12:02:10 <rafi> #startmeeting 12:02:10 <zodbot> Meeting started Wed Oct 7 12:02:10 2015 UTC. The chair is rafi. Information about MeetBot at http://wiki.debian.org/MeetBot. 12:02:10 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 12:02:44 <rafi> #info agenda for todays meeting can be found here https://public.pad.fsfe.org/p/gluster-community-meetings 12:02:53 <rafi> #topic Roll Call 12:03:12 <rafi> let'us see who all are here with us for this week meeting 12:03:40 * hagarth here (partially) 12:03:55 * lpabon here 12:03:57 * overclk is kind of here.. 12:04:02 * kkeithley is here 12:04:18 * raghu is here 12:04:32 <rafi> let'us wait some more time :) 12:05:35 <rafi> not many people for today 12:05:41 <hagarth> rafi: go ahead, folks usually join at various points in time. 12:05:43 <rafi> let's move on 12:06:04 <kkeithley> some people are in Dublin at LinuxCon, probably busy. 12:06:15 <rafi> #topic Action Items from last week 12:06:25 <rafi> I'll be skipping any item that's been marked as done. 12:06:30 * rjoseph is late 12:06:41 <rafi> #topic kshlm to check back with misc on the new jenkins slaves 12:06:57 <rafi> looks like kshlm is not here 12:07:04 <rafi> what about misc ? 12:07:30 <rafi> any one have update on this ? 12:08:01 <rafi> I will carry forward this for next week 12:08:10 <hagarth> rafi: yes, that makes sense. 12:08:16 <rafi> #action kshlm to check back with misc on the new jenkins slaves 12:08:19 <rafi> next one 12:08:35 <rafi> #topic krishnan_p will add information about GlusterD-2.0 to the weekly news 12:09:05 <rafi> I haven't checked the weekly news 12:09:17 <rafi> is this done already ? 12:09:18 <hagarth> we did not add it in the last edition 12:09:28 <rafi> okey 12:09:58 <rafi> BTB we decided to have gluster-news by bi-weekly, right ? 12:10:11 <hagarth> rafi: that's right 12:10:46 <rafi> cool, let'us check this on next week meeting 12:10:58 <rafi> #action krishnan_p will add information about GlusterD-2.0 to the weekly news 12:11:11 <rafi> #topic krishnan_p and atinmu will remind developers to not work in personal repositories, but request one for github.com/gluster 12:11:51 <rafi> atin updated me that they will do this by this week 12:12:06 <rafi> so let's move this for next week 12:12:18 <rafi> #action krishnan_p and atinmu will remind developers to not work in personal repositories, but request one for github.com/gluster 12:12:44 <rafi> #topic krishnan_p will send an email to the -devel list about merging the glusterd-2.0 work into the main glusterfs repo 12:13:14 <jdarcy> For work that's not disruptive to existing functionality, I'd even suggest working directly off gerrit master. 12:13:49 <hagarth> +1 12:13:55 <rafi> jdarcy +1 12:14:19 <kkeithley> that's what branches are for! 12:14:20 <lpabon> just my two cents -- this is confusing when the repo is in Github.. cannot use pull requests, and must send a patch using gerrit.. 12:15:22 <lpabon> i know for those living, breathing gerrit it makes perfect sense.. but for others, it seems weird 12:15:24 <JustinClift1> If only there was a way to have patches to GitHub automatically become Gerrit ones... 12:15:43 <rafi> As kkeithley and jdarcy suggested how about creating a new branch in gerrit itself and work on that 12:16:25 * rastar joins now 12:16:28 <kkeithley> you can push into a branch in github. Only when you want it to finally be merged into gluster do you need to use gerrit and rfc.sh 12:16:52 <jdarcy> If you want your changes to be invisible to the majority of the Gluster developer community so you're not bothered by actual review feedback, by all means do them on Github. 