Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2014-05-21

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

All times shown according to UTC.

Time Nick Message
00:30 yinyin_ joined #gluster-dev
00:57 bala joined #gluster-dev
01:46 bala joined #gluster-dev
02:05 yinyin- joined #gluster-dev
02:50 badone joined #gluster-dev
02:55 baojg joined #gluster-dev
03:02 baojg joined #gluster-dev
03:07 baojg_ joined #gluster-dev
03:10 baojg joined #gluster-dev
03:17 baojg_ joined #gluster-dev
03:18 yinyin_ joined #gluster-dev
03:21 bharata-rao joined #gluster-dev
03:41 kshlm joined #gluster-dev
03:44 kanagaraj joined #gluster-dev
03:45 badone joined #gluster-dev
03:52 itisravi joined #gluster-dev
04:06 shubhendu joined #gluster-dev
04:08 kanagaraj joined #gluster-dev
04:14 ppai joined #gluster-dev
04:28 yinyin_ joined #gluster-dev
04:28 kanagaraj joined #gluster-dev
04:38 badone joined #gluster-dev
04:52 bala joined #gluster-dev
04:59 baojg joined #gluster-dev
05:02 hagarth joined #gluster-dev
05:02 spandit joined #gluster-dev
05:05 kdhananjay joined #gluster-dev
05:08 ndarshan joined #gluster-dev
05:10 lalatenduM joined #gluster-dev
05:23 nishanth joined #gluster-dev
05:31 aravinda_ joined #gluster-dev
05:35 kanagaraj joined #gluster-dev
05:35 yinyin_ joined #gluster-dev
05:50 raghu joined #gluster-dev
06:04 yinyin_ joined #gluster-dev
06:04 yinyin_ joined #gluster-dev
06:19 aravindavk joined #gluster-dev
06:41 ppai joined #gluster-dev
06:51 edward1 joined #gluster-dev
07:21 edward3 joined #gluster-dev
07:54 xavih joined #gluster-dev
08:23 bharata-rao joined #gluster-dev
08:44 bharata-rao joined #gluster-dev
08:46 xavih joined #gluster-dev
08:47 kdhananjay joined #gluster-dev
08:50 kdhananjay1 joined #gluster-dev
08:59 xavih joined #gluster-dev
09:01 rgustafs joined #gluster-dev
09:02 nishanth joined #gluster-dev
09:10 bharata-rao joined #gluster-dev
09:28 _Bryan_ joined #gluster-dev
09:45 aravindavk joined #gluster-dev
09:54 bharata-rao joined #gluster-dev
10:02 ira joined #gluster-dev
10:03 aravindavk joined #gluster-dev
10:05 JustinClift ndevos: As a data point, it turns out that weird mock xz error failure was due to pyliblzma now being needed on the build host.  (asked the mock guys)
10:10 JustinClift kkeithley: You around?
10:14 spandit joined #gluster-dev
10:17 bharata-rao joined #gluster-dev
10:19 edward1 joined #gluster-dev
10:21 Humble JustinClift, Hi .. r u interested to take this bug ? https://bugzilla.redhat.com/show_bug.cgi?id=1086766 ?
10:21 glusterbot Bug 1086766: low, urgent, ---, rwheeler, NEW , Add documentation for the Feature: Libgfapi
10:22 Humble :)
10:23 ndevos Humble: yeah, I think JustinClift was writing something for libgfapi :)
10:23 Humble yep .. :)
10:28 kanagaraj joined #gluster-dev
10:30 JustinClift ndevos: Yeah, I mean to get to that.  It's behind "get the regression testing infrastucture working" in priority though.
10:31 JustinClift Oh, for 3.5.1?
10:31 Humble JustinClift, yep :)
10:31 JustinClift No chance of that.
10:32 ndevos JustinClift: do you have anything, that we can use? maybe an introduction or something?
10:32 JustinClift ndevos: No
10:32 ndevos JustinClift: hmm...
10:34 Humble hmm...  then somehow I have to start ...
10:34 Humble :(
10:34 ndevos Humble: just a very short introduction, pointing to example usage in samba/vfs_glusterfs and nfs-ganesha would do - mention the main header files for reference
10:38 Humble hmmm..
