Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster, 2013-11-29

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

All times shown according to UTC.

Time Nick Message
00:14 glusterbot New news from newglusterbugs: [Bug 1032438] frame_fill_groups intermittently fails to populate frame->root->groups correctly <http://goo.gl/HMvI0V>
00:33 RicardoSSP joined #gluster
01:19 harish joined #gluster
01:25 khushildep joined #gluster
02:12 dylan_ joined #gluster
02:12 dusmantkp_ joined #gluster
02:16 harish joined #gluster
02:20 skered- joined #gluster
02:44 dylan_ joined #gluster
02:56 saurabh joined #gluster
02:58 vshankar joined #gluster
03:12 davidbierce joined #gluster
03:33 RameshN joined #gluster
03:33 davinder2 joined #gluster
03:34 vpshastry joined #gluster
03:34 sgowda joined #gluster
03:38 shubhendu joined #gluster
03:38 kanagaraj joined #gluster
03:42 jag3773 joined #gluster
03:43 skered- joined #gluster
03:52 kshlm joined #gluster
03:57 itisravi joined #gluster
04:08 shyam joined #gluster
04:25 mohankumar joined #gluster
04:27 hagarth joined #gluster
04:31 satheesh joined #gluster
04:32 vpshastry left #gluster
04:54 meghanam joined #gluster
04:54 meghanam_ joined #gluster
04:59 hagarth joined #gluster
05:00 bala joined #gluster
05:02 satheesh joined #gluster
05:02 nshaikh joined #gluster
05:02 MiteshShah joined #gluster
05:06 mohankumar joined #gluster
05:07 shylesh joined #gluster
05:10 dusmantkp_ joined #gluster
05:19 raghu joined #gluster
05:20 shruti joined #gluster
05:34 CheRi joined #gluster
05:41 vpshastry joined #gluster
05:46 spandit joined #gluster
05:51 bala joined #gluster
05:52 shubhendu joined #gluster
05:54 dusmantkp_ joined #gluster
05:59 shyam joined #gluster
06:00 hagarth joined #gluster
06:18 hagarth joined #gluster
06:28 davinder2 joined #gluster
06:35 aravindavk joined #gluster
06:45 dylan_ joined #gluster
06:45 bulde joined #gluster
06:49 ndarshan joined #gluster
06:50 shyam joined #gluster
06:51 anands joined #gluster
06:51 sgowda joined #gluster
06:52 shubhendu joined #gluster
06:56 ababu joined #gluster
06:58 vimal joined #gluster
07:04 dusmantkp_ joined #gluster
07:05 morse joined #gluster
07:16 MarkR_ joined #gluster
07:19 hagarth joined #gluster
07:22 krypto joined #gluster
07:31 spandit joined #gluster
07:38 micu joined #gluster
07:39 schrodinger joined #gluster
07:40 johnmwilliams joined #gluster
07:41 GabrieleV joined #gluster
07:41 VerboEse joined #gluster
07:42 krypto can some one tell me the difference between "peer in cluster" and "accepted peer request",both are showing as connected
07:46 glusterbot New news from newglusterbugs: [Bug 1030098] Development files not packaged for debian/ubuntu <http://goo.gl/uCV2Eu> || [Bug 1034489] SELinux is preventing /usr/sbin/glusterfsd from 'create' accesses on the fifo_file fifo. <http://goo.gl/r8vdIx> || [Bug 1031328] Gluster man pages are out of date. <http://goo.gl/4e1PjH>
07:50 MarkR__ joined #gluster
07:51 MarkR__ Remco: I filed a bug report, https://bugzilla.redhat.co​m/show_bug.cgi?id=1036009
07:51 glusterbot <http://goo.gl/OKrt2o> (at bugzilla.redhat.com)
07:51 glusterbot Bug 1036009: high, unspecified, ---, amarts, NEW , Client connection to Gluster daemon stalls
07:52 psharma joined #gluster
07:56 sgowda joined #gluster
07:58 ctria joined #gluster
07:58 Robin_ joined #gluster
07:59 Robin_ hello i have some problems with gluster loosing file permissions, can someone help on this topic
08:02 hybrid5121 joined #gluster
08:04 eseyman joined #gluster
08:07 dylan_ joined #gluster
08:07 ngoswami joined #gluster
08:08 jtux joined #gluster
08:14 tjikkun joined #gluster
08:15 Robin_ hello i have some problems with gluster loosing file permissions, can someone help on this topic
08:16 glusterbot New news from newglusterbugs: [Bug 1036009] Client connection to Gluster daemon stalls <http://goo.gl/OKrt2o>
08:19 vpshastry joined #gluster
08:24 Robin_ hello gluster is losing all my file permissions and i see only ????
