14:00:39 <ddieterly> #startmeeting freezer 14:00:39 <openstack> Meeting started Thu May 5 14:00:39 2016 UTC and is due to finish in 60 minutes. The chair is ddieterly. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:40 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:42 <openstack> The meeting name has been set to 'freezer' 14:01:07 <ddieterly> attendance please 14:01:09 <slashme> Hello ddieterly :) 14:01:12 <ddieterly> o/ 14:01:18 <EinstCrazy> Hello 14:01:26 <ddieterly> hello ice cubes 14:01:26 <szaher> Hello ddieterly 14:02:07 <slashme> Let's wait five minutes for the others 14:02:43 <ddieterly> sure 14:03:58 <ddieterly> while we are waiting, if you have an item for the agenda, please update it 14:04:07 <ddieterly> https://etherpad.openstack.org/p/freezer_meetings 14:05:52 <ddieterly> can we start? 14:06:08 <slashme> Go ahead :) 14:06:21 <ddieterly> ok, the agenda looks very slim 14:06:37 <ddieterly> maybe we could get some highlights from the summit? 14:06:49 <ddieterly> any important take aways? 14:07:22 <ddieterly> i heard a lot of people interested in a comprehensive backup/restore solution 14:07:42 <ddieterly> i heard a lot about smaug for better or worse 14:08:01 <ddieterly> did anything come out of the design meeting on friday? 14:08:29 <ddieterly> slashme are the notes available from the sessions on the web? 14:09:20 <slashme> We have seen a lot of interest from people regarding all the subject we talked about: 14:09:21 <slashme> - Backup/restore of OpenStack Infra (and as you said comprehensive solution) 14:09:21 <slashme> - Backup/restore as a Service 14:09:21 <slashme> - Disaster Recovery 14:09:21 <slashme> - Better integration with deployment systems 14:09:56 <slashme> Notes from all the sessions are available here: 14:09:56 <slashme> I'll clean that up and send a recap to the mailing list as soon as possible 14:10:12 <slashme> OS Infra backup: https://etherpad.openstack.org/p/freezer_austin_session_backup_os_infra 14:10:22 <slashme> BaaS: https://etherpad.openstack.org/p/freezer_austin_session_baas 14:11:16 <slashme> The Disaster Recovery etherpad stayed empty: https://etherpad.openstack.org/p/freezer_austin_session_disaster_recovery 14:11:35 <slashme> Friday contributors meetup: https://etherpad.openstack.org/p/freezer_austin_session_contributors_meetup 14:12:05 <slashme> The last one contains notes about smaug, Ironic, ... 14:12:20 <ddieterly> i'll take some time today and read those 14:12:34 <ddieterly> sorry i had to leave early on friday 14:12:47 <ddieterly> did the smaug guy finally yield the floor? 14:12:51 <slashme> Overall, we have seen a large interest aroud freezer, which is a very good sign. 14:12:59 <slashme> Yes, He left just after you 14:13:11 <ddieterly> great timing ;-) 14:13:16 <slashme> Conclusion was: 14:14:17 <slashme> Let's arrange a specific meeting with some Freezer devs and some Smaug devs to discuss how integration of the tweo solutions could happen. 14:14:56 <ddieterly> did you get his name by any chance? 14:16:19 <ddieterly> what are your thoughts on how smaug relates to freezer? 14:16:37 <slashme> I gave him my card, so I expect an email in the next few days. 14:17:29 <ddieterly> i guess i should do this 14:17:37 <ddieterly> #topic austin summit recap 14:18:49 <ddieterly> slashme any thoughts on smaug? 14:19:09 <slashme> Well, as we tried to describe in the etherpad, Smaug is trying to achieve what we called "tenant backup". The ability to backup/restore OpenStack resources that don't live outside of OpenStack, that are only metadata (like neutron networks, Keystone users, ... ) 14:19:55 <ddieterly> do you think there is an opportunity to work with that project? 14:20:36 <slashme> When we thought about that in the past, we decided not to engage in doing that for now because it represent a big amount of work to map all the calls to all OpenStack api's 14:21:40 <slashme> We could either leverage smaug to backup this kind of data or provide a plugin so smaug can leverage freezer when doing classic data backups. 