We have a customer reporting poor quality calls when they come to us via a particular provider. The SIP traces look perfectly normal both on the ingress to us and egress to another telco. No additional sip messages after the call has been answered until the BYE is received. However in the asterisk logs I am getting this :-

2014-02-05 13:45:03 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:04 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:04 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:04 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:05 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:05 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:05 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

2014-02-05 13:45:06 | C-00108c80] rtp_engine.c: — Locally bridging SIP/vmpubopensips1-001dfd43 and SIP/magrathea2-001dfd44

Any idea what could be causing this?
I am running asterisk 11.2-cert2.

I am going to get call redirected via our test box and turn on full verbosity in the logs and capture a full tcpdump but any ideas would be welcome.

Thanks Gareth