Camelia, the Perl 6 bug

IRC log for #gluster, 2013-02-23

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

All times shown according to UTC.

Time Nick Message
00:17 xian1 left #gluster
00:20 p3rror joined #gluster
00:29 p3rror please I run "sudo glusterfs-volgen -n storage --raid=1 192.168.56.101:/home/storage/ ..."
00:30 p3rror and I've got raid type (1) and number of volumes (1) invalid
00:33 p3rror also I've run mount command
00:33 p3rror sudo mount -t glusterfs /etc/glusterfs/storage-tcp.vol /data/storage
00:34 p3rror when i try to access /data/storage I get ls: cannot access storage: No such file or directory
00:40 sheneman joined #gluster
00:43 sheneman Hello - this probably seems obvious, but how do I add a new node to an existing set of active peers?  I'd like to expand a 2-node cluster to a 3-node cluster.  With "gluster peer probe xx.xx.xx.xx", I get:  $ xx.xx.xx.xx is already part of another cluster.  How do you add new nodes/peers?
00:44 Cenbe joined #gluster
01:06 JoeJulian p3rror: Please use the ,,(latest) version.
01:06 glusterbot p3rror: The latest version is available at http://goo.gl/zO0Fa . There is a .repo file for yum or see @ppa for ubuntu.
01:06 JoeJulian p3rror: Also reference ,,(rtfm)
01:06 glusterbot p3rror: Read the fairly-adequate manual at http://goo.gl/E3Jis
01:07 JoeJulian ~cloned server | sheneman
01:07 glusterbot sheneman: I do not know about 'cloned server', but I do know about these similar topics: 'cloned servers'
01:07 JoeJulian ~cloned servers | sheneman
01:07 glusterbot sheneman: Check that your peers have different UUIDs ('gluster peer status' on both). The uuid is saved in /var/lib/glusterfs/glusterd.info - that file should not exist before starting glusterd the first time. It's a common issue when servers are cloned. You can delete the /var/lib/glusterfs/peers/<uuid> file and /var/lib/glusterfs/glusterd.info, restart glusterd and peer-probe
01:07 glusterbot again.
01:09 sheneman Ah, I figured it out.  You have to probe peer from within the existing cluster to the new node.  Sorry, I didn't see that in the documentation anywhere....
01:10 JoeJulian sheneman: It's for security purposes. You don't want some random machine adding itself to the trusted pool.
01:10 sheneman Makes perfect sense.  Thanks!
01:10 JoeJulian You're welcome.
01:12 raven-np joined #gluster
02:06 Humble joined #gluster
02:23 Humble joined #gluster
02:39 tomsve joined #gluster
02:42 Humble joined #gluster
03:07 pipopopo joined #gluster
03:14 m0zes_ joined #gluster
03:17 Humble joined #gluster
03:36 Humble joined #gluster
03:52 Humble joined #gluster
04:14 Humble joined #gluster
04:35 dbruhn joined #gluster
04:36 hagarth joined #gluster
04:40 Humble joined #gluster
04:58 Humble joined #gluster
05:19 Humble joined #gluster
05:56 mohankumar joined #gluster
05:56 Humble joined #gluster
06:34 Humble joined #gluster
06:37 glusterbot New news from newglusterbugs: [Bug 914874] Enhancement suggestions for BitRot hash computation <http://goo.gl/HOsJm>
06:52 _pol joined #gluster
07:33 edong23 joined #gluster
07:38 _pol joined #gluster
07:53 flrichar joined #gluster
08:20 jiffe1 joined #gluster
08:22 _pol joined #gluster
08:54 WildPikachu joined #gluster
09:11 raven-np joined #gluster
09:12 _pol joined #gluster
09:18 Humble joined #gluster
09:53 timothy joined #gluster
10:04 Humble joined #gluster
10:48 theron joined #gluster
11:06 ramkrsna joined #gluster
11:06 ramkrsna joined #gluster
11:13 Humble joined #gluster
11:18 disarone joined #gluster
11:37 raven-np joined #gluster
11:49 raven-np1 joined #gluster
11:58 Humble joined #gluster
12:32 bala joined #gluster
12:46 Humble joined #gluster
13:04 mohankumar joined #gluster
13:05 Humble joined #gluster
13:18 p3rror joined #gluster
13:27 tryggvil joined #gluster
13:52 tomsve joined #gluster
13:56 theron joined #gluster
14:05 lh joined #gluster
14:05 lh joined #gluster
14:33 r2 joined #gluster
15:02 hagarth joined #gluster
15:15 lh joined #gluster
15:15 lh joined #gluster
15:17 r2 hey all, i'm seeing very high cpu usage on my 2 replicated gluster nodes, and some strange errors.  i'd appreciate any help you could offer.
15:17 r2 nodes are 3.3
15:17 r2 i've started seeing this message recently: [2013-02-23 15:14:28.829504] I [server3_1-fops.c:252:server_inodelk_cbk] 0-datavol-server: 1493: INODELK (null) (--) ==> -1 (No such file or directory)
