On Thursday 15 May 2003 13:33, Daniel Sparrman wrote:
> Hi
>
> This is a solution where you would benefit from using HACMP.
The TSM server on the primary site is a HACMP node.
> You could have to clustergroups, each containing a TSM server, some disks
> and some network adapters. Each clustergroup would have one preferred
> node, but would also be able to move to the other node.
>
> Only thing to note, is, you have to have different TCPPorts one the two
> TSM servers, as they could be running on the same machine in the event of
> a disaster.
>
> You then use AIX mirroring, or TSM DB mirroring, to mirror the databases
> between the two sites(preferably using AIX mirroring to two different disk
> subsystems, one subsystem on each site).
>
> I have two configurations like this running today, and it works like a
> charm.
I can't use HACMP because the 2 sites are connected with a small (1mbps or so)
link. I know you can buy disk subsystems who can replicate the data to an
other site, but that's not an option. We have SCSI, SAN or a fastT for
storage.
Stef
--
stef.coene AT docum DOT org
"Using Linux as bandwidth manager"
http://www.docum.org/
#lartc @ irc.oftc.net
|