Core Issue
Cisco processors have timers that guard against certain types of system hangs. The CPU periodically resets a watchdog timer. The watchdog timer basically controls the time of each process. If the timer is not reset, a trap occurs. If a process is longer than it should be, the watchdog timer is used to escape from this process.
There are two main types of watchdog timeouts. The first type is usually caused by a software problem and is reported in one or both of these ways:
The second type of watchdog timeout is usually due to a hardware problem and is reported in one or both of these two ways:
Resolution
Once you have identified a watchdog timeout message, refer to the Field Notices for your platform and all the components installed in that system for known critical hardware issues.
If a new module has been recently installed, first try to remove it to determine whether it is causing the watchdog timeout. If the watchdog timeout messages persists, try to re-seat all removable components. If the watchdog timeout continues, at this point there are no Field Notices for your hardware. If no new module has been recently installed, replace the main processor board.
On high-end platforms, the processor board is a separate card, such as the NPE-400 or RSP8. On low-end platforms (such as the Cisco 1700, 2500, 4000, 2600, or 3600 series), the motherboard cannot be shipped separately. In this case, you have to replace the chassis itself.
For more information on Watchdog Timeout crashes, refer to Troubleshooting Watchdog Timeouts.