PowerEdge 1955 Memory corruption issue

Hi,

 

We have a month old PowerEdge 1955 Blade servers. 10 in each enclosure running Win 2003 R2. We use it as build machines. For the last couple of weeks, some of the servers reboot due to some memory problems and the system becomes unresponsive until I remotely log in to the machine. When I log in remotely, the error message I get is that Windows has recovered from a serious error and has rebooted.

When I analyze the dump logs with WinDbg, it tells me one of the 3 reasons below

1. due to MODULE_NAME: mvfs50 -- which looks like a ClearCase issue

2. due to naiavf5 -- which looks like some issue with McAFee program running

3. due to ntkrpamp.exe --- which again looks like a memory problem. And this is the most frequent message I get.

 

All the machines are located at a remote site so I cannot run the memtest or system diagnostic tests as we don't have a digital KVM.

Could you please let me know what kind of tests I can perform remotely to see where the problem is?

Also, I would be going to the remote site soon and it will be a great help if anyone has any suggestions in deciding what tests to run to figure out whether there is a problem with the memory module or any other software.

 

Thanks very much for your time.

My apologies for the long post.

 

Sincerely,

mozzo