12:00:31 <kkeithley> #startmeeting
12:00:32 <zodbot> Meeting started Wed Oct 28 12:00:31 2015 UTC.  The chair is kkeithley. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:00:32 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
12:00:56 <kkeithley> #topic Roll Call
12:00:59 * partner waves
12:01:04 * rastar is here
12:01:54 * tigert is here
12:02:14 * overclk is around
12:02:44 <kkeithley> let's wait another minute and then we'll get started
12:02:54 * kshlm is here
12:03:19 * ira is around here somewhere.
12:03:23 * poornimag here
12:03:42 <kkeithley> I think we have quorum
12:03:58 <kkeithley> AIs from last week
12:04:05 * skoduri is here
12:04:10 <kkeithley> #topic kshlm to check back with misc on the new jenkins slaves
12:04:31 <kshlm> I've not done either of my AIs. Sorry.
12:04:32 * atinm is here
12:04:43 <kkeithley> okay, next week?
12:04:55 <kshlm> Yes.
12:05:03 <kkeithley> #action kshlm  to check back with misc on the new jenkins slaves
12:05:08 * jiffin is here
12:05:14 <kkeithley> #topic krishnan_p and atinmu will remind developers to not work in personal repositories, but request one for github.com/gluster
12:05:37 <kkeithley> atinm?
12:05:40 <atinm> kkeithley, its not done yet
12:05:53 <kkeithley> by next week?
12:05:53 <atinm> kkeithley, I will do that this week
12:05:57 <atinm> kkeithley, sure
12:06:01 <kkeithley> #action krishnan_p and atinmu will remind developers to not work in personal repositories, but request one for github.com/gluster
12:06:02 <kshlm> If this was w.r.t glusterd2, we've moved the repo under the gluster org.
12:06:21 <kkeithley> I think it's a general thing, not just for glusterd 2.0
12:06:25 <atinm> kshlm, yes, but not sure about how is dht2 going now
12:06:44 <kkeithley> #topic ndevos send out a reminder to the maintainers about more actively enforcing backports of bugfixes
12:06:46 <atinm> I think dht2 has decided to contribute in main repo itself
12:07:04 <atinm> anyways I will send a mail on this, thanks
12:07:05 <overclk> atinm, as of now we're using shyams repo in gerrithub
12:08:11 <atinm> overclk, ok
12:09:15 <atinm> kkeithley, I've not seen any mail from ndevos on this topic
12:09:42 <kkeithley> and he's 50% on PTO today. Let's leave this open for next week
12:09:58 <kkeithley> #topic skoduri, poornimag and obnox_ to post SDC trip report on gluster-devel
12:10:46 <kkeithley> skoduri?
12:10:47 <poornimag> obnox had mentioned he is working on it, can be moved for next week
12:11:03 <kshlm> I think I saw obnox posting something on this.
12:11:12 <kshlm> But don't remeber when or where.
12:11:21 <atinm> kshlm is right, its in sme-storage if I am not wrong
12:11:45 <ira> What is sme-storage?
12:11:47 <kkeithley> which is not a community mailing list
12:12:19 <poornimag> yup it was in internal mailing list, but not sent on devel
12:12:44 <kkeithley> #action skoduri, poornimag and obnox_ to forward SDC trip report(s) to gluster-devel
12:12:51 <kkeithley> :-(
12:13:38 <kkeithley> gah, etherpad....
12:13:50 * anoopcs arrives late
12:14:29 <kkeithley> #topic raghu to call for volunteers and help from maintainers for doing backports listed by rwareing to 3.6.7
12:14:36 * kkeithley thinks he has seen that
12:14:53 <kkeithley> raghu: any response?
12:15:02 <kshlm> raghu isn't in today.
12:15:13 <kkeithley> yup, I didn't think I'd seen him
12:15:27 <kshlm> And he hasn't called for volunteers as well.
