We have the following scenario.
30 7940s, running P00303010413
3660 gateway with a single PRI, running 12.2.8T5
2 system Call Manager cluster, running 3.1.3(a)spF
Our problem is the following:
When the WAN experiences an outage the current calls are maintained and all expected featurs work like DID etc..., the phones say that they are in fallback mode and everything is good. However, when the ATM network resumes service, the phone re-register with Call Manager and active calls are dropped. We did the testing several months ago and everything was working fine. We recently updated the phone load, as some users were reporting echo between IP phones. Either coincidentally or for some other reason our testing proves the fallback to the cluster does not work.