16:03:15 <jgriffith> #startmeeting cinder
16:03:16 <openstack> Meeting started Wed Aug 28 16:03:15 2013 UTC and is due to finish in 60 minutes.  The chair is jgriffith. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:03:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:03:19 <openstack> The meeting name has been set to 'cinder'
16:03:22 <bswartz> jgriffith: every time
16:03:29 <kmartin> hello
16:03:32 <jgriffith> bswartz: makes ya wonder if I do it on purpose :)
16:03:37 <winston-d> jjacob512: good. thx for joining.
16:03:38 <bswartz> I think you do
16:03:42 * jgriffith makes a note to submit a code change to IRC
16:03:43 <zhiyan> hi
16:03:48 <jjacob512> thanks
16:04:08 <jgriffith> Ok, short agenda, but I'm sure we'll make up for it like we usually do: https://wiki.openstack.org/wiki/CinderMeetings
16:04:10 <avishay> bswartz: wait till he tries to end the meeting... ;)
16:04:19 <jgriffith> avishay: :)
16:04:28 <jgriffith> #topic blueprints
16:04:31 <bswartz> avishay: maybe we can modify the bot before the meeting ends
16:04:44 <avishay> bswartz: :)
16:05:06 <jgriffith> So I just wanted to give a heads up that for I release, I'm thinking we should go back to being pretty strict with our BP's
16:05:20 <jgriffith> That means having *real* details of what the BP is
16:05:24 <jgriffith> Including a specs page
16:05:28 <jgriffith> and a user story etc
16:05:46 <jgriffith> guess jmh_ didn't like that
16:06:08 <avishay> haha
16:06:28 <DuncanT-> ug, sorry I'm late
16:06:39 <jgriffith> DuncanT-: you dind't miss much
16:06:43 <jgriffith> That's all I had on that
16:06:57 <avishay> Sounds good
16:06:58 <jgriffith> It'll sort of fall into place when folks submit BP's
16:07:08 <bswartz> jgriffith: I can't see any counter argument, other than laziness, so +1
16:07:08 <jgriffith> avishay: you ready?
16:07:15 <jgriffith> bswartz: HA!
16:07:22 <avishay> jgriffith: for?
16:07:23 <DuncanT-> I'm a supporter, even if I'll probably be bitten by it
16:07:23 <jgriffith> bswartz: indeed, I think that is the only counter
16:07:32 <jgriffith> avishay: DOH!
16:07:36 <jgriffith> I meant winston-d
16:07:47 <winston-d> yup
16:07:48 <jgriffith> avishay: sorry
16:08:01 <jgriffith> #topic Feature submission exception
16:08:04 <avishay> we're very similar in both name and appearance, i could understand the confusion
16:08:08 <jgriffith> haha!
16:08:14 <jgriffith> and a is really close to q
16:08:17 <jgriffith> w
16:08:19 <jgriffith> geeesh
16:08:20 <avishay> :)
16:08:25 <avishay> if you loop around, yes
16:08:27 <jgriffith> I obviously can't type this morning
16:08:37 <jgriffith> avishay: look at your keyboard :)
16:08:41 <avishay> :)
16:08:45 <winston-d> so jjacob512 recently submitted a patch/driver for Dell's back-end _after_ feature freeze.
16:08:51 <jjacob512> yes
16:09:05 <jgriffith> winston-d: I've been talking with the Dell folks for a while on this
16:09:12 <winston-d> jgriffith: ok
16:09:21 <jgriffith> winston-d: I planned on granting the exception if there's no arguments from you or others
16:09:33 <winston-d> im fine with that
16:09:38 <jgriffith> cool
16:09:48 <jgriffith> Dell and VmWare were the two that talked to me last month
16:09:50 <winston-d> the driver looks pretty simple for now.
16:09:53 <avishay> it's OK
16:10:15 <avishay> the driver looks to be in decent shape, but exception handling was pretty non-existant if i remember correctly
16:10:19 <DuncanT-> Does it make the minimum feature list for the release?
