Thanx to the Following for a Quick and Descriptivew Response which helped me
in going in the right direction
Mike DeMarco <email@example.com>
"Scott D. Yelich" <firstname.lastname@example.org>
Craig Russell <email@example.com>
Seth Rothenberg <SROTHENB@montefiore.org
My Original Prb:
> hi Guru's
> Description : I have got one A1000 attached to two E250's Machines
> (libra & gemini). Both the Machines have access to E-250 Machines ie.
> the Same Partions on A1000 are locally mounted on the Machines.
> Prb: When the write is done by the user on Server Libra on A1000 it is
> not Reflected on the Server Gemini mount point of same slice, until and
> unless the slice is Re-mounted. I want that the write done from machine
> one on a slice be immediately reflected on to the other machine mount
> for the same slice.
> This is Urgent. I will summarize the PRb
> Munish Sarin
Solution ( Only one Decisive answer)
This is not possible
Can not be done! The only way to flush the cache would be to unmount and
re-mount the filesystem!
This cannot be done mounting straight through Solaris.
The inode changes are made and tracked at the OS
level. Once the file system is mounted that
information is cached and not referenced from disk
unless a change is made. Since the OS on gemini did
not make the change as far as it is concerned the disk
You need to look at a clustering solution.
Online Retail Partners, LLC
I understand what you are doing, it is not supported.
UNIX does not support having two computers access the same disk
at the same time. What UNIX does support is NFS mount. One system
would own the disk, and would export or share the disk with the second
If you want to do this, you need to use the command share (e.g., on libra),
mountd and nfsd must be running on this host.
The second host eg, gemini) then uses a command like
mount -o rw,soft libra:/users /users
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:02 CDT