| *** markvoelker has joined #openstack-meeting-cp | 00:24 | |
| *** markvoelker_ has joined #openstack-meeting-cp | 00:25 | |
| *** markvoelker has quit IRC | 00:29 | |
| *** markvoelker_ has quit IRC | 00:30 | |
| *** notstevemar is now known as stevemar | 01:57 | |
| *** Niham has joined #openstack-meeting-cp | 04:52 | |
| *** sheel has joined #openstack-meeting-cp | 05:21 | |
| *** sigmavirus24_awa has quit IRC | 05:27 | |
| *** belmoreira has joined #openstack-meeting-cp | 06:43 | |
| *** ttx has quit IRC | 06:56 | |
| *** ttx has joined #openstack-meeting-cp | 06:57 | |
| *** bauzas has joined #openstack-meeting-cp | 07:02 | |
| *** sdake_ has quit IRC | 08:10 | |
| *** sdake has joined #openstack-meeting-cp | 08:11 | |
| *** vgridnev has quit IRC | 09:14 | |
| *** vgridnev has joined #openstack-meeting-cp | 09:19 | |
| *** Niham has quit IRC | 09:26 | |
| *** _amrith_ is now known as amrith | 10:04 | |
| *** amrith is now known as _amrith_ | 10:33 | |
| *** Niham has joined #openstack-meeting-cp | 10:40 | |
| *** sdague has joined #openstack-meeting-cp | 10:46 | |
| *** pgreg has joined #openstack-meeting-cp | 10:54 | |
| *** sdake has quit IRC | 11:04 | |
| *** Niham has quit IRC | 11:04 | |
| *** _amrith_ is now known as amrith | 11:14 | |
| *** pgreg has quit IRC | 12:49 | |
| *** sigmavirus24 has joined #openstack-meeting-cp | 13:12 | |
| *** piet has joined #openstack-meeting-cp | 13:23 | |
| *** xyang1 has joined #openstack-meeting-cp | 13:30 | |
| *** belmoreira has quit IRC | 13:32 | |
| *** sheel has quit IRC | 13:35 | |
| *** sdake has joined #openstack-meeting-cp | 13:56 | |
| *** sdake_ has joined #openstack-meeting-cp | 14:00 | |
| *** sdake has quit IRC | 14:02 | |
| *** belmoreira has joined #openstack-meeting-cp | 14:17 | |
| *** sheel has joined #openstack-meeting-cp | 14:20 | |
| *** superdan is now known as dansmith | 14:32 | |
| *** sdake_ has quit IRC | 14:38 | |
| *** markvoelker has joined #openstack-meeting-cp | 14:56 | |
| *** sdake has joined #openstack-meeting-cp | 14:59 | |
| *** amrith is now known as _amrith_ | 15:15 | |
| *** piet has quit IRC | 15:20 | |
| *** belmoreira has quit IRC | 15:39 | |
| *** xyang1_ has joined #openstack-meeting-cp | 16:12 | |
| *** xyang1 has quit IRC | 16:13 | |
| *** xyang1_ is now known as xyang1 | 16:13 | |
| *** xyang1 has quit IRC | 16:17 | |
| *** xyang1 has joined #openstack-meeting-cp | 16:17 | |
| *** geguileo has joined #openstack-meeting-cp | 16:50 | |
| ildikov | #startmeeting cinder-nova-api-changes | 17:00 |
|---|---|---|
| openstack | Meeting started Mon May 23 17:00:08 2016 UTC and is due to finish in 60 minutes. The chair is ildikov. Information about MeetBot at http://wiki.debian.org/MeetBot. | 17:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 17:00 |
| *** openstack changes topic to " (Meeting topic: cinder-nova-api-changes)" | 17:00 | |
| openstack | The meeting name has been set to 'cinder_nova_api_changes' | 17:00 |
| ildikov | scottda ildikov DuncanT ameade cFouts johnthetubaguy jaypipes takashin alaski e0ne jgriffith tbarron andrearosa hemna erlon mriedem gouthamr ebalduf patrickeast smcginnis diablo_rojo gsilvis | 17:00 |
| scottda | Hi | 17:00 |
| scottda | ildikov: Just pinged you in Cinder :) | 17:00 |
| ildikov | scottda: :) | 17:00 |
| geguileo | Hi | 17:01 |
| patrickeast | hi | 17:01 |
| ildikov | I started to answer there, but then I switch to this channel | 17:01 |
| hemna | hey | 17:01 |
| hemna | is this our meeting slot now? | 17:02 |
| hemna | I can't keep it straight | 17:02 |
| ildikov | hemna: yes :) | 17:02 |
