more fun with RPCs (rpcecho cli/srv testing)

[jra at dp.samba.org]

On Tue, Apr 22, 2003 at 05:25:10PM +0100, Ronan Waide wrote:
> Using tpot's new rpcecho client and server; NT4SP6a and Samba HEAD:
> 
> [root at gonzo source]# rpcclient qaznt -U Administrator%password -c "echodata 5639" 
> [root at gonzo source]# rpcclient qaznt -U Administrator%password -c "echodata 5640"
> result was NT_STATUS_UNSUCCESSFUL
> 
> Increasing the data size further gets you NT_PIPE_BUSY and finally
> Call timed out errors.
> 
> Using the patches I posted last week, I get a bit more data down the pipe:
> 
> [root at gonzo source]# rpcclient qaznt -U Administrator%password -c "echodata 15558"
> [root at gonzo source]# rpcclient qaznt -U Administrator%password -c "echodata 15559"
> cli_pipe: return critical error. Error was Call timed out: server did not respond after 10000 milliseconds
> result was NT_STATUS_UNSUCCESSFUL
> 
> The numbers aren't random; I fudged about with various sizes until I
> found the size at which the thing breaks.
> 
> I'm pretty certain the breakpoint for my patches is due to the fact
> that I messed with the final SMBTrans packet it sends; certainly in
> the ethereal trace that's the only packet that doesn't get
> acknowledged. I think this gives a good indication that the patches I
> have are on the right track, though.

I haven't forgotten these patches, I'm working on another bug at the
moment - they're in my queue....

Thanks,

	Jeremy.