Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2017-05-19

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

All times shown according to UTC.

Time Nick Message
01:01 gyadav joined #gluster-dev
01:07 fam_away joined #gluster-dev
01:07 kkeithley joined #gluster-dev
01:08 ndk_ joined #gluster-dev
01:08 bfoster joined #gluster-dev
01:11 glustin joined #gluster-dev
01:11 loadtheacc joined #gluster-dev
01:21 tdasilva- joined #gluster-dev
01:22 gyadav joined #gluster-dev
01:38 mchangir joined #gluster-dev
01:49 ilbot3 joined #gluster-dev
01:49 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/
02:13 pranithk1 joined #gluster-dev
02:19 gem joined #gluster-dev
03:03 nigelb kkeithley: Don't setup python-bugzilla on download.gluster.org
03:03 nigelb kkeithley: I've made it a Jenkins job instead.
03:29 aravindavk joined #gluster-dev
03:38 smit joined #gluster-dev
03:41 prasanth joined #gluster-dev
03:44 gem joined #gluster-dev
03:48 riyas joined #gluster-dev
04:03 atinm joined #gluster-dev
04:06 apandey joined #gluster-dev
04:12 ppai joined #gluster-dev
04:13 gyadav joined #gluster-dev
04:29 poornima joined #gluster-dev
04:33 smit joined #gluster-dev
04:41 itisravi joined #gluster-dev
04:45 nbalacha joined #gluster-dev
04:55 pkalever joined #gluster-dev
04:58 jiffin joined #gluster-dev
04:59 itisravi hi nbalacha, are you not backporting the nodeuuid dht change to the 3.8 branch?
04:59 nbalacha itisravi, I thought it was the last 2 releases
04:59 sanoj joined #gluster-dev
04:59 nbalacha does it need to be backported to 3.8?
05:00 itisravi nbalacha: not sure. I thought just 3.11 was enough because it is a rather new design but since the dht change got merged in 3.10, I sent the afr fix also.
05:00 itisravi nbalacha: and if 3.10 has it maybe 3.8 can have it too :)
05:00 nbalacha itisravi, I dont think we need to backport to 3.8 :)
05:00 itisravi ok
05:01 nbalacha actually now I'm wondering if I should have done it for 3.10
05:01 itisravi nbalacha: yeah maybe we shouldn't have.
05:02 itisravi nbalacha: anyway if we have a consensus that it should not go in a stable branch, we can revert it and abandon afr, ec changes.
05:02 skumar joined #gluster-dev
05:02 nbalacha itisravi, is the afr change merged?
05:02 nbalacha because the dht change does not affect the current behaviour
05:02 itisravi nbalacha: not merged.
05:03 itisravi nbalacha: oh then let me check with pranith once today.
05:03 nbalacha joined #gluster-dev
05:03 itisravi apandey: ^^
05:06 Saravanakmr joined #gluster-dev
05:20 pranithk1 joined #gluster-dev
05:22 skoduri joined #gluster-dev
05:29 ndarshan joined #gluster-dev
05:41 hgowtham joined #gluster-dev
05:41 pranithk1 joined #gluster-dev
05:50 nbalacha joined #gluster-dev
05:51 sona joined #gluster-dev
05:59 apandey_ joined #gluster-dev
06:00 kdhananjay joined #gluster-dev
06:00 apandey_ joined #gluster-dev
06:00 rafi joined #gluster-dev
06:15 nbalacha joined #gluster-dev
06:16 prasanth joined #gluster-dev
06:27 nbalacha_ joined #gluster-dev
06:37 Shu6h3ndu joined #gluster-dev
06:48 nbalacha joined #gluster-dev
06:52 kotreshhr joined #gluster-dev
06:57 apandey__ joined #gluster-dev
06:57 itisravi joined #gluster-dev
07:06 ashiq joined #gluster-dev
07:30 rafi joined #gluster-dev
07:37 ashiq joined #gluster-dev
07:43 nishanth joined #gluster-dev
08:02 sanoj joined #gluster-dev
08:06 nbalacha joined #gluster-dev
08:15 nbalacha joined #gluster-dev
08:21 sona joined #gluster-dev
08:27 ashiq joined #gluster-dev
09:08 pkalever joined #gluster-dev
09:16 prasanth joined #gluster-dev
09:16 prasanth joined #gluster-dev
09:20 percevalbot joined #gluster-dev
09:28 rafi1 joined #gluster-dev
09:34 nbalacha joined #gluster-dev
09:39 dr_bob_bob joined #gluster-dev
09:41 dr_bob_bob Hi :-)
09:56 dr_bob_bob Maybe there is someone in this channel that is willing to help me with a two brick gluster that I set up in a test enviorement that doesnt stop healing? I would appreciate it a lot. Version is 3.10.1
09:57 dr_bob_bob oh sorry, worng channel
10:11 ndevos hah! I could not figure out the magic script format for systemtap with gluster, google shows a nice first hit on "systemtap gluster fuse" pointing to my own presentation O_o
10:12 mchangir :)
10:22 nbalacha apandey__, ping
10:23 apandey__ nbalacha: Hi
10:23 nbalacha apandey__, are you able to reproduce the crash on your system?
