MGX 8850 re-booting

[rajak]

Hi Everyone,

We have several 8850 with 2nos (Pri and sec) PXM and 2nos (Pri and Sec) SRM, the rest of the service module consiting of equal number of RPM and FRSM card, 13 each. We frequently come accross re-booting of this system, we primarily use this system for aggregating voice circuit in our E1 from remote location.

The first question,is, this fully loaded MGX 8850 how many mins (approx) it would take to be ready for the voice calls?

What would be cause of re-boot?

Should we consider something in our design?

Though we have quite a few question on this, we would highly value your inputs, comments on this.

Thanks in advance,

Raja

[a-vazquez]

You can try disabling Cisco Express Forwarding CEF, if you had enabled it expliclitly. I heard there were some problems using CEF with MGX devices. As you are mentioning about constant reboots, it has more probability of hardware problem. We had a similar problem and got our devices replaced.

[rajak]

Thanks Vazquez, we will try disabling the CEF and let you know of the results

[smoosa]

Raja,

Please describe "rebooting" . Unlike some routers, the MGX's should gracefull switchover from one PXM to the other.

Now for every switchover, the logs or corefiles that are logged will provide some clue as to what the switchover reason was.

What release are you running on the PXM ?

what does "dsperrhist 7"

dsperrhist 8

show you ?

Thanks

Shajith

[rajak]

Thanks for your reponse Shajith,

Unfortunately, we have changed the PXM cards, when we come across the re-booting of this MGX we will make sure would take the error logs and the core dump.

By the way, be-booting I mean, even after having the standby PXM in place, we dont seem to have the switching happening on a failure, or as mentioned in the datasheet the standby doesnt take over in 400 msecs, the node goes down for about 10 to 12 mins ... which is something we coudnt justify at time...