Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2015-03-27

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

All times shown according to UTC.

Time Nick Message
02:21 nishanth joined #gluster-dev
03:01 spandit joined #gluster-dev
03:23 shubhendu joined #gluster-dev
04:03 kanagaraj joined #gluster-dev
04:13 atinmu joined #gluster-dev
04:17 ppai joined #gluster-dev
04:33 anoopcs joined #gluster-dev
04:36 ppai joined #gluster-dev
04:38 jiffin joined #gluster-dev
04:42 nishanth joined #gluster-dev
04:48 rafi joined #gluster-dev
04:51 lalatenduM joined #gluster-dev
04:55 ndarshan joined #gluster-dev
05:16 aravindavk joined #gluster-dev
05:18 Manikandan joined #gluster-dev
05:18 ashiq joined #gluster-dev
05:19 hagarth joined #gluster-dev
05:20 hgowtham joined #gluster-dev
05:20 ppai joined #gluster-dev
05:20 soumya_ joined #gluster-dev
05:21 kasturi joined #gluster-dev
05:25 overclk joined #gluster-dev
05:35 nkhare joined #gluster-dev
05:40 kdhananjay joined #gluster-dev
05:48 vikumar joined #gluster-dev
05:56 atinmu joined #gluster-dev
06:01 anrao joined #gluster-dev
06:12 hagarth joined #gluster-dev
06:13 soumya_ joined #gluster-dev
06:31 atinmu joined #gluster-dev
06:49 lalatenduM_ joined #gluster-dev
06:53 atinmu joined #gluster-dev
07:09 nkhare joined #gluster-dev
07:18 raghu joined #gluster-dev
07:20 kdhananjay1 joined #gluster-dev
07:20 gem joined #gluster-dev
07:20 hgowtham joined #gluster-dev
07:26 anrao joined #gluster-dev
07:27 atinmu joined #gluster-dev
08:03 bala joined #gluster-dev
08:06 nkhare joined #gluster-dev
08:11 spandit joined #gluster-dev
08:26 hchiramm joined #gluster-dev
08:40 lalatenduM joined #gluster-dev
08:42 hgowtham joined #gluster-dev
08:44 pranithk joined #gluster-dev
08:59 kshlm joined #gluster-dev
09:11 anrao joined #gluster-dev
09:23 bala joined #gluster-dev
09:35 ppai joined #gluster-dev
09:40 hagarth joined #gluster-dev
09:57 ira joined #gluster-dev
10:03 anoopcs joined #gluster-dev
10:03 bala joined #gluster-dev
10:16 ndevos rafi: could you have a look at bug 1201484 ?
10:16 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1201484 high, high, ---, bugs, NEW , glusterfs-3.6.2 fails to build on Ubuntu Precise: 'RDMA_OPTION_ID_REUSEADDR' undeclared
10:17 rafi ndevos: kkeithley has sent a patch for the same
10:17 rafi ndevos: http://review.gluster.org/#/c/10014/
10:17 ndevos did he? hmm, would have been nice to see that in the bug report :)
10:19 rafi ndevos: i will update :)
10:19 rafi ndevos: can you please review the patch ?
10:20 ndevos rafi: sure, thanks!
10:20 rafi ndevos++ kkeithley++
10:20 glusterbot rafi: ndevos's karma is now 99
10:20 glusterbot rafi: kkeithley's karma is now 60
10:22 ndevos rafi++
10:22 glusterbot ndevos: rafi's karma is now 4
10:22 ndevos rafi: what version of the rdma libs are needed for 3.7? we should have that as a dependency listed somewhere (and in the glusterfs.spec.in)
10:23 rafi ndevos: i think vijay has given the dependency in for 3
10:23 rafi *3.7
10:24 rafi ndevos: RDMA_OPTION_ID_REUSEADDR is declared librdmacm-devel
10:24 ndevos rafi: yes, and the buildrequires in the .spec has librdmacm-devel, but without a version
10:25 ndevos rafi: also Ubuntu uses librdmacm-devel for building, but Precise obviously has a too old version
10:26 rafi ndevos: ok. i will check and tell you
10:27 ndevos rafi: or, maybe we should add a check in configure.ac to see if RDMA_OPTION_ID_REUSEADDR is defined in the header file?
