*** stvnoyes has quit IRC | 00:51 | |
*** stvnoyes has joined #openstack-meeting-cp | 00:51 | |
*** gouthamr has joined #openstack-meeting-cp | 01:14 | |
*** gouthamr has quit IRC | 01:41 | |
*** mars has joined #openstack-meeting-cp | 01:59 | |
*** mars has quit IRC | 01:59 | |
*** stevemar has quit IRC | 02:21 | |
*** stevemar has joined #openstack-meeting-cp | 02:21 | |
*** bswartz has quit IRC | 04:03 | |
*** gouthamr has joined #openstack-meeting-cp | 04:23 | |
*** bswartz has joined #openstack-meeting-cp | 04:31 | |
*** bswartz has quit IRC | 04:32 | |
*** bswartz has joined #openstack-meeting-cp | 04:36 | |
*** bswartz has quit IRC | 04:44 | |
*** bswartz has joined #openstack-meeting-cp | 04:54 | |
*** greghaynes has quit IRC | 06:52 | |
*** greghaynes has joined #openstack-meeting-cp | 06:59 | |
*** greghaynes has quit IRC | 07:04 | |
*** mars has joined #openstack-meeting-cp | 07:08 | |
*** greghaynes has joined #openstack-meeting-cp | 07:15 | |
*** edtubill has joined #openstack-meeting-cp | 07:18 | |
*** edtubill has quit IRC | 07:20 | |
*** hogepodge_ has joined #openstack-meeting-cp | 07:24 | |
*** gouthamr has quit IRC | 07:32 | |
*** hogepodge_ has quit IRC | 07:37 | |
*** mars has quit IRC | 09:05 | |
*** mars has joined #openstack-meeting-cp | 09:05 | |
*** quick has joined #openstack-meeting-cp | 10:14 | |
*** quick has quit IRC | 10:50 | |
*** quick has joined #openstack-meeting-cp | 11:18 | |
*** sdague has joined #openstack-meeting-cp | 12:22 | |
*** mars has quit IRC | 12:58 | |
*** xyang1 has joined #openstack-meeting-cp | 13:44 | |
*** edtubill has joined #openstack-meeting-cp | 14:22 | |
*** diablo_rojo_phon has joined #openstack-meeting-cp | 15:15 | |
*** diablo_rojo has joined #openstack-meeting-cp | 15:37 | |
*** gouthamr has joined #openstack-meeting-cp | 15:47 | |
*** ducttape_ has joined #openstack-meeting-cp | 15:52 | |
*** ducttape_ has quit IRC | 15:55 | |
*** jkomg has joined #openstack-meeting-cp | 16:05 | |
*** ducttape_ has joined #openstack-meeting-cp | 16:34 | |
*** mriedem has joined #openstack-meeting-cp | 16:46 | |
*** ducttape_ has quit IRC | 16:58 | |
ildikov | #startmeeting cinder-nova-api-changes | 17:00 |
---|---|---|
openstack | Meeting started Mon Jan 16 17:00:15 2017 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 DuncanT ameade cFouts johnthetubaguy jaypipes takashin alaski e0ne jgriffith tbarron andrearosa hemna erlon mriedem gouthamr ebalduf patrickeast smcginnis diablo_rojo gsilvis xyang1 raj_singh lyarwood | 17:00 |
mriedem | o/ | 17:00 |
smcginnis | o/ | 17:01 |
ildikov | hi :) | 17:01 |
ildikov | let's wait a bit hoping we will have more people around | 17:02 |
ildikov | ok, lets dive in | 17:03 |
ildikov | we have a few patches up for review | 17:03 |
ildikov | in the server side spec there's a comment from patrickeast regarding terminate_connection and ordering | 17:03 |
ildikov | #link https://review.openstack.org/#/c/414632/11/cinder/volume/manager.py | 17:03 |
ildikov | line 4542 | 17:04 |
ildikov | mriedem: did we agree on anything on the Nova side about what the ordering of things is when we detach? | 17:04 |
mriedem | umm | 17:04 |
mriedem | as in today or in the future? | 17:04 |
mriedem | i thought we had like one detach call now? | 17:04 |
mriedem | in the future | 17:04 |
ildikov | not in yet and there is one attachment-delete call | 17:05 |
mriedem | from what i remember of johnthetubaguy's spec, we detach with cinder api and then get something back, and pass that to os-brick's disconnect_volume | 17:05 |
ildikov | although my understanding is that Nova will still call out to os-brick for instance | 17:05 |
mriedem | and the issue was if disconnect on the host fails | 17:05 |
hemna | mornin | 17:06 |
ildikov | hemna: morning :) | 17:06 |
mriedem | since we don't have like a terminate_connection - disconnect - os-detach (ack) flow anymore | 17:06 |
smcginnis | jgriffith: Are you around? | 17:06 |
* diablo_rojo_phon sneaks in the back of the room | 17:07 | |
ildikov | mriedem: ok, I need to re-read that part | 17:09 |
* jgriffith shows up late | 17:09 | |
