strange nlockmgr behaviour on a computer

[PARANT]

hello,

I write a script checking on my lan ( mainly HPUX 10.20, 11.00 and 11i with some SUN's ) whether the computers are nfs client or not.

To do that, i check by rpcinfo if there is a version of nlockmgr ( from 1 to 4 ) on all machines.

This works for all machines in the lan except one ( os: 11i ), the rpcinfo gives only

program vers proto port service
100000 4 tcp 111 rpcbind
100000 3 tcp 111 rpcbind
100000 2 tcp 111 rpcbind
100000 4 udp 111 rpcbind
100000 3 udp 111 rpcbind
100000 2 udp 111 rpcbind

Is it quite normal ? and if so is there a magic way to check nfs client over a network ??

Thanks for your answers

[Simon Hargrave]

I believe the nlockmgr is only there if filesystems are actually mounted, not if the client is "running", therefore it is possible that a server is capable of being an NFS client but this RPC service does not show.

The easiest way to check if a server is an nfs client is to check for the "biod" process. This is a Block I/O Daemon, used by the NFS client to cache block read/writes via NFS. If this process(es) is running, then the server is an NFS client.

[PARANT]

i use rpcinfo for two main reasons:

no need tuo use rsh stuff
biod does not exist on solaris 6

In fact, for my pb, i begin to check with rpcinfo and if i dont have succeed, i rsh and i lokk after the biod processes. I thought rpcinfo was a nice ( and universal among Unices flavors ), but i think a am wrong