| hemna | seems to change every week | 17:02 |
| ildikov | hemna: sorry for the many variations | 17:02 |
| hemna | nah it's cool. | 17:02 |
| hemna | :) | 17:02 |
| *** piet has joined #openstack-meeting-cp | 17:02 | |
| ildikov | hemna: last week's slot was meant to be the one, but then it turned out we don't have a channel for that slot anymore... | 17:02 |
| hemna | coolio | 17:03 |
| ildikov | hemna: and next week Monday is a US holiday, so we need to find another slot or sync up separately... | 17:03 |
| hemna | so do we have an agenda today? | 17:03 |
| ildikov | I hope we have jgriffith around | 17:03 |
| ildikov | to discuss his part as I haven't seen patches from him but I might missed it | 17:03 |
| hemna | man that'd be great to see that patch(s) | 17:04 |
| ildikov | hemna: I also checked check_attach a bit last week | 17:04 |
| ildikov | and I found that in case of attaching a volume at boot time Nova does not call reserve_volume | 17:04 |
| hemna | w t f | 17:05 |
| hemna | seriously? | 17:05 |
| patrickeast | lol | 17:05 |
| hemna | ugh, lets follow that up | 17:05 |
| hemna | that needs to get fixed asap | 17:05 |
| ildikov | I raised an exception from the reserve call and attach at boot time went through and then detach and the attach after that failed there | 17:05 |
| ildikov | I mean I had this dumb way of testing, but anyway, the point is the same | 17:06 |
| hemna | yah | 17:06 |
| ildikov | I think if we can have that fixed than check_attach can be removed from there too | 17:06 |
| ildikov | s/than/then/ | 17:06 |
| ildikov | maybe where they call check_attach first we can call reserve volume there | 17:07 |
| ildikov | let's follow that up after the meeting | 17:07 |
| hemna | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1288 | 17:07 |
| hemna | that's the one I thought should be changed to a reserve_volume call | 17:08 |
| hemna | I haven't tried it yet though | 17:08 |
| ildikov | yeah, I meant this one | 17:08 |
| ildikov | I think this method is called only in that flow | 17:08 |
| hemna | in the boot from volume flow ? | 17:08 |
| ildikov | yeap | 17:08 |
| ildikov | otherwise they just call attach and that creates the BDM | 17:09 |
| hemna | ok, so do you think I should try the same thing I did here: https://review.openstack.org/#/c/315789/ ? | 17:09 |
| ildikov | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L3101 | 17:09 |
| ildikov | that and also call reserve_volume | 17:10 |
| hemna | ildikov, so for that one, my patch changes it already | 17:10 |
| ildikov | as otherwise no one will | 17:10 |
| hemna | https://review.openstack.org/#/c/315789/2/nova/compute/api.py | 17:10 |
| ildikov | right, I meant that for other attach cases this method is called here, but for boot from volume they use the _validate_bdm, etc | 17:11 |
| hemna | *smh* | 17:11 |
| hemna | ok, so at some point in the boot from volume workflow there has to be a call to initialize_connection and attach | 17:12 |
| ildikov | regarding the reserve call, I would guess we need to check that what if anything goes wrong, I did not analyse the full call sequence... | 17:12 |
| hemna | I'll see if I can track those down | 17:12 |
| ildikov | I didn't look for initialize_connection, next time I will be smarter | 17:13 |
| hemna | I can't comprehend why nova has copy/paste code for doing the same operation. | 17:13 |
| hemna | mind boggling | 17:13 |
| ildikov | do you have any other case I should check? | 17:13 |
| ildikov | agreed, good that we're doing some cleanup | 17:13 |
| ildikov | so normal attach has reserve, live migration is a special case anyway | 17:14 |
