Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2016-07-29

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

All times shown according to UTC.

Time Nick Message
00:56 hchiramm_ joined #gluster-dev
01:30 hagarth joined #gluster-dev
01:56 hchiramm_ joined #gluster-dev
02:15 spalai joined #gluster-dev
02:32 spalai left #gluster-dev
02:56 hchiramm_ joined #gluster-dev
03:25 pranithk1 joined #gluster-dev
03:38 nigelb We have a gerrit issue. Being tracked here: https://bugzilla.redhat.co​m/show_bug.cgi?id=1361400
03:38 glusterbot Bug 1361400: high, high, ---, nigelb, ASSIGNED , Github login issues on Gerrit
03:38 nigelb Everyone is affected. Please try not to logout. You will be locked out.
03:45 magrawal joined #gluster-dev
04:04 spalai joined #gluster-dev
04:06 nigelb And Gerrit is back online :)
04:09 nishanth joined #gluster-dev
04:10 nbalacha joined #gluster-dev
04:14 aspandey joined #gluster-dev
04:17 spalai left #gluster-dev
04:17 suliba joined #gluster-dev
04:18 spalai joined #gluster-dev
04:23 itisravi joined #gluster-dev
04:33 mchangir joined #gluster-dev
04:33 nbalacha joined #gluster-dev
04:33 penguinRaider joined #gluster-dev
04:35 hchiramm_ joined #gluster-dev
04:40 poornimag joined #gluster-dev
04:40 itisravi joined #gluster-dev
04:42 suliba joined #gluster-dev
04:44 asengupt joined #gluster-dev
04:44 atinm joined #gluster-dev
04:52 atinm joined #gluster-dev
05:02 jiffin joined #gluster-dev
05:11 kshlm joined #gluster-dev
05:18 sanoj joined #gluster-dev
05:23 poornimag joined #gluster-dev
05:29 sakshi joined #gluster-dev
05:32 rafi joined #gluster-dev
05:32 devyani7_ joined #gluster-dev
05:34 aravindavk joined #gluster-dev
05:37 Bhaskarakiran joined #gluster-dev
05:39 prasanth joined #gluster-dev
05:42 mchangir joined #gluster-dev
05:49 kotreshhr joined #gluster-dev
05:52 shubhendu_ joined #gluster-dev
05:57 sanoj joined #gluster-dev
05:58 rastar joined #gluster-dev
06:00 ankitraj joined #gluster-dev
06:01 ankitraj joined #gluster-dev
06:05 ndarshan joined #gluster-dev
06:08 hgowtham joined #gluster-dev
06:09 mchangir joined #gluster-dev
06:11 kdhananjay joined #gluster-dev
06:11 Apeksha joined #gluster-dev
06:19 kotreshhr joined #gluster-dev
06:26 rafi nigelb: I couldn't search in gluster.readthedocs.io
06:26 rafi nigelb: Do you know any outage ?
06:28 spalai joined #gluster-dev
06:28 spalai1 joined #gluster-dev
06:29 nigelb rafi: not that I know of.
06:30 Manikandan joined #gluster-dev
06:30 ashiq joined #gluster-dev
06:30 nigelb rafi: ah, seems to be an RTD outage.
06:32 rafi nigelb: oh okey
06:33 rafi nigelb: any idea, when will it be back online ?
06:35 nigelb No idea. I suspect their search backend is down.
06:35 nigelb I'll poke people on IRC
06:44 msvbhat joined #gluster-dev
06:50 Muthu_ joined #gluster-dev
06:51 Manikandan joined #gluster-dev
06:56 pranithk1 joined #gluster-dev
06:58 mchangir joined #gluster-dev
07:01 itisravi joined #gluster-dev
07:10 karthik_ joined #gluster-dev
07:21 kotreshhr joined #gluster-dev
07:22 pranithk1 joined #gluster-dev
07:31 mchangir joined #gluster-dev
07:36 hchiramm_ joined #gluster-dev
07:42 poornimag joined #gluster-dev
08:29 Guest17640 joined #gluster-dev
08:30 Guest17640 left #gluster-dev
08:38 magrawal_ joined #gluster-dev
08:40 mchangir joined #gluster-dev
08:53 aravindavk joined #gluster-dev
09:05 gem joined #gluster-dev
09:13 aspandey joined #gluster-dev
09:24 Apeksha joined #gluster-dev
09:27 mchangir joined #gluster-dev
09:31 nishanth joined #gluster-dev
09:39 rraja joined #gluster-dev
09:56 mchangir ndevos, would http://review.gluster.org/14957 be an acceptable item for 3.8.3 ?