12:17:25 <rastar> or use the draft feature in gerrit 12:17:42 <lpabon> jdarcy: correct. If reviews come in through gerrit, then i suggest using gerrit WIPs and Drafts. 12:17:47 <kkeithley> I meant you can do simple git push into a branch in git.gluster.org. Only when you want it to finally be merged into gluster do you need to use gerrit and rfc.sh 12:17:59 <hagarth> you also miss CI help while working directly in github and not on gerrit 12:18:25 <lpabon> does rfc.sh support Draft option? 12:18:36 <lpabon> i know that gerrit and git-review do 12:18:52 <hagarth> if somebody posts a patchset worth several KLOC without any context, I am going to block those patches certainly ;) 12:18:57 <hagarth> lpabon: use rfc as the topic 12:18:59 * msvbhat arrives late to the meeting 12:19:09 <rafi> DO we need a poll to make a final call ? :) 12:19:57 <hagarth> reviewing patches worth several KLOC is a major PITA to say the least. 12:20:36 <lpabon> at least there is a benefit from using the CI system 12:20:45 <rafi> Seems like gerrit is the best home for Glusterd2.0 12:20:53 <hagarth> lpabon: yes 12:21:29 <rastar> hagarth: by draft I think jdarcy and lpabon meant the hidden draft option in gerrit 12:21:40 <rastar> the change is not even visible to others 12:21:41 <lpabon> rastar: correct 12:21:51 <lpabon> unless they search for Draft 12:22:07 <rastar> and you can choose to show it a specific list of people 12:22:16 <lpabon> it is a way used in openstack .. they can also specify a topic if they want it visible 12:22:31 <hagarth> rastar, lpabon: why hide something? because it is not mature for review yet? 12:22:40 <lpabon> yes 12:22:52 <lpabon> Work in progress like 12:22:56 <lpabon> Not worth reviewing 12:22:58 <hagarth> maybe we can have a different topic, possibly WIP 12:23:06 <rastar> I remember using it before but cant remember if I used git-review or ./rfc 12:23:17 <lpabon> hagarth: using WIP is the norm in openstack 12:23:23 <rafi> ok 12:23:43 <rafi> is this a right time to create branch for glusterd2.0 ? 12:23:51 <rastar> it should be a trivial patch for ./rfc.sh to get that done 12:23:54 <kkeithley> I barely have time to look at my real work, never mind looking at WIP, Draft, or other immature code. Who cares if it's not ready. Just use gerrit 12:24:34 <lpabon> kkeithley: Yes, that is what we mean. In gerrit, send a patch, but notify reviewers that the change is still in-progress 12:24:35 <hagarth> rafi: no branches please 12:24:54 <hagarth> rafi: have a separate directory if need be 12:24:55 <rafi> hagarth: okey 12:25:42 <jdarcy> The one caveat I'd have about using Gerrit with topic=rfc (or similar) for complex changes is that we'd need to be smarter about when we trigger Jenkins runs. Having the queue clogged with big stacks of RFC/WIP/whatever patches would be bad. 12:26:02 <rafi> #topic krishnan_p will send an email to the -devel list about merging/sending the glusterd-2.0 work into the gerrit 12:26:13 <rafi> moving into next topic 12:26:22 <rafi> sorry 12:26:25 <hagarth> jdarcy: agree., let us discuss this on -devel 12:26:31 <jdarcy> hagarth: +1 12:26:44 <rafi> #action krishnan_p will send an email to the -devel list about merging/sending the glusterd-2.0 work into the gerrit 12:27:02 <rafi> #topic poornimag to send a mail on gluster-devel asking for volunteers to backport glfs_fini patches to release-3.5 12:27:08 <rafi> I guess it is done, 12:27:28 <rafi> I remember seeing one mail regarding this 12:28:18 <lpabon> jdarcy: sounds like we need elastic compute ;-) 12:28:53 <rafi> moving into the next topic 12:29:10 <rafi> #topic kkeithley will reply to his previous email, confirming that End-Of-Life bugs will be closed 12:29:18 <rafi> kkeithley: your turn ? :) 12:29:40 <kkeithley> I've closed some, continuing to close remaining 12:30:20 <rafi> kkeithley: Do you have any script to close those bugs ? 