08:24 Robin_ can sombody help?
08:24 samppah what glusterfs version you are using and what kind of setup you have?
08:25 Robin_ the latest 3.4.1
08:26 Robin_ and we have 4 nodes where the bricks and 12 clients
08:26 Robin_ the bricks nodes are also clients
08:26 Robin_ we have around 70 TB on gluster fs
08:27 Robin_ on the servers runs ubuntu 12.0.4
08:31 abyss^ I'd like to move first gluster server to another localization (whole VM, so id disc etc will not change) and turn off the second gluster server, then prepare new gluster server and I'd like to sychronize the first server with that third server. Can I find any tutorial how to do it in polite way?:)
08:32 ndevos Robin_: when file permissions are shown as ???? you probably have a ,,(split-brain)
08:32 glusterbot Robin_: To heal split-brain in 3.3+, see http://goo.gl/FPFUX .
08:32 _polto_ joined #gluster
08:32 _polto_ joined #gluster
08:33 muhh joined #gluster
08:35 Robin_ ok thanks, i will tried it. I allready did it in the past and it didn't help. We had to rebuild the volume from scratch. But now i happens again, so iw ill retried it
08:35 abyss^ On gluster 3.4 you can easily check if you have split-brain: gluster volume heal split-brain or smth like this:)
08:38 vpshastry1 joined #gluster
08:39 abyss^ Robin_: let's me know when you check it. I had the same problem but it wasn't split-brain. I just remount clients... I had some network issue.
08:41 andreask joined #gluster
08:41 Robin_ hmmm, when i run gluter volume heal gluster info split-brain i have a whole list with always the same gfid
08:42 Robin_ i launched the heal command
08:43 Robin_ how do i know of the split-brain is healed?
08:48 spandit joined #gluster
08:49 anands joined #gluster
08:53 shubhendu joined #gluster
09:02 muhh_ joined #gluster
09:03 Robin_ i have one file that has a split-brain, but is nearly the whole filesystem that is loosing the file permissions, can this?
09:04 CheRi joined #gluster
09:04 hchiramm__ joined #gluster
09:07 samppah Robin_: is gluster vol heal volName info' showing other files?
09:07 Robin_ yes
09:09 achuz joined #gluster
09:12 samppah Robin_: is the file list in heal info constant or does it keep changing? ie. is files added and removed from that list?
09:15 Robin_ it keep changing but seems like it is always the same file comming back and away
09:17 StarBeast joined #gluster
09:20 eclectic_ joined #gluster
09:20 davinder4 joined #gluster
09:22 Cenbe_ joined #gluster
09:25 basic- joined #gluster
09:26 Kins_ joined #gluster
09:27 basic` joined #gluster
09:27 Robin_ abyss^: can you tell about the network issue
09:27 foster joined #gluster
09:32 bver joined #gluster
09:33 hchiramm__ joined #gluster
09:33 bala1 joined #gluster
09:34 verdurin joined #gluster
09:35 bver \leave
09:35 vpshastry1 joined #gluster
09:44 geewiz joined #gluster
09:48 _polto_ joined #gluster
09:48 _polto_ joined #gluster
09:51 hchiramm__ joined #gluster
09:52 shubhendu joined #gluster
09:53 anands joined #gluster
09:54 gdubreui joined #gluster
09:55 spandit joined #gluster
10:04 Robin_ abyss^: can you tell about the network issue
10:09 bver joined #gluster
10:19 abyss^ Robin_: priv :)
10:29 bver left #gluster
10:30 bala joined #gluster
10:35 geewiz_ joined #gluster
10:38 rastar joined #gluster
10:41 edward2 joined #gluster
11:09 _polto_ joined #gluster
11:09 calum_ joined #gluster
11:11 Alpinist joined #gluster
11:20 Divad joined #gluster
11:23 helmo left #gluster
11:45 anands joined #gluster
11:52 Divod Hello, I am a colleague of Robin_ and we are severely held back by the permissions issue. We have one volume, in which all files belong to the same user and the same group, and all have extended access ACLs set with setfacl. The volume consists of 12 bricks of ~25T, on 6 servers, distributed/replicated, resulting in ~150TB. Before, it was half the size and got 100% full. We removed some files, freeing up a few Ts, and added 6 bricks
11:52 Divod After a successful expansion, files and dirs started to lose all permissions (to OOO or 1000). Problem looked very much like http://thr3ads.net/gluster-users/2013​/08/2670154-Files-losing-permissions Permissions could only fixed by chmod'ing broken files. After a week trying to find out and not beeing able to solve the problem, we removed the volume, removed .gluster dirs from the bricks, removed all extended attributes,
11:52 glusterbot <http://goo.gl/d25DB3> (at thr3ads.net)
11:53 Divod and ACLs using http://linuxsysadm.wordpress.com/2013​/05/16/glusterfs-remove-extended-attr​ibutes-to-completely-remove-bricks/ recursively.