10:29 rafi ndevos: i guess that will be a good choice,
10:29 ndevos rafi: you want to send a patch for that?
10:31 rafi ndevos: i'm not at all good in GNU libtools
10:31 rafi :)
10:31 rafi ndevos: though ,i will try to send a patch
10:37 ndevos rafi: I think AC_CHECK_DECL could be used for checking a #define, but I cant remember that I used it before
10:38 ndevos rafi: ah, there is a similar check in configure.ac for an LVM function, I guess that should work for RDMA_OPTION_ID_REUSEADDR too
10:40 rafi ndevos: thanks, i will check
10:43 ndevos rafi: something like this, but with the correct header file and version in the error: http://paste.fedoraproject.org/203720/45294714
10:45 rafi1 joined #gluster-dev
10:47 rafi1 ndevos: Thank you very much
10:47 rafi1 i will send a patch updating the error message
10:49 soumya_ joined #gluster-dev
10:52 atinmu joined #gluster-dev
10:55 Manikandan joined #gluster-dev
11:02 anoopcs joined #gluster-dev
11:06 lalatenduM_ joined #gluster-dev
11:12 Manikandan joined #gluster-dev
11:17 rafi joined #gluster-dev
11:18 atinmu joined #gluster-dev
11:29 Manikandan joined #gluster-dev
11:34 aravindavk joined #gluster-dev
11:35 ndevos hagarth: is there anything I can do to get http://review.gluster.org/9947 merged?
11:55 gem joined #gluster-dev
11:57 rafi ndevos: librdmacm-devel-1.0.15
11:57 rafi ndevos: the change was included in the above version
11:58 ndevos rafi: okay, thanks, can you update that in the glusterfs.spec.in (with a note in the %changelog) and post that configure.ac change for it too?
11:59 ndevos rafi++ does not need to be today anymore, in case your weekend is starting :)
11:59 glusterbot ndevos: rafi's karma is now 5
11:59 rafi ndevos: np, i will do both changes today
11:59 rafi ndevos: thanks for your help ndevos++
12:00 glusterbot rafi: ndevos's karma is now 100
12:00 hagarth ndevos: will add it to my review queue for tonight :)
12:00 ndevos hagarth++ thanks
12:00 glusterbot ndevos: hagarth's karma is now 46
12:00 rafi ndevos: 100th karma given by me :)
12:01 ndevos rafi: no problem, glad you're looking into it!
12:01 ndevos rafi: hehe, yeah, dont forget it :D
12:07 kkeithley ndevos: I'm also puzzled by why https://bugzilla.redhat.com:443/show_bug.cgi?id=1201484 wasm
12:07 glusterbot Bug 1201484: high, high, ---, kkeithle, POST , glusterfs-3.6.2 fails to build on Ubuntu Precise: 'RDMA_OPTION_ID_REUSEADDR' undeclared
12:07 kkeithley wasn't updated when I submitted the patch
12:10 vikumar joined #gluster-dev
12:18 ndevos kkeithley: spurious failures!
12:21 lalatenduM_ joined #gluster-dev
12:22 Manikandan joined #gluster-dev
12:28 firemanxbr joined #gluster-dev
12:29 kkeithley ndevos: you think spurious failures are why the BZ wasn't updated?
12:37 kkeithley ndevos, rafi: do we really want to completely disable RDMA just because the rdma stack doesn't have REUSEADDR?
12:37 kkeithley OTOH, I guess I don't care that much, the number of systems with too old rdma stacks is probably pretty small
12:39 aravindavk joined #gluster-dev
12:43 rafi kkeithley: we had a similar option set for socket
12:43 rafi kkeithley: that's why i put the same for rdma also
12:47 kanagaraj joined #gluster-dev
12:54 firemanxbr joined #gluster-dev
12:57 kkeithley Yes, well, REUSEADDR is a used a lot (100%?) on every server's listening socket for TCP.
12:59 kkeithley But it's really only the server(s) that suffer from lack of REUSEADDR.  Maybe we can simply disable building and/or packaging the server in this situation?
