ITfromZX81
Programmer
Hi everyone,
Here's a fun problem:
We have one user at a remote site - he is setup with a 4602 IP Phone which connects directly to a router. He does not have an IPO Appliance in his office, just one phone.
He is then treated as an extension on the main office but with a different prefix so that we can redirect the call to his location via an IP subnet.
The problem is that his phone stops working after about an hour. His computer works fine and is hooked up to the same router. If he unplugs and plugs his phone back in it works again for about an hour.
I created a duplicate setup in my test lab and I get the same problem.
We use a Watchguard V60L router in our main office, and this contains a tcp_idle_timout setting to stop a connection if there is no traffic, which I had set to 3600 seconds, as recommended on their site(the default 3 minute setting tended to timeout all kinds of things). Watchguard, however, suggests that the keepalive messages between tcp peers instead of forcing it, no doubt to lower the open connections.
Is there any way to tell an IP 4602 to do this? I don't really like the idea of increasing the tcp_idle_timeout to a huge number, and this would still be a stopgap, not a solution.
Thanks,
ITfromZX81
Here's a fun problem:
We have one user at a remote site - he is setup with a 4602 IP Phone which connects directly to a router. He does not have an IPO Appliance in his office, just one phone.
He is then treated as an extension on the main office but with a different prefix so that we can redirect the call to his location via an IP subnet.
The problem is that his phone stops working after about an hour. His computer works fine and is hooked up to the same router. If he unplugs and plugs his phone back in it works again for about an hour.
I created a duplicate setup in my test lab and I get the same problem.
We use a Watchguard V60L router in our main office, and this contains a tcp_idle_timout setting to stop a connection if there is no traffic, which I had set to 3600 seconds, as recommended on their site(the default 3 minute setting tended to timeout all kinds of things). Watchguard, however, suggests that the keepalive messages between tcp peers instead of forcing it, no doubt to lower the open connections.
Is there any way to tell an IP 4602 to do this? I don't really like the idea of increasing the tcp_idle_timeout to a huge number, and this would still be a stopgap, not a solution.
Thanks,
ITfromZX81