I hope someone has some idea as to why this might be happening. So far, neither Microsoft or Hewlett Packard are able to come up with an answer on why Native SQL Server Dumps writing to a UNC path, across Gigabit Ethernet from multiple SQL Servers is failing (totally at random) when dumping to an HP Storage Server NAS appliance.

This is not an error that can easily be reproduced and it appears to happen totally at random. One common thread is that this is happening to several SQL Servers, all of which are pointing to the same destination point. The main error being generated is a Operating system error 64(The specified network name is no longer available.) message. So far, the Microsoft-recommended fixes for this are not working. We are also seeing Error 1450s as well.

My associates and I are totally baffled as to why this is happening. The HP Storage Server that is being used does not seem to be generating any errors indicating that I/O, Memory or Network pressure exists. The Gigabit Interface(s) are not over-utilized and there are no errors reported at the Network Switch level either.

I would like to know if others out there have experienced this issue with backing up about 2.5+ Terabytes of SQL Data across multiple servers to a single NAS appliance.

All suggestions are appreciated.


Marty B.