I am working on a program that calls RPC calls, where the client and server are running on the same computer. I want the client to have a timeout for his calls to the server, because we have problems with the fact that he does not respond to waiting on the server.
To solve this problem, we used an asynchronous RPC call and canceled it if time is up.
This works fine in XP, but now I find crashes in Vista, and I have a test program that can play these crashes every time (only in Vista, not in XP). This seems to be a bug on the Microsoft side. The call stack from the broken thread does not include calls to our code. Here is the information from the crash dump:
Call stack:
rpcrt4!_SEH_epilog4_GS+0x3979
rpcrt4!LRPC_CCALL::AbortCall+0x68
rpcrt4!LRPC_BASE_CCALL::HandleCancelMessage+0x6b
rpcrt4!LRPC_CCALL::HandleCancelMessage+0x42
rpcrt4!LRPC_CASSOCIATION::IoCompleted+0x341
rpcrt4!ProcessLrpcComplete+0x3b
rpcrt4!LOADABLE_TRANSPORT:Stick out tonguerocessIOEvents+0x212
rpcrt4!ProcessIOEventsWrapper+0xd
rpcrt4!BaseCachedThreadRoutine+0x5c
rpcrt4!ThreadStartRoutine+0x1e
kernel32!BaseThreadInitThunk+0xe
ntdll!__RtlUserThreadStart+0x23
ntdll!_RtlUserThreadStart+0x1b
Problem:
(ee4.139c): Access violation - code c0000005 (first/second chance not
available)
eax=01e0fcc4 ebx=002f8930 ecx=002f87e4 edx=002f87fc esi=002f87b0 edi=00000000
eip=76f08d9e esp=0200fe2c ebp=0200fe34 iopl=0 nv up ei pl nz na pe nc
cs=001b ss=0023 ds=0023 es=0023 fs=003b gs=0000 efl=00010206
rpcrt4!_SEH_epilog4_GS+0x3979:
76f08d9e a300000000 mov dword ptr ds:[00000000h],eax
ds:0023:00000000=????????
, - ? , - , RPC ?
Jeremy H