Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2016-08-29

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

All times shown according to UTC.

Time Nick Message
01:37 penguinRaider joined #gluster-dev
01:41 EinstCrazy joined #gluster-dev
01:42 EinstCrazy joined #gluster-dev
02:01 EinstCrazy joined #gluster-dev
02:18 EinstCrazy joined #gluster-dev
02:22 spalai joined #gluster-dev
02:32 spalai joined #gluster-dev
02:32 hagarth joined #gluster-dev
02:36 spalai joined #gluster-dev
03:18 magrawal joined #gluster-dev
03:35 kshlm joined #gluster-dev
03:42 kdhananjay joined #gluster-dev
03:46 itisravi joined #gluster-dev
03:58 kdhananjay1 joined #gluster-dev
04:01 riyas joined #gluster-dev
04:06 atinm joined #gluster-dev
04:16 kdhananjay joined #gluster-dev
04:17 shubhendu joined #gluster-dev
04:29 shubhendu joined #gluster-dev
04:30 nbalacha joined #gluster-dev
04:33 nishanth joined #gluster-dev
04:38 bkunal joined #gluster-dev
04:42 jiffin joined #gluster-dev
04:43 sanoj joined #gluster-dev
04:54 ndarshan joined #gluster-dev
04:59 karthik__ joined #gluster-dev
05:02 rafi joined #gluster-dev
05:04 ankitraj joined #gluster-dev
05:10 loadtheacc joined #gluster-dev
05:21 kdhananjay joined #gluster-dev
05:22 atalur joined #gluster-dev
05:34 aravindavk joined #gluster-dev
05:39 karthik_ joined #gluster-dev
05:43 skoduri joined #gluster-dev
05:48 mchangir joined #gluster-dev
05:49 ramky joined #gluster-dev
05:52 loadtheacc joined #gluster-dev
05:57 ndarshan joined #gluster-dev
06:03 aspandey joined #gluster-dev
06:07 Saravanakmr joined #gluster-dev
06:10 Bhaskarakiran joined #gluster-dev
06:11 ppai joined #gluster-dev
06:14 rjoseph kshlm: http://review.gluster.org/#/c/15073/ Can you please review it and provide an ack?
06:17 spalai joined #gluster-dev
06:18 prasanth joined #gluster-dev
06:19 kshlm rjoseph, Looked good to me. Merged the change.
06:21 karthik_ joined #gluster-dev
06:28 rastar joined #gluster-dev
06:29 Muthu_ joined #gluster-dev
06:39 spalai joined #gluster-dev
06:42 devyani7 joined #gluster-dev
06:43 asengupt joined #gluster-dev
06:48 overclk_ itisravi: ping (?)
06:49 itisravi overclk_: was in a meeting sorry
06:54 devyani7 joined #gluster-dev
06:54 * post-factum pings those who could help with valgrind+massif for gluster profiling
06:58 ankitraj joined #gluster-dev
07:09 gem joined #gluster-dev
07:10 ashiq joined #gluster-dev
07:12 k4n0 joined #gluster-dev
07:13 spalai joined #gluster-dev
07:22 kotreshhr joined #gluster-dev
07:28 msvbhat joined #gluster-dev
07:29 sanoj joined #gluster-dev
07:35 EinstCrazy joined #gluster-dev
07:50 EinstCra_ joined #gluster-dev
08:10 hgichon joined #gluster-dev
08:10 hgichon_ joined #gluster-dev
08:25 Debloper joined #gluster-dev
08:32 rjoseph kshlm++ Thanks for merging the patch
08:32 glusterbot rjoseph: kshlm's karma is now 106
08:34 aspandey_ joined #gluster-dev
08:37 itisravi joined #gluster-dev
08:38 atalur joined #gluster-dev
08:39 pur joined #gluster-dev
09:02 itisravi_ joined #gluster-dev
09:13 poornima joined #gluster-dev
09:14 poornima ndevos, have addressed your comments on upcall patches http://review.gluster.org/#/c/15313/ and http://review.gluster.org/#/c/15301/ can you please take a look at it when possible
09:27 penguinRaider joined #gluster-dev
09:38 kdhananjay joined #gluster-dev
09:40 aspandey_ joined #gluster-dev
09:42 rastar joined #gluster-dev
09:46 kdhananjay joined #gluster-dev
09:51 atinm joined #gluster-dev
09:54 kshlm ndevos, In the nightly_rpms ci.centos.org job, the status and console output pages just link to the project page for the child jobs.