10:40 JustinClift Humble: What happened to your blog?
10:40 Humble its been synced with glusterfs
10:40 JustinClift Humble: Didn't that have an intro? http://humblec.com/play-libgfapi-python-bindings/
10:41 Humble ahhhh.. checking ..
10:41 JustinClift humblec.com doesn't seem to be working for me atm
10:41 Humble something wrong
10:41 Humble yep.. thanks for pointing out..
10:41 JustinClift ;)
10:41 Humble :)
10:41 JustinClift Gah.
10:42 JustinClift Rackspace have rolled their version of mock back to a prior release, probably due to other people hitting that xz problem (guessing).
10:42 JustinClift That makes testing epel-7 harder.
10:44 ndevos JustinClift: 'yum --enablerepo=*testing* update mock'  ?
10:53 shubhendu joined #gluster-dev
10:53 JustinClift ndevos: Good idea.  Just checked though, and they've killed the 1.1.38 rpm completely out of their mirror (which is pre-configured in VM's).
10:54 JustinClift I'll just have to do some sed stuff to use the general CentOS mirror list instead I guess.
10:54 ndevos JustinClift: wow, thats quite a thing to do >:-(
10:55 JustinClift Also opened a support ticket with them, to point out the solution (install pyliblzma) to the likely reason they pulled the rpm.
10:55 JustinClift This kind of stuff just takes up way to much of every day... ;)
10:55 JustinClift s/to much/too much/
11:07 ndevos Humble: when you 'backport' changes, could you make sure to replace the Change-Id?
11:08 ndevos Humble: you can include references to the upstream/master patch by marking some of the lines with a >
11:09 Humble ndevos, oh.. yeah..
11:09 ndevos Humble: I'm mostly doing it like http://review.gluster.org/7720 - but am not aware of any backporting guidelines or best practises
11:09 glusterbot Title: Gerrit Code Review (at review.gluster.org)
11:11 Humble ndevos, neither me :) .. wil keep in mind..
11:12 ndevos Humble: I think we should put some advise on the wiki somewhere, maybe linked from the bug lifecycle as they are a little related
11:12 ndevos lalatenduM: ^ got an opinion about documenting advise for backporting fixes?
11:13 Humble ndevos, yep
11:15 lalatenduM ndevos, yup, we should put it in wiki,
11:16 lalatenduM ndevos, the way I have done in the past is , go to local branch of a particular branch, apply cherry picked patch from master (merged), run rfc.sh
11:17 Humble lalatenduM, that will make the change id remain same..
11:17 lalatenduM ndevos, Humble I think the above step automatically includes gerrit review link of master
11:18 Humble lalatenduM, yes
11:18 Humble but change id will be same as master..
11:18 ndevos lalatenduM: I recommend to do a 'git commit --amend' after the cherry-pick, make sure to quote the change-id and related review tags and add a sign-off
11:18 ppai joined #gluster-dev
11:19 lalatenduM ndevos, agree with the sign off part, any specific reason for a diff chaneg id?
11:19 ndevos lalatenduM: it is very confusing to have different patches with the same change-id, I think a change-id should be unique
11:20 Humble ndevos, isnt it good to have same change id in diff. branches.. so that we can better track ?
11:20 lalatenduM ndevos, how do other projects do it
11:20 ndevos lalatenduM: I'm not sure if it breaks things if the change-id is on different branches, but it makes it impossible to reference a single patch change-id
11:20 lalatenduM ndevos, yeah agree with Humble
11:21 ndevos Humble: I'm not sure how Gerrit uses the change-id, and what if a patch does not apply cleanly?
11:21 ndevos Humble: any idea how it is done for ovirt projects?
