NFS/rpcbind kicking my ass

Hi,

	I'm having a hell of a time with NFS/rpcbind.  All of a sudden, out
of the blue, the rpcbind on my NFS server (5.3-p10) just stops answering.
I do have it under libwrap control, but I have in my hosts.allow :

ALL: localhost MACHINENAME MACHINEIP [::1] : allow
rpcbind: MACHINE1 MACHINE2 MACHINE3 [::1] CLASS.C.IP. : allow
portmapper: MACHINE1 MACHINE2 MACHINE3 [::1] CLASS.C.IP. : allow

	So if ANYTHING, the local machine should be able to do what it
needs. A sockstat looks clean.... 

	I've restarted rpcbind 5 times, nfsd, mountd, etc..... And I
still can't contact it.... Ideas?

			Thanks, Tuc