11:53 glusterbot <http://goo.gl/eFTf5u> (at linuxsysadm.wordpress.com)
11:53 Divod After this, we were able to create the volume again, no data was lost. ACLs were put back, and for one week we didn't have a problem. Now, files start to lose their permissions again randomly (although at a much lower rate) and a sudo file access is enough to fix it on a client.
11:53 Divod Logs are quiet, except for an occasional node007: [2013-11-29 10:12:04.267225] W [afr-self-heald.c:946:_process_entries] 0-gluster-replicate-4: /path/to/file: No gfid present skipping
11:54 Divod We didn't dare to start the data migration rebalance yet, so most of the data is on half of the bricks, but there's still at least 3T free on every brick.
11:55 hchiramm__ joined #gluster
11:57 ndarshan joined #gluster
12:04 itisravi_ joined #gluster
12:13 saurabh joined #gluster
12:13 andreask joined #gluster
12:14 davinder4 joined #gluster
12:17 dylan_ joined #gluster
12:25 cyberbootje hi
12:25 glusterbot cyberbootje: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer an answer.
12:26 ctria joined #gluster
12:27 hchiramm__ joined #gluster
12:27 itisravi joined #gluster
12:28 mohankumar joined #gluster
12:43 cyberbootje i have some big trouble with the gluster storage, anyone here whoo can help a bit?
12:55 MiteshShah joined #gluster
12:59 45PAAQU7K joined #gluster
12:59 hchiramm__ joined #gluster
13:01 jtux joined #gluster
13:03 diegows joined #gluster
13:03 diegows_ joined #gluster
13:13 social cyberbootje: please be more specific
13:14 _polto_ joined #gluster
13:21 cyberbootje well
13:22 cyberbootje just now, the connection to the storage seem to go offline, transport endpoint not connected
13:23 cyberbootje this happend while i tried to copy some data on the storage itself, to a backup dit elsewhere on the partition
13:25 cyberbootje dir*
13:25 cyberbootje also, sometimes the client conencted to gluster(virtual machine host) suddenly reboots for no reason at all
13:30 ctria joined #gluster
13:43 _polto_ joined #gluster
13:45 vpshastry joined #gluster
13:47 LoudNoises joined #gluster
13:54 brimstone joined #gluster
13:58 social some gluster logs? version of gluster? exact reasons why you think this is gluster issue? did you check dmesg on all nodes for hw/sw failure/oom-kill?
14:11 social Divod: ping, what version of gluster?
14:12 _polto_ joined #gluster
14:12 _polto_ joined #gluster
14:17 harish joined #gluster
14:23 dbruhn joined #gluster
14:30 psyl0n joined #gluster
14:50 Divod social: glusterfs 3.4.1 built on Sep 27 2013 19:46:44
14:54 calum_ joined #gluster
15:12 geewiz joined #gluster
15:14 dylan_ joined #gluster
15:21 social Divod: I know about several issues with 3.4.1 related to acls and distribute but nothing you mention :/
15:25 Divod social: is is correct that when we removed .gluster directories from the bricks and removed all extended attrs and acls, and recreated the volume we have started frm a clean slate?
15:27 krypto any recommended volume type for storing virtual disk images
15:36 Alpinist joined #gluster
15:40 dbruhn krypton, what are your storage requirements?