16:10:25 <winston-d> if it was 5k+ LOC, i'll be hesitate
16:10:31 <kmartin> sounds good, as long as it has unit test and meets the feature list for H
16:10:38 <jgriffith> Just to be clear, this isn't a review session
16:10:46 <jgriffith> The question was just is it accepted for us to review
16:10:47 <jjacob512> yes , it does, also was able to test it on devstack
16:10:50 <avishay> if it was 5K+ lines of code i'd -3 it :)
16:11:04 * jgriffith wants a -3 button!
16:11:16 <avishay> ok, so we let it through, ok by me
16:11:29 <jgriffith> avishay: to be explicit, we review it!
16:11:36 <jgriffith> avishay: we don't let it through :)
16:11:38 <jungleboyj> :-)
16:11:40 <kmartin> avishay: : you must really be special to get a -3 , what color is that X?
16:11:40 <DuncanT-> I'm happy to review it, with a 'but be early next time' note
16:11:44 <avishay> jgriffith: you know what i meant
16:11:55 <jungleboyj> kmartin: It would have flames around it.
16:11:56 <jgriffith> avishay: :)  I did
16:12:02 <jgriffith> flames!
16:12:03 <jgriffith> nice!
16:12:04 <avishay> kmartin: the patch gets deleted, and the user account destroyed and banned
16:12:08 <jungleboyj> :-)
16:12:17 <kmartin> avishay: nice
16:12:34 <jgriffith> winston-d: anything else?
16:12:35 * winston-d wants -3 button as well!
16:12:46 <jgriffith> winston-d: we're going to have to create one of those
16:12:56 <winston-d> jgriffith: nope, i guess that's the only exception?
16:13:01 <jgriffith> DuncanT-: will get a -4 button
16:13:12 <avishay> jgriffith: noooooo!!
16:13:12 <winston-d> hah
16:13:24 <jgriffith> winston-d: the only other one will be the retype patch if I get to it today :)
16:13:24 <winston-d> -4 will destory gerrit?
16:13:24 <DuncanT-> What does -4 do? Geolocate your IP and nuke from orbit?
16:13:28 <kmartin> sounds like a summit topic "a flaming -3 button"
16:13:34 <jungleboyj> 3d Burning X for -4?
16:13:36 <avishay> OK, enough tomfoolery :)
16:13:43 <jgriffith> winston-d: but yes, I believe everything is in, and I've given up on the local scheduling patch for this release
16:13:54 <jgriffith> Who's Tom?
16:13:56 <avishay> Any other meeting topics?
16:14:01 <jgriffith> avishay: yes!
16:14:11 <jgriffith> #topic reviews and rechecks
16:14:13 <winston-d> jgriffith: retype, ok. let me knwo if you need help with scheduler
16:14:16 * jungleboyj is laughing
16:14:16 <jgriffith> Ok....
16:14:25 <jgriffith> so for those that don't read the ML
16:14:46 <jgriffith> Please please please take the time to look at the output/console on failed jenkins and gate jobs
16:15:00 <jgriffith> we're getting into a habbit of blindly typing "recheck no bug"
16:15:04 <jgriffith> when there is clearly a failure
16:15:23 <jgriffith> and even worse I've noted a few cases this week where the failure was introduced by the patch
16:15:26 * winston-d hides
16:15:39 <jgriffith> it wasn't even an intermittent in tempest
16:15:51 <jgriffith> also... note
16:16:18 <jgriffith> I pushed an add of a check conf file is up to date
16:16:35 <eharney> i'm very glad we have that now
16:16:38 <jgriffith> If you started your branch more than a week or so ago you may want to rebase
16:16:39 <jungleboyj> +2
16:16:51 <jgriffith> and not only rebase, but actually run_tests.sh :)
16:17:25 <jgriffith> if you need help with rebasing let me know, via IRC or email and I can help
16:17:32 <bswartz> test before checking in? what are you, a QA guy?
16:17:40 <jgriffith> bswartz: LOL
16:17:50 <jgriffith> Ok...
16:17:53 <jgriffith> ummmm
16:18:02 <jgriffith> I think that's about it for me
16:18:09 <vincent_hou> add test cases for newly added public methods.
