On Fri, 2 May 2003, Andrew Sundelin wrote:
> I just configured a test based on PolyMix-4 and I set the abort
> level of both the Servers and the Robots to 0.0%. Yet, I still see a
> reasonable percentage of RST TCP connections on the link.
>
> Is there something else that would be causing RSTs in the Polygraph
> traffic?
Race conditions in HTTP persistent connections may cause connection
resets, I guess. To double check that theory, try turning persistent
connections off (set pconn_use_lmt to const(1), for example). Note
that these race conditions are normal in real world.
> Is setting this abort_prob to 0.0% the way I should be trying to
> turn this off?
If you want to prevent transactions from being aborted, then yes. If
you want to prevent connection resets, then you may have to get rid of
persistent connections.
HTH,
Alex.
--
| HTTP performance - Web Polygraph benchmark
www.measurement-factory.com | HTTP compliance+ - Co-Advisor test suite
| all of the above - PolyBox appliance
This archive was generated by hypermail 2b29 : Mon Feb 06 2006 - 12:00:26 MST