12:16:21 <kkeithley> #topic kshlm to clean up 3.7.4 tracker bug
12:16:34 <kkeithley> #action  kshlm to clean up 3.7.4 tracker bug by next week for sure
12:16:48 <kshlm> I'm doing it now. I will be done by next week.
12:16:58 <kkeithley> kewl
12:17:07 <kkeithley> #topic hagarth to post a tracking page on gluster.org for 3.8 by next week's meeting
12:17:16 <kkeithley> etherpad says that's in progress
12:17:28 <kkeithley> #topic rafi to setup a doodle poll for bug triage meeting
12:17:40 * kkeithley just saw rafi left the room
12:17:51 <kkeithley> so no update on that probably. I haven't seen a poll
12:17:58 * obnox joins late...
12:18:17 <obnox> (sorry - confused by daylighy saving time changes in .de)
12:18:31 <kkeithley> #action rafi to setup a doodle poll for bug triage meeting
12:19:26 <kkeithley> obnox: no worries. 14:00 CEST is now 13:00 CET?
12:20:19 * kkeithley updates times on the etherpad
12:21:15 <obnox> kkeithley: yep
12:21:20 <kkeithley> obnox: please forward your SDC trip report to gluster-devel list
12:21:44 <obnox> kkeithley: right. will do. didn't manage before last week's pto - apologies
12:21:59 <kkeithley> #topic rastar and msvbhat to publish a test exit criterion for major/minor releases on gluster.org
12:22:25 <kkeithley> rastar: ?
12:22:34 <rastar> This is not done yet. I am working on it and will work with msvbhat to finish by next week
12:22:46 <kkeithley> #action rastar and msvbhat to publish a test exit criterion for major/minor releases on gluster.org
12:23:24 <kkeithley> #topic hagarth to review http://review.gluster.org/#/c/12210/
12:23:29 <kkeithley> this is FOP sampling
12:23:34 <kkeithley> etherpad says partially done
12:23:48 <kkeithley> #action hagarth to finish review http://review.gluster.org/#/c/12210/
12:24:11 <kkeithley> #topic atinm to send a monthly update for 4.0 initiatives
12:24:29 <kkeithley> did we set a date, e.g. end of every month?
12:24:48 <atinm> kkeithley, first week of every month is a right time
12:25:11 <kkeithley> so we'll see one next week?
12:25:17 <atinm> kkeithley, this time we need to send updates for last couple of months since we missed to send September's updates
12:25:21 <atinm> kkeithley, yes
12:25:25 <kkeithley> excellent
12:25:45 <kkeithley> #action atinm to send a monthly update for 4.0 initiatives, including summarize last couple of months
12:26:05 <kkeithley> #topic GlusterFS 3.7
12:26:33 <kkeithley> are we going to get 3.7.6 on the 30th?
12:26:46 <rastar> I sent a mail on this today
12:26:51 <rastar> with tracker bug
12:27:04 <kkeithley> okay. Are there even any patches for it?
12:27:11 <rastar> requests from people suggest that 3 days might not be enough for them
12:27:23 <rastar> 42 patches merged and 33 in review
12:28:14 <kkeithley> 33 _more_ in review? Wow
12:28:24 <kkeithley> on top of the 42 already merged?
12:28:41 <rastar> yes, not all have been added to tracker so I am guessing priority is not so high
12:28:49 <rastar> I am expecting around 10 more for sure
12:29:25 <rastar> Should I wait till next wednesday for tagging or go ahead with tagging this weekend?
12:29:27 <kshlm> rastar, I suggest you ask the contributors of those changes to add their bug to the release-blocker list if they want it in the release.
12:29:38 <kkeithley> that's certainly worthy of making the release. The rest in the queue for 3.7.7 then.
12:29:44 <kshlm> Don't wait for any not in the list.
12:29:57 <rastar> kshlm: ok, have requested in mail today
12:30:16 <kshlm> Set a hard deadline for changes that are absolutely required.
12:30:25 <rastar> Ok :)
12:30:31 <kshlm> rastar, I haven't read it yet.
