I apologize for the late summary.
I received many helpful suggestions. Thank you to all who replied.
Here's a summary of the suggestions I have received:
1) Make sure that the time on the machines are synced up
(this was done, it fixed one of the many problems I had
months ago). I'm using ntp.
2) Possibly add "noac" to the vfstab entry to improve locking. I
have yet to try this.
3) Get Pine 4.x. Many people have reported improvements with
this version. I immediately installed it. It is a little
bit better but it didn't fix my problem completely.
4) It was recommended that I start more nfsd servers (is this
necessary in Solaris 2.x?)
I've performed a truss on the pine process. The slowdown happens mostly
when it scans the mailbox, so it may indicate NFS performance problems.
I'm going to try the "noac" option and investigate maybe using IMAP to
access the mailboxes on the nfs server from the shell machine.
Thanks again to all who replied... My original post is below:
>From admin1@berkshire.net Wed Oct 14 06:36:51 1998
Date: Wed, 7 Oct 1998 07:09:24 -0400 (EDT)
From: Admin Staff <admin1@berkshire.net>
To: Sun Managers <sun-managers@sunmanagers.ececs.uc.edu>
Subject: Pine Problems again
Greetings Sun Managers,
I'm running a network of Sun UltraSparc 1170E and 2170E machines
interconnected via a Cisco Catalyst 2916 100mbps full duplex switch. An
ultra2 serves as my NFS server. It serves, among other things, the mail
spool to my shell and mail machines, which are part of the above described
network. All machines are Solaris 2.6 that have had their recommended
patches applied.
When using Pine 3.96 on the shell server, accessing the mail spool via
NFS, I'm experiencing a lot of hangs, especially in large mailboxes. A
typical example of a hang would be as soon as I open a message, I get a
"**" in the upper left hand corner and it hangs for atleast a solid
minute. I cannot scroll up or down the message or return to the index.
Other hangs occur when moving mail to the read mail folder on exit.
Pine works very fast when launched on the machine that physically has the
mail spool mounted.
In the past, I've written a message to this group asking for assistance on
this. I also received some "me-toos" after posting. Someone recommended
that I change the NFS mount for the mail spool to version 2, which I did.
This improved things greatly but still left the annoying hangs described
above.
My vfstab entry for the mail spool on the client machine has the
following options:
vers=2,proto=udp,actimeo=0
I was wondering if anyone else has come to a fix of this problem. TIA for
any advice you could give. I'll be sure to post a summary.
-Jason
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:50 CDT