ildikov | jgriffith: we're talking about patrickeast's comment and detach/disconnect in general | 17:09 |
ildikov | jgriffith: I mean the one in line 4542 here: https://review.openstack.org/#/c/414632/11/cinder/volume/manager.py | 17:10 |
jgriffith | looking | 17:11 |
jgriffith | oh, yeah; so I can change that | 17:11 |
hemna | yah patrick is correct | 17:11 |
jgriffith | but I have a fundamental problem with the FC stuff at this ponit | 17:12 |
jgriffith | point | 17:12 |
smcginnis | jgriffith: Haven't you always? ;) | 17:12 |
jgriffith | Patrick may be correct, but the FZM stuff is obviously completely f'd up | 17:12 |
hemna | the FCZM code will remove the zone if the target map is passed back in terminate_connection | 17:12 |
jgriffith | it uses it's own interpretation of things and ignores every other driver in the code base | 17:12 |
jgriffith | it's frankly super annoying | 17:13 |
jgriffith | but anyway | 17:13 |
hemna | fczm code doesn't care about non FC drivers, and only looks for the FC related information | 17:13 |
jgriffith | hemna yes, but an interface is an interface! | 17:13 |
hemna | we are changing the terminate_connection behavior, so we need to adjust things accordingly | 17:13 |
jgriffith | You can't just choose to inherit from a base class and change it however you feel in this case | 17:13 |
jgriffith | hemna I'm not aruging this here, I'll work on addressign the FZM stuff | 17:14 |
hemna | some FC drivers don't inherit from the FC base class | 17:14 |
jgriffith | nuf said | 17:14 |
hemna | I've tried many times to get folks to adopt it | 17:14 |
hemna | but there are several FC drivers that are combined with iSCSI drivers | 17:14 |
hemna | instead of having 2 drivers iSCSI and FC, they combine them both into 1 driver | 17:15 |
hemna | :( | 17:15 |
hemna | so there is that too | 17:15 |
jgriffith | hemna yeah and frankly that's a good thing, again I don't want to air this out in this meeting | 17:15 |
jgriffith | not the appropriate time or place | 17:15 |
hemna | ok, just sharing information. | 17:15 |
hemna | since it's relevant | 17:15 |
ildikov | jgriffith: is the plan for attachment_delete to return a True/False to indicate whether there's a shared connection or not? | 17:17 |
jgriffith | ildikov that *was* the plan yes | 17:17 |
jgriffith | ildikov that had been the discussed proposal for a while now, but that's not going to work with the FZM stuff | 17:18 |
jgriffith | at least, not the way it is currently | 17:18 |
ildikov | do we have alternatives? | 17:18 |
jgriffith | not really | 17:18 |
jgriffith | I mean, no real alternatives other than fixing the mess | 17:19 |
jgriffith | and making it work for both. Ideally making things consistent again | 17:19 |
hemna | wait | 17:19 |
hemna | so | 17:19 |
ildikov | so the current plan is to fix the mess and stick to the above plan? | 17:19 |
hemna | the decorator is on terminate_connection in the driver | 17:19 |
hemna | that code has already been executed by the time _connection_terminate returns | 17:20 |
patrickeast | jgriffith: I kinda like the idea of either changing the fczm code to be smarter with the wrapper function return value or a new driver API that can just like list shared connections or something | 17:20 |
hemna | the drivers are only supposed to return the target map if there are no more connections from the backend to the nova host | 17:20 |
hemna | so I'm not sure anything is wrong here | 17:21 |
hemna | the zoning has already been done by the time that return call is made | 17:21 |
patrickeast | hemna: yep, the fczm can just return a book correctly based on the map it got | 17:21 |
hemna | and drivers are already tracking if there are connections left or not. | 17:21 |
patrickeast | Bool* | 17:21 |
jgriffith | patrickeast I liked "book" better :) | 17:22 |
patrickeast | Lol | 17:22 |
hemna | the zoning code just returns what it got from the driver | 17:22 |
hemna | so I don't see a problem here | 17:22 |
hemna | I don't see a conflict | 17:24 |
hemna | or I'm just slow | 17:24 |