| *** sdake has quit IRC | 17:15 | |
| hemna | http://paste.openstack.org/show/498223/ | 17:15 |
| hemna | that's the entire list of check_attach references | 17:15 |
| hemna | fwiw | 17:15 |
| hemna | lots of tests referencing it | 17:15 |
| hemna | anyway, I'll try testing my patch against he boot from volme and modify the check_attach case there to call check_availability_zone and reserve_volume instead | 17:16 |
| ildikov | there is a fake check_attach code as well, which is the copy of the normal basically | 17:16 |
| hemna | and see how that goes | 17:16 |
| ildikov | that code is a nightmare | 17:16 |
| johnthetubaguy | +1 to the above comments, that seems crazy | 17:17 |
| hemna | check_attach should go away IMHO | 17:17 |
| johnthetubaguy | where is the cut and paste again? | 17:17 |
| ildikov | I meant line 33 for my previous comment | 17:17 |
| * johnthetubaguy reads more of scrollback | 17:17 | |
| *** sdake has joined #openstack-meeting-cp | 17:17 | |
| hemna | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1287-L1291 | 17:17 |
| hemna | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L3097-L3099 | 17:18 |
| hemna | those should be the same thing IMHO | 17:18 |
| hemna | but should be changed to.... | 17:18 |
| ildikov | johnthetubaguy: the base of this discussion is that the BFV case with existing Cinder volume does not call reserve_volume | 17:18 |
| hemna | https://review.openstack.org/#/c/315789/2/nova/compute/api.py | 17:18 |
| hemna | that | 17:18 |
| johnthetubaguy | ah... | 17:18 |
| ildikov | so we thought to add that missing call and also remove check_attach from there as well | 17:20 |
| johnthetubaguy | a new method in volume_api to do those calls? | 17:20 |
| hemna | possibly yah | 17:20 |
| hemna | I had thought of refactoring those 2 into a single place | 17:20 |
| hemna | if all goes well :) | 17:20 |
| johnthetubaguy | yeah, that makes sense | 17:21 |
| johnthetubaguy | seems like reserve and check should always both we done? | 17:21 |
| hemna | possibly | 17:21 |
| hemna | live migration is a special case though | 17:21 |
| hemna | :( | 17:21 |
| johnthetubaguy | oh, true | 17:22 |
| *** sdake has quit IRC | 17:22 | |
| johnthetubaguy | its always a special case | 17:22 |
| johnthetubaguy | thats why its so dammed broken | 17:22 |
| hemna | yup | 17:22 |
| hemna | pain | 17:22 |
| hemna | ok so baby steps | 17:22 |
| ildikov | but in general the less state check in Nova is the better | 17:22 |
| hemna | ildikov, +1 | 17:22 |
| ildikov | and use reserve what it's for | 17:23 |
| johnthetubaguy | yeah, +1 | 17:23 |
| ildikov | do we have any other special case like live migration? | 17:23 |
| hemna | I'm shocked that BFV doesn't use reserve | 17:23 |
| ildikov | or it's the only, but big monster? | 17:23 |
| hemna | I'd like to confirm that | 17:23 |
| ildikov | I haven't checked how it is done when Nova creates the volume | 17:23 |
| johnthetubaguy | so I wonder if we should get you lovely cinder folks to write a very quick doc on how we *should* use your API? we could put that in our devref thingy? | 17:23 |
| scottda | ildikov: I think we need to look at shelve_offload, as cinder state if fudged there as well. | 17:23 |
| hemna | heh | 17:24 |
| johnthetubaguy | ildikov: so migrate, and shelved, evacuate, resize | 17:24 |
| ildikov | just add an exception to the client call and it will get through or I'm a magician... | 17:24 |
| scottda | johnthetubaguy: That's a good idea. | 17:24 |
| hemna | I think jgriffith created a nice blog about it and was working on a devref for it as well | 17:24 |