12:32:01 <kkeithley> Use best judgment about waiting. If something is really critical.  Otherwise let's try to stick to the schedule. Packagers have $dayjob and we don't want them to get bogged down
12:32:42 <rastar> thanks
12:32:52 <rastar> here is the archive link for mail http://www.gluster.org/pipermail/gluster-devel/2015-October/047020.html
12:33:02 <rastar> I will update in the same chain
12:33:09 <kkeithley> thanks
12:33:18 <kkeithley> anything else on 3.7?
12:33:40 <rastar> status change for tierin?
12:33:48 <rastar> *tiering
12:34:02 <rastar> I see 21 patches merged for it
12:34:09 <atinm> I think we have been getting complaints about 3.7.4 to 3.7.5 upgrades
12:34:15 <atinm> is anyone looking into it?
12:35:02 <kkeithley> are we expecting a status change for tiering?
12:35:31 <atinm> If I am not wrong I have seen atleast three different users reporting issue post upgrade, so if we don't fix it we would continue to have the same issue in 3.7.6, a word of caution for rastar
12:35:39 <rastar> one of the patches removed the experimental warning for attach tier, hence the question
12:36:47 <kkeithley> is there a ten word summary of the issue?
12:37:05 <kkeithley> or just that things didn't work smoothly?
12:37:33 <rastar> here is the summary from what I understood
12:37:59 <rastar> Upgrading a subset of cluster to 3.7.5 leads to issues with glusterd commands
12:38:13 <rastar> they fail with staging failed error
12:38:21 <kkeithley> has anyone opened a bugzilla for it?
12:38:30 <atinm> rastar, yes, that's what I remember
12:39:06 <atinm> I don't think we have a BZ yet for this
12:39:25 <kkeithley> since we haven't triaged bugs in about three weeks...
12:39:55 <kkeithley> would someone volunteer to open a BZ, or see if one already exists?
12:40:08 <kkeithley> would someone like to volunteer? Please?
12:40:09 <rastar> I will do that .
12:40:32 <atinm> rastar, thank you
12:40:34 <kkeithley> #action rastar will open a BZ for 3.7.5 upgrade issue with glusterd commands
12:40:40 <kkeithley> rastar++
12:40:56 <atinm> and we need volunteers to analyze the BZ as well :)
12:41:08 <kkeithley> yes, there's that too. ;-)
12:41:44 <kkeithley> and what do we need to sort out for tiering status, anything?
12:42:23 <kkeithley> hearing nothing....
12:42:27 <kkeithley> anything else for 3.7?
12:42:44 <kkeithley> no?
12:42:52 <kkeithley> #topic GlusterFS 3.6
12:43:50 <kkeithley> we shipped 3.6.6.  Is there a new tracker for 3.6.7?
12:44:23 <kshlm> There is glusterfs-3.6.7
12:44:35 <kshlm> https://bugzilla.redhat.com/show_bug.cgi?id=glusterfs-3.6.7
12:45:02 <kkeithley> yup, I've updated the etherpad
12:45:25 <kkeithley> who's on deck for doing 3.6.7?
12:45:46 <kshlm> raghu is the release-maintainer for 3.6, so he should be doing it.
12:46:20 <kshlm> Should we to release-managers for all releases?
12:47:03 <kkeithley> is there a difference between -maintainer and -manager?
12:47:32 <kshlm> As I understand, a release-maintainer manages the whole release-branch and does releases on that branch.
12:48:09 <kshlm> For the 3.7 release branch, we started doing release-managers, who would just take up the responsibilities for on release.
12:48:34 <kkeithley> yup. If we need to do that for the 3.6 and 3.5 branches, then we should.
12:49:00 <kkeithley> anyway, we don't seem to have any update on 3.6.
12:49:46 <kkeithley> moving on then
12:49:52 <kkeithley> #topic GlusterFS 3.5
12:49:53 <kshlm> Yup. The only thing to be done with 3.6 seems to be rwareing's list of backports
12:50:36 <kkeithley> ndevos: ?
