Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2016-08-31

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:33 EinstCrazy joined #gluster-dev
01:47 ilbot3 joined #gluster-dev
01:47 Topic for #gluster-dev is now Gluster Development Channel - http://gluster.org | For general chat go to #gluster | Patches - http://review.gluster.org/ | Channel Logs - https://botbot.me/freenode/gluster-dev/ & http://irclog.perlgeek.de/gluster-dev/
02:03 EinstCra_ joined #gluster-dev
02:24 EinstCrazy joined #gluster-dev
02:36 spalai joined #gluster-dev
02:37 spalai joined #gluster-dev
03:05 magrawal joined #gluster-dev
03:31 kdhananjay joined #gluster-dev
03:39 aspandey joined #gluster-dev
03:51 shubhendu joined #gluster-dev
03:52 itisravi joined #gluster-dev
03:59 kdhananjay1 joined #gluster-dev
04:07 atinm joined #gluster-dev
04:08 riyas joined #gluster-dev
04:08 kdhananjay joined #gluster-dev
04:11 sankarshan_ joined #gluster-dev
04:17 sankarshan_ joined #gluster-dev
04:17 k4n0 joined #gluster-dev
04:40 ndarshan joined #gluster-dev
04:43 ppai joined #gluster-dev
04:44 shubhendu joined #gluster-dev
04:51 nbalacha joined #gluster-dev
04:54 ndarshan joined #gluster-dev
05:05 karthik_ joined #gluster-dev
05:14 Byreddy joined #gluster-dev
05:21 prasanth joined #gluster-dev
05:28 aspandey joined #gluster-dev
05:32 msvbhat joined #gluster-dev
05:34 ankitraj joined #gluster-dev
05:50 skoduri joined #gluster-dev
05:53 Bhaskarakiran joined #gluster-dev
05:55 nbalacha joined #gluster-dev
05:55 hgowtham joined #gluster-dev
05:56 Saravanakmr joined #gluster-dev
05:58 ppai joined #gluster-dev
05:59 Muthu_ joined #gluster-dev
06:00 ashiq joined #gluster-dev
06:03 ndarshan joined #gluster-dev
06:07 mchangir joined #gluster-dev
06:18 ramky joined #gluster-dev
06:18 msvbhat joined #gluster-dev
06:24 guest joined #gluster-dev
06:26 kotreshhr joined #gluster-dev
06:29 shubhendu joined #gluster-dev
06:32 csaba joined #gluster-dev
06:39 spalai joined #gluster-dev
06:40 Manikandan_ joined #gluster-dev
06:41 rraja joined #gluster-dev
06:41 guest Is it possible to use bd xlator with crypt xlator?
06:45 spalai joined #gluster-dev
06:47 kshlm joined #gluster-dev
06:50 msvbhat joined #gluster-dev
06:58 nbalacha do we support the UFS filesystem for GLusterfs bricks on FreeBSD?
06:58 nbalacha From the FreeBSD chmod page:
06:58 nbalacha The FreeBSD VM system totally ignores the sticky bit (S_ISVTX) forexe-
06:58 nbalacha cutables.On UFS-based file systems (FFS,LFS) the stickybit mayonly
06:58 nbalacha beset upon directories.
06:58 nbalacha that completely breaks the linkfile implementation for DHT
06:59 rastar joined #gluster-dev
07:03 an_ joined #gluster-dev
07:11 nbalacha joined #gluster-dev
07:17 pur joined #gluster-dev
07:18 ndevos nbalacha: oh, thats a shame, maybe that is the reason why many FreeBSD users run on ZFS
07:19 devyani7 joined #gluster-dev
07:20 nbalacha ndevos, I have a testfile which when run as root seems to set the T bit on files even in UFS. But I could not figoure out how to update and build the port after including these changes.
07:20 nbalacha but I am reluctant to recommend UFS bricks with this limitation
07:21 poornima joined #gluster-dev
07:21 ndevos nbalacha: yes, I agree that the T-bit is rather crucial
07:23 poornima nigelb, Can you please abort the netbsd regressions:  https://build.gluster.org/job/netbsd7-regression/432/    https://build.gluster.org/job/netbsd7-regression/430/console   https://build.gluster.org/job/netbsd7-regression/426/
07:23 poornima nigelb, they are hung for long time, thanks
07:24 kshlm poornima, nigelb is away today.
07:25 poornima kshlm, oh ok, would you be able to kill them?