| johnthetubaguy | that would be prefect | 17:24 |
| *** harlowja has joined #openstack-meeting-cp | 17:24 | |
| *** itisha has joined #openstack-meeting-cp | 17:25 | |
| ildikov | johnthetubaguy: migrate is live migrate or Cinder migrate there? | 17:25 |
| johnthetubaguy | so I think its when the server moves, we have issues | 17:25 |
| johnthetubaguy | let me find a doc on those, one second... | 17:25 |
| hemna | johnthetubaguy, http://docs.openstack.org/developer/cinder/devref/attach_detach_conventions.html | 17:25 |
| hemna | :) | 17:25 |
| johnthetubaguy | http://developer.openstack.org/api-guide/compute/server_concepts.html | 17:25 |
| johnthetubaguy | hemna: ah, coolness, we might want to add something in the Nova docs, that just link to your doc :) but that might be overkill! | 17:26 |
| ildikov | hemna: nice, we could add a ref to your diagrams there as well | 17:26 |
| hemna | johnthetubaguy, I think that'd be nice. | 17:26 |
| johnthetubaguy | so the api docs cover server moves, we have very silly names | 17:26 |
| johnthetubaguy | live-migrate is what you know | 17:26 |
| johnthetubaguy | migrate is cold migrate | 17:26 |
| hemna | server move = live migration ? | 17:27 |
| johnthetubaguy | i.e. turn it off, move it, turn it on | 17:27 |
| johnthetubaguy | yeah, live-migration is move the server, but with only a slight pause and no reboot | 17:27 |
| ildikov | johnthetubaguy: yeah, right, I forgot you call it migrate as well, my bad :) | 17:27 |
| johnthetubaguy | yeah, live migrate and cold migrate is what we often call it now | 17:27 |
| johnthetubaguy | now we have dead parrot migrate (I totally just made that up) | 17:27 |
| johnthetubaguy | its called evacuate | 17:28 |
| ildikov | johnthetubaguy: how much are these things tested? | 17:28 |
| johnthetubaguy | when you host is dead, you need to move your server | 17:28 |
| johnthetubaguy | ildikov: honestly, its a WIP | 17:28 |
| johnthetubaguy | ildikov: migrate is OK, live-migrate is only tested in the experimental queue, we can't make it pass enough yet | 17:28 |
| hemna | is cold migrate = shelve/unshelve ? | 17:28 |
| johnthetubaguy | nope | 17:28 |
| hemna | ok | 17:28 |
| johnthetubaguy | lets do evacuate | 17:28 |
| johnthetubaguy | your host is head | 17:29 |
| johnthetubaguy | you can't get to it | 17:29 |
| johnthetubaguy | its in a skip | 17:29 |
| johnthetubaguy | but you want your VM back | 17:29 |
| johnthetubaguy | so you use evacuate | 17:29 |
| johnthetubaguy | (resurrect is another name we came up with, since we created that API call) | 17:29 |
| johnthetubaguy | does that make any sense? | 17:29 |
| hemna | yah | 17:29 |
| ildikov | I know the operation | 17:29 |
| hemna | thanks :) | 17:29 |
| johnthetubaguy | sweet | 17:29 |
| ildikov | but what happens with the volume there? | 17:29 |
| johnthetubaguy | it gets re-attached on the destination host | 17:30 |
| johnthetubaguy | we don't have access to the old host to do anything there | 17:30 |
| johnthetubaguy | we technically share that code with rebuild | 17:30 |
| ildikov | but then from volume perspective it's like a migrate? | 17:30 |
| scottda | so volume does not need to be "reserved", i.e. set to attaching, it should already be attached and in-use | 17:30 |
| johnthetubaguy | I am not sure if we call terminate correctly | 17:30 |
| johnthetubaguy | ildikov: kinda, yeah | 17:30 |
| johnthetubaguy | scottda: yeah, thats the key bit I guess | 17:31 |
| johnthetubaguy | the detach is harder, as we can't call os-brick on the old host, its dead | 17:31 |
