Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2014-05-20

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

All times shown according to UTC.

Time Nick Message
00:33 tdasilva joined #gluster-dev
00:58 yinyin joined #gluster-dev
01:22 awheeler joined #gluster-dev
01:45 bala joined #gluster-dev
02:44 yinyin joined #gluster-dev
03:06 kdhananjay joined #gluster-dev
03:13 kshlm joined #gluster-dev
03:39 kanagaraj joined #gluster-dev
03:45 shubhendu joined #gluster-dev
03:51 itisravi joined #gluster-dev
04:11 bharata-rao joined #gluster-dev
04:11 ppai joined #gluster-dev
04:33 lalatenduM joined #gluster-dev
04:46 atinmu joined #gluster-dev
05:16 nishanth joined #gluster-dev
05:20 bala joined #gluster-dev
05:22 hagarth joined #gluster-dev
05:34 ndarshan joined #gluster-dev
05:44 raghu joined #gluster-dev
05:53 hagarth joined #gluster-dev
05:53 aravindavk joined #gluster-dev
06:08 kanagaraj joined #gluster-dev
06:36 nishanth joined #gluster-dev
06:37 hchiramm_ joined #gluster-dev
06:40 bala2 joined #gluster-dev
06:45 aravindavk joined #gluster-dev
06:48 shubhendu joined #gluster-dev
06:52 ndarshan joined #gluster-dev
06:53 hagarth joined #gluster-dev
07:03 hchiramm_ joined #gluster-dev
07:20 shubhendu joined #gluster-dev
07:21 ndarshan joined #gluster-dev
07:22 nishanth joined #gluster-dev
07:24 bala joined #gluster-dev
07:31 aravinda_ joined #gluster-dev
07:32 hagarth1 joined #gluster-dev
07:36 foster joined #gluster-dev
07:43 ndevos ping raghu: when do you think you have a patch for bug 1099270 , if it is soon, should it be included in 3.5.1?
07:43 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1099270 high, unspecified, ---, rgowdapp, ASSIGNED , Gluster 3.5.0 NFS server crashes under load
07:58 edward1 joined #gluster-dev
08:14 edward2 joined #gluster-dev
08:16 JustinClift ndevos: Is your http://review.gluster.org/#/c/7786/ (tests/rpm) to fix regression test failure in rpm.t?
08:16 glusterbot Title: Gerrit Code Review (at review.gluster.org)
08:17 ndevos JustinClift: yeah, for some reason the 'make dist' tarball fails to include install-sh, or at least ./configure fails to find that :-/
08:17 JustinClift Asking because I'm seeing a bunch of rpm.t failures this morning in a set of VM's kicked off, and I remember someone saying the other day that they're looking into something like that.
08:17 JustinClift Cool.
08:17 JustinClift I'll kick off a few VM's with that CR applied, and see if they pass.
08:17 ndevos JustinClift: hmm, not sure if that is related....
08:17 JustinClift :)
08:17 JustinClift k
08:19 ndevos JustinClift: I hoped to not need ./autogen.sh in the .spec, but that keeps failing... the updated rpm.t copies the mock build logs to /build/install/var/log so that Jenkins pulls them in an archive after a failure
08:21 JustinClift ndevos: As a thought, would adding install-sh to the EXTRA_DIST line in Makefile.am do the trick instead?
08:22 JustinClift Or is the problem that install-sh isn't in the build root at all, so isn't there to be pulled in?
08:22 ndevos JustinClift: that should not be needed, autoconf/make should do that automatically - and it gets included in the tar.gz on my systems, or manual runs on build.gluster.org....
08:23 * ndevos has no idea why it fails in the test case
08:23 JustinClift ;)
08:25 JustinClift 8 out of 10 of the VM's launched this morning have failed with rpm.t as the only problem.  2 VM's still going (nearly finished).  That's actually a really good result.
08:25 JustinClift With rpm.t fixed, it could mean master is in ok state for regression test reliability. (more famous last words) :>
08:26 ndevos JustinClift: got any logs why rpm.t failed? could it be a network issue while downloading the rpms from a fedora mirror?
08:27 JustinClift In every case it's:
08:27 JustinClift ./tests/basic/rpm.t                             (Wstat: 0 Tests: 5 Failed: 1) Failed test:  5
08:27 JustinClift Bah.  Newlines killed in IRC.
08:27 JustinClift Failed test is 5.  I'm not sure what test 5 is atm.
08:27 JustinClift Haven't yet looked.
08:29 JustinClift Hmmm, rpm.t is the one you added code to capture debug logs for... I'll kick off a run with debug mode enabled, and see what shows up in the logs
08:30 JustinClift Takes about 90 mins for the regression tests to finish though, so it's kind of a fire and check it later thing. ;)
08:30 JustinClift s/fire/fire off/
08:31 ndevos JustinClift: test 5/6 run parallel, its either building for epel-5-x86_64 or epel-6-x86_64, unless you have the epel-7-x86_64 config enabled
08:32 JustinClift Just checked.  It's nuking the epel-7-x86_64 mock config before running.
