I have installed two Exchange 2013 SP1 servers on Windows Server 2008 R2 SP1 Enterprise, called mail01 & mail02, both have CAS & Mailbox roles installed, and configured DAG. There are four active mailbox database copies on mail01 server, and four passive copies on mail02.
I have read that I should backup the active database copies with Windows Backup, so I setup Windows Backup on mail01, and backup the volume D: where all database and log files are located with VSS Full Backup selected, the backup destination is a network shared folder on another server. I scheduled the backup running in every night.
Next day when I check the database status, I noticed that all databases were activitied on mail02. Then I check the event logs on mail01, and found that the server was crashed and automatically restarted at about 1 hour after the backup started.
The BugCheck event ID 1001 shows the error code: 0x000000f4 (0x0000000000000003, 0xfffffa8013a9c390, 0xfffffa8013a9c670, 0xfffff8000277f270).
In addition, there were some Exchange services stopped unexpectedly before the server crash, such as MSExchangeRepl, MSExchangeDelivery, and Cluster Service.
I also noticed that if I run the backup on mail01 when it holds only passive database copies, the backup will finish without any error, and the server will not crash.
I also tested the backup on mail02 when it holds all active database copies, mail02 crashed and restarted with the same blue screen error.
The hardware of two servers are not identical, one is IBM x3650, and the other is IBM x3250. The two servers are all running fine except for this one issue.
What may cause this issue, and what should I do to troubleshoot? Any help will be greatly appreciated.
-Steven