Description

You might see the following error on the logs "XMCHIP_CMERROR_HSL2_LU0_RTT_PAR_PROTECT_FSET_REG_DETECTED_IBUF_MEM_E" for the FPC and all interfaces down for the reported card.

Symptoms

  • An alarm will be active on the device:

example@router1-re0> show system alarms 

1 alarms currently active

Alarm time        Class Description

2024-01-03 16:10:18 PST Major FPC 0 Major Errors

 

  • The FPC and pics show online

example@router1-re0> show chassis fpc pic-status 

Slot 0  Online    MPC3E NG PQ & Flex Q              

 PIC 0 Online    10x 1GE(LAN) -E SFP

 PIC 1 Online    10x 1GE(LAN) -E SFP

 PIC 2 Online    10X10GE SFPP

Slot 1  Online    MPC3E NG PQ & Flex Q              

 PIC 0 Online    10x 1GE(LAN) -E SFP

 PIC 1 Online    10x 1GE(LAN) -E SFP

 PIC 2 Online    10X10GE SFPP

Slot 5  Online    MPC5E 3D 2CGE+4XGE               

 PIC 0 Online    2X10GE SFPP OTN

 PIC 1 Online    1X100GE CFP2 OTN

 PIC 2 Online    2X10GE SFPP OTN

 PIC 3 Online    1X100GE CFP2 OTN

 

 

Solution

Parity error: Within a computer system, electrical or magnetic interference from internal or external causes can cause a single bit of memory to flip to the opposite state spontaneously. This event makes the original data bits invalid and is known as a parity error.

This message is due to a transient hardware memory parity error in the LU block on the XMCHIP ASIC. A transient hardware issue causes this error. A single message occurrence in syslog, not seen in combination with other events, indicates a one-time transient hardware problem. However, multiple continuous occurrences indicate persistent underlying issues.
 
  • Check the fabric summary:

example@router1-re0> show chassis fabric summary 

Plane  State  Uptime

 0   Online  189 days, 22 hours, 55 minutes, 32 seconds

 1   Online  189 days, 22 hours, 55 minutes, 32 seconds

 2   Online  189 days, 22 hours, 55 minutes, 32 seconds

 3   Online  189 days, 22 hours, 55 minutes, 32 seconds

 4   Spare  189 days, 22 hours, 55 minutes, 31 seconds

 5   Spare  189 days, 22 hours, 55 minutes, 28 seconds

 6   Spare  189 days, 22 hours, 55 minutes, 31 seconds

 7   Spare  189 days, 22 hours, 55 minutes, 25 seconds

 

  • Check the chassis fabric plane
  • All plane fabrics related to the affected FPC will be disabled:

example@router1-re0> show chassis fabric plane      

Fabric management PLANE state

Plane 0

 Plane state: ACTIVE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 1

 Plane state: ACTIVE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 2

 Plane state: ACTIVE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 3

 Plane state: ACTIVE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 4

 Plane state: SPARE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 5

 Plane state: SPARE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok        

   FPC 5

     PFE 0 :Unused

     PFE 1 :Unused

Plane 6

 Plane state: SPARE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Links ok

     PFE 1 :Links ok

Plane 7

 Plane state: SPARE

   FPC 0

     PFE 0 :Fabric Disabled

   FPC 1

     PFE 0 :Links ok

   FPC 5

     PFE 0 :Unused

     PFE 1 :Unused

  

  • Check the hsl2 statistics and look for CRC errors:

example@router1-re0> request pfe execute command "show hsl2 statistics" target <fpc_x>

 

  • Wait 5-10 seconds and run the command again:

example@router1-re0> request pfe execute command "show hsl2 statistics" target <fpc_x>

 

  • If no CRC errors are seen on the hsl2 statistics then proceed to reboot the affected FPC

example@router1-re0> request chassis fpc slot x offline

example@router1-re0> request chassis fpc slot x online

 

  • The FPC should come back online and alarms should be cleared out
  • The card should remain under monitoring for a few days
  • If no similar error is seen again the issue can be considered a transient hardware error
  • No RMA is required with a single occurrence.

 

Modification History

2024-01-03 : Article Created