10:23 apandey__ nbalacha: I have not tried..
10:24 nbalacha apandey__, can you give it a shot?
10:24 nbalacha if it is reproducible can you try the patch I have posted?
10:24 apandey__ nbalacha: sure
10:25 nbalacha apandey__, thanks
10:34 ndevos pranithk1, xavih, apandey__: I'm having a problem where ec fails to unwind when releasing a lock of a file, any hints on where I should be looking?
10:35 pranithk1 ndevos: statedump may be?
10:35 ndevos I can see dht_lk_cbk called in the non-ec case, but not when I run the same test on ec
10:35 ndevos pranithk1: maybe, but it doesnt tell me a lot... statedumps are often like magic for me
10:36 * ndevos restarts the nfs-server to get a clean statedump
10:44 ndevos pranithk1: statedump is at https://bugzilla.redhat.com/attachment.cgi?id=1280370 - ideas welcome :)
10:45 ndevos (the volume is called "disperse", that may be unfortunate)
10:46 sona joined #gluster-dev
10:47 pranithk1 ndevos: seems like it is stuck on inodelk on that file
10:47 ashiq joined #gluster-dev
10:48 ndevos pranithk1: I think so too, wireshark shows inodekl calls, but no replies
10:49 pranithk1 ndevos: You need to capture brick statedumps to see where it is stuck
10:50 ndevos pranithk1: oh, well, I can see some inodelk replies in wireshark, let me verify is all have been handled
10:52 pranithk1 ndevos: okay
10:53 ndevos actually, I think all inodelk calls have replies... and then shd stamps through my tcpdump and it gets really messy
10:54 Saravanakmr joined #gluster-dev
10:58 pranithk1 ndevos: give me statedumps of bricks, I can take a look
11:01 ndevos pranithk1: here you go! https://bugzilla.redhat.com/attachment.cgi?id=1280371&action=edit
11:04 pranithk1 ndevos: there are no active/blocked locks
11:04 pranithk1 ndevos: is the application still hung?
11:05 ndevos pranithk1: ok, good, that is my understanding as well
11:05 ndevos pranithk1: yes, it is still hanging, not change there - I can do an other nfs-server statedump if you like
11:06 pranithk1 ndevos: yes, that would be good
11:07 ndevos argh! the gluster-nfs server crashed due to some weird reconnect notification that I am trying to debug...
11:08 nh2 joined #gluster-dev
11:09 apandey joined #gluster-dev
11:09 ndevos pranithk1: I'll get back to this later, just need to prevent gnfs from crashing when locks are not released
11:10 pranithk1 ndevos: okay
11:10 pranithk1 ndevos: let me know statedump once the unlocks from brick are over, I can help with the next parts
11:11 pranithk1 ndevos: In case I am not available, you can probably learn how to debug it using: https://github.com/gluster/glusterfs/blob/master/doc/debugging/statedump.md
11:12 pranithk1 ndevos: checkout the last section
11:12 ndevos pranithk1: yes, I'll capture all the statedump, but from what I can see, it really is stuck in ec somewhere
11:12 ndevos thanks!
11:13 pranithk1 ndevos: okay
11:17 kotreshhr joined #gluster-dev
11:19 pkalever joined #gluster-dev
11:21 apandey nbalacha: I am not able to reproduce it on my laptop. I tried different things and also with those EC patch for which regression failed but could not recreate this issue.
11:23 Saravanakmr joined #gluster-dev
11:25 sthakkar joined #gluster-dev
11:40 kkeithley nigelb++, I don't see that the untriaged bug report gets sent anywhere, e.g. to gluster-devel@gluster.org
11:40 glusterbot kkeithley: nigelb's karma is now 53
11:41 vbellur joined #gluster-dev
11:41 kkeithley nigelb: and is scheduled somehow to run automatically once a week?
11:43 vbellur joined #gluster-dev
11:44 vbellur joined #gluster-dev
11:45 vbellur joined #gluster-dev
11:49 vbellur joined #gluster-dev
11:51 vbellur joined #gluster-dev
11:54 vbellur joined #gluster-dev
11:55 ndevos kkeithley: looks cron-like as a trigger in https://review.gluster.org/#/c/17325/4/build-gluster-org/jobs/untriaged-bugs.yml
11:55 kkeithley okay
11:55 vbellur joined #gluster-dev
11:56 ndevos kkeithley: you can probably run it manually through https://build.gluster.org/job/untriaged-bugs/ too :)
11:56 kkeithley indeed.
11:58 kkeithley it looks like it's supposed to send email go gluster-devel, but I don't see that any actually made it
11:58 kkeithley Or maybe they're sitting a queue waiting to be approved and white-listed
12:01 kkeithley s/email go/email to/
12:03 nbalacha joined #gluster-dev
12:07 nbalacha joined #gluster-dev
12:18 rafi joined #gluster-dev
12:33 nigelb kkeithley: I never triggered it to go to gluster-devel. I was testing it against my own address first to tweak out the content.
12:34 kkeithley nigelb: okay
12:39 pkalever joined #gluster-dev
12:43 nigelb I've just brought down gerrit to debug and fix a hang.
12:48 vbellur joined #gluster-dev
12:49 sanoj joined #gluster-dev
12:54 amarts joined #gluster-dev
12:56 nigelb Gerrit should be back online now.
12:57 skumar joined #gluster-dev
13:04 shyam joined #gluster-dev
13:07 gyadav joined #gluster-dev
13:08 kkeithley nigelb, ndevos: I don't find the cppcheck job or script in the build-jobs repo.
13:09 kkeithley where do I look to see that?
13:09 kkeithley And is it running on every commit, or daily?
13:09 kkeithley just daily?
13:09 kkeithley where does the output go?
13:09 kkeithley (and should I retire the cppcheck I'm doing?)
13:10 kkeithley And what about clang compile and analyze — where do we stand with that?
13:16 nigelb It doesn't run properly yet.
13:16 nigelb I've been working on it.
13:21 prasanth joined #gluster-dev
13:34 Saravanakmr joined #gluster-dev
13:34 vbellur joined #gluster-dev
13:38 ndevos xavih: in case you want to follow up on the ec-lock issue pranith and I were looking into earlier, https://bugzilla.redhat.com/show_bug.cgi?id=1444515#c13 should have some details
13:38 glusterbot Bug 1444515: unspecified, urgent, ---, ndevos, NEW , [GNFS+EC] Unable to release the lock when the other client tries to acquire the lock on the same file
13:38 ndevos xavih: pranith is not online anymore, and I'll continue with him on this later - unless you have a strong interest in it
13:39 ndevos but, I'll be out for a while...
