[c-nsp] High CPU Temperature on c7201

Peter Rathlev peter at rathlev.dk
Wed Jun 20 10:54:20 EDT 2012


On Wed, 2012-06-20 at 13:07 +0100, Steve McCrory wrote:
> We are having problems with one of our LNS devices dropping packets,
> albeit in small numbers, due to ignored errors. These are the diags we
> have captured: 
> 
> router#sh int g0/1 | i error
>      337 input errors, 0 CRC, 0 frame, 0 overrun, 337 ignored
>      2 output errors, 0 collisions, 1 interface resets
> 
> router#sh controllers g0/1 | i resource
>   rx_crc_error = 0, rx_too_big = 0, rx_resource_error = 337

https://supportforums.cisco.com/docs/DOC-2337

...
> My questions - Could the high CPU Die temperature be attributed to the
> ignored errors we are seeing?

I generally wouldn't think so, but maybe the CPU is throttled because of
the heat.

> What would normally cause high CPU temperature when the ambient
> temperature recored by the inlet and outlet appears normal?

If the thermal grease has degrade because of age you might see this.
Dust could be another reason; everything that could obstruct the airflow
inside the device. Take a look inside for anomalies.

> For the record, we have other devices of the same model (c7201)
> performing the same duty with the same IOS and handling a much larger
> load,

Same load in packets per second or in bits per second? The former is the
real bottleneck.

-- 
Peter




More information about the cisco-nsp mailing list