I have five completely different W2K Pro machines (a mix of desktops and notebooks from different vendors) that occasionally generate the following BSOD when our application running on these machines is accessing a large (10-200 MB) flat file on a W2K Server. The BSOD is as follows:
*** STOP: 0x0000000A (0x0D01800A, 0x00000002, 0x00000000, 0x80415025)
*** Address 80415025 base at 80400000, DateStamp 3ad7ad60 – ntoskrnl.exe
Our application makes heavy use of the MFC CFile class to access this large file. It utilizes the following methods: Open (with attributes of “share” and “DenyNone”), Close, SeekToEnd, SeekToBegin, Seek, and Read.
The closest thing I can find in the MS Knowledge Base is article Q165456 ("The NTFS file system driver is attempting to perform I/O to a fragmented file and does not correctly clear a required field, causing either a STOP 0xA or a deadlock condition, which causes the process to stop responding.") but that article says that it applies only to NT4 (not W2K). Furthermore, it says that the problem was resolved in an NT4 SP.
For us, this problem occurs randomly but occurs across all five different W2K Pro machines - some SP1 and some SP2.
From a W2K Server standpoint, the problem has been observed on the W2K Pro machines when accessing two different servers (Compaq ML350 and ML370 G2 both running SP2) in different geographical locations (Maryland and Michigan) on completely independent LANs.
Because the problem has been seen on all W2K Pro machines in two different geographical locations accessing different W2K Server machines, I don't think that the problem is related to the local hardware or network. Rather, it seems to be related to W2K itself.
We are about to go into production with this system and this occasional, unpredictable BSOD is a real problem.