[Exim] Stuck processes. How do I proceed?

I'm using exim 3 on a Solaris server.
I've searched the mailing list archive without finding an answer to this.

Users' mail is delivered to an NFS mounted mailbox.
Following an NFS server crash, inbound mail is now queueing up on our
mailhub. I've tried restarting exim but the fault persists. The fault is
that if I try to deliver a message exim reports "spool file is locked".
I can see the .lock file is created at this time but unless it is
removed by hand, it remains.

It looks like I have a problem with stuck processes. Although the
FAQ/mailing list archive have allowed me to identify that the problem is
that the process that is delivering the message has 'somehow got stuck',
there doesn't seem to be any advice on how to proceed. I don't want to
lose the incoming mail but would like to know if I can safely remove any
files from the spool dir before retrying delivery.

The local_delivery transport section contains:
local_delivery:
driver = appendfile
file = $home/mailbox
create_file = inhome
use_fcntl_lock = true
use_lockfile = true
require_lockfile = false

I added the folowing without any effect:
lock_retries = 10
lockfile_timeout = 30m


Please reply to e.a.sandland@??? for obvious reasons!

Liz
--
Liz Sandland
e.a.sandland@???