10:01 ndevos mchangir: not really, unless it fixes some critical bugs that can not be addressed otherwise
10:01 mchangir hmm
10:01 pranithk1 joined #gluster-dev
10:02 ndevos mchangir: it looks like a normal maintenance/cleanup to me, not something that fixes bugs
10:02 mchangir ndevos, that's right
10:04 ndevos mchangir: we have started some documentation for backport criteria, see https://public.pad.fsfe.org/p/g​lusterfs-release-process-201606
10:04 mchangir ndevos, I'l take a look
10:14 mchangir ndevos, ok, I've read up the release process ... so it seems that my patch could get into mainline and make it to 3.10.0 since it won't make it to 3.9.0 either
10:15 ndevos mchangir: ok, 3.9 needs the change ready before the end of august
10:19 pranithk1 joined #gluster-dev
10:24 penguinRaider joined #gluster-dev
10:25 misc ndevos: would you be opposed into getting bugs.cloud.gluster.org moved to a different server, and potentially under a different name ?
10:25 misc (provided we keep the name for backward compat)
10:26 misc (for a short period, like 6 months)
10:26 ndevos misc: no problem what so ever, you can even put it on an othere server that hosts a website already
10:27 msvbhat joined #gluster-dev
10:27 misc I almost want to move it to openshift v3, but there is no regular rebuild of container feature (yet)
10:28 misc (and also, no openshiftv3 supported yet)
10:28 mchangir ndevos, are upstream branches created directly from mainline ?
10:28 ndevos misc: the website itself does not change much, only the generate json blob needs regular refereshing
10:28 ndevos mchangir: yes
10:29 ndevos mchangir: so when your change is merged before 3.9 gets branched, it'll be included automatocally
10:29 mchangir ndevos, ok
10:30 gem joined #gluster-dev
10:36 msvbhat joined #gluster-dev
10:37 skoduri joined #gluster-dev
10:52 spalai joined #gluster-dev
11:10 nigelb misc: Can we automatically only apply security updates?
11:10 misc nigelb: that was a security update
11:10 nigelb Oh.
11:11 nigelb It was in the updates stream on another server than security update.
11:11 misc see https://lwn.net/Articles/695067/
11:11 misc and to be fair, no, I want automated update
11:11 misc we had too much issue in the past with "we can't update, it will break"
11:11 nigelb I don't want EVERYTHING updated automatically without trying it on staging first.
11:12 misc nigelb: not gonna happen, I am not gonna spend more time classifying updates
11:12 Bhaskarakiran joined #gluster-dev
11:12 nigelb er, isn't that an option in yum-updates already?
11:12 nigelb I know the debian equivalent has that option.
11:13 nigelb I've used that extensively in the past.
11:13 misc I am still against it
11:13 nigelb how is updating with testing a good idea?
11:14 misc well, it depend on the testing
11:14 misc if that's manual testing, that's no
11:14 misc if you manage to get automated testing, maybe
11:14 nigelb all we have to do is.. does the app still work.
11:14 nigelb for one application.
11:14 misc nigelb: in this case, the application was still working
11:15 nigelb actually, testing login would be something critical.
11:15 nigelb all the build nodes can have automatic updates.
11:15 misc so you need a github account and some webbased broser testing
11:16 misc and I also assume a github account not using 2 FA
11:16 nigelb it takes about 10 minutes :|
11:16 misc nigelb: to automate ?
11:16 nigelb to test manually.
11:16 misc yeah, but I do not want manual test
11:17 misc and today, the login is what broke, tomorow, what if it was some others things ?
