[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]
Re: nfs options
- From: "Cheryl L. Southard" <cld astro caltech edu>
- To: "Derek T. Yarnell" <derek cs umd edu>
- Cc: taroon-list redhat com
- Subject: Re: nfs options
- Date: Wed, 22 Sep 2004 09:15:31 -0700
Hi Derek,
We've been seeing this issue since June when we upgraded the kernel and
nfs-utils modules on our RHEL WS 3 Update 2 computers. Processes on 10 of our
linux clients will lock up trying to read files via NFS from our solaris
mail server.
What seems to help is if we add "rsize=8192,wsize=8192" to our options
in /etc/fstab. So now the /etc/fstab line for our solaris:/var/mail line
looks like this on our linux client:
solaris:/var/mail /var/spool/mail nfs rsize=8192,wsize=8192 0 0
Now, I think 8192 is way too small and we can go as high as 32768, but I
haven't tested out that number yet.
Hope this helps.
Cheryl
On Tue, Sep 21, 2004 at 07:29:48PM -0400, Derek T. Yarnell wrote:
> What do you guys use as options to connect to either solaris nfs servers
> or netapps by default?
>
> I am having some wierd issues with programs that are run out of
> /usr/local (nfs mounted from a solaris host) with the default options
> and ro.
>
> I seem to be able to get the device waits to stop if I use
>
> ro,soft,intr,proto=tcp
>
> This is a mount in fstab that is always mounted.
>
> Thanks for the info.
>
> --
> Derek T. Yarnell
> UNIX System Administrator
> Computer Science Department
> University of Maryland
>
>
>
>
> ------------------------------
--
Cheryl Southard
cld astro caltech edu
[Date Prev][Date Next] [Thread Prev][Thread Next]
[Thread Index]
[Date Index]
[Author Index]