16:18:19 <jgriffith> just give priority to the reviews that are up and have BP's targetted for H3
16:18:30 <jgriffith> we have 17 in review last I checked
16:18:34 <jgriffith> vincent_hou: ?
16:18:36 <kmartin> sounds like a plan
16:18:48 <hemna> oh the sample conf is being generated and checked now like nova?
16:18:50 <jgriffith> vincent_hou: that should be a given no?
16:18:58 <eharney> re: review priority
16:19:07 <jgriffith> eharney: yes?
16:19:16 <eharney> is http://status.openstack.org/reviews/#cinder considered a somewhat useful list?  (other than it lying about -1s and -2s)
16:19:25 <avishay> jgriffith: sample conf is being generated automatically or only checked?
16:19:35 <jgriffith> eharney: hmm... personally I don't use that
16:19:40 <eharney> it tries to prioritize them
16:19:44 <eharney> yeah i just realized it was there recently
16:19:51 <jgriffith> eharney: I just look at https://review.openstack.org/#/q/status:open+cinder,n,z
16:19:54 <avishay> eharney: it also lies about priorities because not all blueprints are properly scheduled
16:19:57 <jgriffith> eharney: I don't trust parsers :)
16:20:01 <eharney> avishay: ah. hrm.
16:20:02 <avishay> i use what jgriffith uses
16:20:02 <vincent_hou> jgriffith: I meant the patches to be submitted.
16:20:52 <jgriffith> vincent_hou: yeah, they shouldn't land without unit tests
16:21:02 <jgriffith> Oh, I did have something else
16:21:05 <jgriffith> 2 things actually
16:21:12 <jgriffith> #topic volume ACL's
16:21:30 <jgriffith> I thought we resolved this
16:21:41 <jgriffith> but there was a discussion in my history this morning that would indicate otherwise
16:21:51 <jgriffith> so there are two patches for this:
16:21:55 <zhiyan> jgriffith: i'm not sure why, but seems my readonly-attach change (#38322) not in that review list..
16:22:27 <jgriffith> zhiyan, looking again
16:23:16 <jgriffith> zhiyan: we'll sort that, I think it's ready to land if we can get one more core to look at it anyway
16:23:21 <winston-d> 
16:23:28 <winston-d> jgriffith: retype, ok. let me knwo if you need help with scheduler
16:23:55 <zhiyan> jgriffith: thanks not push, I just confused.
16:23:58 <winston-d> lsdf
16:24:06 <jgriffith> winston-d: yes, thanks maybe I'll double check something with you after the meeting
16:24:16 <jgriffith> ooops
16:24:19 <avishay> jgriffith: i think thingee wanted to take a look at read-only, it's fine by me
16:24:23 <jgriffith> #topic Read Only attach
16:24:30 <jungleboyj> zhiyan: I will take a look as well.
16:24:42 <zhiyan> jungleboyj: thx
16:24:43 <jgriffith> avishay: since he's on vacation we'll need to move forward
16:24:48 <avishay> jgriffith: aahhhh
16:24:52 <jgriffith> he's enjoying debachery in the desert
16:24:53 <zhiyan> oh, ok
16:25:00 <jgriffith> so this one...
16:25:00 <hemna> mmm debachery
16:25:04 <jgriffith> two cnflicting patches
16:25:22 <jgriffith> zhiyan: 's patch: https://review.openstack.org/#/c/38322
16:25:42 <jgriffith> and Anastasia's patch: https://review.openstack.org/#/c/34723/
16:25:45 <zhiyan> avishay: thanks for you support. and winston-d :)
16:26:02 <jgriffith> I believe we agreed to go with the implementation zhiyan submitted no?
16:26:08 <zhiyan> jgriffith: yes, actually i'm also not sure which way you prefer..
16:26:25 <jgriffith> zhiyan: well my review should answer that for you
16:26:30 <avishay> we should just do a git merge of the two
16:26:37 <jgriffith> avishay: ha!