11:17 misc are we gonna do a full test of gerrit for every upgrade ?
11:17 nigelb all I want to make sure is that people can do their jobs.
11:17 nigelb So the only things to test are - login and review.
11:18 nigelb And push code.
11:18 nigelb Two things.
11:18 misc that's 3
11:18 nigelb Fine, 3 things.
11:18 nigelb I think it's okay to spend 10 minutes to test an update instead of locking people out for half a day.
11:20 misc ok so let's decide that the gerrit server os is not under osas control and you can do what you wish
11:20 aravindavk joined #gluster-dev
11:36 nigelb Right now any discussion is moot since we don't have an official staging server.
11:36 nigelb Once we do have that in place, I'd like to revisit this.
11:37 misc the solution is likely to have first better moniroting
11:38 misc and better coverage
11:39 misc we suffered a lot more from lack of security than from gerrit outage due to a once a year java update
11:40 nigelb Is there a chance we can get a nagios server? I'd like to catch some of the issues on netbsd faster than we get complaints from failed builds.
11:41 misc a chance, yes, but then, I have to stop working on automating the deployment of VM on the 2 new hosts
11:41 nigelb Not instantly.
11:42 misc then yes, that's on the roadmap since a long time
11:42 misc justin did found someone to install zabbix, didn't worked out, but yeah
11:43 misc but we would need to also make sure the netbsd builder are in the cfg mgmt system, so find a solution to the "/usr is read-only" issue, or me to work on the workaround I wanted to add
11:45 nigelb My idea is to redo how we use playbooks
11:45 nigelb have two playbooks for everything.
11:45 nigelb and use them with an -l flag
11:45 nigelb one playbook for centos machines and one for netbsd machines.
11:46 nigelb as opposed to one role per playbook.
11:46 nigelb er one playbook per role
11:46 misc nope, automation depend on that
11:46 misc but we can surely try to make 1 playbook for netbsd and 1 for centos, and do some magic for base
11:48 misc provided that only netbsd use that feature, we can lock and unlock /usr at the start of 'base' and 'jenkins'
11:50 nigelb hrm, we should do a weekly infra call to talk about what we're working on.
11:51 nigelb Or every 2 weeks
11:51 penguinRaider joined #gluster-dev
11:54 mchangir joined #gluster-dev
11:55 shubhendu_ joined #gluster-dev
11:56 ira joined #gluster-dev
12:05 kotreshhr joined #gluster-dev
12:12 mchangir joined #gluster-dev
12:30 misc nigelb: yeah, I was pondering on a irc meeting or something
12:30 misc also, i think that we can make a playbook that include all the others, so you can do deploy using -l if you want, but we keep existing automation on commit as deployed
12:31 nigelb I don't mind an IRC meeting. We have a channel for it anyway.
12:31 mchangir joined #gluster-dev
12:32 misc well, irc meetings are a bit slower
12:32 misc you can compare how much people can talk 1h vs what we can type
12:33 nigelb I'm in favor of short meetings, but IRC meetings are logged
12:33 nigelb So we don't have to take notes separately
12:33 misc now, non text meetings are annoying, you can't grep them, not everybody is able to follow due to ton of reasons like bad hearings, bad network, thick accent from one participants, etc, so that's not really as accessible as i would like
12:34 misc nigelb: short meetings, sure :) we are 2 engineers speaking of technical stuff where we have opinions, there is no way this is gonna be short :p
12:35 nigelb lol
12:35 misc and if we do short, there is no way we cna justify a multi days retreat in a sunny place
12:42 misc grmbl, so I start a test of VM installation before going to lunch
12:42 misc and... it block on me clicking "ok"
12:43 nigelb misc: will you be copying gerrit VM as is?
12:43 nigelb I was wondering if we could upgrade to el7...
12:44 nigelb and add more space. That's what's blocking me on setting up backups.
12:45 atinm joined #gluster-dev
12:46 spalai left #gluster-dev
12:49 mchangir joined #gluster-dev
12:49 misc nigelb: I am planning on doing indeed just a copy
12:50 misc upgrade to EL7 is basically a reinstallation
12:50 misc add more space, this can be done, just add a disk ?
