Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2015-03-20

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

All times shown according to UTC.

Time Nick Message
00:17 rjoseph|afk joined #gluster-dev
00:33 badone__ joined #gluster-dev
03:32 Apeksha joined #gluster-dev
04:03 atinmu joined #gluster-dev
04:19 kanagaraj joined #gluster-dev
04:23 kdhananjay joined #gluster-dev
04:25 ppai joined #gluster-dev
04:32 ndarshan joined #gluster-dev
04:41 badone__ joined #gluster-dev
04:46 anoopcs joined #gluster-dev
04:46 rafi joined #gluster-dev
04:46 spandit joined #gluster-dev
04:53 kshlm joined #gluster-dev
04:55 anrao joined #gluster-dev
04:56 ppp joined #gluster-dev
04:57 lalatenduM joined #gluster-dev
05:13 lalatenduM joined #gluster-dev
05:15 Apeksha joined #gluster-dev
05:22 lalatenduM joined #gluster-dev
05:25 ashiq joined #gluster-dev
05:27 Manikandan joined #gluster-dev
05:27 Manikandan_ joined #gluster-dev
05:32 pranithk joined #gluster-dev
05:39 topshare joined #gluster-dev
05:40 aravindavk joined #gluster-dev
05:47 overclk joined #gluster-dev
05:56 nkhare joined #gluster-dev
06:02 vimal joined #gluster-dev
06:02 raghu joined #gluster-dev
06:05 an joined #gluster-dev
06:07 gem joined #gluster-dev
06:11 kshlm joined #gluster-dev
06:13 foster joined #gluster-dev
06:15 gem joined #gluster-dev
06:16 hgowtham joined #gluster-dev
06:18 ashiq joined #gluster-dev
06:20 hchiramm joined #gluster-dev
06:36 hgowtham joined #gluster-dev
06:37 Manikandan joined #gluster-dev
06:38 ashiq joined #gluster-dev
06:40 deepakcs joined #gluster-dev
06:48 an joined #gluster-dev
07:15 hchiramm joined #gluster-dev
07:33 hchiramm joined #gluster-dev
07:35 ppai joined #gluster-dev
07:44 badone_ joined #gluster-dev
08:14 ashiq joined #gluster-dev
08:18 topshare joined #gluster-dev
08:31 hagarth joined #gluster-dev
08:33 bala joined #gluster-dev
08:44 hchiramm joined #gluster-dev
09:11 ira joined #gluster-dev
09:30 ppai joined #gluster-dev
09:40 badone_ joined #gluster-dev
10:21 topshare joined #gluster-dev
10:31 rjoseph joined #gluster-dev
10:50 rafi1 joined #gluster-dev
10:53 ppai joined #gluster-dev
11:04 hagarth joined #gluster-dev
11:15 firemanxbr joined #gluster-dev
11:16 rjoseph joined #gluster-dev
11:32 ppai joined #gluster-dev
11:57 bala joined #gluster-dev
11:58 rafi joined #gluster-dev
12:05 hchiramm ndevos++
12:05 glusterbot hchiramm: ndevos's karma is now 96
12:13 hchiramm anrao++
12:13 glusterbot hchiramm: anrao's karma is now 2
12:25 ppai joined #gluster-dev
12:32 kdhananjay joined #gluster-dev
12:33 topshare joined #gluster-dev
12:34 kdhananjay joined #gluster-dev
12:35 rjoseph joined #gluster-dev
12:47 xavih joined #gluster-dev
12:47 hagarth joined #gluster-dev
12:47 topshare joined #gluster-dev
12:52 ppai joined #gluster-dev
12:56 topshare joined #gluster-dev
12:56 dlambrig joined #gluster-dev
12:57 topshare joined #gluster-dev
12:58 rafi joined #gluster-dev
13:16 hchiramm joined #gluster-dev
13:20 dlambrig joined #gluster-dev
13:28 _Bryan_ joined #gluster-dev
13:41 lalatenduM hchiramm++
13:41 glusterbot lalatenduM: hchiramm's karma is now 24
13:59 rafi joined #gluster-dev
14:15 topshare joined #gluster-dev
14:16 topshare joined #gluster-dev
14:17 pranithk1 joined #gluster-dev
14:17 shyam joined #gluster-dev
14:28 ppai joined #gluster-dev
14:37 topshare joined #gluster-dev
14:47 ndevos JustinClift: http://build.gluster.org/job/smoke/14417/console failed on a new slave?
