I've been following up on my problem, which I previously described as:
I've concluded that the Netgear router (FVS318) performing the NAT is corrupting the outgoing RTP packets. Traces confirmed that the BudgeTone is sending them out with a UDP checksum of 0 but the next hop after the Netgear router they are set to a non-zero value (an incorrect one). Asterisk is never even seeing the packets because the kernel is recognizing them as corrupt and dropping them, hence the recvfrom() "Resource temporarily unavailable" errors in rtp.c.
Here is Netgear's response:
---------------------------- Original Message ---------------------------- Subject: RE: Webform contact request [#<removed>] From: [EMAIL PROTECTED] Date: Fri, January 23, 2004 7:36 pm To: <removed> --------------------------------------------------------------------------
SIP VOIP phones do not work with netgear routers. The router will always set a value in the checksum.
Regards
<removed> Netgear Support [EMAIL PROTECTED]
Please help us serve you better by clicking here mailto:[EMAIL PROTECTED] if you would like to provide any other valuable feedback. (Note: this feedback is not sent to an agent so you will not receive a reply.)
--------------------------------------------------------------------------
Not exactly what I call stellar customer service!
I realize this may not be the best solution -- actually, it's probably not even a good solution -- but has anyone experimented with using the Linux SO_NO_CHECK setsockopt() option? It looks like it could be used to ignore the checksums for the RTP packets.
Owen
Time to dump the Netgear router. That's an unacceptable answer for a router vendor to say "Oh, well, for this MAJOR protocol we're going to simply corrupt those packets so they're unusable." What!?
JT _______________________________________________ Asterisk-Users mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-users To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