| hemna | so nova will call initialize_connection on a new host then, when the volume is already 'in-use' | 17:31 |
| scottda | and we don't want Nova to check the volume state, because it won't be correct for non-multi-attach case | 17:31 |
| johnthetubaguy | hemna: I think so | 17:31 |
| hemna | ok | 17:31 |
| hemna | yet another case where the connector information should be updated | 17:31 |
| ildikov | yeah, like migrate | 17:31 |
| hemna | ildikov, yup | 17:31 |
| johnthetubaguy | yeah, its very similar | 17:32 |
| scottda | johnthetubaguy: Yes, detach won't work for Nova, we'll want to call cinder force-detach (which needs fixing after jgriffith patches) | 17:32 |
| johnthetubaguy | scottda: yeah, that sounds correct for evacuate (although wrong for rebuild/migrate) | 17:32 |
| johnthetubaguy | which might be tricky, and rebuild and evacuate are mostly common code, but lets ignore that for now | 17:32 |
| ildikov | I just wanted to ask whether we could call force-detach :) | 17:33 |
| johnthetubaguy | hehe, I think we should be able to fix that | 17:33 |
| scottda | ildikov: Force detach won't work properly until Cinder saves the connector info... | 17:33 |
| hemna | scottda, +! | 17:33 |
| scottda | Which is will soon with jgriffith patches. | 17:33 |
| hemna | +1 | 17:33 |
| johnthetubaguy | scottda: +1 | 17:33 |
| johnthetubaguy | cool beans, we should do shelved? | 17:33 |
| ildikov | scottda: right, I got it now :) | 17:33 |
| johnthetubaguy | so that migrate, with a really long pause, and possible a delete before the resume | 17:34 |
| scottda | johnthetubaguy: Yes, please proceed with shelved :) | 17:34 |
| johnthetubaguy | the idea is this: | 17:34 |
| ildikov | johnthetubaguy: please educate us :) | 17:34 |
| johnthetubaguy | user wants to keep all their resources | 17:34 |
| johnthetubaguy | but they don't want to pay for compute resources when the instance isn't used | 17:34 |
| johnthetubaguy | (i.e. over night) | 17:34 |
| johnthetubaguy | and what a really quick resume, where possible, when they do want to use it | 17:34 |
| johnthetubaguy | its a bit messy | 17:35 |
| johnthetubaguy | but the instance is basically not really on a host, once it is selved (offloaded) | 17:35 |
| johnthetubaguy | the offloaded bit is about snapshotting any local disks, so you can resume on a different host) | 17:35 |
| hemna | but the volume is still 'in-use' | 17:35 |
| johnthetubaguy | yeah, volume stays in-use | 17:35 |
| johnthetubaguy | it shouldn't be attached to any other place | 17:36 |
| scottda | so similar to evacuate, we need to skip state checking | 17:36 |
| johnthetubaguy | ... now there is a spec to detach disks from a shelved instance, which we currently don't support, but thats not approved yet, AFAIK | 17:36 |
| johnthetubaguy | scottda: +1 | 17:36 |
| scottda | johnthetubaguy: I believe that under the covers, os-brick is called to actually disconnect the disk when we shelve, does that sound right? | 17:37 |
| scottda | and a new connection occurs during un-shelve. | 17:37 |
| johnthetubaguy | yes, that does sound correct | 17:37 |
| johnthetubaguy | yeah, its assumed to be on a new host, and you want to clean up the old host | 17:37 |
| johnthetubaguy | not sure how well shelve is tested yet, I don't remember if the tempest stuff landed or not | 17:38 |
| johnthetubaguy | but thats the general intent | 17:38 |
| scottda | So, there's no messy force-detach issues, or dangling connection. Just a basic disconncect/re-connect in brick and dealing with weird state management. | 17:38 |
| johnthetubaguy | thats my understanding of it | 17:38 |