16:26:39 <zhiyan> jgriffith: but from former your comments, seems i'm on the right way :)
16:26:46 <jgriffith> avishay: you get to resolve the conflicts
16:26:53 <jgriffith> avishay: and I'll assign all bugs to you
16:26:57 <avishay> jgriffith: :)
16:27:15 <vincent_hou> jgriffith: i am also waiting for ur answer. Anastasia asked me review hers.
16:27:23 <vincent_hou> :-)
16:27:23 <jgriffith> I'd like DuncanT- avishay and winston-d to look at those two patches and vote please
16:27:35 <jgriffith> vincent_hou: my answer is zhiyan 's patch
16:27:39 <winston-d> jgriffith: ok
16:27:45 <DuncanT-> I vote for zhiyan's patch
16:27:46 <jgriffith> vincent_hou: it's simpler, less complicated IMO
16:27:51 <avishay> jgriffith: can you quickly highlight the high-level differences?
16:27:53 <jgriffith> and still does the job
16:28:01 <vincent_hou> yeah, zhiyan explained that to me.
16:28:08 <avishay> jgriffith: i'm familiar with zhiyan's not anastasia's
16:28:11 <jgriffith> that's left as an exercise to the student :)
16:28:17 <winston-d> i'll take a look at ACL
16:28:19 <avishay> i hate homeowkr :(
16:28:23 <avishay> homework
16:28:25 <jgriffith> zhiyan: or Annastasia can do that if they want
16:28:34 <DuncanT-> Anastasia's seems to be more ACL type stuff munged into R/O
16:28:36 <jgriffith> but not right now :)
16:28:40 <winston-d> i thought that was to solve a different problem?
16:28:45 <jgriffith> the ACL stuff worries me
16:28:54 <jgriffith> I'm not very comfortable with it for some reason
16:29:17 <jungleboyj> jgriffith: +1 seems like that is something to get in earlier in a release cycle.
16:29:27 <jungleboyj> IMHO
16:29:31 <jgriffith> jungleboyj: I'd agree
16:29:39 <jgriffith> jungleboyj: I'm also concerned about confusion for end users
16:29:45 <jgriffith> and most of all quota management
16:29:55 <DuncanT-> I still don't understand the ACL code right now, and that is a bad sign for something being merged this late in a cycle
16:30:14 <jgriffith> Ok, so I think the concensus is we defer it
16:30:19 <jgriffith> avishay: opinion?
16:30:20 <jungleboyj> Seems like a good idea, just bad timing.  Is there anyone begging for it right now?
16:30:24 <jgriffith> winston-d: ?
16:30:31 <avishay> jgriffith: agree
16:30:32 <jgriffith> just mirantis folks
16:30:54 <jgriffith> Ok
16:30:58 <jungleboyj> He he, I am sure my folks won't be too far behind.  :-)
16:30:59 <winston-d> agree
16:31:05 <vincent_hou> This feature is demanded by Annastasia and her team.
16:31:18 <jgriffith> 'demanded'
16:31:28 <hemna> heh
16:31:35 <vincent_hou> She used to ask me for the idea and design doc.
16:32:14 <winston-d> vincent_hou: that was your BP?
16:32:49 <winston-d> vincent_hou: no, it is not.
16:32:50 <vincent_hou> I proposed this topic in Portland.
16:33:53 <jgriffith> vincent_hou: so what is that your'e suggesting?
16:34:15 <vincent_hou> Since there is no clear use cases for it, I decided to pause it.
16:34:41 <winston-d> so i vote for zhiyan's patch for it's simplicity
16:35:00 <vincent_hou> jgriffith: i sent her the design, and she said that what they need and would like to implement it.
16:35:35 <jgriffith> vincent_hou: fine, but everybody here feels uncomfortable with merging it at this stage
16:35:38 <winston-d> any mirantis folks in here?
16:35:41 <jgriffith> vincent_hou: do you have a proposal?
16:35:59 <vincent_hou> I decided not to merge it now.
16:36:07 <winston-d> i'm curious in their use cases for ACL.
