New question + Summary I:Sun 5100 and NFS

From: Michael DeSimone (michael@desimone.net)
Date: Mon Oct 30 2000 - 15:52:53 CST


Original post at the end.

My two responses pointed out that I was suffering a hardware, specifically,
disk problem. After some research on Sunsolve I found that there were a
couple of patches for the 5100 that fixed these errors, they were not
actually hardware but OS/kernel related.

Now I no longer get any of the original errors. But my NFS through put is
amazingly slow. Over 3 hours to do 200 MB, all on the same network & switch
with no other traffic. I have gone through the Answer book and implemented
almost all of the NFS performance tuning suggestions with no noticeable
improvement. I do have very heavy write traffic to the NFS server. Any
additional suggestions for improving performance?

I also see the following errors. Where my program is supposed to be trying
to write is world writable, the entire path. It is writing multiple files on
an NFS mounted directory that is owned by the same user & UID on both the
server and client, and the entire path to that directory is world writable.
Is there a way to find out where it is trying to write that is getting
denied & what these errors mean, I haven't been able to find anything yet.

Oct 30 12:04:47 mep-cn1 NFS3 write error on host pslab01: Permission denied.
Oct 30 12:04:47 mep-cn1 (file handle: 80000e 2 a0000 30873 6505e83c a0000
1bc143 1c7ae0b5)
Oct 30 12:04:47 mep-cn1 nfs_bio: cred is not kcred
Oct 30 12:04:47 mep-cn1 NFS3 write error on host pslab01: Permission denied.
Oct 30 12:04:47 mep-cn1 (file handle: 80000e 2 a0000 30876 6a771b64 a0000
1bc143 1c7ae0b5)
Oct 30 12:04:47 mep-cn1 nfs_bio: cred is not kcred

Thanks again in advance,
Michael DeSimone
Computers & Stuff

Original Question:
I have inherited a Sun 5100 with 14 36GB drives attached to a 220R via
fiber. I am using Veritas on 12 of the disks in a Raid 0+1 arrangement.
Everything on same subnet and same switch with little or no network traffic
other then what I am doing. Most of the files I am moving are rather large,
100+ MB to over a GB. Locally everything works great. When trying to mount
this partition via NFS on the client I get NFS timeouts. On the 220 I get
the following errors:

Oct 25 11:22:24 mybox Unix: WARNING:
/pci@1f,2000/SUNW,ifp@1/ssd@w21000020376c5bfd,0 (ssd12):
Oct 25 11:22:24 pslab01.ps. SCSI transport failed: reason
'incomplete': retrying command
Oct 24 20:21:14 mybox Unix: WARNING:
/pci@1f,2000/SUNW,ifp@1/ssd@w21000020376c5bfd,0 (ssd12):
Oct 24 20:21:14 pslab01.ps. SCSI transport failed: reason
'timeout': retrying command

Using the NFS troubleshooting/tuning guide at
http://www.princeton.edu/~unix/Solaris/troubleshoot/nfs.html and the Sun
Answer Books
I don't really see anything wrong by those guidelines.

Is there anything else I should be looking for? Is there anything I should
change no matter what my nfsstat reports? Is it just stupid trying to move
files this large via NFS? Would going to a NetApp help?

S
U BEFORE POSTING please READ the FAQ located at
N ftp://ftp.cs.toronto.edu/pub/jdd/sun-managers/faq
. and the list POLICY statement located at
M ftp://ftp.cs.toronto.edu/pub/jdd/sun-managers/policy
A To submit questions/summaries to this list send your email message to:
N sun-managers@sunmanagers.ececs.uc.edu
A To unsubscribe from this list please send an email message to:
G majordomo@sunmanagers.ececs.uc.edu
E and in the BODY type:
R unsubscribe sun-managers
S Or
. unsubscribe sun-managers original@subscription.address
L To view an archive of this list please visit:
I http://www.latech.edu/sunman.html
S
T



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:20 CDT