12:51:43 <kkeithley> I guess the 50% of ndevos isn't monitoring this channel.  No news then
12:51:56 <kkeithley> moving on
12:52:04 <kkeithley> #topic GlusterFS 3.8
12:52:08 <kkeithley> Any status here
12:52:10 <kkeithley> ?
12:52:53 <kkeithley> I guess not.
12:52:56 <kkeithley> moving on
12:53:02 <kkeithley> #topic GlusterFS 4.0
12:53:14 <kkeithley> how about now, any status?
12:53:23 <atinm> I will talk about it
12:53:33 <kkeithley> excellent
12:53:42 <overclk> me too (after atin)
12:54:24 <kkeithley> okay, the floor is yours
12:54:28 * shyam raises his hand, to resolve http://review.gluster.org/#/c/12321/ relating to 4.0/Next
12:54:40 <atinm> GlusterD 2.0 team was working on volume-create end to end flow and peer commands ReST API, we have also started writing the API doc and the same should be up for review in a day or two
12:55:08 <atinm> I will also put up the design document for review in a week or two
12:55:16 <atinm> that's about GlusterD 2.0
12:55:38 * kshlm will be afk for a couple of minutes
12:55:47 <kkeithley> #action atinm will also put up the GlusterD 2.0 design document for review in a week or two
12:56:06 <overclk> shyam, I'll take a look at #12321
12:56:25 <shyam> overclk: It is more than a look, are we decided on 'experimental' in master?
12:56:42 <shyam> Wanted to put that discussion to rest, or into motion (as you see it)...
12:57:15 <overclk> shyam, to decide we could probably start out a mail of gluster-devel and see what the community has to say.
12:57:38 <shyam> overclk: Already done, and mostly in agreement. So what needs to be done to move this patch along?
12:57:49 <overclk> shyam, ah ok! I recall now.
12:59:13 <overclk> shyam, need more reviews? more eyes on the patch?
12:59:13 <shyam> Well if the mail discussion on devl ended in agreement, I guess a request to devel to push this change along would be the next step, others?
12:59:42 <atinm> shyam, +1
12:59:49 <shyam> overclk: sort of, we need some eyes, and close on it, so that things can start appearing on master...
13:00:16 <overclk> shyam, that's the reason I told I'll have a look :)
13:01:21 <kkeithley> Is there an action we need to take then?
13:01:35 <shyam> Ok, so I will ping devel again on this and overclk will review the patch.
13:01:46 <overclk> yep
13:02:27 <kkeithley> #action overclk to review http://review.gluster.org/#/c/12321/
13:02:36 <kkeithley> anything else on 4.0?
13:03:02 <shyam> overclk: Want to update on DHT2?
13:03:11 <overclk> shyam, sure.
13:03:38 <overclk> myself and Shyam have started out prototype (well it's more than that) and things are going out well
13:04:14 <overclk> we were able to get lookup() working and kind of heading towards other ops such as create, mkdir.
13:04:35 <overclk> So, by next week, things would look more interesting
13:04:51 <kkeithley> will a demo be possible sometime soon?
13:05:31 <shyam> kkeithley: We intend to get a demo when we have *sufficient* FOPs in place
13:05:37 <overclk> Also, every Thursday (if not every then 2nd), we plan to have an hangout/BJ session
13:05:44 <shyam> We will show smaller recorded demos as above ^^^
13:05:50 <kkeithley> kewl
13:06:01 <overclk> So, anyone interested are free to participate.
13:06:33 * shyam puts his hand down now...
13:06:34 <overclk> This is starting from next Thursday, correct Shyam?
13:06:46 <shyam> overclk: yes
13:06:52 <shyam> (maybe this Friday ;) )
13:06:58 <shyam> but that would be a trial run
13:07:13 <kkeithley> is it bluejeans, or g'hangouts? Or both?