09:54 kshlm Is there a way to link to the actual job?
09:54 kshlm For eg, https://ci.centos.org/view/Gluster/job/gluster_nightly-rpm-builds/171/
09:55 kshlm The links are all to glusterd_build-rpms project.
09:55 kshlm Not to the actual jobs, which would make it much easier to get rpms.
09:57 kshlm nigelb, Do you think this could be achieved? ^
10:01 penguinRaider joined #gluster-dev
10:07 karthik_ joined #gluster-dev
10:11 nigelb kshlm: I want to talk to ndevos when he's around about whether we need the job on centos at all.
10:11 nigelb we do the same thing on build.gluster.org.
10:12 nigelb I could just set it to email maintainers + patch author trivially.
10:13 ndevos kshlm: you can get the nightly RPMs with the .repo files from http://artifacts.ci.centos.org/gluster/nightly/
10:13 misc so, quick question, would people be against a reboot of server on the 1st ?
10:13 misc (ie, after the release) ?
10:13 kshlm nigelb, The nightly job on centos-ci generates and maintains a yum repo, which the rpm build jobs on build.g.o don't.
10:14 ndevos nigelb: we also want to provide packages in the different distributions, and preferably not on download.gluster.org
10:14 kshlm ndevos, I just found out that repo is created by the nightly job.
10:15 kshlm misc, I don't have any problem with that.
10:15 kshlm 3.7.15 will be tagged and released tomorrow.
10:15 ndevos kshlm: yeah, is that a problem?
10:15 misc kshlm: yeah, just wanted to make sure that release is ok, and no last minute issue
10:16 kshlm You'll need to check with 3.9 maintainers about their status. 3.9 is to be branched at the end of this month ie. in 2 days.
10:16 kshlm ndevos, Nope. That's actually better.
10:17 misc kshlm: yup, gonna send a email on infra as well
10:17 misc (just too busy with gerrit and ovirt right now, trying to be tactful on ansible code review)
10:17 kshlm But there is no easy way to get to the repo from the nightly build job status page.
10:19 ndevos kshlm: hmm, I thought it was listed on the nightly build page in the docs... but I cant even find the page now :-/
10:21 penguinRaider joined #gluster-dev
10:21 anoopcs ndevos, Was the intentional that you retained the version on which glusterfs-3.8.4 was filed against to 3.8.2?
10:22 anoopcs https://bugzilla.redhat.com/show_bug.cgi?id=glusterfs-3.8.4
10:22 glusterbot Bug glusterfs: could not be retrieved: InvalidBugId
10:22 ndevos anoopcs: there wasnt a 3.8.4 version, if there is now, you can change it :)
10:23 anoopcs ndevos, Did you mean 3.8.3?
10:23 * ndevos would be more than happy to just have a 3.8 version
10:24 ndevos anoopcs: 3.8.3 has been released already? the tracker for 3.8.4 should be against version 3.8(.4)
10:25 anoopcs That's the most recent version available in bugzilla and I thought we used to do so.. If not feel free to change it back to appropriate version
10:26 ndevos oh, it really does not matter what .minor version is used, I only check for 3.8 anyway
10:27 nigelb ndevos: huh, how's that connected to build.gluster.org vs ci.centos.org?
10:27 nigelb Does the Centos CI job push to distributions?
10:28 anoopcs ndevos, Ok.
10:28 ndevos nigelb: the intention is to provide the packages to other users in the centos-ci, for example nfs-ganesha
10:29 nigelb Ah.
10:29 ndevos nigelb: our Jenkins was too overloaded, and nobody maintained it, so the centos team offered their assistance
10:30 ndevos nigelb: we can also run real hardware multi-host tests there, and when Glusto has tests to run, we plan to do so nightly
10:31 nigelb ndevos: Yes, I plan to actually get glusto tests working on centos ci soon, even if it's as simple as install gluster and see if it starts up.
10:31 nigelb The question now is, now that we do maintain our jenkins what goes where.
10:31 nigelb I'm not going to change anything for now.
10:32 ndevos well, all the heavy testing would be done in the centos-ci, that has access to real hardware
10:32 nigelb "real hardware"? Isn't it all VMs anyway?
10:32 ndevos no, *real* hardware
10:33 ndevos https://wiki.centos.org/QaWiki/PubHardware
10:33 ndevos and installed cleanly for each test
10:34 nigelb Sure but when we want to run a test, when we get a VM from this *real* hardware, don't we?
