Re: NFS fails after 40 minutes of use

Richard B. Johnson (root@chaos.analogic.com)
Sun, 21 Feb 1999 20:51:50 -0500 (EST)


On Sun, 21 Feb 1999, Simon Kirby wrote:

> Try mounting each remote filesystem separately...Eg:
>
> [sroot@alfie:/dock]# cat unix_mount
> #!/bin/tcsh -f
>
> mount -t nfs -o ro ${*}:/ /dock/$*/live || exit -1
> echo " ${*}:/ mounted on /dock/$*/live/"
> foreach a (`cat /dock/$*/live/etc/mtab | \
> egrep '^[^ ]+ /[^ ]+ ext2 ' | cut -f2 -d' '`)
> mount -t nfs -o ro,rsize=8192,wsize=8192,hard,intr,timeo=6,retrans=9 \
> ${*}:$a /dock/$*/live$a/ || exit -1
> echo " ${*}:$a mounted on /dock/$*/live$a/"
> end
>
> :
>
> I was told this is necessary in order to avoid inode hash clashes. I was
> having a problem where as soon as it stat()ed or touch a certain remote
> file, the mount point would instantly turn into that file. Apparently
> mounting each remote filesystem separately after mounting the root avoids
> this problem (somehow).
>
[SNIPPED...]

Thanks. I really don't need a work-around since I would not normally
back up file-systems that way. I sure hope that nobody thinks that
this observed behavior is correct. We need to understand that this
is a bug.

FYI I normally do `tar -clf` (stay in the same file-system) locally
to a local tape-drive. The information presented was to show the
bug in a repeatable manner.

Cheers,
Dick Johnson
***** FILE SYSTEM WAS MODIFIED *****
Penguin : Linux version 2.2.1 on an i686 machine (400.59 BogoMips).
Warning : It's hard to remain at the trailing edge of technology.
Wisdom : It's not a Y2K problem. It's a Y2Day problem.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/