[OpenSIPS-Users] DNS failover on 6xx reply

[Bogdan-Andrei Iancu]

Salut Ion,

based on what RFC 3263 and RFC 3261say, the DNS failover is trigger (at 
signalling level) only by local 408 (local timeout) and received 503.

The 603 does not trigger DNS failover (at least according to the specs).

Regards,
Bogdan

Ion Constantinescu wrote:
> Hello,
>
> Help / advice on the following issue would be much appreciated.
>
> We are running a setup where a front-end opensips box is relaying calls 
> to two or more end-client HMP machines (e.g., asterisk).
>
> We are using DNS/SRV records to failover between the end-client machines 
> - which seems to be working fine when the actual HMP service is 
> completely stopped.
>
> In some cases tough, the HMP service is running in a particular state 
> where it is returning 603 replies which in turn seem to be breaking the 
> DNS failover protocol.
>
> Note: in a separate setup we are successfully running a 
> transaction/dispatcher based failover configuration for which we needed 
> the following:
> modparam("tm", "disable_6xx_block", 1)
>
> So the question is - is there any way to deal with 6xx response codes 
> such that the DNS failover mechanism works (perhaps by using a setting 
> similar to the one above) ?
>
> Thank you,
> Ion
>
>
> _______________________________________________
> Users mailing list
> Users at lists.opensips.org
> http://lists.opensips.org/cgi-bin/mailman/listinfo/users
>
>