| johnthetubaguy | evacuate is the only one with the force, I think... | 17:38 |
| johnthetubaguy | I think thats the main move operations all covered | 17:39 |
| johnthetubaguy | oh, right, live-migrate is slightly different to the other migrates | 17:39 |
| smcginnis | johnthetubaguy: So for shelve, were you saying we need to detach the volume but still keep it in-use? | 17:40 |
| johnthetubaguy | smcginnis: yes | 17:40 |
| johnthetubaguy | its possible we just don't call terminate_connection thinking about it | 17:40 |
| ildikov | I guess it's only disconnected from the host | 17:40 |
| johnthetubaguy | ildikov: +1 | 17:40 |
| johnthetubaguy | ildikov: at least, thats the intent | 17:40 |
| johnthetubaguy | its logically still attached to the instance | 17:41 |
| ildikov | johnthetubaguy: otherwise the state would change and you don't want to have the volume available | 17:41 |
| johnthetubaguy | +1 | 17:41 |
| scottda | I need to move to IRC_on_phone mode, sorry folks, mostly Read-only from here on out.... | 17:42 |
| ildikov | scottda: any updtes on the migrate test? | 17:42 |
| ildikov | scottda: on the Cinder migrate ones I meant | 17:42 |
| johnthetubaguy | are there any other questions like the above ones I can help with? | 17:43 |
| scottda | ildikov: sorry, no update. | 17:44 |
| ildikov | johnthetubaguy: I wonder where we will update the host info for the shelve case | 17:44 |
| ildikov | scottda: no worries, I just wanted to check | 17:45 |
| johnthetubaguy | ildikov: so there is no host, until you do unshelve, and we call initialize_connection again | 17:45 |
| hemna | ildikov, any time initialize_connection is called the connector passed in has the host and it should be updated | 17:45 |
| johnthetubaguy | I guess for local volumes, that messes things up, but otherwise, that should be OK? | 17:46 |
| hemna | I'm hoping jgriffith's change has this in place as part of his patch(es) | 17:46 |
| ildikov | johnthetubaguy: ah, ok, I wasn't sure we have the initialize_connection call | 17:46 |
| johnthetubaguy | ildikov: I think we should have | 17:46 |
| ildikov | ok, cool, I'll check if I get there | 17:46 |
| ildikov | johnthetubaguy: can you also take a look at the multiattach spec? | 17:47 |
| ildikov | next week is the deadline if I'm not mistaken | 17:47 |
| johnthetubaguy | ildikov: for unshelve, its doing this: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L4205 | 17:48 |
| johnthetubaguy | ildikov: yeah, sorry, I must look at that again, ASAP | 17:48 |
| hemna | I think we are stuck waiting for jgriffith's patches no ? | 17:48 |
| johnthetubaguy | I have a quite a few specs in that list, I am afraid, but hope to get through more of those tomorrow | 17:48 |
| johnthetubaguy | we should be able to merge the spec before the patches, or is there something undecided we are depending on? | 17:49 |
| ildikov | I think we should be able to conclude on the spec | 17:50 |
| johnthetubaguy | cool | 17:51 |
| ildikov | at least in my opinion, but it would be nice to get a heads up if there's anything missing | 17:51 |
| johnthetubaguy | totally, I will try hit that soon | 17:51 |
| johnthetubaguy | ildikov: you totally have permission to keep bugging me every morning till I review it | 17:52 |
| ildikov | I tried to handle things like jgriffith's patches as an identified dependency, but not part of the multiattach spec technically | 17:52 |
| johnthetubaguy | ildikov: that sounds correct | 17:52 |
| johnthetubaguy | we could call the other stuff a bug fix | 17:52 |
| johnthetubaguy | ish | 17:52 |