13:43 major joined #gluster-dev
13:50 shyam joined #gluster-dev
13:53 pkalever joined #gluster-dev
13:56 pranithk1 joined #gluster-dev
13:58 sthakkar_ joined #gluster-dev
14:20 kotreshhr joined #gluster-dev
14:24 nh2 kkeithley: can you confirm whether the Ubuntu packaging should be fixed in 3.10.2? I still get the same unexpanded autoconf macros
14:24 kkeithley yes it should be (as I said on a couple of occasions)
14:25 kkeithley it has https://review.gluster.org/17298 included
14:26 nh2 kkeithley: the v3.10.2 git tag does not include that commit
14:26 kkeithley that's correct. It's added as a patch
14:27 kkeithley in the .dsc or .source_changes
14:28 nh2 kkeithley: then I might be misunderstanding something. Am I supposed to be able to build correctly on Ubuntu from the v3.10.2 git tag?
14:28 kkeithley only if you add https://review.gluster.org/17298
14:29 kkeithley like the packages in the PPA
14:32 nh2 kkeithley: I find it confusing that the 3.10.2 PPA doesn't seem to be built from the v3.10.2 tag (or at least some history based on top of it), what is the idea behind this approach?
14:33 kkeithley it is built from the v3.10.2 tar file, and has a patch added. Pretty standard packaging stuff.
14:33 nh2 kkeithley: and the tar file is not built from something on top of the v3.10.2 tag?
14:34 kkeithley the tar file comes from the jenkins release job which is from the v3.10.2 tag
14:35 kkeithley the release job drops the tar file at http://bits.gluster.org/pub/gluster/glusterfs/src/
14:36 kkeithley which is then copied to https://download.gluster.org/pub/gluster/glusterfs/3.10/3.10.2/ in this particular case
14:36 shyam joined #gluster-dev
14:36 nh2 kkeithley: ah, thanks for explaining. So the autoconf fix was made only specifically for the ubuntu/debian packages? I'm asking also because other distributions seem to have the same problem with the autoconf macros
14:37 kkeithley The Fedora and SuSE packages have the patch. The CentOS packages don't have the patch. ndevos can confirm
14:39 nh2 kkeithley: ah OK, so it doesn't break on some OSs, interesting. Do you know if it is planned that some 3.10.* point release will eventually get a commit that will fix it equally on all platforms (i.e. in git)?
14:40 kkeithley I certainly expect https://review.gluster.org/17298 to be merged in time for 3.10.3
14:40 nh2 kkeithley: OK, thank you!
14:41 kkeithley csaba fixed it in master but didn't do a backport in time for 3.10.2. Thus I did it.
14:43 csaba thanks kkeithley! btw as of the original change, I answered your comment of Apr 27 on May 5 (already post merge).
14:44 kotreshhr left #gluster-dev
14:50 nbalacha joined #gluster-dev
14:57 Saravanakmr joined #gluster-dev
15:01 gyadav joined #gluster-dev
15:01 vbellur joined #gluster-dev
15:01 shyam joined #gluster-dev
15:08 amarts joined #gluster-dev
15:08 wushudoin joined #gluster-dev
15:09 gyadav_ joined #gluster-dev
15:32 riyas joined #gluster-dev
15:48 jiffin joined #gluster-dev
15:51 shyam joined #gluster-dev
15:52 vbellur joined #gluster-dev
16:00 sona joined #gluster-dev
16:01 atinm joined #gluster-dev
16:12 skoduri joined #gluster-dev
16:16 vbellur joined #gluster-dev
16:19 pranithk1 joined #gluster-dev
16:53 nbalacha joined #gluster-dev
17:12 sona joined #gluster-dev
17:18 bwerthmann joined #gluster-dev
17:30 pkalever joined #gluster-dev
17:35 amarts joined #gluster-dev
17:47 vbellur joined #gluster-dev
17:47 shyam joined #gluster-dev
17:54 gyadav_ joined #gluster-dev
18:04 pkalever joined #gluster-dev
18:10 shyam joined #gluster-dev
18:14 vbellur joined #gluster-dev
19:12 shyam joined #gluster-dev
19:34 bwerthmann joined #gluster-dev
19:37 shyam left #gluster-dev
21:03 shyam joined #gluster-dev
21:25 shyam left #gluster-dev

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