15:17 r2 in addition to these: [2013-02-23 15:14:54.826678] E [afr-self-heal-data.c:1311:afr_sh_data_open_cbk] 0-datavol-replicate-0: open of <gfid:8c8d2885-5762-4f95-9f16-49518974e740> failed on child datavol-client-1 (No such file or directory)
15:38 Humble joined #gluster
15:42 bala1 joined #gluster
15:43 raven-np joined #gluster
15:53 sara joined #gluster
15:57 sara hello all, i have problem with gluster, on client machine 100% cpu, gluster 3.3.1, serv php pages, problem after update, i try change version, reconfigured, et. problem on distrib and replica configuration
15:57 sara everyone know kow fix this?
16:06 elyograg sara: the way php apps generally work is almost worst-case for gluster. there are things you can do, though.  http://joejulian.name/blog/optimizi​ng-web-performance-with-glusterfs/
16:06 glusterbot <http://goo.gl/uDFgg> (at joejulian.name)
16:08 sara elyograg: i optimalzie this
16:08 sara and read this page, still CPU 100%
16:08 sara i use cache, load balancer, eaccelelator
16:08 sara etc, etc
16:08 sara problem is after update
16:11 elyograg i don't know what to do.  it's saturday morning in the US, 8 AM for JoeJulian, who would be one of the primary people to help you.
16:13 sara thx that u try, i too dont know where i can find help
16:13 sara :(
16:13 JoeJulian What process is using all the cpu?
16:15 sara glsuterfs
16:15 sara 12432 root      20   0 1100m 821m 1712 S 100.4  5.2   1327:28 glusterfs
16:15 sara on client machine by openvz
16:15 sara mounted by standard glusterfs client by fuse
16:15 sara problem only client machine
16:16 JoeJulian Have you looked in the client log for clues?
16:17 sara yes, logs lokk ok
16:24 JoeJulian fpaste.org the last ~100 lines of your client log
16:24 sara ok
16:28 sara http://fpaste.org/F7Go/
16:28 glusterbot Title: Viewing Paste #279822 by sara (at fpaste.org)
16:30 sara this is on client machine
16:30 JoeJulian right
16:32 sjoeboo_ joined #gluster
16:32 JoeJulian So does eaccellerator reduce the number of stat calls like apc does? It doesn't appear to from my cursory examination.
16:33 sara i try move eaccelelator cach to normal disc
16:33 sara cpu stil 100%
16:33 sara eaccelelator have shm config
16:34 Humble joined #gluster
16:34 sara i think this is ni problem with eaccelelator
16:35 JoeJulian Wasn't really my question, though. The question was about whether it checks every time to see if the php script has changed.
16:36 sara http://fpaste.org/tAWw/
16:36 glusterbot Title: Viewing Paste #279823 by sara (at fpaste.org)
16:36 sara i dont see eaccelelator, its on cache ram
16:37 sara moust images as cache by proxy, but before update work ok
16:37 sara beforue update stystem on masternode and version gluster
16:37 sara i try older version glsiter still CPU 100%
16:39 JoeJulian Ok, so it wasn't broken before upgrading. Did break after upgrading. Won't go back to pre-upgrade performance after downgrade. Does that sound right?
16:40 sara i try, back to 3.3.0
16:40 sara still CPU 100%
16:40 sara i on the some time update system Centos, but o cant reupdate
16:40 sara i test 3.4.0 stil high load
16:41 sara now i have 3.3.1 stable
16:41 JoeJulian Are your bricks ext4?
16:41 sara yes
16:41 JoeJulian @ext4
16:41 glusterbot JoeJulian: Read about the ext4 problem at http://goo.gl/PEBQU
16:41 JoeJulian I'm surprised it works at all.
16:41 sara one moment i see
16:41 sara :)
16:42 sara hmm
16:42 sara i have openvz
16:42 JoeJulian In fact, I had dismissed that as the problem because it still worked! :O
16:42 sara can use this on openvz?
16:43 JoeJulian I use kvm or xen for virtualization. I don't know what openvz does or doesn't do.
16:43 sara ok, thx u for information, i try fix this problem if really et4 problem, very very thx for u ;)
16:46 sara by ;)
16:58 Humble joined #gluster
17:00 elyograg JoeJulian: I was hoping you'd be asleep.  of course i wasn't either and hand't been for a few hours ... ;)
17:00 JoeJulian I was hoping I was too. ;)
17:25 Humble joined #gluster
17:38 jclift_ joined #gluster
17:49 Humble joined #gluster
18:59 Humble joined #gluster
19:23 jclift_ joined #gluster
19:36 _pol joined #gluster
20:23 _pol joined #gluster
20:27 tomsve joined #gluster
21:45 jclift_ joined #gluster
22:03 lh joined #gluster
22:41 GLHMarmot joined #gluster
23:29 disarone joined #gluster
23:33 mezgani_ joined #gluster
23:45 mezgani__ joined #gluster
23:48 mezgani_ joined #gluster
23:53 dbruhn__ joined #gluster

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