Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2016-02-18

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

All times shown according to UTC.

Time Nick Message
01:17 EinstCrazy joined #gluster-dev
01:45 baojg joined #gluster-dev
02:18 EinstCrazy joined #gluster-dev
02:48 ilbot3 joined #gluster-dev
02:48 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/
02:57 josferna joined #gluster-dev
02:57 Humble joined #gluster-dev
03:08 primusinterpares joined #gluster-dev
03:14 skoduri joined #gluster-dev
03:21 overclk joined #gluster-dev
03:45 baojg joined #gluster-dev
03:49 ashiq_ joined #gluster-dev
03:49 EinstCra_ joined #gluster-dev
03:55 vimal joined #gluster-dev
03:56 gem joined #gluster-dev
04:01 EinstCrazy joined #gluster-dev
04:03 itisravi joined #gluster-dev
04:04 atinm joined #gluster-dev
04:09 nbalacha joined #gluster-dev
04:10 shubhendu joined #gluster-dev
04:14 mchangir joined #gluster-dev
04:25 baojg joined #gluster-dev
04:26 nishanth joined #gluster-dev
04:31 kanagaraj joined #gluster-dev
04:32 kshlm joined #gluster-dev
04:34 jiffin joined #gluster-dev
04:36 sakshi joined #gluster-dev
04:42 hgowtham joined #gluster-dev
04:44 hgowtham joined #gluster-dev
04:44 vimal joined #gluster-dev
04:44 ppai joined #gluster-dev
04:44 kdhananjay joined #gluster-dev
04:45 Saravanakmr joined #gluster-dev
04:48 aspandey joined #gluster-dev
04:55 Manikandan joined #gluster-dev
05:14 asengupt joined #gluster-dev
05:17 ggarg joined #gluster-dev
05:19 sakshi joined #gluster-dev
05:21 pppp joined #gluster-dev
05:23 Apeksha joined #gluster-dev
05:27 anoopcs hagarth, Can you please take a look @ http://review.gluster.org/#/c/13401/ ?
05:30 josferna joined #gluster-dev
05:30 rafi joined #gluster-dev
05:31 kotreshhr joined #gluster-dev
05:35 asengupt joined #gluster-dev
05:44 poornimag joined #gluster-dev
05:55 kdhananjay joined #gluster-dev
06:03 Saravanakmr joined #gluster-dev
06:05 Bhaskarakiran joined #gluster-dev
06:06 karthikfff joined #gluster-dev
06:07 rafi joined #gluster-dev
06:09 skoduri joined #gluster-dev
06:22 atalur joined #gluster-dev
06:33 skoduri joined #gluster-dev
06:37 Humble joined #gluster-dev
06:49 overclk joined #gluster-dev
06:51 baojg joined #gluster-dev
06:58 rastar kshlm++
06:58 glusterbot rastar: kshlm's karma is now 60
06:58 rastar kshlm: thank you for "skip test on netbsd for 3.6 branch" fix :)
07:15 anoopcs overclk++
07:15 glusterbot anoopcs: overclk's karma is now 18
07:16 baojg joined #gluster-dev
07:35 asengupt joined #gluster-dev
07:55 EinstCra_ joined #gluster-dev
08:04 Saravanakmr joined #gluster-dev
08:27 ndarshan joined #gluster-dev
08:34 Manikandan joined #gluster-dev
08:46 rraja joined #gluster-dev
08:53 asengupt joined #gluster-dev
09:02 ira joined #gluster-dev
09:08 skoduri joined #gluster-dev
09:09 Saravanakmr joined #gluster-dev
09:11 Apeksha joined #gluster-dev
09:19 poornimag joined #gluster-dev
09:19 EinstCrazy joined #gluster-dev
09:28 ppai joined #gluster-dev
09:50 sakshi joined #gluster-dev
09:54 skoduri_ joined #gluster-dev
10:03 kdhananjay joined #gluster-dev
10:05 nishanth joined #gluster-dev
10:05 ira joined #gluster-dev
10:14 nbalacha joined #gluster-dev