| ildikov | johnthetubaguy: thanks :) I might use it, but hopefully I will not need :) | 17:52 |
| ildikov | johnthetubaguy: +1 | 17:53 |
| ildikov | many of the discussions today sound like a cleanup and fixing issues we have hidden | 17:53 |
| johnthetubaguy | yeah | 17:53 |
| smcginnis | ildikov: +1 | 17:53 |
| ildikov | ok, we have 7 minutes left from the today's slot | 17:54 |
| johnthetubaguy | honestly, most of this comes from neither side understand what the other is wanting to do | 17:54 |
| johnthetubaguy | so great we are cleaning this up now :) | 17:54 |
| ildikov | is there anything we should discuss? | 17:54 |
| johnthetubaguy | I was looking at the devref stuff in cinder, I think we might want a more API focused doc | 17:54 |
| ildikov | johnthetubaguy: my thinking as well :) | 17:54 |
| johnthetubaguy | that includes os-brick in the discussion, maybe | 17:54 |
| smcginnis | johnthetubaguy: I do like the suggestion of a devref in Nova for some of this. It's duplication, but hopefully it saves this from happening again once it's cleaned up. | 17:54 |
| johnthetubaguy | actually, that would work | 17:55 |
| johnthetubaguy | do the nova focused one in Nova | 17:55 |
| johnthetubaguy | i.e. here are the patterns | 17:55 |
| ildikov | smcginnis: johnthetubaguy: I think we can also add hemna's diagrams | 17:55 |
| johnthetubaguy | if we do create server, delete server, live-migrate server, evacuate and shelve, we should be there | 17:55 |
| johnthetubaguy | yeah, diagrams are good to make this clearer | 17:55 |
| ildikov | not as an API ref, but if someone would like to figure it out what's in the background it's pretty useful | 17:56 |
| johnthetubaguy | yeah | 17:56 |
| scottda | I'll put it on my list to put up a patch in Nova devref | 17:57 |
| johnthetubaguy | it goes the core reviews something to go double check when we can't remember the details next time | 17:57 |
| johnthetubaguy | s/ws/wers/ | 17:57 |
| johnthetubaguy | scottda: awesome, thank you | 17:57 |
| johnthetubaguy | nothing more from me for this week :) | 17:58 |
| ildikov | I think we have home work with these server moving actions and fix the reserve_volume in BFV | 17:58 |
| ildikov | I will try to hunt down jgriffith as soon as I can | 17:58 |
| johnthetubaguy | yeah, feel free to ping me in channel if later in the week, if thats useful | 17:58 |
| ildikov | johnthetubaguy: cool, thanks | 17:59 |
| ildikov | anything else from anyone before we close? | 17:59 |
| ildikov | then, thanks for the good chat! | 18:00 |
| johnthetubaguy | +1 | 18:00 |
| scottda | Thanks everyone | 18:00 |
| ildikov | I learnt a lot today :) | 18:00 |
| smcginnis | ildikov: Thanks again for arranging this. | 18:00 |
| ildikov | oh one thing | 18:00 |
| ildikov | next Monday is US holiday, so I will ask around when we could have a quick sync | 18:00 |
| scottda | I'm out until Wed next week | 18:00 |
| hemna | I think next week is going to be a bit of a loss for some | 18:01 |
| hemna | I might be taking time off next week too. | 18:01 |
| johnthetubaguy | actually, its a UK holiday too | 18:01 |
| ildikov | thanks for the info, Matt is out too at the beginning | 18:01 |
| johnthetubaguy | I forget about those | 18:01 |
| johnthetubaguy | by which I mean, I forget they are coming up | 18:01 |
| ildikov | ok, then let's aim for having a short chat mid next week | 18:02 |
| scottda | cool | 18:02 |
| johnthetubaguy | that works | 18:02 |
| ildikov | johnthetubaguy: we need to get the spec landed, otherwise we should be good | 18:02 |
| johnthetubaguy | yeah, +1 | 18:02 |