15:00 * JustinClift looks
15:00 JustinClift Damn
15:00 topshare joined #gluster-dev
15:00 JustinClift That's not a new slave
15:01 JustinClift Something _still_ seems to be changing the perms for /var/lib/glusterd, and I don't know what. :(
15:02 JustinClift ndevos: I have no idea how this could possibly fail like that: http://fpaste.org/200569/26863736/
15:03 JustinClift Ugh
15:03 JustinClift Someone's added a "sudo yum -y install cmockery2 cmockery2-devel" without any kind of comment or info why/when/why
15:03 JustinClift Wtf?
15:03 JustinClift :(
15:04 ndevos haha, and cmockery2 is not needed anymore for the master branch... it might still be a dependency for 3.6?
15:05 JustinClift It's not going to hurt to leave it in... it just pisses me off that people aren't following ultra-basic foundational teamwork practises, especially when they know it's important
15:05 JustinClift And the lines right before them are both commented, date stamped, etc.  So, it's not like there's no reminder
15:05 JustinClift I think I'll get a coffee and try to un-pissed-off myself ;)
15:06 JustinClift Anyway, the only explanation I can see for that failure... is that something in build.sh itself is altering /varlib/glusterd perms
15:07 JustinClift Hmmm, I'm going to add some friggin echo $? statments in that script, so we can see if something's failing when creating dirs, setting perms, or whatever
15:14 JustinClift Meh, maybe I just woke up on the wrong side of the bed today. ;)
15:14 JustinClift ndevos: This time is passed: http://build.gluster.org/job/smoke/14418/console
15:14 JustinClift Same slave :/
15:14 JustinClift No functional change in the script either. ;)
15:16 * ndevos shrugs
15:18 misc JustinClift: changing the permission ?
15:20 * misc check
15:20 misc mhh no, salt also do 755 on /var/lib/glusterd
15:20 misc also, saved slave20 today, was full, not in salt
15:20 misc full due to log
15:24 JustinClift misc: Yeah, slave20 filled up yesterday or the day before I think.  That's the one being left alone so overclk can investigate the cores
15:24 misc JustinClift: oups :/
15:24 misc I might have cleaned that
15:24 JustinClift It's going to be one of those days isn't it?
15:24 JustinClift ;)
15:25 misc cannot be worst than yesterday
15:25 misc fighting against freebsd
15:25 misc I guess i should go back on way to automate jenkins slave creation
15:25 JustinClift Hmmm, I left a note in Jenkins about it I think, but Jenkins seems to still have tried to start it, then overwritten note about stuff
15:25 JustinClift Yeah, I'm going to ignore all this, and get back to attacking Gerrit.
15:27 JustinClift I've just connected slave20 again in Jenkins, now that it's got free space and all.  Might as well have it be useful. :)
15:34 JustinClift misc: On an entriely unrelated note, the domain "bad.jobs" is available
15:34 JustinClift Could so CNAME that to eg ORCL, MSFT, etc ;)
15:35 JustinClift Or even APPL :D
15:48 ndevos kkeithley_: review+merge http://review.gluster.org/9957 ?
15:48 ndevos oh, wait, its not verified yet...
16:05 dlambrig joined #gluster-dev
16:24 JustinClift misc: http://pir.ate.ninja
16:25 misc ah ah
16:26 JustinClift You haven't read The Martian have you?
16:28 misc nope
16:28 JustinClift No worries.  Pirate Ninja's feature prominently in the last 1/3 of the book. ;)
16:29 misc I wonder how that appear in the book, as it doesn't seems exactly what would happen
16:30 misc ( or that's not the good one I found on the web )
16:30 JustinClift ;)
16:42 ndevos JustinClift: again, awkward smoke test failure: http://build.gluster.org/job/smoke/14422/console
16:43 JustinClift YEah, just noticing that
16:43 JustinClift Something is weird
16:43 JustinClift It failed a bunch of them :/
16:45 ndevos JustinClift: yeah, it tried to delete some cores, and got permission denied?
16:48 JustinClift I've just made the chmod and chown at the script start be recursive
16:48 JustinClift Maybe that'll fix it
16:48 JustinClift Just told slave22 to run a smoke test, so lets see if that helps
16:52 kkeithley_ all the regression test runtimes are now in millis?
