This article describes HSL2 links and focuses on the following syslog message and how to address it:
fpc Cmerror Op Set: HSL2: HSL2 Fault MQCHIP(0)-chan-rx-62 <= IXCHIP(0)-chan-tx-3 - set
HSL2: The PFE’s ASICs communicate with each other and with the fabric chips at the physical layer via HSL2 (High Speed Link version 2) links.
When issues are detected in the HSL2 links between the PFE's ASICs, an alarm similar to the following will be active:
NEO> show chassis alarms 1 alarms currently active Alarm time Class Description 2020-05-13 15:42:50 PST Major FPC 0 Major Errors - HSL2 Error code: 0x200001
The HSL2 also allow communication between the PFE's ASICs and the fabric chips. An alarm similar to the following could also become active:
NEO> show chassis alarms 2 alarms currently active Alarm time Class Description 2020-05-13 11:06:30 PST Minor Check CB 1 Fabric Chip 1 2020-05-13 11:06:30 PST Major Fault CB 2 Fabric Chip 0
For both scenarios, the most common reason of failure is CRC errors. Cyclic redundancy check is one of the most popular methods of error detection for digital signals.
However, the errors could be increasing. In that case, if no FPC action has been configured for the major alarm, manual intervention is needed.
To verify if the errors are increasing, login to the PFE shell and obtain the HSL2 statistics of the FPC (the output below was shortened):
NEO> start shell pfe network fpc0 NPC0(NEO)# show hsl2 statistics Cell Received (last) CRC Errors (last) ------------------------------------------------------ MQCHIP(0) channel statistics : MQCHIP(0)-chan-rx-0 <= QXCHIP(0)-chan-tx-1 99520353703489 (13939454) 0 (0) MQCHIP(0)-chan-rx-2 <= LUCHIP(0)-chan-tx-2 758842696983198 (106278017) 0 (0) MQCHIP(0)-chan-rx-3 <= LUCHIP(0)-chan-tx-3 758842696983198 (106278017) 0 (0) MQCHIP(0)-chan-rx-7 <= remote/unknown 6776073636 (1876871644) 0 (0) MQCHIP(0)-chan-rx-8 <= remote/unknown 6776073636 (1876871644) 0 (0) MQCHIP(0)-chan-rx-14 <= remote/unknown 6938103514 (1918669588) 0 (0) MQCHIP(0)-chan-rx-15 <= remote/unknown 6938103514 (1918669588) 0 (0) MQCHIP(0)-chan-rx-21 <= remote/unknown 6938118207 (1918669836) 0 (0) MQCHIP(0)-chan-rx-22 <= remote/unknown 6938118207 (1918669836) 0 (0) MQCHIP(0)-chan-rx-28 <= remote/unknown 6776078710 (1876871932) 0 (0) MQCHIP(0)-chan-rx-29 <= remote/unknown 6776078710 (1876871932) 0 (0) MQCHIP(0)-chan-rx-62 <= IXCHIP(0)-chan-tx-3 283631445184166 (39743389) 2 (0)
To clear the errors, restart the affected PFE#0, which is in control of PIC#0 and PIC #1.
> show chassis hardware Hardware inventory: Item Description Chassis MX480 CB 1 Enhanced MX SCB FPC 1 MPCE Type 2 3D EQ CPU MPCE PMB 2G MIC 0 3D 20x 1GE(LAN) SFP PIC 0 10x 1GE(LAN) SFP Xcvr 0 SFP-T Xcvr 1 SFP-T PIC 1 10x 1GE(LAN) SFP Xcvr 8 SFP-SX Xcvr 9 SFP-SX MIC 1 3D 4x 10GE XFP PIC 2 2x 10GE XFP PIC 3 2x 10GE XFP > request chassis mic mic-slot 0 fpc-slot 1 offline fpc 1 mic 0 offline initiated, use "show chassis fpc pic-status 1" to verify
Refer to the technical documentation for details on r equest chassis mic
The errors should stop. If they continue, contact your JTAC Representative to troubleshoot further.
If the alarm is related to a Fabric chip, refer to the following KB article to troubleshoot further:
KB23173 - Resolution Guide - MX - Troubleshoot Fabric Planes in Fault or Check State [juniper.net]