11:21 Humble ndevos, cross-checking
11:21 lkoranda joined #gluster-dev
11:22 lalatenduM ndevos, Humble should I create a wiki page for the backport guide lines
11:23 ndevos lalatenduM: yeah, if you can get one started, I can add more details to it
11:23 ndevos lalatenduM, Humble: openstack suggests to keep the same change-id: https://wiki.openstack.org/wiki/StableBranch#Workflow
11:23 glusterbot Title: StableBranch - OpenStack (at wiki.openstack.org)
11:25 Humble ndevos, yep.. change id remanins the same
11:25 Humble as long as those are in diff. branches its fine
11:25 lalatenduM ndevos, Humble yeah
11:25 Humble thats what followed in other projects..
11:25 ndevos Humble: yeah, looks like it, I'm fine with it too then :D
11:25 Humble thanks :)_
11:27 ndevos Humble: but it would be good to add the git-commit-hash of the patch in the master branch, and comment out the BUG line (and reviewed by?) in case there is a different bug
11:31 skoduri joined #gluster-dev
11:31 Humble in that case , adding more details to the commit message is better..
11:37 lalatenduM ndevos, Humble , I have created the page , add the guidelines now http://www.gluster.org/community/documentation/index.php/Backport_Guidelines
11:37 glusterbot Title: Backport Guidelines - GlusterDocumentation (at www.gluster.org)
11:39 Humble lalatenduM, thanks.. will add ..
11:47 lalatenduM ndevos, Humble , I have put the steps in http://www.gluster.org/community/documentation/index.php/Backport_Guidelines#Backport_Guidelines_For_Stable_Branch
11:47 glusterbot Title: Backport Guidelines - GlusterDocumentation (at www.gluster.org)
11:47 lalatenduM kindly review it
11:47 ndevos lalatenduM: thanks!
11:49 lalatenduM Humble, ndevos, we may need to add exact git commands so that new contributor can do a back port with ease
11:49 lalatenduM hagarth, ^^
11:49 ndevos lalatenduM: editing it already ;)
11:49 lalatenduM ndevos, :) awesome
11:53 lalatenduM ndevos, sorry I have not completed http://www.gluster.org/community/documentation/index.php/Bug_triage#Tracking_bugs till now
11:53 glusterbot Title: Bug triage - GlusterDocumentation (at www.gluster.org)
11:54 ndevos lalatenduM: no problem :)
11:57 kkeithley ndevos, JustinClift: I'm running nightly coverity and cppcheck on master/$HEAD and release-3.5/$HEAD on an internal machine. I'd like to rsync the results out to a public web server, e.g. bits.gluster.org (a.k.a. build.gluster.org). (And soon add nfs-ganesha, samba, and ceph too.) What do you think about running rsyncd on bits.gluster.org.
11:58 kkeithley Or should we put it somewhere else? E.g. a dedicated rackspace server?
11:58 kkeithley hagarth: ^^^
11:59 lalatenduM kkeithley, regarding Coverity , I am little unsure how these runs will help us, we are using Coverity scan website as of now for tracking Coverity issues
11:59 kkeithley community visibility?
12:00 ndevos kkeithley: I've got a job setup on build.gluster.org to ssh to download.gluster.org and download nightly builds there - not sure what the difference with bits.gluster.org is
12:00 lalatenduM kkeithley, I am think two Coverity scan results might confuse community,
12:00 kkeithley maybe if people in the community see them they might try to fix some of them.
12:00 kkeithley bits.gluster.org == build.gluster.org
12:01 ndevos ah
12:01 lalatenduM kkeithley, check my latest mail on Coverity scan on gluster devel
12:04 ppai joined #gluster-dev
12:11 tdasilva joined #gluster-dev
12:13 itisravi joined #gluster-dev
12:14 ndevos Humble, lalatenduM: I'm done with the backport guideline page, the next one is free to edit it :)
12:15 lalatenduM ndevos, excellent :)
12:15 JustinClift kkeithley: I've no objection to having rsyncd run on bits.gluster.org.  As long as it's secured ok. :)
12:16 JustinClift kkeithley: It could be useful for anyone wanting to pull down stuff.
12:16 ndevos lalatenduM: I dont like it that it is not a numbered list and each item got bold printed, but the numbering didnt really work when adding the commands :-/
12:17 lalatenduM ndevos, will fix it
12:18 ndevos lalatenduM: I tried (hard!), good luck!
