To reiterate Col's statement of that this could be just about anything:
SAN attached cluster, W2K3x64 SP2.
As part of a migration process, I had to disconnect a number of connections to SAN attached drives on the active SQL cluster node.
At exactly the same time as the disconnect of the SAN drive paths happened, the application event log listed a number of ID19019, all pointing to a network connectivity issue:
------------------------------
[sqsrvres] printODBCError: sqlstate = 08S01; native error = 40; message = [Microsoft][SQL Native Client]TCP Provider: The specified network name is no longer available.
[sqsrvres] printODBCError: sqlstate = 08S01; native error = 40; message = [Microsoft][SQL Native Client]Communication link failure
------------------------------
No SQL cluster resources go down though.
How could this happen? Well here's my take on it:
Windows is humming along, gets wind of a few devices it knows about going AWOL, has to engage the hardware management subsystem to go and verify who's there and who not and make status change notifications to other subsystems as necessary...
This process holds up network communications at low (driver) level for a few split seconds too, making it look like there was a problem with the link, but it's actually not.
Rgs,
K28.5