Avaya Media Processing Server Series System (Software Release 2.1) Manual de usuario descargar pdf (Pagina 283)

Configuration Procedures and Considerations

# P0602477 Ver: 3.1.11 Page 283

Verifying N+1 Functionality

This section assumes that the MPS cluster is fully configured, all nodes are up and

running, the secondary is running its standby configuration, and the secondary has

performed at least one successful file synchronization for each primary.

This test should be executed multiple times for primaries with varying numbers of

MPS components (if available), as well as with different configurations of auto-

failback commented out, and set to hard in the trip.cfg files on the primary.

Failover

Initiate a failover from one of the primaries in one of the following ways:

• Unplug the Ethernet cable from the MPS/node that connects that MPS/node to

the TMS LAN (this requires that the MPSs are configured to communicate

with each other over the private AVAYA LAN (which is the same as the TMS

LAN)).

• Shutdown the MPS/node.

• Remove power from that MPS/node (this is a rather drastic test which may

result in file system corruption resulting in the need for user intervention to

bring the node back up).

• Stop SRP on the Primary MPS/node (via services on windows and

/etc/rc3.d/S20vps.startup script on Solaris).

Depending on the method used to initiate the failover and the health check interval it

can take 60+ seconds before the secondary attempts to take control of the TMS(s).

Follow these steps to verify that the failover was successful:

1. On the secondary, verify that SRP and all COMMON and MPS

components are in the running state using:

vsh> srp -status

2. On the secondary, verify that the following alarm appears for every MPS

component of the failed primary (the following is just an example of the

alarm. The "#mps.10" is really "#mps.N", where N is the primary MPS

component number that failed over).

Make sure that the number of these alarms exactly matches the number of

primary MPS components (extra or too few alarms are an indication of a

problem).

Fri Jun 10 12:21:41 <trip> 10001 Severity 1 Comp #mps.10/tmsi01

Entering ACTIVE state for #mps.1

In the first line, #mps.10 is the standby MPS.

In the second line, #mps.1 is the failed primary MPS.

3. Verify that the secondary is running the correct applications for the

primary it replaced.

1 2 ... 278 279 280 281 282 283 284 285 286 287 288 ... 305 306

Sin comentarios

Avaya Media Processing Server Series System (Software Release 2.1) Manual de usuario Pagina 283