I am in the process or replacing three EOL 3800 voice gateways (H323) with 2911. I am having issues with one location and curious on feedback.
I replaced the router 2 weeks ago. Everything was fine until earlier this week. The circuit went hard down from the routers perspective, seeing remote alarm and the controller was down. Initially I thought it was a carrier issue so I initiated a ticket with them. They were able to test clear to their NIU (I know....we have all heard this before...) and tossed it back to me. I went on site and unplugged the cable, put in a loop back, plugged it back in and it came up. Hmm, I guess it just needed to be reset?
Well, As soon as I left, the circuit went down again. This time, I logged and verified the controller was down/remote alarm then did a shut/no shut on the controller and voice port and it came back up for 15 minutes or so. I limped by monitoring this for most of the day while I continued to troubleshoot.
Fortunately, I left the EOL 3800 configured and racked directly next to the 2911 so I connected with 3800 with the same cables and monitored for an hour with no issues. As soon as I move everything back to the 2911, the controller takes a dive after a few calls. Unable to take anymore outages, I moved back to the 3800 and all is well for 24hrs.
During the troubleshooting process, I saw a few bugs that could be somewhat similar to my issue so I upgraded which did not help. I opened a TAC case but the engineer advised that it is a line issue and using the 3800 is not a valid test to rule out hardware/software issues on the 2911. The configuration is identical on both. The cabling is the same. They are sitting right next to each other. He asked that I tweek the cablelenght settings on the controller, but I have already had enough outages that I cannot afford to try various lenght settings.
So....I am leaning towards pushing for an RMA on both the chassis and the hwic card. This is actually the second card I have used on this router. The first one was out of box failure, as soon as it was connected to the router, the controller would take hundreds of errors within seconds and would never establish a connection so was RMA'd under a different case. TAC is pushing back on the RMA.
Any advice here? Running c2900-universalk9-mz.SPA.154-3.M8, the same code I am running at a few other locations with no issues.
Thanks for any input
All seems to lead to a chassis issue but, is you can, I would keep a debug isdn q921 and stop the debug when the circuit goes down.
Please post it here the result.
I will suggest to first check the output of the command "show controller e1 <port>" and monitor which error (slips, error secs etc..) is increasing. The type of error count increasing can tell you what the actual issue is. Some of the errors can cause the interface to reset itself or go down. Please clear the counters "clear counter" before you start monitoring.
These kind of issues normally happens if there is issue with cable (hope you have tried to change cable), issue with DSP or interference.