16:00:02 #startmeeting Cinder 16:00:03 Meeting started Wed Jul 26 16:00:02 2017 UTC and is due to finish in 60 minutes. The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:05 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:07 The meeting name has been set to 'cinder' 16:00:09 Hello. 16:00:12 hi 16:00:13 <_alastor_> o/ 16:00:16 Nihao 16:00:18 o/ 16:00:18 hi 16:00:23 smcginnis: lol 16:00:26 :) 16:00:27 hi!! 16:00:32 Hi 16:00:32 o/ 16:00:36 o/ 16:00:37 hi 16:00:40 @! 16:00:40 Coutesy ping 16:00:40 <_pewp_> jungleboyj ( ^_^)/ 16:00:41 dulek duncant eharney geguileo winston-d e0ne jungleboyj jgriffith thingee smcginnis hemna xyang1 tbarron scottda erlon jbernard _alastor_ bluex karthikp_ patrickeast dongwenjuan JaniceLee cFouts Thelo vivekd adrianofr mtanino karlamrhein diablo_rojo jay.xu jgregor lhx_ rajinir wilson-l reduxio wanghao thrawn01 chris_morrell watanabe.isao,tommylikehu mdovgal ildikov wxy viks ketonne abishop 16:00:47 sivn 16:00:51 Hi 16:00:52 o/ 16:00:54 Hi 16:01:01 DuncanT! :) 16:01:05 hello 16:01:09 hi 16:01:09 <_pewp_> hemna (♦亝д 亝)ノ 16:01:55 #topic Announcements 16:02:00 Hello :) 16:02:02 The end is near. 16:02:04 #link https://wiki.openstack.org/wiki/CinderMeetings#Next_meeting 16:02:16 https://releases.openstack.org/pike/schedule.html 16:02:24 o/ 16:02:28 hi 16:02:30 Tomorrow is milestone 3. 16:02:33 * jungleboyj pictures smcginnis in a sandwich board reporting this. 16:02:38 lhx_: Welcome! :) 16:02:49 jungleboyj: Yay, my intention worked. 16:03:14 :-) 16:03:18 Please let me know if there are any critical python-cinderclient patches that need to get through. 16:03:20 smcginnis, nice to see you the first time :D 16:03:30 lhx_: You too! 16:03:42 We just have today to get anything else through. 16:03:58 We had a cinderclient release last week, but we've had a few more patches make it through since then. 16:04:05 http://gph.is/1pr6Izn 16:04:16 I'm going to request another release later today to pick up those. 16:04:30 jungleboyj: ;) 16:04:55 A few other things with the schedule... 16:04:58 smcginnis: I hope, you don'te forget python-brick-cinderclient-ext too :) 16:05:10 e0ne: Yep, absolutely! :) 16:05:19 e0ne: Thanks for the reminder though. 16:05:31 This is feature freeze as well. 16:05:34 smcginnis: you're welcome 16:06:00 So no more feature patches after tomorrow unless the core team feels it's important to still get through. 16:06:11 But I'm hoping we can limit that as much as possible. 16:06:39 I've updated the etherpad - still a few specs and things we agreed on that are not complete. 16:06:48 #link https://etherpad.openstack.org/p/cinder-spec-review-tracking Review focus 16:06:52 Overall not too bad thoguh. 16:06:55 *though 16:07:15 And Soft String Freeze. 16:07:47 After tomorrow, we should not be making any changes to translatable strings to help the i18n team have any kind of chance of actually translating them. 16:08:15 I think this has changed from the past, but we are supposed to request on the ML if we want to merge string changes after tomorrow. 16:08:26 Let's just try to avoid them if at all possible. 16:08:39 Should be better now that it's just the exception strings and not all log messages. 16:09:21 Should be much easier now :) 16:09:38 It is also the community goal completion deadline. 16:09:59 Thanks to e0ne, we have some patches out there for the uWSGI goal. 16:10:12 But still not merged, so I'm hoping someone looks at those soon. 16:10:24 Out of our control, unfortunately. 16:10:42 smcginnis: we need at lease devstack patch to be merged 16:11:00 #link https://review.openstack.org/#/c/441266/ 16:11:27 So main points: 1) Let me know if we have any cinderclient patches I need to wait on, 2) help get through priorities from the etherpad, 3) try not to make string changes. 16:11:34 s/at lease/at least 16:11:39 e0ne: +1 16:11:59 OK, just a couple more announcement items... 16:12:01 #link https://etherpad.openstack.org/p/cinder-ptg-queens Planning etherpad for PTG 16:12:11 Any and all ideas for discussion at the PTG, please add there. 16:12:17 Whether you are able to attend or not. 16:12:21 * e0ne still waits for a visa renewval 16:12:30 And if you are planning on attending, please add your name to that so we can get an idea. 16:12:38 e0ne: I hope that goes well. 16:13:08 smcginnis: Is P3 a deadline for back port bugs as well? 16:13:12 If you are planning on attending, please register so they can plan overall: 16:13:15 #link https://www.eventbrite.com/e/project-teams-gathering-denver-2017-tickets-33219389087 PTG registration 16:13:39 tommylikehu: No, support cycles have different deadlines. So a little more time there. 16:13:50 tommylikehu: Not that we should wait if there is anything to backport. ;) 16:14:06 smcginnis: ok 16:14:07 TSP for the PTG is still open for another week I believe. 16:14:17 tommylikehu: This page shows when the support phase dates are: 16:14:18 https://releases.openstack.org/ 16:14:31 tommylikehu: it's a feature freeze and client freeze deadline 16:14:32 diablo_rojo_phon: Excellent. Good to bring up. 16:14:41 * lhx_ looking forward to attending, but tsp application failed 16:14:52 If you need financial support to attend, please look into the travel support program. 