16:36:08 <avishay> ok i think we're all in agreement
16:36:15 <jgriffith> yep, let's move on
16:36:26 <jgriffith> The only other thing I have (really this time, I promise)
16:36:32 <jgriffith> Please take a look at: https://review.openstack.org/#/c/40660/
16:36:47 <jgriffith> Not my comment
16:36:53 <jgriffith> s/Not/Note/
16:37:21 <jgriffith> and with that... I'm finally going to shut-up :)
16:37:23 <DuncanT-> I don't understand it either - for tables with no uud (like flavor) yes, but we don't have any of those...
16:37:30 <jgriffith> #topic open-floor
16:37:39 <hemna> I don't get that patch
16:37:52 <jgriffith> DuncanT-: boris did point out a few tables like service etc
16:37:56 <hemna> I don't see the need for a unique constraint on a flag for deleted
16:38:00 <jgriffith> DuncanT-: but I don't see the issue personally
16:38:17 <jgriffith> DuncanT-: he gave a pretty convoluted example of how to have a race there
16:38:29 <DuncanT-> jgriffith: on #openstack-cinder ?
16:38:29 <jgriffith> guitarzan: fully agreed with him so maybe he could explain :)
16:38:32 <guitarzan> metadata key values
16:38:34 <jgriffith> DuncanT-: yes
16:38:38 <guitarzan> are a good example
16:38:46 <DuncanT-> jgriffith: I'll read the scroll back
16:38:53 <kmartin> any updates on the multi-attach patch? https://review.openstack.org/#/c/42886/
16:38:55 * jgriffith votes we drop integer ID's everywhere anyway
16:38:59 <avishay> i didn't understand the patch either...i was hoping thingee would come back and save the day :)
16:39:26 <jgriffith> kmartin: sadly I think it's dead for this release
16:39:37 <zhiyan> jgriffith: it is not ready
16:39:44 <avishay> looks that way unfortunately
16:39:48 <kmartin> jgriffith: ok, maybe first part of I
16:39:51 <zhiyan> kmartin: probably, sorry for that
16:39:58 <hemna> :(
16:40:10 <kmartin> zhiyan: no problem
16:40:20 <zhiyan> kmartin: yes we will do that out in early I. thanks
16:40:31 <zhiyan> jgriffith: if you like it :)
16:40:46 <jgriffith> zhiyan: I like it and we need it
16:40:50 <DuncanT-> guitarzan: But a new metadata K/V pair will get a new ID, so what's the problem?
16:40:50 <jgriffith> zhiyan: just not this late
16:41:00 <zhiyan> jgriffith: nod
16:41:04 <guitarzan> DuncanT-: the idea is that the key needs to be unique
16:41:09 <kmartin> zhiyan: may want to move it to WIP
16:41:18 <guitarzan> and it could be enforced by the db
16:41:23 <jgriffith> zhiyan: kmartin good point ^^
16:41:31 <zhiyan> kmartin: indeed
16:41:36 <jgriffith> zhiyan: kmartin or even abandon and bring it back to life after H is out
16:41:43 <DuncanT-> guitarzan: So this is just about increasing the DB enforced constraints, nothing else?
16:41:46 <guitarzan> yes
16:41:51 <guitarzan> at least I guess so
16:42:14 <zhiyan> jgriffith: i prefer mark it WIP if you ok. and kmartin.
16:42:18 <DuncanT-> guitarzan: Ok, that's the first I've heard of that. People kept talking about race conditions
16:42:21 <guitarzan> changing the deleted flag type is necessary due to the soft deletes
16:42:32 <guitarzan> yeah, a db uniqueness constraint prevents the races too
16:42:47 <kmartin> zhiyan: +1 from me as WIP
16:42:50 <guitarzan> at least that's what I understand the "race condition" part of it to mean
16:42:56 <winston-d> so i vote for zhiyan's patch for it's simplicity
16:43:01 <DuncanT-> guitarzan: I'll read the scrollback
16:43:14 <zhiyan> winston-d: thank you
16:43:22 <guitarzan> DuncanT-: it descended into the weeds, like most of our discussions
16:43:31 <avishay> another winston? :)
16:43:42 <kmartin> zhiyan: just add a note with your plans as the note for the WIP
16:43:47 <jgriffith> winston-1: thanks
16:43:51 <winston-1> i really need DuncanT-'s list of slow freenode server.