10:16 gem joined #gluster-dev
10:19 kotreshhr joined #gluster-dev
10:29 atalur joined #gluster-dev
10:51 poornimag joined #gluster-dev
11:11 kkeithley1 joined #gluster-dev
11:16 atalur joined #gluster-dev
11:37 foster joined #gluster-dev
11:45 kdhananjay1 joined #gluster-dev
11:49 itisravi joined #gluster-dev
11:57 asengupt joined #gluster-dev
11:59 EinstCrazy joined #gluster-dev
12:00 atalur joined #gluster-dev
12:00 ppai joined #gluster-dev
12:24 mchangir joined #gluster-dev
12:24 hgichon joined #gluster-dev
12:25 asengupt joined #gluster-dev
12:34 lpabon joined #gluster-dev
12:56 kanagaraj joined #gluster-dev
13:09 EinstCrazy joined #gluster-dev
13:14 nishanth joined #gluster-dev
13:15 shubhendu joined #gluster-dev
13:22 poornimag joined #gluster-dev
13:24 ira joined #gluster-dev
13:42 hchiramm joined #gluster-dev
13:46 rafi1 joined #gluster-dev
13:54 shubhendu joined #gluster-dev
13:56 nishanth joined #gluster-dev
14:13 nbalacha joined #gluster-dev
14:27 raghu joined #gluster-dev
14:28 raghu kshlm: there? for this patch http://review.gluster.org/#/c/13453/ what other things are needed for me to see submit button?
14:28 kshlm Checking.
14:28 kshlm It needs the netbsd-regression+1 flag.
14:28 kshlm Just comment 'recheck netbsd' on the review.
14:29 kshlm For 3.6 you should almost instantly get the +1
14:30 raghu where is 'recheck netbsd'?
14:30 kshlm You type it.
14:30 kshlm I'll do it now.
14:30 kshlm Check the review now.
14:31 kshlm You just have to add a comment like that to re-trigger a job, 'recheck (netbsd|centos|smoke)'
14:31 raghu ahh ok
14:31 kshlm And now you have a 'Submit' button available.
14:31 raghu thanks :)
14:31 raghu kshlm++
14:31 glusterbot raghu: kshlm's karma is now 61
14:32 kshlm You're welcome.
14:32 kshlm You should read our mailing-lists though.
14:32 raghu kshlm: well, there were lot of things. I got confused
14:33 kshlm Hehe, yeah.
14:33 kshlm We did do a lot of changes.
14:33 raghu and if netbsd is -1, then cant we override it with 'recheck netbsd' comment?
14:33 kshlm The comment retriggers the job.
14:34 kshlm But first make sure that the failure is genuine.
14:34 kshlm Don't spam the build servers.
14:34 raghu kshlm: ok. And by default for release-3.6, netbsd give +1 automatically whenever triggered?
14:34 kshlm Yes,
14:34 kshlm I did that yesterday, when we were talking.
14:34 raghu well, for release-3.6 I have to do it as netbsd regressions are not considered
14:35 kshlm You don't have to do anything for newer changes or rebasing existing changes on release-3.6
14:35 kshlm NetBSD will now run, and give an immediate +1.
14:36 kshlm rastar had the job for release-3.6 for sometime as it kept failing.
14:36 Bhaskarakiran joined #gluster-dev
14:37 raghu kshlm: Well, if I rebase the ones which have got acks and passed other regressions, I will have to wait for them to finsih those tests again. Its better to 'recheck netbsd' for those changes
14:37 kshlm The regression flags should get carried over, but yeah the comment retriggering is easier.
14:38 raghu yes. Atleast for the patches that are in the review queue I will do that. Otherwise its difficult. I will be away from Feb 19 to Feb 22.