12:30:40 <kkeithley> nope, I don't think there's that many. 12:30:49 <kkeithley> figuring out how to script it would probably take longer 12:30:57 <rafi> :) 12:30:59 <kkeithley> I'm guessing 12:31:04 <rafi> that's right 12:31:36 <rafi> I will mark this AI as done 12:32:05 <rafi> #topic kkeithley will close all the EOL'd bugs with a note 12:32:15 <kkeithley> someone who already knows how to do bugzilla queries could probably do it in their sleep ;-) 12:32:54 <rafi> :) that helps for future also 12:33:51 <rafi> kkeithley: since not every bug is closed, I will keep this AI 12:33:58 <kkeithley> yeah 12:34:05 <rafi> #action kkeithley will close all the EOL'd bugs with a note 12:34:10 <rafi> next topic 12:34:31 <rafi> #topic jdarcy (and/or others) will post version of the NSR spec "pretty soon" 12:35:14 <rafi> I don't remember seeing a mail about the same 12:35:22 <rafi> jdarcy: any update ? 12:35:36 <jdarcy> I'll bug Avra about his part. I'm still writing the journal part. 12:36:28 <rafi> coool, Do you guys need more than one week or one is enough ? 12:36:40 <rafi> jdarcy: ^ 12:36:49 <jdarcy> One should be enough, especially on top of the three or four we've already had. ;) 12:37:27 <rafi> #action jdarcy avra will post version of the NSR spec "pretty soon" 12:37:31 <rafi> thanks jdarcy 12:37:52 <rafi> #topic kshlm to clean up 3.7.4 tracker bug 12:38:55 <rafi> since kshlm is not here carrying this for next week 12:39:13 <rafi> #action kshlm to clean up 3.7.4 tracker bug 12:39:23 <rafi> #topic overclk will get the dht-scalability doc in glusterfs-specs update to the latest design 12:39:36 <rafi> overclk: any update ? 12:39:37 <overclk> rafi, I sent out a pull request 12:39:42 <overclk> on github.. 12:40:05 <rafi> overclk: great, thanks for the update 12:40:09 <overclk> rafi, it still needs some more work as there are missing parts. 12:40:21 <overclk> rafi, so, I'll be sending out other parts too (later). 12:40:38 <rafi> overclk: okey 12:40:59 <rafi> overclk: I will put a AI to keep track of this 12:42:00 <rafi> #action overclk will send another patch to enhance latest dht-scalability design doc in glusterfs-specs 12:42:18 <rafi> #topic poornimag and skoduri will write a trip report for SDC 12:42:47 <rafi> I guess they haven't done this yet 12:43:03 <rafi> will move to the next week 12:43:09 <rafi> #action poornimag and skoduri will write a trip report for SDC 12:43:22 <rafi> #topic obnox to write up a report on his trip to SDC. 12:43:29 <rafi> obnox_: any update ? 12:43:58 <rafi> #action obnox to write up a report on his trip to SDC 12:44:09 <rafi> #topic pranithk will update the community on 3.7.5 release plans. 12:45:19 <rafi> #info glusterfs-3.7.5 will be released by end of the day mostly 12:45:55 <hagarth> rafi: cool update, thanks! 12:46:04 <rafi> #action pranithk will send out a mail to the community on 3.7.5 release and updates 12:46:13 <rafi> moving on 12:46:25 <rafi> #topic raghu to call for volunteers and help from maintainers for doing backports listed by rwareing to 3.6.7 12:46:45 <rafi> raghu: any updates ? 