14:22:01 <ddieterly> ok, so i could see that project asking for functionality from freezer in the future 14:22:48 <ddieterly> slashme anything else from the summit you want to share? 14:22:55 <daemontool_> o/ 14:23:19 <ddieterly> daemontool_ anything from the summit you want to share with the team? 14:23:27 <szaher> slashme: Can we consider smaug as a engine for freezer ? 14:23:28 <daemontool_> yes 14:23:31 <yangyapeng> :) 14:23:42 <daemontool_> yes to ddieterly don-t know about smaug 14:23:45 <slashme> szaher: Yes that's one of the possibilities 14:23:50 <daemontool_> but yes at first 14:23:52 <daemontool_> :) 14:23:52 <daemontool_> so 14:23:56 <daemontool_> couple of guys 14:24:04 <daemontool_> requested to integrate with tripleo and ironic 14:24:08 <daemontool_> to do image based backup 14:24:13 <daemontool_> for distributions like rh 14:24:17 <daemontool_> I think is a great idea 14:24:31 <daemontool_> one othe thing is 14:24:48 <daemontool_> that we need to speed up the publication of the dr related code 14:25:01 <daemontool_> I-m creating the freezer-dr repo today 14:25:22 <daemontool_> s/-/\'/ :) 14:26:23 <daemontool_> can I go on or is there any topic that is currently being dicussed_ 14:26:24 <daemontool_> ? 14:26:33 <ddieterly> the topic is the summit 14:26:37 <daemontool_> ok 14:26:49 <ddieterly> we can add dr to the agenda 14:27:05 <ddieterly> there, done 14:27:07 <daemontool_> ty 14:27:33 <ddieterly> ok, want to move on to DR? 14:28:10 <ddieterly> anything else on the summit 14:28:22 <ddieterly> everyone have a good time? 14:28:52 <ddieterly> daemontool_ hope you used the party pass that i gave you well ;-) 14:28:58 <daemontool_> brilliant time 14:28:59 <daemontool_> ty 14:29:10 <daemontool_> I made it count ;) 14:29:21 <ddieterly> i knew if anyone could, it would be you 14:29:39 <ddieterly> #topic disaster recovery 14:29:49 <ddieterly> daemontool_ you have the floor 14:30:53 <ddieterly> some of the feedback on frescos blueprint from dan dyer was that there was a lot out of scope for a backup/restore solution 14:31:48 <ddieterly> i got that feedback on friday from 2 people in the contributors meeting before slashme arrived and the smaug discussion started 14:34:13 <ddieterly> daemontool_ you there? 14:36:00 <daemontool_> ddieterly, yes 14:36:05 <daemontool_> sorry 14:36:12 <daemontool_> so we need to create the freezer-dr repo 14:36:20 <daemontool_> and upstream the code we have 14:36:37 <daemontool_> following up our previous mettings 14:36:54 <daemontool_> now, I understand we need to place the bp onto the freezer-specs repo 14:36:56 <daemontool_> but the specs repo 14:37:03 <daemontool_> is not ready yet 14:37:05 <daemontool_> ? 14:37:22 <slashme> I'll try to prepare it before EOW 14:37:35 <daemontool_> ok ty 14:38:07 <ddieterly> anything else? 14:38:10 <daemontool_> that's all from me 14:38:28 <daemontool_> anyway this stream is being mainly followed by Fresco, ddieterly 14:39:02 <ddieterly> ok, ty 14:39:06 <ddieterly> next topic? 14:39:07 <szaher> daemontool_: are you going to submit a request to create the repo 14:39:13 <daemontool_> szaher, yes 14:39:21 <daemontool_> we need to do that sooner than later 14:39:23 <daemontool_> hopefully today 14:39:28 <daemontool_> or tomorrow 14:39:36 <szaher> daemontool_: Good, ty 14:39:49 <daemontool_> then upstream 14:39:51 <daemontool_> :( 14:39:52 <daemontool_> :) 14:40:29 <ddieterly> #topic pending blueprints 14:41:00 <ddieterly> i'm not sure which are the most active/pressing blueprints that people are working on 14:41:20 <ddieterly> daemontool_ do you have a bp for dedup? 14:41:22 <slashme> ddieterly: I guess it is thoses about refactoring 14:41:36 <slashme> dedup is not yet at the BP stage 14:42:00 <ddieterly> do we need a discussion around dedup? 