12:20 hagarth JustinClift: +1
12:20 lalatenduM ndevos, check the page , just edited one command "
12:20 lalatenduM * Git clone the GlusterFS code
12:20 lalatenduM git clone ssh://username@review.gluster.org/glusterfs</code>" do you like it?
12:21 ndarshan joined #gluster-dev
12:23 ndevos lalatenduM: the problems is that the numbering breaks that way, and the <code> part is not really indented
12:35 lalatenduM ndevos, agree, check the page now and let me know
12:36 shubhendu joined #gluster-dev
12:36 ndevos lalatenduM: I like that more, but you dont happen to know how to remove the bullet-point in front of the command?
12:37 ndevos lalatenduM: we should have a markdown format for the wiki :D
12:38 lalatenduM ndevos, :), lect me try if I can remove the bullet point
12:38 ndevos lalatenduM: I dont mind the formatting too much, it's more important to have something and can point others to it
12:48 awheeler joined #gluster-dev
12:53 lalatenduM ndevos, I think you will like the page now :)
12:54 ndevos lalatenduM: hehe, yeah that looks good!
12:54 lalatenduM ndevos, cool :)
12:54 Humble ndevos, looks good
12:56 ndevos lalatenduM: you want to send an email to the devel ML about it?
12:58 ndevos Humble, lalatenduM: care to review backports http://review.gluster.org/7829 and http://review.gluster.org/7830 ?
12:58 glusterbot Title: Gerrit Code Review (at review.gluster.org)
12:58 lalatenduM ndevos, sure will send it gluster devel
12:58 ndevos lalatenduM: great, thanks!
12:59 lalatenduM ndevos, checking for the back ports
13:08 lalatenduM ndevos, sent the mail on gluster-devel
13:08 JustinClift kkeithley: Should we update the rpm.t to test the last 3 epel versions?
13:08 JustinClift (at least in master)
13:08 * JustinClift just tested it locally, and it worked fine on CentOS 6.5
13:12 JustinClift That's weird...
13:12 JustinClift Has anyone seen this error when uninstalling rpms compiled from git?
13:12 JustinClift /sbin/ldconfig: relative path `0' used to build cache
13:12 JustinClift warning: %postun(glusterfs-libs-3.5qa2-0.510.git768a4ad.el6.x86_64) scriptlet failed, exit status 1
13:12 JustinClift Ahhh, warning, not error.
13:13 JustinClift rpms still uninstalled
13:15 ndevos JustinClift: hmm, no? what does 'rpm -q --scripts -p glusterfs-libs-3.5qa2-0.510.git768a4ad.el6.x86_64.rpm' show?
13:18 kkeithley JustinClift: last three being 5, 6, .... and 7? Once RHEL/CentOS 7 ship then I think it gets more interesting. I don't feel like it's a priority to test 7 just yet. But if it's easy and it works, then sure.
13:24 lalatenduM ndevos, http://review.gluster.org/7829 looks like a straight forward backport
13:24 glusterbot Title: Gerrit Code Review (at review.gluster.org)
13:27 JustinClift ndevos: I'll have to check again later.  Already nuked those rpms.
13:28 JustinClift kkeithley: So far, testing only on CentOS 6.5 host, epel-5,6,7 are working fine in master branch (once dependencies on the host are installed).
13:28 * JustinClift will spin up a Fedora 20 instance in a bit and see how that goes too
13:30 jobewan joined #gluster-dev
13:32 ndk joined #gluster-dev
13:32 ndevos lalatenduM: both backports are pretty straight forward, but I prefer a 2nd pair of eyes before merging the changes in 3.5
13:34 ndevos JustinClift: I'm more in favor to get the regressions tested on el-6.5 and not on fedora yet, maybe add tests for fedora at a later stage?
13:35 ndevos JustinClift: also, some builders can have problems building fedora packages in mock, it might require mock and kernel updates :-/
13:41 lalatenduM ndevos, yup
13:46 JustinClift ndevos: no worries. ;)
13:47 ndevos JustinClift: just helping you with your scheduling and planning :P
13:47 JustinClift k, testing epel-5, 6, 7 works fine atm with both release-3.5 and master branches
13:47 JustinClift This is good
13:48 ndevos oh really, 3.5 on epel-7 too?