16:52 ndevos hey, yes, I noticed that too
16:52 dlambrig joined #gluster-dev
16:58 JustinClift Ugh
16:58 JustinClift On the one hand, we've really improved our scalability...
16:59 JustinClift kkeithley_: smoke tests, not regression?
16:59 JustinClift I'm not seeing regression weirdness just at the moment?
17:00 JustinClift Ok, the recursive chmod/chown may have fixed it.  Trying a few more runs
17:05 kkeithley_ regression tests!
17:05 kkeithley_ times in millis isn't broken or bad, per se. Just noticing the change
17:06 JustinClift Hmmm.  Point me at one?
17:06 JustinClift I guess I'm being blind?
17:06 kkeithley_ well, any of them. But http://build.gluster.org/job/rackspace​-regression-2GB-triggered/5700/console
17:07 JustinClift Ahhh
17:07 JustinClift You mean "xxx ms" on the tests
17:07 JustinClift Ahhh
17:07 kkeithley_ correct
17:07 JustinClift I thought you were meaning the regression runs were finishing in xxx ms .  eg the whole thing
17:07 kkeithley_ nope
17:07 JustinClift eg something busted in the regression scripting
17:07 kkeithley_ if only
17:07 JustinClift Goitcha
17:09 JustinClift kkeithley_: The ms is an improvement isn't it?
17:10 kkeithley_ it's just different. I'm just noticing the difference
17:10 JustinClift k
17:10 kkeithley_ Was it an intentional change?
17:11 kkeithley_ I'm less/not concerned if it was deliberate. If it wasn't deliberate then I'm raising an eyebrow at it.
17:18 overclk joined #gluster-dev
17:21 hagarth joined #gluster-dev
17:23 JustinClift kkeithley_: It's not something I know about
17:23 JustinClift kkeithley_: I just looked at the recent history for the tests subdir and not seeing anything obvious in the commit messages there
17:25 kkeithley_ I'm puzzled.  I submitted http://review.gluster.org/9955 (about five hours ago.) For a while there was a rackspace-netbsd7-regression-triggered queued waiting for a machine to run it. Now it's gone from the queue.  I've not seen a netbsd7 or regular regression run come up in the running tests.
17:25 kkeithley_ wrt millis run times.... hmmmmmmm
17:25 misc so... let's see if slave75.cloud.gluster.org , the first VM automatically deployed by salt, work
17:27 JustinClift kkeithley_: Manu or myself have just been zapping the netbsd7 regression tests out of the queue in the last few days, as they build up chronically
17:27 JustinClift kkeithley_: I haven't touched any today, but it could have been Manu
17:27 anrao joined #gluster-dev
17:29 kkeithley_ Oh, I wasn't suspecting foul play or anything. Just wondering if I'm looking at all the right places to see jobs that are running, because I'm obviously not seeing everything there is. Usually I can find the regression runs for the changes I've submitted.
17:30 kkeithley_ Just a bit of weirdness is all.
17:30 overclk hagarth, things look good w/ bitrot patches in gerrit.
17:30 kkeithley_ Like the secret elephant burial ground
17:30 kkeithley_ graveyard
17:30 kkeithley_ ;-)
17:32 JustinClift Shhh.  We don't talk about the secret elephant burial ground in open channels...
17:32 hagarth overclk: fantastic!
17:34 kkeithley_ but the secret jenkins job graveyard we can talk about.
17:36 JustinClift Yeah, that's not real secret ;)
17:37 ndevos really, are all regression tests failing now?
17:39 misc ( so cloud75 failed to install ... )
17:39 lalatenduM joined #gluster-dev
17:41 JustinClift Gah
17:41 * JustinClift looks at that now
17:43 JustinClift ndevos: Well there are regression failures happening
17:43 JustinClift ndevos: Some of them seem like legit bugs though
17:44 JustinClift eg glfs function definition problem
17:44 JustinClift (also failed the smoke test I think)
17:44 JustinClift Doesn't look infrastructure related
17:44 ndevos JustinClift: which one is that?
17:45 JustinClift http://build.gluster.org/job/rackspace​-regression-2GB-triggered/5711/console
17:45 JustinClift Add no, wrong error
17:45 JustinClift Just a standard C coding problem in glfs
17:46 ndevos I wonder if people actually compile and tests their changes...
