![]() |
![]() HP OpenVMS Systemsask the wizard |
![]() |
The Question is: WHere can I find an explanation of how a cluster failover is handled? I am specifically in need of an axplanation of the way tcp/ip address and mac addresses (which are cached in the local user systems arp cache) are handled. for example if i am connected to a cluster via an alias and the actual ip address/mac address of my target system disappears, then how do i reconnect given that my arp cache is 'dirty', and points to a mac address which is no longer available....? Where are these things handled? Is this documented. A prospect I am selling to needs an accurate description of this please!> many thanks Dave The Answer is : You are seeking information on how the TCP/IP Services for OpenVMS (UCX) product or other IP package handles a cluster, and apparently are not specifically interested in how cluster failover operates. In general, when the system fails underneath a connection, be it DECnet or IP, the application must reconnect back into the cluster. Using the alias capabilities, the TCP/IP or DECnet software will distribute any incoming connection to an available node -- the application software does not generally have to trouble itself with the lower-level implementation details of the cluster alias. For information beyond what is included in the UCX documentation on the cluster alias and associated processing, please contact the pre-sales technical support organization.
|