08:32 JustinClift So, it's doing 5/6 in parallel then I guess.
08:32 JustinClift Let's see what the logs show up when the run finishes.
08:37 hchiramm_ joined #gluster-dev
09:15 kdhananjay left #gluster-dev
09:15 kdhananjay joined #gluster-dev
09:15 kdhananjay left #gluster-dev
09:23 JustinClift ndevos: Gah.  Found it.  http://fpaste.org/103356/05777771/
09:23 glusterbot Title: #103356 Fedora Project Pastebin (at fpaste.org)
09:23 JustinClift ndevos: I forgot it only nukes that EPEL7 thing for the release-3.4/3.5 repox, as that patch to allow things to work on master was merged a while ago.
09:24 ndevos JustinClift: I thought you were not running the epel-7...
09:24 JustinClift So, it looks like it's failing on the EPEL7 bit.
09:24 JustinClift Yeah, I was wrong.  It's testing EPEL6 and 7 after all.
09:24 JustinClift I forgot I leave that running for master, as master's supposed to be able to handle it.  Doh!
09:24 ndevos JustinClift: maybe you need to install the xz package on the builder?
09:24 JustinClift Yeah, pretty likely.
09:25 JustinClift This is how I learn about new dependencies. ;)
09:26 ndevos I've not enabled nightly builds for 3.5/3.4 on epel-7, I dont think they build without changes to the .spec :-/
09:27 JustinClift np
09:27 JustinClift I had epel-7 disabled for 3.4/3.5, pending CR 7564 getting merged.  Just checked, and it is merged now, so I should prob test if it works.
09:27 JustinClift If it does, we should probably get rpm.t to test the last 3 epel's instead of last 2.
09:28 JustinClift Oh that's not good.  xz is already installed.
09:29 JustinClift Dammit, time to investigate properly.
09:33 kdhananjay joined #gluster-dev
09:41 hchiramm_ joined #gluster-dev
09:48 aravinda_ joined #gluster-dev
10:18 yinyin joined #gluster-dev
10:23 ira joined #gluster-dev
10:34 ndevos JustinClift: with your multi-machine-regression-framework, would it be possible to add one VM and run a client app, kill the VM, and detect the result on a storage server or other client?
10:35 ndevos JustinClift: use-case: a client obtains a lock on a file, client crashes (like kernel panic), the lock should get released in X seconds and an other client gets the lock
10:37 bala joined #gluster-dev
10:43 spandit joined #gluster-dev
11:06 _Bryan_ joined #gluster-dev
11:25 kdhananjay joined #gluster-dev
11:36 JustinClift ndevos: Um, not really atm.  The multi-machine-regression-framework thing I was previously going to work on (different from present thing) never got turned into actual code
11:37 JustinClift ndevos: I have some idea on how to do it, using the orchestration tools in rackspace would be easiest.  But no, there's nothing written for it atm
11:38 ndevos JustinClift: ok :-/
11:38 JustinClift Too many things to do, not enough time. :(
11:38 * JustinClift needs to be more efficient at everything I guess
11:47 kkeithley1 joined #gluster-dev
12:03 edward1 joined #gluster-dev
12:14 tdasilva left #gluster-dev
12:15 kshlm joined #gluster-dev
12:17 itisravi joined #gluster-dev
12:20 JustinClift ndevos: It seems to be a mock problem, not a glusterfs problem.  http://fpaste.org/103385/05883291/
12:20 glusterbot Title: #103385 Fedora Project Pastebin (at fpaste.org)
12:21 JustinClift ndevos: I'm tossing up between doing more work to isolate it (eg rolling back mock version, etc), vs just nuking the epel-7 stuff for master as well for now.
12:22 JustinClift The above fpaste is from a local CentOS 6.x VM here (not rackspace VM), which I used to use for building on.  It's an old CentOS snapshot as of March 21 2014.
12:23 JustinClift Meh, I think I'll just stop epel-7 being used in rpm.t for now.
12:49 hagarth joined #gluster-dev
13:17 ndevos JustinClift: on that CentOS6 where you updated mock, did you check if xz was installed?
13:18 ndevos JustinClift: also, the root.log might be helpful, that should have been saved in /var/lib/mock/epel-7-x86_64/result
13:20 ndevos JustinClift: any idea whats up with the gluster-devel mailinglist archives?
13:23 tdasilva joined #gluster-dev
13:23 awheeler joined #gluster-dev
13:26 dlambrig_ joined #gluster-dev
13:34 JustinClift ndevos: Archives?
