runaway lockd problems (SunOS 4.1.3)

[rwing!pat () ole cdac com (Pat Myrto)]

NFS lockd problems:

I am having difficulties with lockd running away and bringing a system
down to its knees, requiring one to kill it to get sanity back.

A centralized mail system (using a centralized /var/spool/mail via NFS)
was installed on another host.  The mail server uses a local mount to
process mail, but all the other hosts (including the one with the runaway
lockd) access the spool via nfs.

The latest sendmail from uunet is the one in use on the server, and the
bind 4.9 in.named is used to handle DNS.

Is there anyone out there who knows what causes lockd to run away,
and what changes one needs to make (either in mail user agents or sendmail)
to get around this annoyance?  This has shown up intermittently since
the centralized server was set up.

For the moment, lockd was killed.  Impact of leaving it dead are not
known yet, there is no commercial software (site is a public access
internet host with dialup shell access).

Thanks for any info or suggestions.
-- 
pat@rwing  [If all fails, try:  rwing!pat () ole cdac com]  Pat Myrto - Seattle WA
"No one has the right to destroy another person's belief by demanding
empirical evidence."  --   Ann Landers, nationally syndicated advice columnist
and Director at Handgun Control Inc.