Description of problem:

Solaris 10 NFS client hangs while communicating with built-in GlusterFS NFS server.  Filesystem is distributed across 252 peers running RHEL6 and is configured as follows:

nfs.register-with-portmap: off
nfs.port: 38467
nfs.volume-access: read-only
nfs.ports-insecure: on

Each peer NFS-mounts the filesystem locally without problems.  Mounting from Solaris using e.g. 'mount -o proto=tcp,vers=3 nfs://node251:38467/cache /gfs/frames' works as well, but hangs when attempting to access subdirectories 2 levels deep.


Version-Release number of selected component (if applicable):

3.2.5


How reproducible:

always

Steps to Reproduce:
1.
2.
3.
  
Actual results:

# uname -a
SunOS qfs-nfs1 5.10 Generic_147441-07 i86pc i386 i86pc
# mount -o proto=tcp,vers=3 nfs://node251:38467/cache /gfs/frames
# df -h /gfs/frames
Filesystem             size   used  avail capacity  Mounted on
nfs://node251:38467/cache
                        93T    52T    41T    56%    /gfs/frames
# cd /gfs/frames
# ls
A6     S6     VSR2   VSR3   trend
# ls S6
L1           L3           LDAShoft     LDAShoftC02
# ls S6/L1
NFS server node251 not responding still trying



Expected results:


Additional info: