[reSIProcate] repro AssumeFirstHopSupportsOutbound not quite working?
Scott Godin
sgodin at sipspectrum.com
Fri Dec 2 13:30:35 CST 2016
Seems to make sense to disable that error if AssumeFirstHopSupportOutbound
is true.
Scott
On Fri, Dec 2, 2016 at 2:05 PM, Daniel Pocock <daniel at pocock.pro> wrote:
>
>
> I tried the following setup:
>
> UA without RFC5626 support
> |
> Edge Proxy (repro), EnableFlowTokens = true, AssumePath = true
> |
> SBC (reConServer)
> |
> Registration proxy (repro), AssumeFirstHopSupportsOutbound = true
>
>
> The Edge proxy is correctly adding the Path headers
>
> The reConServer does REGISTER forwarding, it adds a Path header for itself
>
> The registration server sees the REGISTER requests and complains "Trying
> to use TLS with an IP-address in your Contact header won't work if you
> don't have a flow. Consider implementing outbound, or putting an FQDN in
> your contact header.". The message comes from dum/ServerRegistration[1]
>
> I temporarily disabled that code and observed that:
>
> a) the repro registration is able to read the Path headers and store them
>
> b) when I send an INVITE, it correctly constructs Route headers for
> reaching the UA and forwards the INVITE to the SBC, the call is successful
>
> Should AssumeFirstHopSupportsOutbound be more generous and help suppress
> this error?
>
> When the flow-token hack is enabled in the edge proxy, should it add
> ";ob" to the Path header, is there a way to make it do so for
> non-outbound clients?
>
> Or should the SBC do something to the Contact header to make repro
> accept it? I tried that too and it is an effective workaround, but I
> felt this was a bit ugly and prefer to use Path rather than modifying
> Contact.
>
> Regards,
>
> Daniel
>
>
> 1.
> https://github.com/resiprocate/resiprocate/blob/master/resip/dum/
> ServerRegistration.cxx#L535
> _______________________________________________
> resiprocate-devel mailing list
> resiprocate-devel at resiprocate.org
> https://list.resiprocate.org/mailman/listinfo/resiprocate-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.resiprocate.org/pipermail/resiprocate-devel/attachments/20161202/95c320e0/attachment.htm>
More information about the resiprocate-devel
mailing list