Is it possible to just simply add another T1/PRI module to Network Slot 2 to bring in another PRI from a different provider for redundancy? We currently have one T1/PRI module in network slot 1, a quad T1/PRI module in option slot 1 (ports 1 and 2 are connected with crossover), and three Octal FXS modules in slots 2,3,4. We are setup in a rabbit-hole configuration. FXS ports 1-15 are in a circular "hunt group" and FXS ports 16-23 map to a specific number. Almost all traffic is incoming. I planned on adding the phone numbers from the new provider to the In#Accept List under the User Term Dial Plan. DTO failover would be responsible for forwarding the lines from one provider phone number to the other. Would this work or would there be problems or conflicts? If this would work am I missing any other setup or configuration?
It should work. It's not uncommon to have multiple incoming trunks capable of feeding the same destination numbers.
The tricky part for inbound is to get the rest of the world to reroute to the second carrier in case of failure of the primary. Depending on the nature of the carrier failure DTO may not always work as planned and SS7 can just blackhole your traffic at the primary carrier. It's generally reliable and the best practical option.
Out of curiosity, what's the reason for tying two T1 slots together with a crossover?
I believe it's required for the "rabbit-hole" configuration we have. In the User Term we direct calls to certain DSO's based on DNIS on T1 Port 1. In the Dedicated Map FXS ports are assigned to those DSO's on T1 Port 2. The crossover connects them. It allows a group of analog lines individually connected to alarm receiver line cards to be available as a group without tying one number to one line card. Maybe there's a way around the crossover in the configuration. I'm definitely not an Adtran expert.
Can both of the Network Slot T1 cards be active at the same time, or does network slot 2 just backup to network slot 1 on failure of slot 1? My goal is to bring in a different carrier on each network slot but have them output to the same FXS ports on the User Term side. It looks like there isn't much additional setup to adding a 2nd network Slot T1 card. I have the new T1 card added in the Network Term Dial Plan. I don't see anywhere you have to map the network term to the user term, it just must be assumed that everything on either network slot is passed to the User Term side. I believe I just need to define the new provider's numbers on the Incoming Number Accept List for the User Term Dial Plan under the corresponding interfaces. Does this sound correct to you?
Absolutely they can both be active. The dial plan routes on destination, so calls arriving inbound from either carrier will go to the matching digit pattern on the user side. You may want to assign different source IDs to the incoming carriers and set the user to accept from either. Then set both carriers to accept only from the source ID of the user side. This prevents the possibility of looping calls inbound from one carrier out to the other.
Thanks for the advice. I wouldn't have thought about the loop scenario. This worked well once I realized the system was looking at the source ID from my FXS port interfaces for outgoing calls. I left my user side interfaces at a source ID of 0. I changed network 1 interface to a source ID of 1 and network 2 interface to a source Id of 2. It works well logically as I can assign inbound numbers from network 1 with a source ID of 1 and inbound numbers from network 2 with a source ID of 2. On the network term I limited outgoing accepted numbers to anything with a source ID of 0 so only a call originating from the user side should be sent outgoing. On the user term I defined the inbound numbers under In#Accept twice. Once with source ID 1 and once with source ID 2 so the numbers will work if forwarded to the other T1/PRI circuit in a DTO failover situation.