13:05 shaunm joined #gluster-dev
13:09 ndevos argh! uuid implementations between different OS's suck!
13:10 ndevos kkeithley: yes, lets not enable single small functions if a library is (too) old, rather disable that function completely
13:11 ndevos kkeithley: maybe REUSEADDR in RDMA could be an exception for it, but the whole RDMA stack would be older on those distributions, and I dont think we can assume it 'just works'
13:16 kkeithley ??? I'm not following you.
13:17 kkeithley whether it works is one thing. I kinda presume it works. The community can tell us otherwise because we can't test every distro on every kind of hardware
13:18 kkeithley not having REUSEADDR is suboptimal, but it in theory the rest works
13:20 kkeithley I think Ubuntu Precise is old enough we could live with that, either by documenting it, and packagers can opt to not package the server(s)
13:27 ndevos kkeithley: I do not think we should encourage users on old distributions to use new functionality that may not be stable, Precise can still package gluster-3.7, just disable rdma there
13:28 kkeithley okay
13:29 ndevos any opinion about bug 1206587 ?
13:29 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1206587 low, low, ---, bugs, NEW , Replace contrib/uuid by a libglusterfs wrapper that uses the uuid implementation from the OS
13:29 ndevos mostly related to http://review.gluster.org/10017
13:32 kkeithley I've wanted to get rid of contrib/uuid (and contrib/fuse*) for a long time and use the system supplied versions
13:33 kkeithley Shhhh, don't tell anyone we have bundled stuff in our Fedora packages.
13:35 kkeithley and if only we could get rid of contrib/qemu
13:36 kkeithley but that's harder
13:38 kkeithley Perhaps using "native" fuse should be a goal for 4.0!
13:39 kkeithley JustinClift++
13:39 glusterbot kkeithley: JustinClift's karma is now 39
13:40 aravindavk joined #gluster-dev
13:45 shyam joined #gluster-dev
13:47 JustinClift ndevos: Just saw on an AMS outage report: "Massive outage in Amsterdam as national power grid overloads
13:47 JustinClift that affects AMS office."
13:47 JustinClift kkeithley: Tx :)
13:49 kkeithley JustinClift: thank you for fixing blog.g.o
13:52 JustinClift np :)
13:52 JustinClift The themes and wp pages I know how to edit :D
13:56 ndevos JustinClift: yeah, but I'm not in the office :)
13:57 JustinClift :)
14:07 shubhendu joined #gluster-dev
14:09 hchiramm JustinClift++ :)
14:09 glusterbot hchiramm: JustinClift's karma is now 40
14:09 hchiramm JustinClift, FJYI 3.4.7beta4 is available at d.g.org.. kkeithley++ will announce it though
14:09 glusterbot hchiramm: kkeithley's karma is now 61
14:15 kanagaraj joined #gluster-dev
14:21 wushudoin joined #gluster-dev
14:26 kkeithley hchiramm++
14:26 glusterbot kkeithley: hchiramm's karma is now 25
14:46 shyam joined #gluster-dev
14:47 kkeithley GlusterFS-3.4.7beta4 is available. Release Notes at http://blog.gluster.org/2015/03/glusterfs-3-4-7beta4-is-now-available-for-testing/
14:47 kkeithley Tentative GA is April 6. Please test at your earliest convenience; time is short.
14:48 anrao joined #gluster-dev
14:56 lpabon joined #gluster-dev
14:56 lpabon quick question.. if a volume has snapshots, can I still delete the volume, or should all the snapshots be deleted first?
14:57 dlambrig1 joined #gluster-dev
14:59 lpabon dlambrig1, kkeithley:  do you know if a volume has snapshots, can I still delete the volume, or should all the snapshots be deleted first?
15:27 soumya_ joined #gluster-dev
15:30 dlambrig1 lpabon: if it works how I think it does, the snapshots share the same data with the source volume, but they are referenced counted and not deleted until the count is 0. So I wold assume gluster would leverage that and allow you to delete the primary even if snapshots exist.