14:42:18 <slashme> Not today. But we definity need to plan for one. 14:42:26 <daemontool_> I think there was some discussion at the midcycle meet up? 14:42:31 <slashme> daemontool_: Yes 14:43:06 <slashme> But nothing good/detailled/valid enough to be transformed into a blueprint 14:43:27 <daemontool_> ok 14:43:33 <slashme> Plan is to start the engineering process during the next few weeks 14:43:36 <ddieterly> what is driving dedup? is there someone asking for it or is it just nice to have? 14:44:02 <ddieterly> sorry for my ingorance 14:44:08 <daemontool_> ddieterly, there was a competitive analysis mentioned 14:44:16 <slashme> So we can discuss the design at the next midcycle and start implementing it as a stretch goal for Newton 14:44:19 <daemontool_> ask to the ps guys they have more information 14:45:19 <szaher> I guess we can write pb for re-factoring and plug-able layers 14:45:44 <ddieterly> #topic mid-cycle meetup 14:45:58 <ddieterly> when/where is the mid-cycle meetup? 14:46:26 <daemontool_> July, Ibiza :( 14:46:27 <daemontool_> :) 14:46:39 <slashme> Nothing is decided yet I think. 14:46:40 <ddieterly> ok, where is ibiza? 14:46:41 <slashme> daemontool_ ;) 14:46:49 <slashme> Spainish island 14:47:17 <ddieterly> so, roughly july, then? 14:47:44 <daemontool_> I don't know, was just an idea 14:47:55 <ddieterly> i guess that would be 'mid-cycle' 14:48:03 <daemontool_> but nothing serious really 14:48:10 <daemontool_> what works bes for you 14:48:13 <daemontool_> best 14:48:16 <daemontool_> good for me 14:48:25 <ddieterly> july is fine for me 14:49:06 <ddieterly> when do we need to decice? 14:49:07 <daemontool_> We can organize it in Ericsson offices if needed 14:49:14 <ddieterly> decide 14:49:21 <ddieterly> the sooner the better 14:49:41 <ddieterly> travel might be a problem for me 14:49:54 <ddieterly> hpe is tightening the purse strings 14:50:55 <slashme> Same thing as last one, I guess, the best solution (cheapest overwall) will be to organize in Galway 14:51:50 <ddieterly> that's fine with me, i just need to get approval 14:52:17 <ddieterly> i'll let management know that travel to galway in july is coming 14:52:22 <ddieterly> see what they say 14:52:47 <ddieterly> i think you should all come to colorado 14:52:56 <slashme> I'd love that :) :) :0 14:53:07 <ddieterly> (you know, marijuana is legal here) 14:53:17 <ddieterly> if that is your thing 14:53:52 <ddieterly> ok, can we change topics? 14:54:16 <slashme> yes 14:54:17 <ddieterly> btw, if i can improve on this chairman thing, please let me know 14:54:31 <slashme> Is there any topics left ? 14:54:31 <ddieterly> communication skills are not really my strong suit ;-) 14:54:37 <ddieterly> reviews 14:54:38 <vannif_> getting to colorado might be dangerous ... we could never make it back to Ireland ;) 14:55:03 <ddieterly> #topic pending reviews 14:55:24 <ddieterly> szaher i'll take a look at that review today and try out the testing loging 14:55:26 <ddieterly> logic 14:55:36 <slashme> Consitency check : https://review.openstack.org/#/c/300080 14:55:50 <daemontool_> vannif_, is alive!! :) 14:55:50 <szaher> ddieterly: Thanks :) 14:56:01 <ddieterly> szaher what are your concerns about it? 14:56:03 <slashme> It also contains some supposed fix for our swift dsvm errors 14:56:17 <clsacramento> ddieterly: szaher: ty 14:56:39 <daemontool_> slashme, side question, does the bug occure without using --incremental during restore with tar? 14:56:44 <szaher> ddieterly: Just want to check if this is Ok and no problems with it for me it's working fine and I tested it 14:56:47 <daemontool_> did you had chance to see that? 14:56:59 <szaher> clsacramento: You welcome :) 14:57:51 <ddieterly> szaher ok 14:58:12 <slashme> daemontool_: Not really 14:58:28 <daemontool_> ok 14:58:35 <daemontool_> I'm curious :) 14:58:40 <ddieterly> slashme any ideas why that incremental restore thing is working on the 2nd try? 14:59:03 <slashme> I found a solution for one of the tar error (Directory not empty) not the other one (No such file or directory) 14:59:17 <slashme> I'm still doing test to try to understand... 14:59:25 <ddieterly> that is strange 14:59:48 <ddieterly> slashme you want to chair the meeting next week? 15:00:17 <ddieterly> let's continue in the freezer room; out of time 15:00:20 <ddieterly> ciao! 15:00:26 <slashme> ddieterly: thatnk you :) 15:00:34 <ddieterly> #endmeeting