TCP/IPv6 Configuration and Management Manual (G06.29+, H06.03+, J06.03+)

Manage the NonStop TCP/IPv6 Subsystem
HP NonStop TCP/IPv6 Configuration and Management Manual—524523-012
6-22
Manage Configuration Errors

Each LNP that a ND6HOSTD is run on also requires that at least one DUAL mode
SUBNET exist in that LNP so that the ND6HOSTD can communicate with the DNS
server.

When using multiple logical-network partitions, each with its own copy of the
ND6HOSTD, the IPv6 DNS entries from each partition have the same hostname
used in the fully qualified domain name in the DNS updates unless the
ND6HOSTD process is started with an option to use a different host name on the
partition.
Manage Configuration Errors
If a configuration error is detected, an EMS message is generated and the process
terminates. If the ND6HOSTD process was configured to be a persistent process, the
persistence manager restarts the process. Manual intervention is required to stop the
ND6HOSTD process and correct the configuration error condition. For information
about these errors, see the Operator Messages Manual.
Ethernet Failover for Nonshared IP Behavior
For a nonshared failover SUBNET, a DNS update is performed for the SUBNET IPv6
prefix and the interface ID of the SUBNET and also for the interface ID of the associate
SUBNET in the failover pair.
Prefix Lifetimes
Each IPv6 prefix received in a router advertisement has a valid lifetime. When the valid
lifetime expires, the ND6HOSTD sends a DNS update record to delete the addresses
associated with the IPv6 prefix.
Periodic Updates
When a DNS add or modify update is performed, the ND6HOSTD does periodic DNS
updates while the address or prefix is still valid. Updates are performed every:
Lifetime - (Lifetime/10)
Where:
Lifetime is the valid lifetime of the prefix in seconds.
Managing Performance
When using NonStop TCP/IPv6, processor utilization measurements of client
applications might tend to exhibit higher numbers because the TCP/IP processing is
now done in the context of the application. In the conventional TCP/IP environment,
this work was attributed to the TCP/IP process. Because of this, Measure analysis
might show an increase in processor utilization by the application process.