13:34 ndevos JustinClift: http://supercolony.gluster.org/pipermail/gluster-devel/2014-May/thread.html seems to be out of date?
13:34 glusterbot Title: The Gluster-devel May 2014 Archive by thread (at supercolony.gluster.org)
13:35 JustinClift ndevos: Damn.
13:35 ndevos JustinClift: it says May 20, but at least my last email isnt there, and I have the feeling some others are missing too...
13:35 JustinClift ndevos: I noticed the other day that it seems to not be getting all of the messages or something, but then I forgot about it
13:36 JustinClift We should probably get someone who's an actual mailing list admin experienced type to look into it
13:36 JustinClift Hmmm, I'll ask misc
13:36 ndevos right, that sounds like the same symptom :)
13:37 hchiramm_ joined #gluster-dev
13:41 JustinClift ndevos: k, misc said he's got time to assist.  I'm going to create an account on the box for him and give him wheel.
13:42 ndevos JustinClift: uhm, yes, whatever?
13:45 lpabon joined #gluster-dev
14:11 Humble tdasilva++
14:13 tdasilva :)
14:18 hagarth Humble: do we need a karma counting bot here? :)
14:19 Humble hagarth, indeed we need :)
14:19 kkeithley_ does dharma override neg karma?
14:20 hagarth glusterbot just needs to have the karma plugin loaded <-- JoeJulian: would this be possible?
14:20 hagarth kkeithley_: dharma - plugin not found :D
14:21 Humble :D
14:21 kkeithley_ ;-)
14:22 wushudoin joined #gluster-dev
14:23 JustinClift ndevos: Misc fixed the archives.  Permissions problem.
14:23 JustinClift ndevos: They should be all good now. :)
14:43 hchiramm_ joined #gluster-dev
14:44 kshlm joined #gluster-dev
14:50 lalatenduM joined #gluster-dev
15:05 hchiramm__ joined #gluster-dev
15:06 ndevos JustinClift: thanks!
15:07 ndevos Humble: I think you closed bug 1086756, but should we not include the documentation in the release-3.5 branch too?
15:07 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1086756 low, urgent, ---, hchiramm, CLOSED CURRENTRELEASE, Add documentation for the Feature: zerofill API for GlusterFS
15:08 hchiramm__ ndevos, yeah.. actually we need ..
15:08 hchiramm__ I will send it soon..
15:08 ndevos hchiramm__: okay, thanks, I'll reopen and assign it to you
15:08 hchiramm__ :) ok..
15:23 lalatenduM joined #gluster-dev
15:23 shubhendu joined #gluster-dev
15:55 JustinClift ndevos: Yeah, that quota bug was showing up in release-3.5 branch at the time too
15:55 JustinClift ndevos: If it shows up again in future testing, I'll create a new 3.5 specific bug for it
15:55 ndevos JustinClift: and you were not running quota.t from the master branch?
15:55 JustinClift I don't remember
15:56 ndevos JustinClift: because, the errors you posted are related to add-brick, and quota.t doesnt do that in release-3.5
15:56 jobewan joined #gluster-dev
15:56 ndevos JustinClift: I just tried to backport that change, but uhm, it doesnt make any sense :)
15:56 JustinClift I could have been running the master branch version.  I remember copying the master version of quota.t to other branches to try, but it's so long ago I don't remember the details
15:57 JustinClift Yeah, I think the thing is that we didn't know the problem was in the test itself
15:57 JustinClift So, copying the test to earlier branches seemed like a valid approach
15:57 JustinClift Which really just copied the problem around to any branch that supports the needed commands :)
15:57 JustinClift So yeah, thinking that through, it's probably not valid for release-3.5 :)
15:58 ndevos \o/
16:04 foster joined #gluster-dev
16:07 lalatenduM joined #gluster-dev
16:21 lalatenduM_ joined #gluster-dev
16:55 wushudoin joined #gluster-dev
16:59 baojg joined #gluster-dev
17:02 bala joined #gluster-dev
17:38 kanagaraj joined #gluster-dev
17:52 hagarth joined #gluster-dev
17:52 lpabon joined #gluster-dev
17:55 dlambrig_ left #gluster-dev
17:56 dlambrig__ joined #gluster-dev
18:38 baojg joined #gluster-dev
19:55 baojg joined #gluster-dev
20:03 edward1 joined #gluster-dev
20:05 robert7811 joined #gluster-dev
20:16 baojg joined #gluster-dev
20:26 edward2 joined #gluster-dev
20:58 badone joined #gluster-dev
21:41 wushudoin joined #gluster-dev
21:50 lpabon joined #gluster-dev
22:10 yinyin joined #gluster-dev
22:41 dlambrig__ left #gluster-dev
23:23 tdasilva left #gluster-dev

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