High CPU due to ECC MSG on 1941

[Shadow-82]

Hi!

Recently I get error on one of my routers:

%PLATFORM-5-ECC_MSG: A corrected single bit error has occurred in L2C Data Cache at location 0x86

Router: Cisco CISCO1941/K9

Image: c1900-universalk9-mz.SPA.152-4.M4.bin

While it happens, CPU skyrockets to 100% utilization & overrun error occurs on physical interfaces

I've read some articles about this kind of error with suggestions:

Software bug - do an upgrade. But it was for VG310/320 running 15.4(3)M1

Hardware problem - 1. reseat DIMM, 2. replace DIMM, 3. RMA whole chassis

Any suggestions for this case?

Thanks in advance

[Palani Mohan · Accepted solution]

Hi there

This message is informational in nature. It is telling you that ECC (Error Correcting Code) is doing what it is supposed to do. Since such a correction would require CPU resources, momentary increase in CPU use is to be expected. Such spike in CPU usage is least likely to cause any service interruption.

If you see this message happening frequently OR if the referenced address location is changing, then consider replacing the memory module. The documentation for handling (removal/insertion) memory module is here

Assuming you have the base config, the part to order/replace would be MEM-1900-512MB.

Kind regards ... Palani