The Question is:
I would like to install additional network cards into Alpha Servers running VMS
7.2-1 and TCP/IP V5.0A. Some of the network cards are to be used by the
application and some will be used by the quorum node for cluster communication.
The questions I have are about the fail-over capabilities of the network cards
in a single
If I purchase dual ported network cards and there is a failure on one of the
circuits would I be correct in assuming that the network card/VMS detects this
failure and enables the other port for use?
If I purchase two separate network cards, and have one active and the other one
in stand-by mode, will the failover be automatic or is this a manual process.
Can I have the same IP address for both cards?
The Answer is :
If the NIC fails, then the platform itself can often encounter an
associated failure -- a bus hang, some corrupted bus traffic, network
problems, etc. If you happen to encounter a non-fatal controller
error (NIC or otherwise), then OpenVMS and the host software can
generally transparently manage operations across all of the available
OpenVMS treats multi-port NICs as multiple individual NICs.
If a multi-port NIC fails, the failure may or may not affect other
ports on the NIC.
Far more to failover and recovery planning must be considered than can
be easily discussed in Ask The Wizard -- topics range from redundant
network servers to non-adjacent cable runs to periodic full failure
testing to component tracking to lighting to power sources, etc.