16:15:07 TSP? 16:15:14 Travel Support Program 16:15:40 Swanson: Or Tea Sipping Protocol. You decide. 16:15:46 :) 16:16:01 Mmmm tea :) 16:16:07 That's all I had for announcements. 16:16:16 #topic New os-brick disconnect parameters 16:16:29 geguileo: Thanks for waiting a week on this. 16:16:38 no problem :-) 16:16:53 basically this is just an announcement 16:17:06 following smcginnis last week announcement of the the os-brick version 16:17:18 now minimum version for cinder is 1.15.1 16:17:31 which has a couple of new parameters which are only useful for iSCSI for now 16:17:40 force and ignore_errors 16:17:51 these params are for disconnect 16:17:58 force means that we don't care if we lose data on disconnect 16:18:13 but we give priority to leaving the system as clean as possible on error 16:18:21 useful when you are creating a vol from image 16:18:23 for example 16:18:35 since on failure data loss is irrelevant, you'll be deleting the vol anyway 16:18:48 ignore errors means you don't want an exception to be raised if there's a failure 16:18:49 Makes a lot of sense. 16:19:01 ignore errors is useful if you are forcing 16:19:17 and for example are detaing a volume that was mounted as read only to back it up 16:19:25 you detach it forcing and ignoring errors 16:19:37 you don't want to fail a backup because the detach of the source volume failed at the end 16:19:55 this was part of the iSCSI refactoring 16:20:08 and there is a patch in Cinder that uses those new parameters 16:20:09 https://review.openstack.org/459454 16:20:29 for now it's only usefull for iSCSI 16:20:46 and that's all I wanted to say 16:20:58 geguileo: Thanks, good for folks to know the change is there. 16:21:22 important for any new code that we add 16:21:55 I do think there are a lot of situations where force makes sense and we really don't care the state of the data since we will just be blowing it away anyway. 16:22:11 geguileo: Thanks for the update. 16:22:15 np 16:22:22 #topic Open Discussion 16:22:26 smcginnis: +1 16:22:35 Short agenda. What color should we paint the shed? 16:22:58 Lime green like the new wolves jersey. 16:22:59 Blue. No. Green. 16:22:59 shed? 16:23:11 e0ne: Bikeshedding time. :) 16:23:18 :) 16:23:41 https://en.wikipedia.org/wiki/Law_of_triviality 16:24:26 OK, if there are no other topics, go forth and review. :) 16:24:43 Well I found something interesting yesterday. 16:25:03 diablo_rojo_phon: Good interesting or bad interesting? 16:25:09 Poking around on an ovo patch for making migration status into an enum. 16:25:14 Bad interesting. 16:25:28 FYI https://bugs.launchpad.net/cinder/+bug/1705407 is a bothersome bug that i'm going to post a fix for soon... pike-3 would be nice but i don't know if it'll make it 16:25:28 Launchpad bug 1705407 in Cinder "Bootable volume creation failed with glance image of 'ISO' format" [High,New] - Assigned to Eric Harney (eharney) 16:25:37 Looks like there a lot of places where target info is being saved into the migration status. 16:25:53 eharney: yeah... 16:26:30 diablo_rojo_phon: The status field is being set to the target info? 16:26:40 When migration status shouldn't be more than a string it has the target address. It was in the core code all the way back to like the start of cinder. 16:26:42 Yeah. 16:27:18 diablo_rojo_phon: do you have a link to the patch nearly? 16:27:19 Yeah and there are checks to make sure that migration status starts with 'target:' 16:27:30 Yeah one sec. 16:27:33 diablo_rojo_phon: That's.... odd. 16:27:55 diablo_rojo_phon: oh.. we have to get rid of these things 16:28:12 I think the migration status field was a sort of magic field 16:28:19 since it wasn't really shown anywhere 16:28:27 I think there were reasons for this at one point 16:30:05 In manager.py line 1619 16:30:24 https://github.com/openstack/cinder/tree/e62cb5eac5ee8074235a855b0ccafa1134eecc19/cinder/volume#L1619 16:31:08 I was gonna go file a bunch of bugs, but I thought I'd let you all know too. 16:31:22 Good to at least raise awareness. 16:31:28 yah looks like that's there on purpose 16:31:46 Sounds like maybe we'll need a new field if we want to be able to switch that to use the enums. 16:31:47 changing that means db migrations 16:31:54 hemna: Yep. :[ 16:32:05 Yeah it is in a fee driver's too. 16:32:15 Solidfire and NEC that I saw so far 16:32:19 Probably not a huge deal, but would take some work. 16:32:32 It's all jgriffith's fault. :) 16:32:34 jgriffith: Any idea why that is there? 16:32:53 actually following the blames back it was avishay.. 16:33:03 Oy 16:33:14 jgriffith: just took advantage in his driver ;) 16:33:15 Oh :( 16:33:24 trying to catch up 16:33:47 I'm sure it was to just save the info somewhere that was easier. 16:34:08 https://github.com/openstack/cinder/blob/e62cb5eac5ee8074235a855b0ccafa1134eecc19/cinder/volume/manager.py#L1619 16:34:14 that link works 16:34:55 Well, probably something we can work out in channel. Thanks for the knowledge sharing diablo_rojo_phon. 16:35:05 Anything else for the meeting while everyone is here? 16:35:33 No problem smcginnis 16:36:44 OK, thanks everyone! 16:36:54 #endmeeting