[Spread-users] BUFFER_TOO_SHORT && endian_mismatch >= 0

[Jeremy Hylton]

>>>>> "JS" == Jonathan Stanton <jonathan at cnds.jhu.edu> writes:

  JS> So I'm not sure if that answers the question. If you are saying
  JS> that endian_mismatch is 0 when you also got a BUFFER_TOO_SHORT
  JS> error then that shouldn't occur I think.

We are specifically seeing a BUFFER_TOO_SHORT return from SP_receive()
where endian_mismatch is >= 0.  Your message seems to confirm that
this should not happen, except that it does ;-).

What should we be doing to further diagnose the problem?  I'll see if
I can ask our user how often this is happening.  We haven't had an
opportunity to reproduce the error in our testing environment.  It
seems to be infrequently occurring, though, so I'm guessing source
code inspection is as likely to find a problem as anything. 

Jeremy