[Gluster-users] Gluster 2.0.1 Locking up entire machine under load

Daniel Jordan Bambach dan at lateral.net
Wed Jun 17 10:45:43 UTC 2009


I am running a test install of GlusterFS 2.0.1 in preparation of  
rolling it out to a couple of production servers.

It is set up as a simple two machine as both client and server (see  
below config), to mirror the filesystem between the machines.

I am finding that if I put it under heavy load, the machine under load  
will hang completely, and need to be hard-rebooted to bring it back up.

Im not seeing any errors from Gluster - is there anything I can do to  
collect more data/ crashlogs?

I am running this on Fedora 7, under VMWare, with the stock Fuse  
version 2.7.3-2.fc7
GlusterFS compiled from source.

One thing I am unsure of is whether my AFR config is ok - on one side  
it is a tcp client, and the other is the local 'brick' Is this ok, or  
do I need to specify another tcp client for the local server to afr  
with?

Many thanks for any help.
D.


volume posix
  type storage/posix
  option directory /home/export
end-volume

volume locks
   type features/locks
   subvolumes posix
end-volume

volume brick
  type performance/io-threads
  subvolumes locks
end-volume

volume server
  type protocol/server
  option transport-type tcp
  option auth.addr.brick.allow *
  subvolumes brick
end-volume

volume latsrv2
  type protocol/client
  option transport-type tcp
  option remote-host latsrv2
  option remote-subvolume brick
end-volume

volume afr
   type cluster/replicate
   subvolumes brick latsrv2
   option read-subvolume brick
end-volume

volume writebehind
   type performance/write-behind
   option cache-size 1MB
   subvolumes afr
end-volume

volume cache
   type performance/io-cache
   option cache-size 128MB
   option priority *.pyc:4,*.html:3,*.php:2,*:1
   subvolumes writebehind
end-volume





More information about the Gluster-users mailing list