patrickeast | It's backwards is all, the fczm needs to return true when it didn't get a request to tear down zones since the volume connection is still in use on that initiator.... I think | 17:27 |
patrickeast | So like empty map means return true, map with stuff to clean means return false | 17:27 |
patrickeast | Assuming my sleepy logic checks out | 17:28 |
jgriffith | patrickeast yeah, so the "absence" of anything currently means "nothing shared, nothing to do" | 17:28 |
patrickeast | Yea | 17:28 |
jgriffith | patrickeast from the new codes perspective that is | 17:29 |
hemna | the fczm just returns what the driver does | 17:29 |
hemna | so, the driver can simply return it, because it already knows there are no more connections or not. | 17:29 |
hemna | the FCZM's job is to add or remove zones only. | 17:29 |
jgriffith | hemna but you have a return in your decorator as well | 17:29 |
hemna | which is a pass through from the driver | 17:30 |
patrickeast | Right, my proposal is that we make it a little more than a pass though | 17:30 |
patrickeast | So that it complies with the new api | 17:30 |
hemna | to do what? | 17:30 |
patrickeast | To return true or false | 17:31 |
patrickeast | Like the new terminate connection should | 17:31 |
jgriffith | patrickeast +1 | 17:31 |
smcginnis | So iscsi drivers will return true/false but FC drivers will return a dict? | 17:31 |
hemna | we are changing every driver to return something different now in terminate_connection ? | 17:32 |
hemna | ?! | 17:32 |
patrickeast | smcginnis: yea, they need to know that the fczm monkeys with things | 17:32 |
smcginnis | patrickeast: Not a big fan of it, but if we can get that to work, I guess I'm fine. | 17:32 |
jgriffith | hemna the idea is to have a consistent interface, which would be kinda nice IMO | 17:32 |
ildikov | jgriffith: +1 | 17:33 |
smcginnis | There's already a presumed assumption that FC driver developers know to use the decorator and what to return for the decorator. | 17:33 |
smcginnis | jgriffith: +1 for consistent interface. | 17:33 |
jgriffith | part of why all of this work and things like multi-attach have been difficult is lack of consistency in interfaces | 17:33 |
jgriffith | everybody did "their own thing" | 17:33 |
patrickeast | They could all just return maps of initiators similar to the fczm | 17:33 |
jgriffith | patrickeast and that's fine as well IMO | 17:33 |
jgriffith | patrickeast just so long as it's consistent and we know what it means | 17:34 |
smcginnis | patrickeast: I kind of like that just because then it is also consistent from the driver implementation side. | 17:34 |
patrickeast | Ehh kinda | 17:34 |
jgriffith | the point for me is I would like to get rid of all the snow-flake implementations of attach/detach | 17:34 |
jgriffith | that was the whole point of reworking these to begin with | 17:35 |
jgriffith | I don't care what you do in the driver of course, but the interfaces and what they are perceived to do should be consistent | 17:35 |
hemna | so, we could also change the way drivers interact with the fczm | 17:35 |
hemna | and have the drivers call the fczm remove code manually | 17:35 |
patrickeast | Everyone has their own idea of hosts and initiators, it might be similar for FC where the spec sorta forces it, but on iscsi and other protocols it might get more weird | 17:35 |
hemna | instead of as a decorator | 17:36 |
patrickeast | hemna: +1 | 17:36 |
hemna | that would make the driver return from terminate_connection consistent | 17:36 |
smcginnis | My preference is always return an initiator map, or have an explicit call for the FC management. | 17:36 |
jgriffith | personally I would like to see the decorator 86'd | 17:36 |
smcginnis | It is almost universal in new driver reviews that we have to point out FC drivers need a decorator and what they need to return for it | 17:36 |
* jgriffith is not a fan | 17:36 | |
patrickeast | I learn towards removing the decorator if we aren't going to modify it transparently | 17:37 |
hemna | this is one of the reasons why I wanted FC drivers to extend the FC base class | 17:37 |
xyang1 | some FC drivers don't use the zone manager | 17:37 |
