Hi,
Should I try something like
nat_uac_test(diff-ip-src-contact || private-contact || diff-ip-src-via ||
diff-port-src-via)?
Any ideas?
Regards,
Ronald
December 15, 2023 at 4:38 PM, r...@rvgeerligs.nl wrote:
>
> Hi
>
> I use opensips 3.4 and I have NAT problems with 2 devices behind the same NAT
> (called party hears nothing).
> The A party is softphone on iPhone (linphone) the B (called)party is
> Polycom310. The other way around works (Polycom calls linphone).
>
> Actually I tested this in two different NAT locations.
> The first location repeatedly works. The second location gives the problem.
> Both locations have changing public IP addresses and DHCP on 192.168 network.
> Both locations have one FX number assigned to them. No SIP ALG active.
>
> I use nat_uac_test(diff-ip-src-contact).
>
> There is a table:
> 1.5.5. nat_uac_test(flags)
> Tries to guess if client's request originated behind a nat. The parameter
> determines what heuristics is used.
>
> Meaning of the flags (string) parameter is as follows:
>
> private-contact - (old 1 flag) Contact header field is searched for
> occurrence of RFC1918 / RFC6598 addresses.
>
> diff-ip-src-via - (old 2 flag) the "received" test is used: address in Via is
> compared against source IP address of signaling
>
> private-via - (old 4 flag) Top Most VIA is searched for occurrence of RFC1918
> / RFC6598 addresses
>
> private-sdp - (old 8 flag) SDP is searched for occurrence of RFC1918 /
> RFC6598 addresses
>
> diff-port-src-via - (old 16 flag) test if the source port is different from
> the port in Via
>
> diff-ip-src-contact - (old 32 flag) address in Contact is compared against
> source IP address of signaling.
>
> diff-port-src-contact - (old 64 flag) Port in Contact is compared against
> source port of signaling
>
> carrier-grade-nat - (old 128 flag) also include RFC 6333 addresses in the
> checks for ct, via and sdp flags.
>
> A CSV of the above flags can be provided, the test returns true if any of the
> tests identified a NAT.
>
> Currently I use old flag 32.
>
> I read that using the equivalent of 19 might help but I dont see that in the
> table.
>
> Any advice is appreciated.
>
> Regards,
>
> Ronald Geerligs
>___
Users mailing list
Users@lists.opensips.org
http://lists.opensips.org/cgi-bin/mailman/listinfo/users