://review.openstack.org/#/c/187719/ ; https://review.openstack.org/#/c/187720/ ; https://review.openstack.org/#/c/188109/ ; https://review.openstack.org/#/c/188110/ | 21:23 |
| *** promulo__ has quit IRC | 21:23 | |
| superdan | vilobhmm: cool, thanks! | 21:23 |
| vilobhmm | np | 21:23 |
| ndipanoff | superdan, so I responded on this https://review.openstack.org/#/c/180637/ | 21:24 |
| ndipanoff | see if it makes any sense | 21:25 |
| ndipanoff | and thanks for looking | 21:25 |
| *** ijw has quit IRC | 21:27 | |
| *** VW has joined #openstack-nova | 21:29 | |
| *** tonytan4ever has joined #openstack-nova | 21:30 | |
| *** claudiub has joined #openstack-nova | 21:30 | |
| *** agireud has quit IRC | 21:30 | |
| *** iamjarvo has quit IRC | 21:31 | |
| *** cbader has quit IRC | 21:32 | |
| *** figleaf is now known as edleafe | 21:34 | |
| *** tonytan4ever has quit IRC | 21:34 | |
| *** zzzeek has quit IRC | 21:37 | |
| *** iamjarvo has joined #openstack-nova | 21:38 | |
| *** Marga_ has joined #openstack-nova | 21:39 | |
| *** Marga_ has quit IRC | 21:39 | |
| *** Marga_ has joined #openstack-nova | 21:40 | |
| *** openstack has joined #openstack-nova | 21:42 | |
| *** rex_lee has joined #openstack-nova | 21:43 | |
| *** amotoki has joined #openstack-nova | 21:44 | |
| *** vladikr has joined #openstack-nova | 21:52 | |
| *** krtaylor has joined #openstack-nova | 21:54 | |
| *** alanf-mc has quit IRC | 22:00 | |
| *** burt has quit IRC | 22:00 | |
| *** marzif_ has quit IRC | 22:01 | |
| *** angdraug has quit IRC | 22:01 | |
| *** matrohon has quit IRC | 22:02 | |
| *** Kennan2 has joined #openstack-nova | 22:02 | |
| *** Kennan has quit IRC | 22:03 | |
| *** thedodd has quit IRC | 22:07 | |
| *** amotoki has quit IRC | 22:11 | |
| *** markvoelker has joined #openstack-nova | 22:12 | |
| *** VW has quit IRC | 22:12 | |
| *** VW has joined #openstack-nova | 22:13 | |
| *** tjones1 has quit IRC | 22:13 | |
| *** tjones1 has joined #openstack-nova | 22:14 | |
| *** angdraug has joined #openstack-nova | 22:16 | |
| *** markvoelker has quit IRC | 22:17 | |
| *** annashen has quit IRC | 22:19 | |
| *** ijw has joined #openstack-nova | 22:22 | |
| *** tjones2 has joined #openstack-nova | 22:23 | |
| *** tjones1 has quit IRC | 22:24 | |
| *** eantyshev has quit IRC | 22:24 | |
| *** otter768 has joined #openstack-nova | 22:25 | |
| *** VW has quit IRC | 22:25 | |
| *** marun has quit IRC | 22:26 | |
| *** ijw has quit IRC | 22:27 | |
| *** xyang1 has quit IRC | 22:29 | |
| *** otter768 has quit IRC | 22:30 | |
| *** alanf-mc has joined #openstack-nova | 22:30 | |
| *** ZZelle has quit IRC | 22:33 | |
| *** gabriel-bezerra has quit IRC | 22:35 | |
| *** tjones2 has quit IRC | 22:36 | |
| *** openstackgerrit has quit IRC | 22:37 | |
| *** iamjarvo has quit IRC | 22:37 | |
| *** openstackgerrit has joined #openstack-nova | 22:37 | |
| *** gabriel-bezerra has joined #openstack-nova | 22:40 | |
| *** ozialien has joined #openstack-nova | 22:49 | |
| *** tjones1 has joined #openstack-nova | 22:53 | |
| *** Marga_ has quit IRC | 22:53 | |
| *** bradjones is now known as bradjones_away | 23:00 | |
| *** cfriesen has joined #openstack-nova | 23:01 | |
| *** ozialien has quit IRC | 23:02 | |
| *** ozialien has joined #openstack-nova | 23:05 | |
| *** angdraug has quit IRC | 23:06 | |
| *** angdraug has joined #openstack-nova | 23:07 | |
| mingy | question: is it preferred to run tests with tox or with run_tests.sh? what are their differences? | 23:08 |
| *** davideagnello has quit IRC | 23:08 | |
| *** hemna is now known as hemnafk | 23:09 | |
| *** nihilifer has quit IRC | 23:11 | |
| *** nihilifer has joined #openstack-nova | 23:12 | |
| *** davideagnello has joined #openstack-nova | 23:14 | |
| *** davideagnello has quit IRC | 23:18 | |
| *** annegentle has quit IRC | 23:18 | |
| *** davideagnello has joined #openstack-nova | 23:18 | |
| *** annashen has joined #openstack-nova | 23:20 | |
| *** bradjones_away is now known as bradjones | 23:20 | |
| *** ijw has joined #openstack-nova | 23:22 | |
| *** destiny has quit IRC | 23:25 | |
| *** sdake has joined #openstack-nova | 23:26 | |
| *** annashen has quit IRC | 23:26 | |
| *** ijw has quit IRC | 23:27 | |
| *** ian_brown has quit IRC | 23:33 | |
| *** ozialien has quit IRC | 23:33 | |
| *** mingy has quit IRC | 23:36 | |
| *** ian_brown has joined #openstack-nova | 23:37 | |
| *** iamjarvo has joined #openstack-nova | 23:38 | |
| *** cfriesen has quit IRC | 23:39 | |
| *** jwcroppe has joined #openstack-nova | 23:41 | |
| *** bradjones has quit IRC | 23:42 | |
| *** bradjones has joined #openstack-nova | 23:42 | |
| *** ian_brown has quit IRC | 23:42 | |
| *** bradjones is now known as bradjones_away | 23:47 | |
| *** ian_brown has joined #openstack-nova | 23:47 | |
| *** tjones1 has quit IRC | 23:50 | |
| *** dboik_ has quit IRC | 23:50 | |
| *** bradjones_away is now known as bradjones|away | 23:52 | |
| *** amotoki has joined #openstack-nova | 23:53 | |
| *** BadCub is now known as BadCub-Hiding | 23:55 | |
| *** sdake_ has joined #openstack-nova | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!