Re: replicate/distribute oddities in 2.0.0 Was Re: rc8

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 


Even with manually fixing (adding or removing) the extended attributes i was never able to get Gluster to see the missing files.  So i ended up writing a quick program that searched the raw bricks filesystem and then checked to make sure the file existed in the Gluster cluster and if it didn't it would tag the file.  Once that job was done i shut down Gluster, moved all the missing files off the raw bricks into temp storage, and then i restarted Gluster and copied all the files back into each directory.  That fixed the missing file problems.

Id still like to find out why Gluster would ignore certain files without the correct attributes.  Even removing all the file attributes wouldn't fix the problem.  I also tried manually coping a file into a brick which it still wouldn't find.  It would be nice to be able to manual copy files into a brick, then set an extended attribute flag which would cause gluster to see the new file(s) and copy them to all bricks after a ls -alR was done.  Or even better just do it automatically when new files without attributes are found in a brick.

thanks,
liam


On Wed, May 6, 2009 at 4:13 PM, Liam Slusser <lslusser@xxxxxxxxx> wrote:

To answer some of my own question, looks like those files were copied using gluster 1.3.12 which is why they have the different extended attributes:

gluster 1.3.12

Attribute "glusterfs.createtime" has a 10 byte value for file
Attribute "glusterfs.version" has a 1 byte value for file
Attribute "glusterfs.dht" has a 16 byte value for file

while gluster 2.0.0 has

Attribute "afr.brick2b" has a 12 byte value for file
Attribute "afr.brick1b" has a 12 byte value for file

I've been unsuccessful on fixing the attributes, can anybody point me in the right direction?

thanks,
liam

On Wed, May 6, 2009 at 12:48 PM, Liam Slusser <lslusser@xxxxxxxxx> wrote:
 Big thanks to the devel group for fixing all the memory leak issues with the earlier RC releases.  2.0.0 has been great so far without any memory issues what-so-ever.

I am seeing some oddities with the replication/distribute translators however.  I have three partitions on each gluster server exporting three bricks - We have two servers.  The gluster clients replicates each brick between the two servers and then i have a distribute translator for all the replicated bricks - basically gluster raid10.

There are a handful of files which have been copied into the gluster volume but since have disappeared, however the physical files exist on both bricks.

(from a client)

[root@client1 049891002526]# pwd
/intstore/data/tracks/tmg/2008_02_05/049891002526
[root@client1 049891002526]# ls -al 049891002526_01_09.wma.sigKey01.k
ls: 049891002526_01_09.wma.sigKey01.k: No such file or directory
[root@client1 049891002526]# head 049891002526_01_09.wma.sigKey01.k
head: cannot open `049891002526_01_09.wma.sigKey01.k' for reading: No such file or directory
[root@client1 049891002526]#


(from a server brick)

[root@server1 049891002526]# pwd
/intstore/intstore01c/gcdata/data/tracks/tmg/2008_02_05/049891002526
[root@server1 049891002526]# ls -al 049891002526_01_09.wma.sigKey01.k
-rw-rw-rw- 1 10015 root 19377712 Feb  6  2008 049891002526_01_09.wma.sigKey01.k
[root@server1 049891002526]# attr -l 049891002526_01_09.wma.sigKey01.k
Attribute "glusterfs.createtime" has a 10 byte value for 049891002526_01_09.wma.sigKey01.k
Attribute "glusterfs.version" has a 1 byte value for 049891002526_01_09.wma.sigKey01.k
Attribute "selinux" has a 24 byte value for 049891002526_01_09.wma.sigKey01.k
[root@server1 049891002526]# attr -l .
Attribute "glusterfs.createtime" has a 10 byte value for .
Attribute "glusterfs.version" has a 1 byte value for .
Attribute "glusterfs.dht" has a 16 byte value for .
Attribute "selinux" has a 24 byte value for .


Nothing in both the client and server logs.  I've tried all the normal replication checks and self-heal such as ls -alR.  If i copy the file back from one of the bricks into the volume it will show up again however it has a 1/3 chance of getting written to the files original location.  So then i end up with two identical files on two different bricks.

This volume has over 40 million files and directories so it can be very tedious to find anomalies. I wrote a quick perl script to search 1/25 of our total files in the volume for missing files and md5 checksum differences and as of now its about 15% (138,500 files) complete and has found ~7000 missing files and 0 md5 checksum differences.

