Question about RTT

Vernon Schryver vjs@calcite.rhyolite.com
Thu Nov 14 14:42:48 UTC 2002


> From: Xavier <xavier@rootshell.be>


> ...
> dcc-1,-     xxxxx xxxxxxxx
> #   xxx.xx.xxx.xx,-        dcc-1       xxx server-ID xxxx
> #      75% of 32 requests ok 1969.89 ms RTT             15 ms queue wait
>
> dcc-2,-     xxxxx xxxxxxxx
> # * xxx.xx.xxx.xx,-        dcc-2       xxx server-ID xxxx
> #     100% of 32 requests ok    2.19 ms RTT              1 ms queue wait
>
> So, all requests are sent to 'dcc-2' and a few minutes later, servers are inverted, dcc-2
> has the lowest RTT! :/
>
> Both servers are on the same LAN and are not overloaded by DCC and/or other traffic.
> Any idea?

I do not understand.
Are you asking why the RTTs differ?  Those numbers suggest that dcc-1
failed to answer 8 of the most recent 32 requests.  Failures to answer
are interpreted as extreme slowness for updating the RTT.  Failures to
answer can be caused by network problems or loading of the computer
running dccd.

Or are you asking why the RTTs change?


Vernon Schryver    vjs@rhyolite.com



More information about the DCC mailing list

Contact vjs@rhyolite.com by mail or use the form.