Subject: Re: NFS default block size
From: Jim (jwadell@gci.net)
Date: Fri Jan 24 2003 - 21:51:03 AKST
Paul Carr wrote:
> I am tring to optimize my NFS setup. According to the doc's, in modern
> kernels (such as 2.4) the default block size for an NFS is set by the
> server at 4096. Trying to make an adjustment on the client side at
> mount does not seem to work for me. (could be a syntax issue) Any ideas
> on how to do it or where on the server side of the setup to adjust the
> block size?
>
> The doc's also indicate that this is probably the single most important
> optimization issue. Anybody actually seen better performance with a
> block size other the the default?
>
> Thanks.
>
> ---------
> To unsubscribe, send email to <aklug-request@aklug.org>
> with 'unsubscribe' in the message body.
It is set in the mount command, if I remember correctly the syntax is
tsize and rsize, but it has been a couple of months since I messed with
it. If noone comes up with the correct syntax, I will check at work on
Monday.
Some Cisco switches are really fussy about full blocks, and we had to cut
them down to 1024
Jim
---------
To unsubscribe, send email to <aklug-request@aklug.org>
with 'unsubscribe' in the message body.
This archive was generated by hypermail 2a23 : Sun Jan 26 2003 - 02:58:25 AKST