On Thu, 17 Oct 2002, Wolfe, Gordon W wrote:

> We have a number of Linux (SuSE 2.4.7) images running under z/VM 4.2.
> These are all connected to the VM TCPIP stack via virtual CTCA's.
>
> If the VM TCPIP stack comes down and back up, is there a way to reconnect
> the linux servers without shutting them down and rebooting?
There are only 2 reasons to reboot a linux: changing the number of
processors or changing your abount of memory. Everything else could be
done without reboot. It is a linux.
In worst case you need to shutdown all ctc-interfaces on that box, remove
the ctc module, reload it and start the interfaces up again.

You see, CTCs are not very reliable. The best virtual network interfaces
at the moment are GuestLans.
The reason for this is, they only "couple" to the interface and not to the
guest at the other side of the interface. For this reason, if a guest
dies, it doesn't matter to the others, except it stays down longer than
the tcp timeout.

And at all I claim there are no reasons to route a linux guest through the
VM TCPIP. ;-)

Greetings
Oliver Paukstadt

+++LINUX++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+++Manchmal stehe ich sogar nachts auf und installiere mir eins....+++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Reply via email to