[Gluster-devel] posix-locks under AFR not working for server+client in one process

Rommer rommer at active.by
Sat Oct 11 14:53:32 UTC 2008


Hello Everybody,

I'm testing glusterfs on two nodes, and found one problem.
I decided to use one process as server and client, cause of
best performance in my case. This is my configuration file:

####### local brick #########
volume posix
  type storage/posix
  option directory /mnt/export
end-volume
volume posix-locks
  type features/posix-locks
  subvolumes posix
end-volume
volume io-thr
  type performance/io-threads
  subvolumes posix-locks
  option thread-count 4
end-volume

#### export local brick #####
volume server
  type protocol/server
  subvolumes io-thr
  option transport-type tcp/server
  option bind-address 192.168.1.12
  option auth.ip.io-thr.allow 192.168.1.*
end-volume

####### remote bricks #######
volume remote
  type protocol/client
  option transport-type tcp/client
  option remote-host 192.168.1.13
  option remote-subvolume io-thr
  option transport-timeout 3
end-volume

########### afr #############
volume afr
  type cluster/afr
  subvolumes io-thr remote
  option read-subvolume io-thr
end-volume
volume wb
  type performance/write-behind
  subvolumes afr
end-volume
volume ra
  type performance/read-ahead
  subvolumes wb
end-volume
#############################

Mounting on both nodes by:
# glusterfs --spec-file=/etc/glusterfs/glfs.vol -n ra /mnt/glusterfs

However, locking doesn't work in that configuration.
I can still lock the same file on both nodes at the same time.

If I use server (glusterfsd) and client (glusterfs) as separated
processes, the same lock tests work perfectly.

Rommer





More information about the Gluster-devel mailing list