hemna | then we could modify the base FC class and everyone gets the changes for free | 17:37 |
hemna | xyang1, yah and those FC drivers don't work :) | 17:37 |
hemna | in a real openstack deployment | 17:37 |
xyang1 | hemna: they work using their own zone mgmt | 17:38 |
smcginnis | Well... we actually use open zoning quite a bit since the array handles isolating the target and initiator. | 17:38 |
smcginnis | So fczm just adds complexity. | 17:38 |
jgriffith | smcginnis +1 same here | 17:38 |
hemna | ok do you want me to look at a follow up patch? | 17:38 |
jgriffith | but I'm still resisting actually supporting it anyway | 17:38 |
jgriffith | :) | 17:39 |
smcginnis | jgriffith: :) | 17:39 |
hemna | well some arrays don't have fabric management | 17:39 |
jgriffith | hemna honestly I'd prefer we have an answer and include it in the proposed patch | 17:39 |
smcginnis | hemna: Right, some I think absolutely need it. | 17:39 |
jgriffith | there have been WAY too many cases where we say "follow up" and it never happens, or doesn't work | 17:39 |
hemna | I don't think it would be hard to adjust things | 17:39 |
jgriffith | leading to more and more 1/2 in 1/2 out impls | 17:40 |
hemna | ok do what you want | 17:40 |
smcginnis | SO what are our options? All return initiator map or explicit FCZM calls in the driver? | 17:40 |
hemna | I was just offering help | 17:40 |
hemna | I'd prefer explicit fczm calls in initiialize_connection and terminate_connection. | 17:40 |
smcginnis | hemna: I kind of lean that way too. Then we can have a cleaner interface and more explicit code in the drivers. | 17:41 |
smcginnis | patrickeast: Was that what you were saying too? | 17:42 |
patrickeast | Yep | 17:42 |
patrickeast | Thats my preference if we are going to make changes to all the FC drivers | 17:43 |
smcginnis | jgriffith: Work for you? I just wonder how we're going to get all the drivers updated, but I guess we will need to either way. | 17:43 |
jgriffith | sure | 17:43 |
patrickeast | IMO just fiddling with decorator return values isn't so bad for an interim solution though, wouldn't require additional driver patches to work with the new APIs | 17:45 |
patrickeast | But if the concern is that it's too confusing I'm OK with not doing it | 17:46 |
ildikov | I just wanted to ask what's the roadmap for this what you all seem to agree on? | 17:46 |
ildikov | just to clarify this for less knowledgable people on the driver side like me :) | 17:47 |
patrickeast | To get all of them updated? | 17:47 |
ildikov | to get to the next step with the new attach/detach API | 17:47 |
ildikov | if everything needs to be updated before anything happens with that than yes | 17:48 |
ildikov | sorry for killing the joy with administrational questions :/ | 17:49 |
patrickeast | Lol | 17:49 |
jgriffith | patrickeast I'm fine with fiddling with the decorator to make everything consistent | 17:49 |
jgriffith | patrickeast and if later we want to dump the decorator and force drivers to impl the method that's fine | 17:50 |
jgriffith | patrickeast my only requirement is consistency | 17:50 |
jgriffith | don't care how that's done right now | 17:50 |
patrickeast | jgriffith: sounds good to me | 17:50 |
jgriffith | I was just confused on what if any initiators *needed* that return info | 17:51 |
jgriffith | if they don't need it anyway, was confused on why it was there :) | 17:51 |
mriedem | so any updates on getting nova to use cinder v3 in the last <10 min? | 17:51 |
ildikov | mriedem: did you see the latest patch scottda uploaded last week? | 17:51 |
ildikov | mriedem: that solves the endpoint not found error | 17:52 |
ildikov | mriedem: although I'm not sure what's the preferred way of setting the api micorversion for the cinderclient for instance | 17:53 |
mriedem | ildikov: do you mean this? https://review.openstack.org/#/c/420201/ | 17:53 |
ildikov | mriedem: yeap, sorry didn't have the link handy | 17:54 |
mriedem | re: the microversion to use, i commented on that here https://review.openstack.org/#/c/385682/3/nova/volume/cinder.py@100 | 17:54 |
mriedem | nova definitely doesn't want to use the latest available | 17:54 |
mriedem | nova needs to say, i want 3.x, is that available? | 17:54 |
mriedem | if nova needs 3.24 and it's not available, then we can't do whatever we needed 3.24 for | 17:54 |
mriedem | the client has to opt into whatever version it's going to use so it can handle backward compat issues | 17:55 |
ildikov | mriedem: did you see the comment about the config option here: https://review.openstack.org/#/c/420201/2/nova/volume/cinder.py ? | 17:55 |
ildikov | line 98 | 17:55 |
mriedem | we don't want a config option | 17:56 |
ildikov | that was my thought too | 17:57 |
ildikov | but wasn't sure how it is supposed to work in general | 17:57 |
mriedem | nova is going to hardcode the microversion it needs | 17:57 |
mriedem | if cinder can provide it, we use it, else we fallback to the old flow | 17:57 |
ildikov | ok, that sounds good to me | 17:58 |
*** ducttape_ has joined #openstack-meeting-cp | 17:59 | |
mriedem | btw http://logs.openstack.org/01/420201/2/check/gate-tempest-dsvm-neutron-full-ubuntu-xenial/e66fd9a/logs/screen-n-api.txt.gz?level=TRACE | 17:59 |
mriedem | so i guess he's around the endpoint issue which is good | 17:59 |
ildikov | I think the get_version patch is not merged yet in the client | 18:00 |
mriedem | we're at time btw | 18:00 |
ildikov | mriedem: https://review.openstack.org/#/c/420119/ | 18:00 |
ildikov | mriedem: yep, we are, tnx | 18:01 |
ildikov | anything urgent for this meeting? | 18:01 |
ildikov | ok thanks everyone, let's catch up on the Cinder channel if there's anything | 18:02 |
smcginnis | ildikov: Thanks. | 18:02 |
ildikov | otherwise chat next week! | 18:02 |
ildikov | #endmeeting | 18:02 |
*** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings" | 18:02 | |
openstack | Meeting ended Mon Jan 16 18:02:33 2017 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 18:02 |
openstack | Minutes: http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2017/cinder_nova_api_changes.2017-01-16-17.00.html | 18:02 |
openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2017/cinder_nova_api_changes.2017-01-16-17.00.txt | 18:02 |
openstack | Log: http://eavesdrop.openstack.org/meetings/cinder_nova_api_changes/2017/cinder_nova_api_changes.2017-01-16-17.00.log.html | 18:02 |
*** ducttape_ has quit IRC | 18:03 | |
*** mriedem has left #openstack-meeting-cp | 18:13 | |
*** jkomg has quit IRC | 18:19 | |
*** jkomg has joined #openstack-meeting-cp | 18:20 | |
*** quick has quit IRC | 18:48 | |
*** ducttape_ has joined #openstack-meeting-cp | 19:15 | |
*** ducttape_ has quit IRC | 19:19 | |
*** ducttape_ has joined #openstack-meeting-cp | 19:25 | |
*** ducttape_ has quit IRC | 19:30 | |
*** diablo_rojo has quit IRC | 19:41 | |
*** diablo_rojo has joined #openstack-meeting-cp | 20:09 | |
*** jaugustine has joined #openstack-meeting-cp | 20:12 | |
*** ducttape_ has joined #openstack-meeting-cp | 20:45 | |
*** stvnoyes has quit IRC | 20:49 | |
*** ducttape_ has quit IRC | 20:49 | |
*** jkomg has quit IRC | 21:21 | |
*** jkomg has joined #openstack-meeting-cp | 21:21 | |
*** jkomg has quit IRC | 21:26 | |
*** jkomg has joined #openstack-meeting-cp | 21:31 | |
*** jkomg has quit IRC | 21:35 | |
*** ducttape_ has joined #openstack-meeting-cp | 21:35 | |
*** ducttape_ has quit IRC | 21:43 | |
*** ducttape_ has joined #openstack-meeting-cp | 21:44 | |
*** ducttape_ has quit IRC | 21:44 | |
*** diablo_rojo has quit IRC | 21:44 | |
*** diablo_rojo_phon has quit IRC | 22:00 | |
*** diablo_rojo has joined #openstack-meeting-cp | 22:18 | |
*** ducttape_ has joined #openstack-meeting-cp | 22:45 | |
*** ducttape_ has quit IRC | 22:49 | |
*** diablo_rojo has quit IRC | 23:02 | |
*** jaugustine has quit IRC | 23:05 | |
*** edtubill has quit IRC | 23:09 | |
*** jkomg has joined #openstack-meeting-cp | 23:10 | |
*** xyang1 has quit IRC | 23:34 | |
*** ducttape_ has joined #openstack-meeting-cp | 23:35 | |
*** knikolla has quit IRC | 23:47 | |
*** knikolla has joined #openstack-meeting-cp | 23:47 | |
*** knikolla has quit IRC | 23:47 | |
*** knikolla has joined #openstack-meeting-cp | 23:48 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!