15:40 psyl0n joined #gluster
15:42 krypto dbruhn fault tolerance with out losing performance
15:43 jag3773 joined #gluster
15:43 dbruhn What are you sub volume sizes? and what are your expected virtual disk file sizes
15:44 krypto vdisk size starts from 20Gb
15:45 dbruhn max size?
15:45 shyam joined #gluster
15:45 krypto 500 Gb
15:47 dbruhn what size are your bricks
15:49 krypto Havent configured disks as bricks, testing in virtual machines each vm with 10 Gb
15:49 krypto total vms
15:50 krypto total 4 vms.Planning to use 550 Gb in production
15:51 krypto dbruhn:is there any equation for finding best volume types,i can see 7 different types in documentation
15:52 dbruhn DHT is the fastest, replication obviously gives you protection, striped should only be used if you have file sizes that will grow beyond the size of the bricks
15:54 bala joined #gluster
15:55 krypto thanks dbruhn, but what is DHT,i am sorry couldn't find  anything in docs
15:55 hchiramm__ joined #gluster
15:55 dbruhn distributed
15:55 krypto yes and HT?
15:55 dbruhn Distributed Hash Table
15:55 krypto voookkk :) thanks
15:55 dbruhn You can mix the types
15:56 dbruhn I use DHT + replication
15:56 dbruhn You want to use what suites your data needs
15:57 krypto thanks dbruhn
15:57 dbruhn np
15:58 krypto one more question  what is the minimum iops u see in your environment
16:12 dbruhn It's completely bound by your system/subsystem
16:12 dbruhn my environment might not be the best example, it's all 15k sas and QDR infiniband
16:13 failshell joined #gluster
16:24 andreask joined #gluster
16:25 zerick joined #gluster
16:31 vpshastry joined #gluster
16:35 dneary joined #gluster
16:35 dneary joined #gluster
16:47 gmcwhistler joined #gluster
16:55 andreask joined #gluster
17:01 aliguori joined #gluster
17:04 hagarth joined #gluster
17:04 dkorzhevin joined #gluster
17:07 davinder4 joined #gluster
17:18 gmcwhistler joined #gluster
17:25 gmcwhistler joined #gluster
17:27 vpshastry left #gluster
17:30 ricky-ti1 joined #gluster
17:35 gmcwhistler joined #gluster
17:44 rotbeard joined #gluster
18:20 calum_ joined #gluster
18:34 andreask joined #gluster
18:37 RedShift joined #gluster
18:45 ricky-ticky joined #gluster
18:46 sroy_ joined #gluster
18:51 MarkR_ joined #gluster
19:06 dylan_ joined #gluster
19:18 andreask joined #gluster
19:50 hchiramm__ joined #gluster
20:12 gmcwhistler joined #gluster
20:18 _polto_ joined #gluster
20:18 _polto_ joined #gluster
20:26 hchiramm__ joined #gluster
20:32 gmcwhistler joined #gluster
21:15 bgpepi joined #gluster
21:53 _polto_ joined #gluster
21:53 _polto_ joined #gluster
22:03 badone joined #gluster
22:07 dylan_ joined #gluster
22:20 shani joined #gluster
22:20 shani hello
22:20 glusterbot shani: Despite the fact that friendly greetings are nice, please ask your question. Carefully identify your problem in such a way that when a volunteer has a few minutes, they can offer you a potential solution. These are volunteers, so be patient. Answers may come in a few minutes, or may take hours. If you're still in the channel, someone will eventually offer an answer.
22:21 shani i have a problem with gluste FS
22:21 shani i have frontend 3 servers and 1 file server
22:21 shani file server using Glusterfs server and 3 are clients
22:21 shani everyday when traffic is too high and I/O RATE is high
22:21 shani i get an error .
22:22 shani transport end is not connected
22:22 shani and i have to remount it again and again
22:22 shani what should i do please tell me
22:23 shani any gluster expert available ?
22:24 shani please i seek help
22:28 failshel_ joined #gluster
22:30 shani there?
22:30 shani anybody ?
22:45 JoeJulian shani, have you looked in the client logs?
23:00 shani yes
23:00 shani i do
23:14 RicardoSSP joined #gluster
23:15 Comnenus joined #gluster
23:44 shani joined #gluster

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