On Wed, 2008-10-22 at 16:29 -0600, Farr, Steven C. wrote: > What we've seen when doing the two PRI gateways as two separate > unmanaged gateways is that if the first one in the list goes down, often > the first couple of calls attempted don't work but it does seem to > figure out what to do after that. Does sipX mark the gateway as "dead" > and ignore it for a while? Or does it still try it for each and every > outbound call? I guess most of my failover testing has been with failing > the PSTN, as opposed to taking the actual gateway offline, which is to > say that the AudioCodes gateway was still there to say "nope, sorry the > PRI is down" as opposed to getting no response at all...
The answer depends on exactly how sipX is configured to pair the two gateways. I suspect that in your case, the call is forked to each gateway. In that case, if an attempt to use the first gateway returns a failure response, sipX will try the second gateway. Most likely, sipX is trying the first gateway every time, but the gateway responds so quickly the caller doesn't notice a delay. Dale _______________________________________________ sipx-users mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-users Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-users