15:38 shaunm joined #gluster-dev
15:47 obnox joined #gluster-dev
15:47 obnox ndevos: here we go :)
15:48 ndevos oh, hello obnox!
15:49 obnox ndevos: next thing now: configure complains about missing sqlite. I note that as something to mention on the website instructions
15:50 ndevos obnox: actually, I thought dlambrig1 and JustinClift made sure to have the sqlite dependency on the build instructions...
15:51 obnox i'll sumbit a patch :)
15:51 obnox also curious to see whether my previous webpage patches have been merged... didn't check
15:51 ndevos obnox: its easier than that, its a wiki!
15:51 obnox ndevos: ah now the build succeeded
15:52 ndevos obnox: okay, thats a good thing, right?
15:52 obnox yay
15:52 ndevos obnox: are you patching something up? or are you just not aware of the nightly builds?
15:53 obnox ndevos: do I really need to make install befor running the tests?
15:53 ndevos but I agree, building glusterfs is nice to practice once every now and then
15:53 ndevos obnox: what tests? the cmocka ones or the regression-tests?
15:53 obnox ndevos: I want to get into the dev process of gluster a bit. Want to do some patches. so i need to locally build + test
15:54 hagarth joined #gluster-dev
15:54 ndevos obnox: I guess you need to 'make install' before running the regression tests, but they take quite a while
15:54 ndevos 1-2 hours, depending on your system
15:55 ndevos I normally build the rpms on my laptop, scp them to an other system for testing
15:55 obnox i was just wondering, because in samba, you can make test without prior make install
15:55 obnox can I run individual tests, i.e. selective?
15:56 obnox samba's testsuite also takes some 2.5 hours atm but  I can select subtests to run
15:56 obnox which is what I do for the dev cycle when working on a specific thing
15:57 obnox hmm. apparently my patches to gluster-docs-project.git have not been merged
15:57 obnox no, haven't
15:59 ndevos hchiramm, lalatenduM_: could you help obnox getting his patches in the gluster-docs-project?
15:59 * lalatenduM_ is in storage sig meeting right now
16:01 ndevos obnox: I always build the rpms, and then run /usr/share/glusterfs/run-tests.sh /usr/share/glusterfs/tests/.....
16:01 ndevos or, well, something like that
16:02 ndevos running the tests is rather destructive for the gluster environment on the system, and there have been broken test-cases that just removed/overwrote files in /etc :-/
16:02 * ndevos is very careful with running those tests
16:03 obnox oh, /me restricts to compile test locally for a start :-D
16:04 ndevos :)
16:04 ndevos lalatenduM_: was that a hint?
16:04 lalatenduM_ ndevos, haha nope
16:05 ndevos :)
16:05 hagarth obnox: you can run a single test unit through prove tests/.../test.t
16:05 obnox hagarth: ah, thanks
16:07 obnox btw: here is my merge request for the docs project:
16:07 obnox https://forge.gluster.org/gluster-docs-project/gluster-docs-project/merge_requests/3
16:07 obnox it basically fixes broken links
16:13 ndevos oh, misc seems to have permissions for that repo too
16:13 bala joined #gluster-dev
16:23 JustinClift kkeithley: https://twitter.com/realjustinclift/status/581489034452697089
16:24 kkeithley booyah.  JustinClift++
16:24 glusterbot kkeithley: JustinClift's karma is now 41
16:28 * JustinClift attempts to highlight the fear aspect for 3.4 users
16:53 deepakcs joined #gluster-dev
17:02 shaunm joined #gluster-dev
17:50 rafi joined #gluster-dev
18:04 vipulnayyar joined #gluster-dev
19:08 shyam joined #gluster-dev
19:34 shyam joined #gluster-dev
19:55 vipulnayyar joined #gluster-dev
19:57 shaunm joined #gluster-dev
20:19 rafi joined #gluster-dev
21:01 rafi joined #gluster-dev
21:30 dlambrig__ left #gluster-dev
21:45 shyam joined #gluster-dev
22:53 dlambrig_ joined #gluster-dev
22:54 dlambrig_ if I want to add a bugzilla bug # to a particular gluster patch, is this possible using review.gluster.org?
23:21 badone joined #gluster-dev

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