| smcginnis | Maybe sync up on Thurday? | 18:02 |
| johnthetubaguy | thats a good plan | 18:02 |
| ildikov | smcginnis: yeap, sounds good | 18:02 |
| scottda | yup | 18:03 |
| ildikov | I hope 1700UTC can work | 18:03 |
| johnthetubaguy | if its quick, that should be OK | 18:03 |
| ildikov | I will send out a heads up mail later this week | 18:03 |
| scottda | thanks ildikov | 18:03 |
| smcginnis | Works for me. We can hijack the cindre channel again if we need to. | 18:03 |
| ildikov | johnthetubaguy: due to the holidays I would assume we can keep it short | 18:04 |
| johnthetubaguy | +1 | 18:04 |
| johnthetubaguy | :) | 18:04 |
| ildikov | smcginnis: cool, thanks, I'll check the channels | 18:04 |
| ildikov | ok, I don't want steal more of your time today :) | 18:04 |
| ildikov | thanks guys! | 18:05 |
| ildikov | #endmeeting | 18:05 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings" | 18:05 | |
| openstack | Meeting ended Mon May 23 18:05:17 2016 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 18:05 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2016/cinder_nova_api_changes.2016-05-23-17.00.html | 18:05 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2016/cinder_nova_api_changes.2016-05-23-17.00.txt | 18:05 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2016/cinder_nova_api_changes.2016-05-23-17.00.log.html | 18:05 |
| *** geguileo has left #openstack-meeting-cp | 18:08 | |
| *** markvoelker has quit IRC | 19:14 | |
| *** markvoelker has joined #openstack-meeting-cp | 19:15 | |
| *** _amrith_ is now known as amrith | 19:31 | |
| *** piet has quit IRC | 19:46 | |
| *** amrith is now known as _amrith_ | 19:46 | |
| *** piet has joined #openstack-meeting-cp | 19:48 | |
| *** rockyg has joined #openstack-meeting-cp | 20:11 | |
| *** rockyg has quit IRC | 20:19 | |
| *** sdake has joined #openstack-meeting-cp | 20:21 | |
| *** itisha has quit IRC | 20:29 | |
| *** sdake has quit IRC | 20:37 | |
| *** _amrith_ is now known as amrith | 20:46 | |
| *** sdake has joined #openstack-meeting-cp | 20:47 | |
| *** sdake_ has joined #openstack-meeting-cp | 20:51 | |
| *** sdake has quit IRC | 20:54 | |
| *** rockyg has joined #openstack-meeting-cp | 21:00 | |
| *** amrith is now known as _amrith_ | 21:01 | |
| *** sheel has quit IRC | 21:45 | |
| *** rockyg has quit IRC | 22:07 | |
| *** rockyg has joined #openstack-meeting-cp | 22:10 | |
| *** piet has quit IRC | 22:11 | |
| *** rockyg has quit IRC | 22:12 | |
| *** sdague has quit IRC | 22:28 | |
| *** tonyb_ has joined #openstack-meeting-cp | 22:48 | |
| *** lbragstad_ has joined #openstack-meeting-cp | 22:51 | |
| *** dolphm_ has joined #openstack-meeting-cp | 22:51 | |
| *** reed_ has joined #openstack-meeting-cp | 22:51 | |
| *** odyssey4me_ has joined #openstack-meeting-cp | 22:51 | |
| *** tbarron_ has joined #openstack-meeting-cp | 22:52 | |
| *** xyang1 has quit IRC | 22:53 | |
| *** ildikov has quit IRC | 22:53 | |
| *** GheRivero has quit IRC | 22:53 | |
| *** odyssey4me has quit IRC | 22:53 | |
| *** lbragstad has quit IRC | 22:53 | |
| *** samueldmq has quit IRC | 22:53 | |
| *** tonyb has quit IRC | 22:53 | |
| *** reed has quit IRC | 22:53 | |
| *** tbarron has quit IRC | 22:53 | |
| *** dolphm has quit IRC | 22:53 | |
| *** patrickeast has quit IRC | 22:53 | |
| *** tbarron_ is now known as tbarron | 22:53 | |
| *** reed_ is now known as reed | 22:53 | |
| *** dolphm_ is now known as dolphm | 22:53 | |
| *** lbragstad_ is now known as lbragstad | 23:33 | |
| *** tonyb_ is now known as tonyb | 23:43 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!