Bob Goddard wrote:
The apparent packet loss you are seeing may be just fine tuning of the routers in question.
This is the conclusion I came to as well; however, with the way PingPlotter works the router is not sending ICMP unreachables but rather ICMP TTL expired responses. In any case, the routers in question may either be:
1) ...intentionally discarding the received UDP "ping" packets (these are not ICMP pings, but rather UDP packets with TTL down to zero when they get to the router), because the router has better things to do.
2) ...throttling the ICMP TTL expired responses to a certain rate per period of time, as you suggest. This would appear as packet loss.
3) ...actually congested, with the received UDP "pings" (and other types of packets) getting discarded on the input side at the rate shown in the data.
I wish there was a way to measure 3) without being affected by 1) and 2).
I agree then, that PingPlotter is not a highly accurate way to measure path quality. Still, though, looking over the data for a couple days now it is easy to see cyclical patterns that go from 1% to 30% (PingPlotter measured) loss, and an easily seen correlation with the voice quality of my outbound Broadvoice calls.
Interestingly enough, switching from a Firefly soft phone on my workstation, using IAX2/ulaw, to an analog phone->TDM400 FXS port right at the Asterisk server has made a big difference. So some of the perceived crappiness was in the soft phone->Asterisk path and was probably being exacerbated by the network loss on the net or at Broadvoice's router.
-Johnathan _______________________________________________ Asterisk-Users mailing list Asterisk-Users@lists.digium.com http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
