OK, The VIO server does two jobs - it connects networking through. and it handles disk I/O.
What I do is in thwo phases:
1. Ethernets are one per VIO Server (I always use dual) and I etherchannel them at the LPAR level. If one of the interfaces goes down thenn we generate a TEC event based on that - just as if it was a physical error.
2. Disks: I allocate whole LUNS through the VIO servers - but in the event of a path to the LUN becoming unavailable then agan teh error log will note this and we generate an event.
3. TEC can the issue an error that one of the admins will go and chack the state of the VIo Server.
We do not do anything on the VIO server itself - will simply monitor the services that it provides to the LPARS based on the eroorlog monitoring already offered by TEC.
__________________
Ross Mather, IBM AIX IT Specialist.
That said anything I say here is my own opinion and not anything that you can ever hold against IBM.
Ohhh and don't forget that I make mistakes too.... |