10:34 ndevos no, it really runs on the physical machines
10:34 ndevos you reserve a machine, and get one cleanly installed with a minimal CentOS 6 or 7 environment
10:35 nigelb That doesn't make sense. Especially considering the number of communities and number of servers each use.
10:35 ndevos once done with testing, you return the machine to the pool
10:35 nigelb But this makes things interesting.
10:35 nigelb I've been wondering how to acquire real hardware for some tests.
10:35 ndevos I think there are ~250 physical machines available, how many do you need?
10:35 nigelb This may solve my problems.
10:35 nigelb 8 :)
10:36 ambarish joined #gluster-dev
10:36 ndevos projects can reserve up to 10 (or 20?) machines at the same time, if we need to increase the limits, we can request that :)
10:37 nigelb Noted.
10:37 nigelb That closes one of my open loops to explore.
10:37 ndevos there are VMs as Jenkins Slaves, but those are like the management heads that reserve machines and start tests, the slaves do not really run tests themselves
10:37 nigelb ah.
10:37 nigelb On that note, I'm going to start the XML -> YML conversion for Centos CI jobs this week.
10:38 ndevos enjoy!
10:38 nigelb Which brings me back to another question.
10:38 nigelb We've talked about moving the jobs to gerrit.
10:38 nigelb Do you prefer I do the move before or after?
10:38 ndevos if you're interested, you can also rewrite the duffy-reservation python scripts to use cicoclient :)
10:39 ndevos I do not really have a preference :)
10:39 nigelb You're among the few people who review the jobs :)
10:39 nigelb Though, I intend to take on most of the load away from you.
10:39 ndevos I dont think I've done much with the centos-ci/gluster jobs lately
10:43 nigelb ndevos: I'm planning on giving Glusto an infra boost by getting it setup to run ASAP on centos CI (I may need help later in the week)
10:44 nigelb Hopefully, by summit time we can talk about what tests we want to write and it should Just Work.
10:45 ndevos nigelb: msvbhat and kshlm started some work in distaf in the centos-ci, hopefully that can be reused and modified for Glusto
10:46 nigelb that's the plan.
10:48 atinm joined #gluster-dev
10:50 ndevos nigelb: I think our Jenkins will keep on doing most of the smoke, regression testing and release building, but not much more
10:51 nigelb ndevos: I'd like it to eventually automate more release work as well.
10:52 ndevos nigelb: yeah, that would be good, there are considerations on doing CBS builds from the CentOS-CI, but I do not know how far that is away
10:53 ndevos nigelb: I dont think it can be automated towards Fedora Koji, no idea about Debian, Ubuntu and OBS builds...
10:53 nigelb ndevos: I'm hoping to have hallway conversations with you and Kaleb on how packaging can be less manual.
10:54 ndevos nigelb: I guess that depends more on the distributions than on Kaleb and me :)
10:54 nigelb Well, if I can get it down to "sign and upload", does that make packgers life easier?
10:55 nigelb Ceph has gone through this. So, I'm planning on talking to their CI folks do things.
10:56 ndevos sure, but there are differences in our upstream .spec that we do not want in the distributions, like the glusterfs-regression-tests sub-package
10:57 ndevos and, someone still needs to upload the tarball+.spec to Fedora/CentOS/... buildsystems, definitely not all of them allow automation
10:58 ndevos integration with other packages in distribution is important, we can not do that when our packages are only on download.gluster.prg
10:58 nigelb Yeah, hence the need for us to talk about how best to approach this.