12:53 nigelb yeah. There's just too little space to copy everything to a folder to run tar on it.
12:53 misc I can do that now if you want, but I suspect it might requires a restart
12:53 misc unless.... I pretend it to be a floppy of 4G
12:54 nigelb Monday maybe.
12:54 nigelb Do not want to make large changes on a friday night.
12:54 nigelb I can file a bug now and announce a downtime for Monday.
12:58 misc we can also make it without downtime
12:58 misc just use a nfs export
12:58 misc if that's just a one off
13:01 ndarshan joined #gluster-dev
13:04 nigelb no, I want to setup a backup cron.
13:04 nigelb Also happy to wait until after migration.
13:06 nigelb misc: Hey, happy sysadmin appreciation day :)
13:10 misc nigelb: thanks !
13:10 misc nigelb: I should say you too
13:10 semiosis joined #gluster-dev
13:20 nigelb :)
13:22 misc so I am almost able to install VM on centos 6 automatically
13:22 misc then if I manage to get networking going
13:23 misc I could install 20 Centos 6 builders on haplometrosis.rht.gluster.org
13:23 misc if all goes well, which will not obviously
13:38 jiffin1 joined #gluster-dev
13:42 msvbhat joined #gluster-dev
13:48 dlambrig joined #gluster-dev
13:50 shaunm joined #gluster-dev
13:54 mchangir dlambrig, patch #14780 is reviewed and available for merge
13:54 dlambrig mchangir: Ok- I will merge it
13:58 nbalacha joined #gluster-dev
14:22 msvbhat joined #gluster-dev
14:48 hagarth joined #gluster-dev
14:56 spalai joined #gluster-dev
14:56 sanoj joined #gluster-dev
14:57 spalai left #gluster-dev
14:58 spalai joined #gluster-dev
14:59 ira joined #gluster-dev
15:11 spalai left #gluster-dev
15:23 spalai joined #gluster-dev
15:23 spalai left #gluster-dev
15:29 ira joined #gluster-dev
15:50 mchangir joined #gluster-dev
15:51 nbalacha joined #gluster-dev
15:54 pranithk1 joined #gluster-dev
15:55 mchangir joined #gluster-dev
16:17 Danishman joined #gluster-dev
16:24 rafi joined #gluster-dev
16:34 mchangir joined #gluster-dev
16:41 wushudoin joined #gluster-dev
16:42 shubhendu joined #gluster-dev
16:45 Acinonyx joined #gluster-dev
16:55 Acinonyx joined #gluster-dev
17:21 kotreshhr joined #gluster-dev
17:21 kotreshhr left #gluster-dev
17:22 shubhendu joined #gluster-dev
17:23 mchangir nigelb, nbslave7h seems to be failing at self-heald.t
17:32 amye mchangir: bug report?
17:32 mchangir amye, eh?
17:33 mchangir ah! :)
17:34 amye I know it's past EOD for him :)
17:35 mchangir EOW too
17:36 misc mchangir: I placed the node offline in the mean time
17:38 mchangir misc, thanks ... my other problem is my "recheck netbsd" hasn't re-triggered a job run for NetBSD after an earlier failed job #18420
17:40 misc mchangir: ok, taht would requires a more torough investigation of jenkins, and no one I am gonna start on a friday evening
17:40 misc so i guess that deserve a 2nd bug report
17:40 mchangir haha
17:41 mchangir looks like the CI Farm goes for a weekend rejoicing too
18:01 flakinson joined #gluster-dev
18:11 spalai joined #gluster-dev
18:47 penguinRaider joined #gluster-dev
18:52 jiffin joined #gluster-dev
19:31 msvbhat joined #gluster-dev
19:36 msvbhat_ joined #gluster-dev
22:40 misc nigelb: so, after one battle against the code, I finally have install from scratch virt install, with a ansible role for centos 6
22:48 amye misc++
22:48 glusterbot amye: misc's karma is now 32
23:51 nigelb misc: neat!

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