12:47:40 <raghu> rafi: I have to talk to dht and afr maintainers. 12:48:09 <rafi> raghu: no issues , will keep this for next week 12:48:13 <raghu> rafi: sure 12:48:18 <rafi> #action raghu to call for volunteers and help from maintainers for doing backports listed by rwareing to 3.6.7 12:48:31 <rafi> #topic All the leads to update community on the outcomes of the design discussions. 12:48:58 <rafi> any update on this ? 12:49:46 <rafi> I think most of them were busy 12:50:28 <jdarcy> In many cases, those discussions will be reflected in the upcoming specs (other AIs). 12:50:48 <jdarcy> Certainly that's true for NSR. 12:50:57 <rafi> we can expect the community discussion updates by next week 12:51:16 <rafi> jdarcy: agreed 12:51:36 <rafi> moving into the last AI item of a very long list :D 12:51:45 <rafi> #topic msvbhat to send out an update on DiSTAF once he pushes all his new changes and creates repo under github.com/gluster 12:52:08 <hagarth> rafi: I think this has happened 12:52:17 <rafi> hagarth: ya 12:52:19 <hagarth> msvbhat is looking for some inputs 12:52:24 <hagarth> so we are good here 12:52:31 <msvbhat> eah, I have sent the mail 12:52:49 <rafi> okey cool, I will mark this as done 12:53:05 <rafi> so that ends out AI from last week 12:53:13 <rafi> #topic GlusterFS 3.7 12:53:29 <hagarth> i think we touched upon 3.7.5 update previously 12:53:38 <rafi> ya 12:53:39 <hagarth> think we can move on here 12:53:44 <rafi> sure 12:53:51 <rafi> #topic GlusterFS 3.6 12:54:13 <rafi> raghu: your turn ? 12:54:51 <raghu> rafi: not much. 3.6.7 will be released on 20th. Waiting for patches. 12:55:22 <rafi> #info glusterfs-3.6.7 will be released on 20th of this month 12:55:35 <rafi> moving into the next topic 12:55:41 <rafi> #topic GlusterFS 3.5 12:56:06 <rafi> since our 3.5 maintainer is participating linux conf , any one else have any update ? 12:56:56 <rafi> I guess nothing is there to discuss for today 12:57:09 <rafi> #topic Glusterfs 3.8 12:57:14 <rafi> hagarth: ? 12:58:19 <hagarth> rafi: etherpad for 3.8 has updates on features 12:58:30 <hagarth> plan is to convert it into a project tracking page on gluster.org 12:58:45 <hagarth> if you have more features planned for 3.8, please include them in the etherpad 12:58:45 <rafi> okey 12:58:57 <hagarth> will help me do a one shot conversion to the project tracking page ;) 12:59:32 <rafi> #info please add any future targeted for 3.8 into the tracking ether pad 13:00:02 <rafi> hagarth: sure 13:00:32 <rafi> please every one take your mic and start speaking :D 13:01:00 <hagarth> rafi: let us move on 13:01:04 <rafi> sure 13:01:19 <rafi> #topic GlusterFS 4.0 13:01:33 <rafi> jdarcy: any update for glusterfs.next ? 13:01:43 <msvbhat> . 13:02:00 <jdarcy> Lots of good discussions last week, mostly reflected in per-subproject AIs. 13:02:38 <rafi> jdarcy: good to hear we are moving fast with lot of passion :) 13:03:04 <rafi> moving into the last topic 13:03:07 <rafi> #topic Open floor 13:03:17 * jdarcy falls in. 13:03:18 <rafi> please bring your own topic here 13:05:10 <rafi> cool, looks like every one are done for today 13:05:14 <rafi> cool 13:05:19 <jdarcy> W00t. 13:06:05 <rafi> Considering the time, I'm going to stop the meeting 13:06:39 <rafi> we are little late to finish the meeting, please every one excuse for me 13:06:53 <rafi> Thanks all for participating 13:07:00 <rafi> #endmeeting