14:00:39 <isviridov> #startmeeting magnetodb
14:00:39 <openstack> Meeting started Thu Dec 18 14:00:39 2014 UTC and is due to finish in 60 minutes.  The chair is isviridov. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:43 <openstack> The meeting name has been set to 'magnetodb'
14:00:51 <isviridov> o/
14:01:04 <nunosantos> o/
14:01:12 <ajayaa> o/
14:01:15 <isviridov> Agenda for today https://wiki.openstack.org/wiki/MagnetoDB/WeeklyMeetingAgenda#Agenda
14:01:25 <isviridov> hello nunosantos ajayaa
14:01:34 <ajayaa> Hi isviridov
14:01:57 <isviridov> Let us start with action items
14:02:16 <isviridov> #topic Go through action items isviridov
14:02:30 <isviridov> achudnovets create a spec for integration with external monitoring systems
14:02:33 <aostapenko> o/
14:02:47 <isviridov> Hey aostapenko
14:03:12 <achudnovets> spec in progress. need some time to finish
14:03:41 <isviridov> achudnovets ok, anything to look at yet?
14:04:40 <isviridov> #link https://blueprints.launchpad.net/magnetodb/+spec/monitoring-api-url-refactoring
14:05:03 <achudnovets> not yet. I've added bp about monitoring api refactoring. Next step is edit monitoring api spec and add spec for integration with external monitoring systems
14:05:21 <openstackgerrit> Illia Khudoshyn proposed stackforge/magnetodb: (WIP) Add backup manager  https://review.openstack.org/141026
14:05:33 <achudnovets> thanks isviridov
14:05:42 <isviridov> achudnovets thanks for update
14:06:00 <isviridov> No other AIs
14:06:16 <isviridov> #topic Collaboration with API-WG about API v2 isviridov
14:06:44 <isviridov> To inform the tema.
14:07:31 <ominakov> hi guys
14:07:35 <isviridov> There is API work group in OpenStack and we are not working with them in checking and improving our API to follow REST and general OpenStack best practices
14:07:40 <ominakov> i'm sorry, i'm late
14:07:55 <isviridov> ominakov it is ok, just be quiet
14:08:12 <isviridov> More details about API-WG https://wiki.openstack.org/wiki/API_Working_Group
14:08:37 <isviridov> They are building smth like guide for API https://github.com/openstack/api-wg
14:09:06 <isviridov> Also there is a meeting today, everybody is wellcome to join
14:09:44 <isviridov> At 16:00UTC #openstack-meeting-3
14:10:26 <isviridov> Anything here to discuss?
14:10:58 <isviridov> Moving forward
14:11:19 <isviridov> #topic Kilo-1 milestone status update isviridov
14:11:48 <isviridov> So, today is kilo-1 milestone release
14:12:30 <isviridov> I'm going to create tag 2015.1.b1 and do release of work we have done.
14:12:40 <charlesw> Is there a release notes?
14:13:23 <isviridov> charlesw yes, based in list of fixed bugs and implemented bps.
14:14:20 <isviridov> #action isviridov do release of 2015.1.b1 and announce in ML
14:14:35 <isviridov> charlesw moving forward?
14:14:49 <charlesw> sure, thx
14:15:07 <isviridov> #topic  Open reviews status isviridov
14:16:40 <isviridov> Any patch we have to discuss?
14:17:24 <isviridov> or catch some core's eyes
14:18:16 <isviridov> charlesw thank you for your comments about documentation.
14:18:27 <isviridov> Ok, let us move forward
14:18:30 <charlesw> np
14:18:40 <isviridov> #topic Migration to new local secondary indices design dukhlov
14:18:56 <isviridov> dukhlov my congrats about it!
14:18:58 <isviridov> !m dukhlov
14:18:58 <openstack> isviridov: Error: "m" is not a valid command.
14:18:58 <[o__o]> You're doing good work, dukhlov!
14:19:19 <charlesw> dukhlov, great job, as always
14:19:21 <isviridov> dukhlov how are we going to integrate it?
14:19:48 <dukhlov> ok, It seems new implementation  now works on our gate
14:20:28 <dukhlov> and all are welcome to review this patch and ask questions
14:21:35 <isviridov> dukhlov any documentation about it?
14:22:02 <ajayaa> isviridov, This could use a blessing. https://review.openstack.org/#/c/124391/
14:22:04 <dukhlov> hm, no:) only bp spec
14:22:53 <isviridov> #action charlesw isviridov dukhlov ikhudoshyn  https://review.openstack.org/#/c/124391/
14:23:04 <dukhlov> What kind of documents would you like to see?
14:23:27 <dukhlov> general concept or some details?
14:23:56 <isviridov> dukhlov it would be great to have some technical concept overview for developers and for promotion of mdb in general.
14:24:27 <dukhlov> like article?
14:25:01 <isviridov> And I think we have to think about upgrade procedure from previous version of storage.
14:25:12 <charlesw> Also, if you can run some performance comparison with existing LSI solution, it will be helpful why we should migrate.
14:25:26 <isviridov> dukhlov some blogpost would be great!
14:25:33 <dukhlov> charlesw, sure
14:26:00 * isviridov the more content about mdb the better: screencast, blogposts...
14:26:35 <charlesw> Is there an official mdb blog?
14:26:57 <dukhlov> for upgrade probably it would be nice to have backup/restore working first
14:27:20 <isviridov> charlesw not, but articles in Symantec, Miranits blogs
14:27:56 <charlesw> Would be nice to have a central place, with links pointing to existing blogs.
14:27:58 <isviridov> dukhlov so no migration script is delivered?
14:28:40 <dukhlov> no now we don't have migration scripts
14:28:50 <isviridov> charlesw yeap, I've just updated wiki yesterday, we have such a place now https://wiki.openstack.org/wiki/MagnetoDB#Presentations_.26_blogposts
14:29:21 <charlesw> isviridov, excellent, thx
14:30:32 <dukhlov> I have in mind one more bp which will change data model and if we really need migration procedure I would suggest make it after that
14:30:52 <isviridov> dukhlov for now it is probably ok.
14:30:54 <dukhlov> it connected with dynamic attributes packing
14:31:38 <isviridov> dukhlov it is not urgent really, but we have to think about it. It can be migration or using old data format for old tables and new one for newly created.
14:31:50 <isviridov> dukhlov would love to see your bp
14:32:14 <dukhlov> isviridov, will do
14:32:27 <isviridov> Ok, moving forward?
14:33:08 <isviridov> #topic Open discussion isviridov
14:33:34 <charlesw> I have one bp in mind, it's connected to this bug: https://bugs.launchpad.net/magnetodb/+bug/1402791
14:33:49 <charlesw> SSL termination
14:34:05 <charlesw> right now it's broken in our deployment
14:35:38 <miqui_> hi ..sorry am late..
14:35:59 <isviridov> miqui_ hi
14:36:04 <miqui_> ..hi..
14:36:32 <isviridov> charlesw do you have any header with public endpoint from SSL proxy?
14:37:15 <charlesw> Not sure, probably not, we need to work with the SSL proxy guys
14:37:36 <isviridov> charlesw I think it is important to fix it, just wondering if you have any ideas how to do it
14:38:00 <charlesw> what you just mentioned is my first choice of implementation
14:38:24 <charlesw> Another option is to have it configured in MDB deployment
14:39:23 <isviridov> Is it possible to have several endpoints?
14:39:55 <charlesw> Not in my deployment.
14:40:13 <charlesw> But somebody may have this requirement
14:40:50 <dukhlov> we can get prefix of our endpoint from keystone
14:41:06 <dukhlov> not from header
14:41:51 <isviridov> dukhlov +1 good idea, specially if Keystone is catalog for OpenStack
14:42:34 <charlesw> but isviridov was just asking if it's possible to have multiple endpoints
14:42:54 <charlesw> If yes, how do you know which one is being used?
14:44:08 <isviridov> charlesw in general it is possible usecase, but for OpenStack it sounds strange with keystone as catalog and endpoints list.
14:44:37 <isviridov> The question is if it is possible to have several enpoints in keystone for the same instance if mdb
14:45:40 <charlesw> I don't see why not
14:45:56 <isviridov> We can start with one configurabel parameter and improve just we have such a case.
14:46:01 <charlesw> using the proxy header can handle this
14:46:19 <dukhlov> I believe we can add extra header with endpoint and use it if it is set
14:46:41 <isviridov> I'm ok with both.
14:47:02 <dukhlov> if not use one of available endpoint from keystone
14:47:30 <charlesw> dukhlov, what do you mean extra header with endpoints? Extra header in where?
14:48:24 <dukhlov> extra header in request for storing original url if some proxy overwrites it
14:49:16 <charlesw> ok, sounds good. I'd think we just need protocol and port
14:49:41 <dukhlov> I think it is not a big problem to configure proxy for adding this extra neader
14:50:12 <charlesw> dukhlov, +1
14:50:44 <charlesw> seems that we have consensus. Try the proxy header approach first, right?
14:51:29 <dukhlov> +1
14:51:30 <isviridov> charlesw +1
14:52:11 <charlesw> perfect, we will do that. I'll write up a bp for that.
14:53:20 <isviridov> Any other topic?
14:54:35 <isviridov> Looks like we are done for today
14:55:51 <isviridov> Thank you for comming
14:55:55 <isviridov> #endmeeting