Cisco: Disable Single Bit ECC errors from being logged in Syslog
Summary: Disable Single Bit Error-Correcting Code (ECC) errors from being logged in Syslog.
Symptoms
Multilayer Director Switch (MDS) 9700 with a DS-X9448-768K9 or DS-X9648-1536K9 line card that had a correctable ECC error.
Or
MDS 9396S has a correctable ECC error on module 1.
Several ECC 1-bit errors were seen on switch in logging in 9700 or 9396 series switches: 2017 Aug 3 14:13:59 the_switch %MODULE-4-MOD_WARNING: Module 1 (Serial number: JAE19280NLX) reported warning fc1/73-80due to single-bit ECC error in device DEV_F16_CMN (device error 0xccc09600)2017 Aug 3 14:13:59 the_switch %CALLHOME-2-EVENT: MODULE_WARNING2017 Aug 5 21:02:12 the_switch %MODULE-4-MOD_WARNING: Module 1 (Serial number: JAE19280NLX) reported warning fc1/57-64due to single-bit ECC error in device DEV_F16_CMN (device error 0xccc07600)2017 Aug 5 21:02:12 NVMB-DC2-S03-ED-DCN-FAB2-SW010 %CALLHOME-2-EVENT: MODULE_WARNING
Exceptions reported in exception logs:exception information --- exception instance 5 ----Module Slot Number: 1Device Id : 204Device Name : F16 Generic DriverDevice Errorcode : 0xccc07600Device ID : 204 (0xcc)Device Instance : 07 (0x07)Dev Type (HW/SW) : 06 (0x06)ErrNum (devInfo) : 00 (0x00)System Errorcode : 0x42b80022 single-bit ECC errorError Type : WarningPhyPortLayer : Fibre ChannelPort(s) Affected : Error Description : F16_MEM1_TM_SAT0_ECC_1BIT_ERR1DSAP : 0 (0x0)UUID : 0 (0x0)Time : Mon Aug 5 21:02:12 2018 (Ticks: 5A78870C jiffies) exception information --- exception instance 6 ----Module Slot Number: 1Device Id : 204Device Name : F16 Generic DriverDevice Errorcode : 0xccc09600Device ID : 204 (0xcc)Device Instance : 09 (0x09)Dev Type (HW/SW) : 06 (0x06)ErrNum (devInfo) : 00 (0x00)System Errorcode : 0x42b80022 single-bit ECC errorError Type : WarningPhyPortLayer : Fibre ChannelPort(s) Affected : noneError Description : F16_MEM1_TM_SAT0_ECC_1BIT_ERR1DSAP : 0 (0x0)UUID : 0 (0x0)Time : Sat Aug 3 14:13:59 2018 (Ticks: 5A75926F jiffies)
Cause
Resolution
Workaround 1
Responses are informational in nature, and the message can be ignored unless there are many repeating errors (hundreds) against the same line card within minutes.
If there are many repeating ECC errors on the same line card, this should be investigated as a hardware concern. If needed, to validate this case, reload the module or the switch and replace the module or switch if the errors continue.
Additional information:
Single-bit ECC errors are not uncommon and are correctable.
Workaround 2
Upgrade to 6.2.21 or later.