10:59 ndevos I know others want the ability for the CentOS SIGs, so that is coming at one point
10:59 ndevos but really, I dont know the situation for other distributions, but I expect several of them need a person to upload/build
11:00 misc I think this can be scripted
11:01 misc the real problem in the case of Fedora is that they want to have the spec in their git
11:01 ndevos and the tarball in the look-aside-cache
11:01 ndevos and uploading to either requires an account with the needed permissions
11:01 misc yeah, the tarball is easy to automate
11:02 misc the git is more tricky, cause someone might change git outside of the automation
11:02 misc and IIRC, this was alaready attempted for olpc
11:02 misc (not without resistance)
11:04 msvbhat joined #gluster-dev
11:04 ndevos if it can be done, that would be nice, but I doubt we'll get a worker-ant account that can automatically do updates
11:04 misc yup, me too
11:05 misc but I also feel this is the right moment to experiment with that, with the modularity project being started
11:05 misc too bad, flock was 1 month ago, we could have discussed that
11:19 Muthu_ joined #gluster-dev
11:20 atinm joined #gluster-dev
11:31 kdhananjay joined #gluster-dev
11:44 ankitraj joined #gluster-dev
11:46 ppai joined #gluster-dev
11:51 shyam joined #gluster-dev
12:14 mchangir joined #gluster-dev
12:21 ppai joined #gluster-dev
12:25 aravindavk joined #gluster-dev
12:27 magrawal joined #gluster-dev
12:42 atinm joined #gluster-dev
12:45 lpabon joined #gluster-dev
12:46 poornima joined #gluster-dev
12:57 shyam joined #gluster-dev
13:01 aravindavk joined #gluster-dev
13:04 hchiramm joined #gluster-dev
13:18 EinstCrazy joined #gluster-dev
13:29 ira joined #gluster-dev
13:44 jiffin1 joined #gluster-dev
13:45 msvbhat joined #gluster-dev
13:55 aravindavk joined #gluster-dev
14:01 atalur joined #gluster-dev
14:02 jiffin1 joined #gluster-dev
14:06 dlambrig joined #gluster-dev
14:08 ashiq_ joined #gluster-dev
14:21 samikshan joined #gluster-dev
14:22 ira joined #gluster-dev
14:22 nbalacha joined #gluster-dev
14:23 lpabon joined #gluster-dev
14:24 lpabon joined #gluster-dev
14:26 lpabon joined #gluster-dev
14:35 atalur joined #gluster-dev
15:02 jiffin ndevos, kkeithley: can u guys please merge http://review.gluster.org/#/c/15105/?
15:05 wushudoin joined #gluster-dev
15:10 kkeithley jiffin: done
15:11 ndevos ah!
15:11 kkeithley ah! ?  Is that a good ah! or a bad ah! ?
15:12 ndevos it's the "ah!" for "you beat me to it"
15:12 kkeithley I'm going to merge all the unused variable fixes that have +2.  Would appreciate it if the maintainers/owners of the remaining patches would please review them
15:14 hchiramm joined #gluster-dev
15:16 kkeithley oh, unless ndevos merges them first
15:18 dlambrig joined #gluster-dev
15:23 ndevos not all of them, only the ones I feel comfortable with or noticed the maintainer did a +2 before the regression tests finished
15:23 ndevos whats left is all yours, I wont do more of them today
15:24 ndevos quite some left, see http://review.gluster.org/#/q/topic:bug-1369124+status:open
15:27 ndevos hagarth: could you abandon http://review.gluster.org/6449 ?
15:29 shaunm joined #gluster-dev
15:30 rafi1 joined #gluster-dev
15:32 kkeithley nearly lunch time. I'll wait until after lunch. That will give anyone who's still around and wants to object a chance.
15:38 msvbhat joined #gluster-dev
15:39 spalai joined #gluster-dev
15:40 ambarish joined #gluster-dev
15:45 hagarth ndevos: done
15:47 shubhendu joined #gluster-dev
16:06 rafi joined #gluster-dev
16:47 shaunm joined #gluster-dev
16:48 kotreshhr left #gluster-dev
17:33 skoduri joined #gluster-dev
17:47 rafi joined #gluster-dev
17:50 jiffin joined #gluster-dev
18:05 atinm joined #gluster-dev
18:17 shaunm joined #gluster-dev
18:17 k4n0 joined #gluster-dev
18:23 rastar joined #gluster-dev
18:25 hagarth joined #gluster-dev
18:50 jiffin joined #gluster-dev
18:59 reapbooster joined #gluster-dev
19:12 devyani7 joined #gluster-dev
19:13 msvbhat joined #gluster-dev
19:21 devyani7 joined #gluster-dev
19:29 post-factum @later
19:29 post-factum glusterbot: help
19:29 glusterbot post-factum: (help [<plugin>] [<command>]) -- This command gives a useful description of what <command> does. <plugin> is only necessary if the command is in more than one plugin. You may also want to use the 'list' command to list all available plugins and commands.
19:29 post-factum glusterbot: help list
19:29 glusterbot post-factum: (list [--private] [--unloaded] [<plugin>]) -- Lists the commands available in the given plugin. If no plugin is given, lists the public plugins available. If --private is given, lists the private plugins. If --unloaded is given, it will list available plugins that are not loaded.
19:30 post-factum @later tell pranithk ping me once you are here
19:30 glusterbot post-factum: The operation succeeded.
20:08 shyam joined #gluster-dev
20:49 ira joined #gluster-dev

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