[Simh] VMS Cluster boot on an unique container

[Matt]

On 20/10/2011 19:16, Boucher, François wrote:
>
> Hi!
>
> I never succeeded to operate a cluster of machines that attaches directly
> to the same disk drives file images, when a first node writes to the
> system
> disk, the other node was not aware of the changes and it ended up in a
> filesystem
> corruption.  Perhaps I did not found the correct way to
> create/configure some sort of
> dual-port scsi devices?
>
The trick is to make sure that all nodes have the same ALLOCLASS
parameter. This will cause there to be multiple devices in the cluster
with the same full name (e.g. $1$DUA0) so VMS will assume they are the
same physical device and the lock manager will take care of concurrent
access. The only catch is that _every_ disk device that has the same
name between two nodes is assumed to be the same physical device. If you
attach them to different containers on the host side then you will get
strange results and probably data corruption.

To boot the secondary nodes with Simh use the R5 parameter to specify an
alternative system root:

sim> boot rq0 /r5:10000000

to boot from SYS1.

Matt

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20111020/31120124/attachment-0002.html>