Re: NFS root monitoring & system optimization

["Robb Main" <mainr@aci.on.ca>]

Thanks, Ken.
That was exactly what I needed - don't know how I missed it before.

For the record, I am using Slackware 7.1, NFS server is v2.2beta46.

From the man page for "NFSD", I used the "-d" option.
I found "-d call" provided data in a format closest to what I want, although
"-d fhcache" also provides similar info.

I did find it interesting that NFS seems to break all file read transactions
into 4KB blocks (pages size?), on my EXT2 NFS mount.  I would have expected
it to trap filesystem requests at a higher level (i.e., OS says read (all)
of the file "libc-2.1.2.so").  However the blocking is probably due to the
way the OS invokes the filesystem, rather than something artificially
introduced by the NFS implementation.

Be warned that these options produce ALOT of log info (600KB+ just to boot
my minimal config).

Cheers,
Robb Main.

----- Original Message -----
From: Ken Yap <ken@nlc.net.au>
To: <mainr@aci.on.ca>
Sent: Thursday, July 27, 2000 9:42 AM
Subject: Re: NFS root monitoring & system optimization


> Can you not monitor it on the exporting machine, say Linux? I know NFSD
> as distributed on my machine is compiled with some debugging turned on
> (annoying) so there must be some hooks in there.

===========================================================================
This Mail was sent to netboot mailing list by:
"Robb Main" <mainr@aci.on.ca>
To get help about this list, send a mail with 'help' as the only string in
it's body to majordomo@baghira.han.de. If you have problems with this list,
send a mail to netboot-owner@baghira.han.de.