Frame-Relay Problems!!! Need advice.

[65fastback]

Our company utilizes a frame-relay network. At our main office, we have a 3600 router for frame-relay communications between us and remote offices. It is configured with several subinterfaces. The link between us and one remote office went down today. The remote office router interface is line is up, protocol is down. However, on our router subinterface corresponding to the remote office, the router indicates that line is down, protocol is down. The service provider tested the line and says that it looks to be our issue. Nothing has been changed on our router, and all other subinterfaces are functional. Could their be a clock rate issue at the frame-relay switch? Any typical causes for this problem? Thanks in advance for any information!

 

[BuckWeet]

If you haven't changed your config at all, and made sure that the remote router is okay, then its a carrier issue, your 3600 is obviously still working with other remote sites... I'd still bet my money on carrier problems, they always will feed you that line of crap, saying its your problem. But in the end its really them...



BuckWeet

 

[marsd]

They probably fscked up the config on their end.
I had a provider mess up the lmi and that killed
the circuit.
Try emailing them the output of sh frame-relay lmi
and sh frame-relay map to demonstrate your issue.
Also, is inverse arp in use?
Inverse arp can get messy under certain conditions.

 

[65fastback]

Thanks for the information! I'm also still leaning toward it being a carrier issue. I will post again with more information if the issue can't be resolved, or if we discover other issues.

Again, I really appreciate the help. I have my CCNA, but I'm not really that experienced dealing with frame relay issues.

 

[KSM]

In subinterface of main router, if pvc related to that subinterface went down, then both line and protocol were down.
Reboot the remote router, then look up sh frame-relay lmi and sh frame-relay pvc. If you see lmi packet received and pvc status is inactive, that means carrier problem.
Ask them to test the line and loopback the smartjack( D-mark).
Carrier always ask you to prove your equipment is ok before they escalate the problem to next level

 

[65fastback]

Thanks again for all the information.

The circuit mysteriously came back up sometime last night. I definitely think it was probably a carrier issue. It would be nice if they could just tell you these things up front...but I guess that is the nature of the game.

 

[stoney66]

Also nextime if it happens again, get the tech on the phone, and setup a loop on your interface (done from the interface config), that way they can look for your loop. If the encaps is hdlc, you will see your local interface go up/up, frame-relay though you wont get up/up but you still should get input packets.

stoney.

 

[jeter]

It had to be a carrier issue on the remote site, because the host and the other sub interfaces were up and up. It is possible the the LMI or the number of channels could have been changed. ANS (Advance network services) have a habit of making a change because they think something is wrong.

What is the lmi type and the number of channels on your circuit? What is the DCLI? The reason I ask is because again if it is set to say Cisco REV1 and not ansi annex D the mysteriously change.
“Reserve your right to think, for even to think wrongly is better than not to think at all”

Fisher CCNA,(CCNP-Routing)
UOP Student BSIT