Fallback Timer Tweaking

[admin_2]

I have a remote site with about 30 7962's on the other side of a T1. At least once a day about half of them go into fallback and register with SRST.

the T1 never takes a hit, and I assume they are just hitting some high latencies and its timing out.

What are the best values for the keepalives?

[rodmont74]

Hi, how much is the latency you see in the network? Before changing keepalive values, I'd propose you monitor if the phones lose network connectivity to the cluster servers.

Maybe a recurring ping during business hour to a single phone or two just to make sure they don't have dropped packets somewhere.

Regards

[Andrew Short]

Did you ever find an answer for this one? I am having the same problems with a site and a set of 7971 phones. The 7941s stay up and connected to their Primary and backup CUCMs, but the 7971s hit SRST.

My 4507 is very near the peak of it's power capacity, and I am wondering if that has anything to do with the weirdness since they take the most power per phone in our mix.