14:39 kshlm @all if no one has any objections, I'll be merging the open pull-requests on glusterfs-patch-acceptance-tests
14:39 raghu and +1 verified is also given by myself only? Or is it automatic?
14:40 kshlm It's not necessary anymore.
14:41 ggarg joined #gluster-dev
14:41 raghu ok. So its automatic then
14:42 ndevos kshlm: release-3.5 also should not run the netbsd regression, did you set it up to +1 for that too?
14:43 kshlm ndevos, Nope. I'll do that right away.
14:43 kshlm ndevos, also who's merging PRs on glusterfs-patch-acceptance-tests?
14:43 ndevos kshlm++ thanks! I dont have any patches for 3.5 at the moment, will see the result when someone posts a backport
14:43 glusterbot ndevos: kshlm's karma is now 62
14:44 ndevos kshlm: no idea who maintains that, I just merge whenever I think things are acceptible, or are running live :)
14:47 raghu kshlm: also, are able to send patches to gerrit? I mean rfc.sh working for you?
14:50 kdhananjay joined #gluster-dev
14:53 kshlm raghu, What's not working?
15:02 kdhananjay joined #gluster-dev
15:03 raghu for some reason, rfc.sh is failing for me saying ssh connection timed out
15:06 kshlm Does 'ssh <gerrit-username>@review.gluster.org' also fail?
15:06 kshlm Can you check your git remote?
15:06 kshlm It's possible you're using review.gluster.com/
15:06 kshlm Since the migration, only review.gluster.org is accessible.
15:07 kshlm And I think it should stay that way.
15:07 kshlm It's been over 4 years, and people shouldn't still be using any *.gluster.com addresses.
15:11 raghu ok
15:11 raghu this is my remote url
15:11 raghu ssh://raghavendrabhat@git.gluster.com/glusterfs.git
15:11 kshlm Change the com to org.
15:12 raghu so it should be git.gluster.org? or gerrit.gluster.org?
15:12 kshlm review.gluster.org
15:12 kshlm git.gluster.org should work as well, but better use the proper one.
15:12 kkeithley_ everything should be gluster.org
15:15 raghu kshlm++
15:15 glusterbot raghu: kshlm's karma is now 63
15:16 kshlm I'm signing off now. I might be back online once I reach home.
15:18 ashiq_ joined #gluster-dev
15:39 Bhaskarakiran joined #gluster-dev
15:41 Bhaskarakiran joined #gluster-dev
15:46 rafi joined #gluster-dev
16:10 josferna joined #gluster-dev
16:12 jiffin joined #gluster-dev
16:16 nishanth joined #gluster-dev
16:20 jiffin joined #gluster-dev
16:33 skoduri joined #gluster-dev
16:40 anoopcs hagarth, Can you please take a look at the following back port https://review.gluster.org/#/c/13401/ ?
16:40 hagarth anoopcs: will do
16:40 anoopcs hagarth, thanks.
17:01 atinm joined #gluster-dev
17:06 jiffin joined #gluster-dev
17:14 kanagaraj joined #gluster-dev
17:20 gem joined #gluster-dev
17:43 Manikandan joined #gluster-dev
18:04 rafi joined #gluster-dev
18:09 luizcpg joined #gluster-dev
18:13 kanagaraj joined #gluster-dev
19:21 alex___ joined #gluster-dev
19:23 alex___ Hi there, I need some help here :D .... We are using Gluster in our web server architecture, with a brick for assets. Each server write to the assets brick, but read from the local (to speed up things). For some reasons, we need to create a "known folder-name" symlink to one of the folders in assets, but to keep the things fast we want to target the local directory (not the brick) so the server keep reading from the local. Does
19:25 alex___ when I say "brick" I mean "cluster"
19:29 kkeithley_ alex___: you must not ever write to the back-end (local) brick. Always write to the mounted (fuse or NFS) volume.