16:44:01 <DuncanT-> guitarzan: It the patch hasn't got a clear cut reason for going in, I'm going to vote against it probably
16:44:03 <winston-d> it's killing me
16:44:42 <zhiyan> kmartin: sure, i will ask Charlie help mark it.
16:44:44 <DuncanT-> winston: I've got no correlation between lag and servers... I'm now blaming our corporate proxy
16:44:51 * jgriffith worries we're getting into a habbit of needless churn and complexity
16:44:52 <guitarzan> DuncanT-: the commit message mentions unique constraints
16:44:58 <winston-d> the vote message was sent 5 mins ago. Grrr.
16:45:04 <guitarzan> so it sounds like that is the goal
16:45:29 <DuncanT-> guitarzan: But why is that a useful goal?
16:45:43 <guitarzan> that's not my question to answer
16:45:48 <jgriffith> haha
16:45:57 <guitarzan> lots of people like data integrity at the db layer
16:46:02 <jgriffith> guitarzan: drops chaos grenads and exits stage-left
16:46:02 <guitarzan> and I can't really argue with them about it
16:46:21 <DuncanT-> There's nothing in the blueprint, and DBAs be crazy
16:46:24 <guitarzan> it *seems* useful to use unique constraints where appropriate
16:46:32 <jgriffith> so there's an thread on ML about the long-term goal here
16:46:50 <jgriffith> uuid, uuid, uuid, uuid, uuid
16:46:50 <hemna> unique constraints on uuids makes sense to me
16:47:11 <avishay> there definite value in making the DB proper, but i need to understand why/how/what..
16:47:16 <hemna> but not sure why it's needed elsewhere
16:47:17 <guitarzan> do unique constraints on metadata keys make sense?
16:47:19 <guitarzan> or service names?
16:47:40 <hemna> I don't think so on metadata keys
16:47:44 <guitarzan> no? why not?
16:47:51 <hemna> why can't there be dupes?
16:47:59 <guitarzan> how do you get to the second copy?
16:47:59 <hemna> who cares, it's metadata
16:48:01 <guitarzan> kv get
16:49:43 <jgriffith> guitarzan: winston-d avishay hemna kmartin ... everyone
16:49:46 <jgriffith> http://lists.openstack.org/pipermail/openstack-dev/2013-August/013877.html
16:50:07 <winston-d> jgriffith: thx
16:50:10 <jgriffith> no need to go in the weeds here, vote via review, or discuss in cinder channel
16:50:13 <jgriffith> let's move on
16:50:15 <joel-coffman> I have a quick question regarding https://bugs.launchpad.net/cinder/+bug/1215962
16:50:18 <uvirtbot> Launchpad bug 1215962 in cinder "Cinder migration 017_add_encryption_information adds two new volume types without warning" [Undecided,Fix committed]
16:50:27 <jgriffith> haha
16:50:30 <jgriffith> joel-coffman: yes
16:50:34 <joel-coffman> is the migration policy to avoid adding anything to the database?
16:51:11 <joel-coffman> caught me by surprise this week...
16:51:17 <hemna> are we adding volume types by default ?
16:51:36 <avishay> DB migrations should add columns and tables, not modify rows
16:51:47 <avishay> *add/remove/modify columns and tables
16:51:51 <guitarzan> well, that's not necessarily true
16:51:53 <eharney> IMO if you need to add types like that by default, it needs to be done by code in Cinder, not DB migrations
16:51:55 <kmartin> agree
16:52:00 <hemna> avishay, unless there is data migration needed due to a change no?
16:52:00 <guitarzan> but there's no reason to add a volume type just for fun
16:52:12 <hemna> guitarzan, +1
16:52:13 <kmartin> these should be removed
16:52:15 <avishay> hemna: sure there are exceptions, but overall, yes
16:52:19 <zhiyan> hemna: yes +1
16:52:32 <DuncanT-> joel-coffman: We migrated a system and suddenly had a new type, that was a surprise....
