I need some help with what would be the best setup for CCM 6 centralized deployment and having multiple t1's to each remote site. HQ has a 7206VXR with T3-PA adapter and 2 channels (2 t1's) setup to each of 10 remote site. We are now planning on running VOIP across the network and am wondering the best method for setting up the t1's.
Should I use eigrp equal cost load balancing (either per-packet or per destianation) or should I setup MLP (multilink PPP) using fragmentation and interleaving?
Can i apply auto qos to the MLP interface? or do i need to create an MPF policy like before auto qos was around?
Any suggestions as to the best approach would be appreciated.
If you are having frame relay connection then MLP would be a better option.MLP provides a method of splitting, recombining, and sequencing datagrams across multiple logical data links. MLP allows packets to be fragmented and the fragments to be sent at the same time over multiple point-to-point links to the same remote address.
The document you suggest only points out how to configure MLP for RTP but doesn't say this is a preferred method over eigrp equal cost load balancing.
What I'm asking is what is the preference for VOICE. The t1's in question will be straight point to point hdlc links. I can configure them with any enacap. i want.
I can't find a definitive cisco design recommendation for multiple t1's with voice. i can configure eigrp for equal cost load balancing using either per-packet round robin or per destination or use MLP.
Does cisco have a design recommendation for my situation?
I personally would configure multilink PPP. This will save you headache when configuring QoS for 10 multilinks instead of 20 serial interfaces.
Are you running any QoS on the network today? My recomendation would be to implement a simple 3 class system protecting your UDP audio/realtime voice, signaling and control protocols. Without any other changes, this will utilize LLQ/priority queue for real-time traffic and keep the data portion of the network roughly the same as it was. Cisco IP phones and devices will correctly mark DSCP, so typically it's just a matter of trusting uplinks and phones on your switchports.
I'd also take a look at call admission control in CUCM. By configuring the location bandwidth for each site to match the BW in your priority queue, you can keep from oversubscribing the link. The LLQ is unique in that if oversubscription does occur, packets will be dropped not queued which of course impacts all active calls.
IntroductionCUCM Routing RulesDial String implementation PolicyCUCM Routing LogicSIP URI Call Routing Analysis+++ Case Study: 1 ++++++ Case Study: 2 +++Conclusion
Over the last few months, I have had the privilege of working on SI...
Are you getting this error “Installer User Interface Mode Not Supported. The installer cannot run in this UI mode. To specify the interface mode, use the -i command-line option, followed by the UI mode identifier. The value UI mode identifiers...