[SSI-devel] SSI-1.1.1-FC2 NFS MOUNTD_PORT doesn't work

[Roger T. <pe...@ho...>]

Hi on SSI-1.1.1-FC2 I've tried unsuccessfully to force mountd to listen on a 
specific port by adding MOUNTD_PORT=4002 in /etc/sysconfig/nfs  I also have 
MOUNTD_NFS_V2=no

If you see below it seems rpc.mountd is trying to bind to the same listening 
port for all nfsd versions.

[root@node1 init.d]# service SSInfs start
(node 1)
[root@node1 init.d]# service SSInfs status
(node 1)
rpc.mountd is stopped
nfsd (pid 70275 70274 70273 70272 70271 70270 70269 70268) is running...
rpc.rquotad (pid 70393 70264) is running...

Nov 30 08:44:18 node1 nfs: Starting NFS services:  succeeded
Nov 30 08:44:18 node1 rpc.rquotad: rpc.rquotad: Cannot bind to given 
address: Address already in use
Nov 30 08:44:18 node1 nfs: rpc.rquotad startup succeeded
Nov 30 08:44:18 node1 nfs: rpc.nfsd startup succeeded
Nov 30 08:44:18 node1 rpc.mountd: Could not bind name to socket: Address 
already in use
Nov 30 08:44:18 node1 nfs: rpc.mountd startup failed

Okay so I stop nfs and restart some stuff below...

[root@node1 init.d]# service portmap restart
(node 1)
Stopping portmapper:                                       [  OK  ]
Starting portmapper:                                       [  OK  ]
(node 2)
Stopping portmapper:                                       [  OK  ]
Starting portmapper:                                       [  OK  ]
[root@node1 init.d]# rpcinfo -p cluster
   program vers proto   port
    100000    2   tcp    111  portmapper
    100000    2   udp    111  portmapper
    100024    1   udp  32768  status
    100024    1   tcp  32784  status
    391002    2   tcp  32788  sgi_fam
[root@node1 init.d]# service SSInfs start
(node 1)
[root@node1 init.d]# rpcinfo -p cluster
   program vers proto   port
    100000    2   tcp    111  portmapper
    100000    2   udp    111  portmapper
    100024    1   udp  32768  status
    100024    1   tcp  32784  status
    391002    2   tcp  32788  sgi_fam
    100011    1   udp   4003  rquotad
    100011    2   udp   4003  rquotad
    100011    1   tcp   4003  rquotad
    100011    2   tcp   4003  rquotad
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
    100021    1   udp  32806  nlockmgr
    100021    3   udp  32806  nlockmgr
    100021    4   udp  32806  nlockmgr
    100021    1   tcp  33201  nlockmgr
    100021    3   tcp  33201  nlockmgr
    100021    4   tcp  33201  nlockmgr
    100005    1   udp   4002  mountd
    100005    1   tcp   4002  mountd
    100005    3   udp   4002  mountd
[root@node1 init.d]# showmount cluster
rpc mount dump: RPC: Unable to receive; errno = Connection refused
[root@node1 init.d]# service SSInfs status
(node 1)
rpc.mountd is stopped
nfsd (pid 70729 70728 70727 70726 70725 70724 70723 70722) is running...
rpc.rquotad (pid 70718) is running...