07:25 kshlm All of them have been stopped.
07:25 poornima kshlm, thanks
07:25 poornima kshlm++
07:25 glusterbot poornima: kshlm's karma is now 108
07:27 asengupt joined #gluster-dev
07:28 poornima ndevos, have addressed your comments on upcall readdirp patch http://review.gluster.org/#/c/15313/ , can you please look at it?
07:29 poornima ndevos, also the io-stats patch http://review.gluster.org/#/c/15193/ thanks
07:32 an_ joined #gluster-dev
07:36 ndevos poornima: do you know if it makes readdirp noticibly slower?
07:37 jiffin joined #gluster-dev
07:41 ndevos poornima: also, http://review.gluster.org/15313 depends on the invalidation added to md-cache, right? has that one been merged yet?
08:32 aravindavk joined #gluster-dev
08:39 kdhananjay joined #gluster-dev
08:40 an_ joined #gluster-dev
08:45 atalur joined #gluster-dev
08:45 amye joined #gluster-dev
08:54 atalur_ joined #gluster-dev
09:10 skoduri joined #gluster-dev
09:18 poornima ndevos, the dependent patches are merged. And i didn't see noticeable slowness but i didn't test it with real workload. Also will be working on getting upcall below io-threads.
09:35 ndevos poornima: ok, thanks!
09:46 guest Is it possible to use bd xlator with crypt xlator?How can I use it?
09:47 an_ joined #gluster-dev
09:53 spalai left #gluster-dev
09:53 spalai joined #gluster-dev
09:56 an__ joined #gluster-dev
10:08 skoduri joined #gluster-dev
10:17 guest Is it possible to have both thin LV and usual LV in a same time in bd xlator?
10:23 an_ joined #gluster-dev
10:26 guest xavih: Is it possible to use bd xlator with crypt xlator?How can I use these two xlator in a same time?
10:31 kshlm guest, AFAIK, no one has ever tried to use those two together.
10:31 kshlm I don't think they will work together either.
10:32 kshlm Also, both of these xlators are currently orphans, as the developers who implemented them have moved on.
10:32 nbalacha ndevos, can you please take a loot at and merge http://review.gluster.org/#/c/15122/ ?
10:32 nbalacha *look
10:33 shyam joined #gluster-dev
10:33 kshlm So, it will be little hard to get support for them, as there is not a lot of experience regarding them.
10:36 guest kshlm: So , you mean nobody
10:37 guest kshlm: answer the reported bugs for crypt xlator?
10:37 kshlm guest, Almost nobody. There may be someone out there in the community using them.
10:37 kshlm guest, From the developers, yes. There is nobody at the moment.
10:38 guest :-(
10:46 guest kshlm: sorry, one more question, Is it possible to have both thin LV and usual LV in a same time in bd xlator?
10:47 atinm joined #gluster-dev
10:47 kshlm Sorry. I don't have much knowledge of the BD xlator.
10:51 msvbhat joined #gluster-dev
11:04 poornima joined #gluster-dev
11:06 hchiramm atalur++ All the best!
11:06 glusterbot hchiramm: atalur's karma is now 16
11:14 atinm joined #gluster-dev
11:20 guest joined #gluster-dev
11:22 dlambrig joined #gluster-dev
11:29 poornima joined #gluster-dev
11:37 aravindavk joined #gluster-dev
11:43 rjoseph kshlm: Review needed for http://review.gluster.org/#/c/15359/
11:45 karthik_ joined #gluster-dev
11:45 kshlm rjoseph, Done.
11:46 rjoseph kshlm++ that was fast :-)
11:46 glusterbot rjoseph: kshlm's karma is now 109
11:48 jiffin1 joined #gluster-dev
11:49 nigelb kshlm: Hey, thanks for grabbing those test failures. there's a few machines in a bad state. I'll get to them now.
11:49 nigelb You need to make sure umount is unstuck when you cancel a job or else it'll hang the next job as well.
11:52 kshlm nigelb, Found that out later, when one of the machines hung again. I just rebooted it.
11:53 kshlm rastar, 15374 can be merged now. Do you want me to merge it as well?
11:55 nigelb woo, two more hung regressions
11:55 nigelb I'm glad I got to my desk a little early for the community meeting.
11:57 rastar Gluster community meeting starts in 5 mins in #gluster-meeting
12:00 kkeithley sigh, lots of printf format string errors are creeping back in.
12:00 nigelb kkeithley: we could just add that to smoke test :)
12:00 rastar kshlm: you have already copied the agenda for today?
12:00 kkeithley that's a great idea
12:00 kkeithley just build on 32-bit
12:01 kshlm rastar, I should have.
12:01 nigelb well, we can get a centos 32bit machine easily.
12:01 rastar kshlm: yes, confirmed.
12:01 kshlm I have done it.
12:01 nigelb hang on netbsd is already 32 bit, isn't it?
12:01 kkeithley lol, yes, lots of choices for 32-bit
12:02 kkeithley netbsd might be 32-bit, but if the test doesn't fail then nobody is going to look at the details
12:02 kkeithley we need a test that will fail if there are fmt string errors
12:02 nigelb Don't we have something like that on centos-ci?
12:02 jdarcy joined #gluster-dev
12:02 nigelb Should I just set it to run as a smoke test?
12:03 jiffin1 joined #gluster-dev
12:04 kkeithley probably we should clean it up before we add something that's going to bring things to a screetching halt
12:04 kkeithley otherwise going to bring
12:05 nigelb If you want to start a thread on gluster-devel about it, I'm happy to back you up on the infra side.
12:05 nigelb I can start it off as non-voting and then get it to vote.
12:06 kkeithley sounds good, do it
12:12 nigelb kkeithley: email away. All yours to +1 and add suggestions :)
12:27 ndevos jdarcy: http://termbin.com/4m55 works for me :)
12:28 jdarcy Yeah, just ran essentially the same test here.
12:28 ira joined #gluster-dev
12:29 jdarcy We could have cleanup look for stray jobs and explicitly kill them.  Would that help?
12:32 ndevos hopefully it'll help, but I guess it should be something we need to do in any case
12:32 nigelb Can we also have something to clean up pre-test.
12:32 nigelb when something is aborted, there's no clean way to get the machine consistent other than a reboot.
12:32 nigelb s/pre-test/before we start running any test/
12:35 kkeithley nigelb: I sent email about fmt string to -devel 10 min ago. Haven't seen it come back yet
12:35 jdarcy https://gist.github.com/jdarcy/9bfc817568211872ce28a6ea62f44dcb
12:36 nigelb lol, I sent one 25 mins ago.
12:36 jdarcy Hairy sed expressions FTW.
12:36 kkeithley nigelb: yeah, it just arrived.
12:37 * kkeithley wonders why it's so slow
12:42 rastar nigelb++
12:42 glusterbot rastar: nigelb's karma is now 29
12:51 nigelb ndevos: A build takes only a few minutes and since it needs 32-bit, we can have just one 32-bit node doing it.
12:52 nigelb Or, more easily, have centos-ci voting on our gerrit.
12:52 nigelb which solves both problems in one go.
12:53 ndevos nigelb: the current strfmt check runs on the build.log from the nightly builds, so it needs some adjustment
12:53 nigelb so build, get log, run a grep.
12:53 nigelb That shouldn't be too hard.
12:53 ndevos yeah, pretty simle
12:53 ndevos *simple
12:54 ndevos build it in mock, so that the correct rpm build-options are used
12:55 nigelb oh excellent.
12:59 shyam joined #gluster-dev
13:03 EinstCrazy joined #gluster-dev
13:06 nigelb kkeithley: when do you leave?
13:06 nigelb I'm technically coming to BLR for not-work :)
13:07 nigelb I'm running http://www.bengalurumarathon.in/
13:07 misc you are harshe to call meeting "not-work"
13:07 nigelb misc: In the current plan, I don't plan to step food in the office.
13:08 kkeithley I depart here the 8th (evening) and arrive early AM on the 10th. Have a good run. I admire you guys that have knees.
13:08 nigelb no, I mean, when you leave BLR
13:08 kkeithley rastar++, jiffin++
13:08 glusterbot kkeithley: rastar's karma is now 45
13:08 glusterbot kkeithley: jiffin's karma is now 50
13:08 kkeithley I'm leaving BLR on the 17th
13:08 nigelb after a marathon, I'm acutely aware of every muscle in my legs.
13:08 nigelb Because they all hurt.
13:09 jdarcy Sounds like work to me.
13:09 jdarcy Got my "13.1 - only half crazy" bumper sticker the other day.  ;)
13:09 nigelb haha
13:09 nigelb The bangalore one is "training"
13:09 nigelb for the real one: http://scmm.procamrunning.in/
13:10 kkeithley I'll stick to riding mountain bike.
13:10 nigelb Heh
13:10 jdarcy On the subject of training . . . gottagobye.
13:17 ndevos nigelb: hmm, I would not want to run a whole marathon, but the 1/2 is possible
13:17 ndevos or the 5km part :)
13:17 nigelb ndevos: there's a half as well.
13:17 nigelb If you want to risk it :D
13:18 ndevos why isnt there a date on the page, or am I just reading over it?
13:18 rastar joined #gluster-dev
13:18 nigelb It's annoyingly on the logo
13:19 nigelb I found this exact thing frustrating when I tried to sign up.
13:19 nigelb 16th Oct.
13:19 ndevos ah, yes, now I see it too
13:19 ndevos but I'm there in september, not october...
13:20 nigelb OH.
13:20 ndevos yeah, and so is kkeithley
13:20 nigelb I didn't think you were coming in the next 2 weeks.
13:22 ndevos well, yes, we are
13:22 * nigelb wonders if it's useful to to visit around then.
13:23 ndevos sure it is! :D
13:30 kkeithley yes, 12-16 september.  Maybe again in early December.
13:38 jiffin1 joined #gluster-dev
13:42 mchangir joined #gluster-dev
13:49 jiffin1 joined #gluster-dev
14:02 rastar joined #gluster-dev
14:05 baojg joined #gluster-dev
14:06 dlambrig joined #gluster-dev
14:09 ira joined #gluster-dev
14:14 kotreshhr left #gluster-dev
14:15 spalai left #gluster-dev
14:23 msvbhat joined #gluster-dev
14:24 Muthu_ joined #gluster-dev
14:31 Manikandan_ joined #gluster-dev
14:39 aravindavk joined #gluster-dev
14:41 kdhananjay joined #gluster-dev
14:51 Manikandan joined #gluster-dev
14:51 jiffin joined #gluster-dev
14:56 nigelb kkeithley: where's the upstream spec files for fedora/centos?
14:57 kkeithley fedora, including EPEL (CentOS) are in fedora dist-git at ssh://username@pkgs.fedoraproject.org/glusterfs
14:57 nigelb aha.
14:57 kkeithley one sec for the CentOS Storage SIG dist-git
14:58 wushudoin joined #gluster-dev
14:58 nigelb Do we do EPEL anymore for newer packages? They all go into Centos Storage SIG, correct?
14:58 kkeithley 3.8 is entirely in CentOS Storage SIG
14:59 nigelb kkeithley: If I make it to Bangalore, can I borrow you for a few hours to talk about packaging?
14:59 kkeithley and in the community meeting last week (and reaffirmed this week) starting with 3.7.15
14:59 kkeithley oh, definitely
15:00 kkeithley email on -devel too
15:12 ashiq joined #gluster-dev
15:18 rafi joined #gluster-dev
15:23 jiffin1 joined #gluster-dev
15:30 kshlm joined #gluster-dev
15:32 msvbhat joined #gluster-dev
15:43 skoduri joined #gluster-dev
15:44 ira joined #gluster-dev
15:45 devyani7 joined #gluster-dev
15:46 EinstCrazy joined #gluster-dev
15:50 shubhendu joined #gluster-dev
15:58 dlambrig joined #gluster-dev
16:30 kkeithley ndevos, nigelb: wrt fmtstr errors, master branch is mostly clean. I did find at least one size_t error that has crept in.  None of the fixes have been back-ported to 3.8 or 3.7.
16:45 kkeithley hmmm. I thought we were keeping sub-packages alphabetized in the .spec file.   But -events is after -geo-rep !
16:49 aravindavk joined #gluster-dev
17:05 spalai joined #gluster-dev
17:10 skoduri joined #gluster-dev
17:22 ashiq joined #gluster-dev
18:17 spalai joined #gluster-dev
18:34 ndevos kkeithley: yeah, that subpackage got sneaked in somehow, I only recently noticed it got built
18:35 kkeithley my mistake, it's after -client-xlators, which is after -server
18:48 mchangir that reminds me ... I'll have to wrap the -events sub-package inside a %if ( 0%{?_build_server} ) ... %endif in a DOWNSTREAM ONLY patch
18:49 msvbhat joined #gluster-dev
19:48 dlambrig joined #gluster-dev
20:25 shyam joined #gluster-dev
21:03 dlambrig joined #gluster-dev
22:08 hagarth joined #gluster-dev

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary