| One of the first things you learned about Frame | | | | What you have to keep in mind about FREEK is |
| is that the LMI also serves as a keepalive, or a | | | | that each and every PVC needs two separate |
| heartbeat - and if three consecutive LMIs are | | | | keepalive processes. Remember, with a PVC, |
| missed, the line protocol goes down. There's a | | | | there's no guarantee that the path taking through |
| limitation to LMI as a keepalive, though. The LMI is | | | | the frame relay cloud to get from R1 to R2 is |
| exchanged only between the DTE and the closest | | | | going to be the same path taken to go back |
| DCE. The LMI is therefore a local keepalive that | | | | from R2 to R1. One process will be used to send |
| does not reflect any possible issues on the | | | | requests for information and handle the responses |
| remote end of the virtual circuit. | | | | to these requests; this is the send side. When the |
| Taking the LMI concept to the next logical level, | | | | send side transmits a keepalive request, a |
| Frame Relay End-To-End Keepalives (FREEK, one | | | | response is expected in a certain number of |
| of the least-heard Cisco acronyms for some | | | | seconds. If one is not received, an error event is |
| reason) are used to verify that | | | | noted. If enough error events are recorded, the |
| endpoint-to-endpoint communications are | | | | VC's keepalive status is marked as down. |
| functioning properly. | | | | |