SUMMARY: Stale NFS file handle with Guardware

From: Carlos Lorence (clorence@sunup.com)
Date: Mon Nov 02 1998 - 05:22:26 CST


Thanks to:

From: Christian Lotz <cl@lotz.de>
Salehi, Michael E Mike.Salehi@usa.xerox.com
Kevin Sheehan u-kevin@veritas.com
Casper Dik casper@holland.Sun.COM
Bruce R. Zimmer bzimmer@all-phase.com

Basically all the answars was:

>make sure, that both hosts are using the same major device numbers
>for their their filesystem/Volume Manager ...

The way to check which mayor/minor numbers is using ls -lL
/dev/rdsk/c1t0d0s0 in both servers, the output should be:

crw-rw---- 1 root root 32, 128 Dec 6 23:23
/dev/rdsk/c1t0d0s0
As seen from the output, the mayor and minor numbers are 32 and 128,
respectively. These values MUST match on both servers for the same device
file in order to fail over NFS transparently.
After modify the instance number in /etc/path_to_inst (something tought and
not recommended if you don't know what you are doing because the system may
hang during bootup if you changed the instance number of a boot disk) i boot
with -rv options and that's it.

Carlos Lorence.



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:51 CDT