13:48 JustinClift Yep
13:48 * ndevos goes and enables it for the next nightly builds
13:49 JustinClift Cool
13:49 JustinClift k, I've kicked off 20 VM's in Rackspace to run the full regression test on master branch
13:50 JustinClift After the weekly meeting I'll go and collect stats
13:50 * JustinClift is hoping for 100% pass
14:13 skoduri joined #gluster-dev
14:18 wushudoin joined #gluster-dev
14:42 ndevos JustinClift: no meeting reminder sent to the lists?
14:42 JustinClift ndevos: Arrgh
14:42 kkeithley are we back to etherpad or still using goo-docs
14:43 JustinClift Yeah, I was just starting looking into that earlier today, then got grabbed for something else.
14:43 JustinClift Hmmm, I think we'll TitanPad it for today and hope it doesn't crash
14:44 JustinClift Gimme a few mins to copy stuff across to the old TitanPad, then I'll send out a reminder
14:59 hagarth joined #gluster-dev
15:00 JustinClift *** Gluster Community Meeting time ***
15:00 JustinClift (in #gluster-meeting)
15:02 JustinClift kkeithley: Thx.  I keep on doing that.  Guess it'll take a few more reptitions for it to sink in for me. ;)
15:02 lalatenduM joined #gluster-dev
15:03 kdhananjay joined #gluster-dev
15:06 kkeithley are we starting?
15:08 kkeithley gah, wrong window
15:08 JustinClift Yeah :)
15:15 lpabon joined #gluster-dev
16:21 lpabon left #gluster-dev
16:22 lpabon joined #gluster-dev
16:28 ndevos kkeithley: do you want to keep bug 1081016 as a blocker for 3.5.1?
16:28 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1081016 unspecified, unspecified, ---, kkeithle, ASSIGNED , glusterd needs xfsprogs and e2fsprogs packages
16:29 xavih kkeithley: Should I change the name of the test script for bug 1099955 on release-3.4 branch ? (from bug-859181.t to bug-1099955.t)
16:29 glusterbot Bug https://bugzilla.redhat.com:443/show_bug.cgi?id=1099955 high, unspecified, ---, vbellur, NEW , self-heal process can sometimes create directories instead of symlinks for the root gfid file in .glusterfs
16:29 ndevos xavih: I normally dont do that (anymore)
16:30 xavih ndevos: ok
16:36 xavih patch updated with new bug-id: http://review.gluster.org/6737
16:36 glusterbot Title: Gerrit Code Review (at review.gluster.org)
16:50 kshlm joined #gluster-dev
16:58 kkeithley xavih: no, let's leave it the same on the master and release-3.x branches
17:01 kkeithley ndevos: wrt 1081016, yes, probably.
17:11 ndk joined #gluster-dev
17:30 hagarth joined #gluster-dev
18:34 edward1 joined #gluster-dev
18:36 wushudoin joined #gluster-dev
18:47 JoeJulian @load karma
18:47 glusterbot JoeJulian: Error: You don't have the owner capability. If you think that you should have this capability, be sure that you are identified before trying again. The 'whoami' command can tell you if you're identified.
18:48 JoeJulian @load karma
18:48 glusterbot JoeJulian: The operation succeeded.
19:07 ira joined #gluster-dev
20:03 tdasilva left #gluster-dev
20:12 semiosis JoeJulian++
20:22 ira joined #gluster-dev
20:30 badone joined #gluster-dev
20:53 badone joined #gluster-dev
21:26 swebb joined #gluster-dev
21:38 edward1 joined #gluster-dev
22:26 kkeithley1 joined #gluster-dev
22:35 portante joined #gluster-dev
22:35 sickill_ joined #gluster-dev
22:36 [o__o] joined #gluster-dev
22:47 nixpanic_ joined #gluster-dev
22:47 nixpanic_ joined #gluster-dev
22:48 foster_ joined #gluster-dev
22:49 hchiramm_ joined #gluster-dev
23:11 kkeithley joined #gluster-dev

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