17:46 ndevos hchiramm ?
17:48 JustinClift Likely varies per person, and which part of day
17:48 JustinClift Even though it shouldn't ;)
17:54 JustinClift I really should turn off my GitHub notifications
17:54 JustinClift They're getting distracting :/
17:58 shaunm joined #gluster-dev
17:58 shaunm joined #gluster-dev
18:35 misc yeah \o/ found and corrected a bug in salt
18:35 misc ( because bash sux, assuming people use LC_ALL=C sux )
18:54 misc just to be sure, where are the gluster rpm built ?
18:54 tdasilva joined #gluster-dev
19:10 JustinClift misc: Depends which ones.  I know there's a job that runs on every CR submission, building rpms for that specific patch set
19:11 JustinClift misc: That job runs on our CentOS rackspace slaves
19:11 JustinClift misc: There's also a semi-manual process for building the official GlusterFS rpms
19:11 JustinClift I'm not sure where that happens.  kkeithley_ and some of the others know, as they do it :)
19:17 misc they are not signed ?
19:17 tdasilva joined #gluster-dev
19:20 tdasilva joined #gluster-dev
19:41 JustinClift No idea.  I think our official releases are
19:41 JustinClift kkeithley_ ndevos: ^?
19:58 ira joined #gluster-dev
20:11 dlambrig joined #gluster-dev
20:51 kkeithley_ JustinClift: yes?
20:55 kkeithley_ JustinClift, misc: the rpm.t in regression tests building rpms. Or has that been superceded by the *rpm* jobs? ndevos has a cron job I think that builds, as well as some "official" nightly builds in COPR.  Official release builds for download.gluster.org are build in Fedora Koji and are signed.
21:09 misc signed on fedora ?
21:11 kkeithley_ JustinClift: oh, that netbsd regression I couldn't find? (http://review.gluster.org/9955) I found it. It ran "fast" because mkdir: /build/install/lib: Permission denied and bailed out.  on nbslave78
21:13 kkeithley_ And its corresponding rackspace-regression-2GB-triggered job failed to clone from the git repo.  ( http://build.gluster.org/job/rackspace-​regression-2GB-triggered/5694/console)
21:13 kkeithley_ sign
21:13 kkeithley_ sigh
21:15 kkeithley_ misc: no, signed by GlusterPackager on download.gluster.org
21:18 kkeithley_ JustinClift: looks like the rackspace-regression-2GB-triggered job is working now, and waiting for the requeued rackspace-netbsd7-regression-triggered to run
21:31 JustinClift kkeithley_: Hmmm, some of those netbsd regression runs got stuck on build.sh
21:31 * JustinClift is cancelling those now
21:34 JustinClift 100% of the ones running were borked on something
21:34 JustinClift So, they've all been aborted
21:35 JustinClift kkeithley_: Yours should be starting about now... ;)
21:44 shyam joined #gluster-dev
21:46 JustinClift kkeithley_: Your job is already unhappy on the netbsd regression run
21:47 JustinClift It may not be your code though.  We've had massive failure rate on NetBSD over the last few days. :(
21:47 JustinClift ... and it's not really NetBSD's fault generally.  (or seems not)
21:47 JustinClift http://build.gluster.org/job/rackspace-net​bsd7-regression-triggered/2177/consoleFull
22:03 dlambrig have the NetBSD machines got sqlite-devel loaded? When we spin up a new one of those machines, will it be loaded?
22:07 JustinClift dlambrig: Yep
22:08 JustinClift Hmmm, for the "new one of those machines"... that I'm not sure of
22:08 JustinClift I guess we get to find out. ;)
22:11 dlambrig I see what seems to be spurious errors on Linux smoke tests…  “regression_2GB”
22:12 tdasilva joined #gluster-dev
22:14 misc JustinClift: there is no sudo on the netbsd slave
22:28 dlambrig left #gluster-dev
22:41 JustinClift misc: Hmmm, you may need to su -
22:41 JustinClift misc: Email Manu for the pw :)
22:41 JustinClift Or ask him to install sudo I guess ;)
22:48 misc JustinClift: well, it was more to see
22:48 misc I could install my own netbsd image if I need to
23:22 JustinClift :)
23:42 badone_ joined #gluster-dev

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