How could i debug this?  I'd image it has something to do with the extended attributes on either the file or parent directory...but as far as i can tell that all looks fine.

thanks,
liam

client glusterfs.vol:

volume brick1a
  type protocol/client
  option transport-type tcp
  option remote-host server1
  option remote-subvolume brick1a
end-volume

volume brick1b
  type protocol/client
  option transport-type tcp
  option remote-host server1
  option remote-subvolume brick1b
end-volume

volume brick1c
  type protocol/client
  option transport-type tcp
  option remote-host server1
  option remote-subvolume brick1c
end-volume

volume brick2a
  type protocol/client
  option transport-type tcp
  option remote-host server2
  option remote-subvolume brick2a
end-volume

volume brick2b
  type protocol/client
  option transport-type tcp
  option remote-host server2
  option remote-subvolume brick2b
end-volume

volume brick2c
  type protocol/client
  option transport-type tcp
  option remote-host server2
  option remote-subvolume brick2c
end-volume

volume bricks1
  type cluster/replicate
  subvolumes brick1a brick2a
end-volume

volume bricks2
  type cluster/replicate
  subvolumes brick1b brick2b
end-volume

volume bricks3
  type cluster/replicate
  subvolumes brick1c brick2c
end-volume

volume distribute
  type cluster/distribute
  subvolumes bricks1 bricks2 bricks3
end-volume

volume writebehind
  type performance/write-behind
  option block-size 1MB
  option cache-size 64MB
  option flush-behind on
  subvolumes distribute
end-volume

volume cache
  type performance/io-cache
  option cache-size 2048MB
  subvolumes writebehind
end-volume

server glusterfsd.vol:

volume intstore01a
  type storage/posix
  option directory /intstore/intstore01a/gcdata
end-volume

volume intstore01b
  type storage/posix
  option directory /intstore/intstore01b/gcdata
end-volume

volume intstore01c
  type storage/posix
  option directory /intstore/intstore01c/gcdata
end-volume

volume locksa
  type features/posix-locks
  option mandatory-locks on
  subvolumes intstore01a
end-volume

volume locksb
  type features/posix-locks
  option mandatory-locks on
  subvolumes intstore01b
end-volume

volume locksc
  type features/posix-locks
  option mandatory-locks on
  subvolumes intstore01c
end-volume

volume brick1a
  type performance/io-threads
  option thread-count 32
  subvolumes locksa
end-volume

volume brick1b
  type performance/io-threads
  option thread-count 32
  subvolumes locksb
end-volume

volume brick1c
  type performance/io-threads
  option thread-count 32
  subvolumes locksc
end-volume

volume server
  type protocol/server
  option transport-type tcp
  option auth.addr.brick1a.allow 192.168.12.*
  option auth.addr.brick1b.allow 192.168.12.*
  option auth.addr.brick1c.allow 192.168.12.*
  subvolumes brick1a brick1b brick1c
end-volume


On Wed, Apr 22, 2009 at 5:43 PM, Liam Slusser <lslusser@xxxxxxxxx> wrote:

Avati,

Big thanks.  Looks like that did the trick.  I'll report back in the morning if anything has changed but its looking MUCH better.  Thanks again!

liam

On Wed, Apr 22, 2009 at 2:32 PM, Anand Avati <avati@xxxxxxxxxxx> wrote:
Liam,
 An fd leak and a lock structure leak has been fixed in the git
repository, which explains a leak in the first subvolume's server.
Please pull the latest patches and let us know if it does not fixe
your issues. Thanks!

Avati

On Tue, Apr 21, 2009 at 3:41 PM, Liam Slusser <lslusser@xxxxxxxxx> wrote:
> There is still a memory leak with rc8 on my setup.  The first server in a
> cluster or two servers starts out using 18M and just slowly increases.
>  After 30mins it has doubled in size to over 30M and just keeps growing -
> the more memory it uses the worst the performance.  Funny that the second
> server in my cluster using the same configuration file has no such memory
> problem.
> My glusterfsd.vol has no performance translators, just 3 storage/posix -> 3
> features/posix-locks -> protocol/server.
> thanks,
> liam
> On Mon, Apr 20, 2009 at 2:01 PM, Gordan Bobic <gordan@xxxxxxxxxx> wrote:
>>
>> Gordan Bobic wrote:
>>>
>>> First-access failing bug still seems to be present.
>>> But other than that, it seems to be distinctly better than rc4. :)
>>> Good work! :)
>>
>> And that massive memory leak is gone, too! The process hasn't grown by a
>> KB after a kernel compile! :D
>>
>> s/Good work/Awesome work/
>>
>> :)
>>
>>
>> Gordan
>>
>>
>> _______________________________________________
>> Gluster-devel mailing list
>> Gluster-devel@xxxxxxxxxx
>> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel@xxxxxxxxxx
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>
>




[Index of Archives]     [Gluster Users]     [Ceph Users]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux