At line 116 and 128 of torture/raw/mux.c, the test case expects the server to respond OK to the 1st NT_CREATE_ANDX request and to respond SHARING_VIOLATION to the 2nd request.
The following is take from MS-SMB 2.2.9: (http://msdn.microsoft.com/en-us/library/cc246334(PROT.13).aspx)
If the server receives more than one SMB_COM_NT_CREATE_ANDX request from a client before it sends back any response, the server may respond to these requests in any order.
The test case should expect server to return an OK and a SHARING_VIOLATION for the two requests in any order. E.g. the server may return SHARING_VIOLATION for the 1st request and OK for the 2nd request.
Also, the test case should not enforce timeout limits on server response time. (such as at line 118) Timeout is not specified in the protocol doc.
Add metze to CC list since he does some work on SMB(2) and torture.
Tim Prouty, since you have worked lately on s4 torture: would you like to fix this?
Steven, would you like to correct the "torture/raw/mux.c" test as stated here?
I'll look into this.
Steven, any news on this - a year has already passed?