lkml.org 
[lkml]   [1999]   [Feb]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [patch] workaround for solaris 2.5.1 and 2.6 FIN bug (ID 4083814)
Andrea Arcangeli wrote:
> >The fix was to make sure that we did *not* retransmit the data, but
> >*do* retransmit the FIN. When Solaris receives the FIN without data
> >it handles it correctly.
>
> I thought about this way to workaround the Solaris bug infact. Maybe I
> missed some bit of TCPRFC, but I thought it was not possible because once
> Solairs didn't seen the FIN flag, it just interpreted the last octect of
> the FIN-full-of-data packet as data while it was a virtual octect. Am I
> wrong here?

Well -- I never saw any evidence of extra bytes in my testing. It seems
unlikely that Solaris thinks that there is an extra byte because TCP
only passes over the sequence number of the first byte of data and
the length of data is computed from the IP length. Also, Solaris
generates
an ACK indicating that it received all the data bytes but not the
virtual FIN control byte.

Philip
--
Philip Gladstone +1 781 530 2461
Axent Technologies, Waltham, MA[unhandled content-type:application/x-pkcs7-signature]
\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.041 / U:2.444 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site