16:52:35 <eharney> hemna: but you can do that in code, not in the migration schema, right?
16:52:43 <joel-coffman> okay
16:53:09 <joel-coffman> what's the *correct* way to create new types that correspond to our front-end encryption schemes that are implemented in Nova?
16:53:17 <hemna> eharney, depends, if the data needs migration prior to code...that's what migration is for IMO.  but I don't think we should be adding volume types by default
16:53:24 <jgriffith> joel-coffman: the admin creates them as part of config
16:53:25 <avishay> admins should create whatever volume types they want in the deployment...i think cinder shouldn't create any at all
16:53:39 <jgriffith> avishay: +1
16:53:50 <bswartz> avishay: +2
16:53:56 <jgriffith> joel-coffman: for example, there's likely to be a number of people who choose not to enable/use encryption
16:54:08 <jungleboyj> avishay: +1
16:54:09 <joel-coffman> jgriffith: any way to automate this process for the types that we *know* should exist
16:54:21 <hemna> jgriffith, +1
16:54:23 <guitarzan> there are no types that should exist everywhere
16:54:27 <avishay> the cinder code should check extra_specs and do as necessary
16:54:28 <joel-coffman> configuration options could enable / disable it
16:54:29 <DuncanT-> Why should encrypted types exist?
16:54:32 <jgriffith> joel-coffman: no, as types are all custom
16:54:37 <jgriffith> joel-coffman: not sure of the problem?
16:54:38 <winston-d> guitarzan: +1
16:54:53 <jgriffith> joel-coffman: cinder type-create luks
16:54:53 <jgriffith> done
16:54:57 <DuncanT-> joel-coffman: We didn't want encrypted types. Not now, maybe never
16:55:05 <avishay> joel-coffman: for example, i may want a type that has encryption, a certain QoS setting, and no compression
16:55:18 <joel-coffman> okay, that makes sense
16:55:20 <avishay> joel-coffman: why should i be stuck with those 2 types?
16:55:34 <avishay> joel-coffman: or i may not want compression at all, and then i need to go and remove the types
16:55:39 <joel-coffman> just trying to understand the environment better
16:55:40 <jgriffith> joel-coffman: going through the code I didn't see where this would cause a problem
16:55:59 <jgriffith> joel-coffman: did I miss something perhaps (ie this being taking out the creation of the types)
16:56:21 <joel-coffman> no, not really a problem -- just a surprise
16:56:31 <jgriffith> joel-coffman: ahh, sorry about that
16:56:45 <jgriffith> joel-coffman: I was looking for you when DuncanT- and I discussed it the other day
16:56:55 <jgriffith> joel-coffman: you weren't around and sadly I forgot to follow up
16:57:05 <jgriffith> joel-coffman: You were also supposed to be added to the review
16:57:10 <joel-coffman> no worries, I'll know better next time (and try to hang out more on IRC)
16:57:14 <joel-coffman> :)
16:57:22 <jgriffith> joel-coffman: sorry for the surprise
16:57:45 <joel-coffman> jgriffith: thanks, being part of that review would have been helpful
16:57:56 <avishay> next time we surprise you we'll make sure to bring confetti :)
16:58:06 <jgriffith> joel-coffman: yeah, that for sure would've been appropriate
16:58:18 <joel-coffman> avishay: ha!
16:58:31 <jgriffith> Ok.. so two topics on the agenda that took 10 minutes and we still filled our hour :)
16:58:37 <jgriffith> dang we're good
16:58:41 <avishay> never fails
16:58:55 <avishay> now try to end the meeting :)
16:59:01 <bswartz> wait for it....
16:59:03 <hemna> lol
16:59:07 <kmartin> waiting...
16:59:14 <jgriffith> #end meeting
16:59:14 <guitarzan> hah
16:59:16 <jgriffith> ha!
16:59:18 <avishay> lol
16:59:19 <winston-d> asd///asdfas
16:59:21 <jgriffith> #endmeeting cinder