19:31 alex___ yes kkeithley_ we write to the mounted volume, but we read from the local to speed-up things, so for example in each server we have a mounted "/cluster/assets" and a local "/shared/assets", we write to /cluster/assets be nginx read from "shared/assets",
19:32 alex___ all works fine...
19:32 kkeithley_ And you need to write shorter lines.  Your long lines are getting cut off
19:33 alex___ the question is if we could create a symlink in "/cluster/assets/testsymlink" pointing to "/shared/assets/testsymlink" without any problem
19:33 alex___ sorry I didnt know
19:36 alex___ shorter version: nginx write to mounted "/cluster/assets" and read from local "/shared/assets/" brick
19:36 kkeithley_ Sure, you can make a any symlink in /cluster/.... (/cluster being the fuse or nfs mounted gluster volume).  Nothing says a symlink has to "resolve" to anything in the volume. If your web server knows what to do with it.
19:37 alex___ could I create a symlink in the mounted cluster targeting a local folder (used in the brick)
19:37 post-factum alex___: symlink target is not restricted, so you can
19:38 alex___ even if the target is a folder inside the local brick ? sure?
19:38 alex___ :D
19:40 post-factum alex___: noone says it will be resolved OK on client side, but you may create it
19:40 post-factum let me rephrase you. you want server-side brick to export server-side local folder to client via network?
19:41 post-factum then no, one cannot do that
19:42 alex___ sorry cause my english is not very good :( Im Spaniard, let me try again...
19:43 alex___ on each server we have a mounted "cluster" volume and a local "copy" (I think the local brick?)
19:43 alex___ so each server write to the mounted cluster and read from their local filesystem
19:44 post-factum dont worry amigo
19:44 alex___ now, for some reason, we need to create a symlink from one folder to another
19:44 post-factum lets start from basics. you have, saying, server1 and server2, that are both glusterfs servers and web servers?
19:45 alex___ yeah
19:45 post-factum so you use glusterfs for keeping assets in sync just between those 2 nodes?
19:45 alex___ server1 and server2, as web front-ends, both with glusterfs, with a mounted "cluster" and a local "copy?brick?"
19:45 alex___ that is
19:45 alex___ they keep assets in sync
19:46 post-factum then yes, you can symlink from gluster mount to local folder because it is all within one node
19:46 post-factum and read from that symlink
19:47 post-factum but in general i guess you are doing it wrong. why not just read from brick directly?
19:47 alex___ because is so much slower
19:47 alex___ (at least for us)
19:48 alex___ in our test, is much faster configure nginx to read from the local copy/brick
19:48 post-factum that is what i am talking about
19:48 post-factum why using symlink to local folder if it is local and you may reach it without symlink?
19:49 alex___ ahhh thats the problem in the "app development part" :D
19:50 post-factum i see. you need a good shotgun instead
19:50 alex___ in assets we have hash generated folders lik "23423ef", "eb234", and so, and for a developing problem, we need a "constant" folder name for some of them
19:51 post-factum alex___: you may keep brick paths consistent across the nodes
19:51 alex___ so there are two solutions, not hashing that folders (is the worst solution), or create a well-known name symlink to that hashed folder and keep it "in sync" from PHP
19:52 post-factum alex___: smth is definitely wrong in your architecture
19:53 alex___ they are consistent, for sure...
19:55 alex___ so finally... we can do this "/cluster/assets/symlinkfolder" -> "/localbrick/assets/symlinkfolder", right=
19:55 alex___ ?
19:56 alex___ anyway
19:56 alex___ thanks post-factum
19:56 alex___ for your time!
19:57 alex___ ;)
19:57 alex___ regards
19:57 post-factum alex___: np
20:11 lalatend1M joined #gluster-dev
21:10 BuffaloCN joined #gluster-dev
21:18 BuffaloCN joined #gluster-dev
21:26 kkeithley1 joined #gluster-dev
21:32 kkeithley1 joined #gluster-dev
23:19 primusinterpares joined #gluster-dev

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