13:07:17 <overclk> ah! yeh and too less time to send an invite out to community ;)
13:07:49 <shyam> I think, bluejeans and recording will be posted (we will get back on that?)
13:07:55 <overclk> kkeithley, whatever works without much fuss...
13:08:08 <kkeithley> lol, do either work without a fuss?
13:08:12 <shyam> :)
13:08:37 <overclk> kkeithley, G'hangouts works for me all the time, not the other one. So, YMMV.
13:08:46 <kkeithley> we'll look for the announce in -devel
13:08:59 <kkeithley> okay, we're already ~10 over. Anything else on 4.0?
13:09:10 <overclk> done from my side.
13:09:34 <kkeithley> one last item. Hopefully quick.
13:09:39 <kkeithley> #topic open floor
13:10:18 <kkeithley> what about having a bot or pseudo/faux email user for gerrit and bugzilla?
13:10:29 <kkeithley> csim: any thoughts re: ^^^
13:10:38 <kkeithley> or anyone else for that matter?
13:10:40 <kshlm> We have one for bugzilla.
13:10:51 <csim> kkeithley: same as for github question on gluster-infra :)
13:10:58 <kkeithley> yes
13:11:13 <csim> ( ie, who get the alias, and what is the use of the said alias )
13:11:15 <kshlm> csim, Yes. I added the same here so we could get some more opinions.
13:11:51 <kshlm> We want to create a bot account on github, so that we can add ssh keys to the bot account and have gerrit replicate to github.
13:11:58 <kkeithley> we have bug@gluster.org as a default assignee.  I'm talking about an email address that can be a pseudo user with a bugzilla account to update BZs.  Right now we use hagarth's account.
13:12:05 <kkeithley> yes, all of that
13:12:11 <kshlm> But there were a couple of questions.
13:12:33 <kshlm> - Who would recieve the emails sent to the bot accounts address
13:12:45 <kshlm> - How do we do 2FA?
13:12:58 <kkeithley> What is 2FA?
13:13:03 <csim> 2 factor authentication
13:13:06 <kshlm> 2 factor authentication
13:13:17 <kkeithley> doh
13:13:23 <csim> but this one was specific to github
13:13:44 <kkeithley> are you saying that a pseudo account has problems with 2FA?
13:14:03 <csim> well, in the case of github, whose phone is the token ?
13:14:07 <kkeithley> e.g. a gemalto token
13:14:24 <csim> it was in the specific github case, and gemalto is not a option
13:14:58 <csim> (and there is only 1 single token on github)
13:14:59 <kkeithley> yes, e.g. = for example.
13:15:28 <csim> the problem is that the owner of the 2nd factor become suddenly a single point of failure
13:15:33 <kkeithley> or exempli gratia, if you're into the Latin
13:15:35 <csim> and that's something we should avoid
13:15:46 <kkeithley> okay
13:15:53 <csim> ie, with a token, who keep the token, etc
13:16:05 <kkeithley> Aren't there other soft token options besides a phone?
13:16:15 <csim> if we could disable the authentication and delegate setting or anything to a group, that would be perfect
13:16:18 <kkeithley> Or they're proprietary
13:16:54 <csim> kkeithley: then we have to distribute the seed around
13:17:07 <csim> I am not sure how "easy" it is to setup
13:17:25 <csim> since (IIRC), github give you a qrcode to scan
13:17:26 <kkeithley> well, this sounds too complicated to resolve now, esp. given we're already 15min over
13:17:48 <kshlm> We can continue this on the gluster-infra thread.
13:18:02 <kshlm> I just put it so that we could discuss it if we had time.
13:18:05 <kkeithley> yes, let's take this to gluster-infra
13:18:10 <kkeithley> sure
13:18:15 <csim> I would rather have discussion on ML :)
13:18:28 <kkeithley> absolutely. ;-)
13:18:38 <kkeithley> anything else then before we wrap?
13:18:52 <kkeithley> going once?
13:18:56 <kkeithley> going twice?
13:19:14 <kkeithley> #endmeeting