Resolve the Port from the Err-Disable Status on the Cisco 8000 Router

Introduction

This document describes how to recover an interface from the status of err-disable on the Cisco 8000 router with Cisco IOS® XR 7.3.15 and later.

Prerequisites 

Requirements

There are no specific requirements for this document.

Components Used

The information in this document is based on these software and hardware versions:

• Cisco 8000 Router
• Cisco IOS® XR release 7.3.15 - 7.8.1

The information in this document was created from the devices in a specific lab environment. All of the devices used in this document started with a cleared (default) configuration. If your network is live, ensure that you understand the potential impact of any command.

Problem

Once an interface is put into the status of "err-disable" due to the physical link error, like CRC (Cyclic Redundancy Check), it cannot be recovered even after the physical layer error is cleared.

This behavior is due to the NPU (Network Processing Unit) performance enhancement introduced by Cisco bug ID CSCwb76264 and Cisco bug ID CSCwc32333 .

An example of this problem on Cisco 8000 router is:

RP/0/RP0/CPU0:8000#show interface des loc 0/4/cpu0 Interface Status Protocol Description -------------------------------------------------------------------------------- ... ... Te0/4/0/35/0 err-disable admin-down ... ...

The syslog messages for it are similar like:

LC/0/4/CPU0:2022 Dec 7 08:38:50.059 UTC: npu_drvr[130]: %FABRIC-NPU_DRVR-3-PLUGGABLE_MODULE_ERROR : [6879] : NPU ASIC (2) Port (35) Interface TenGigE0_4_0_35_0 is disabled due to bad link. Please replace optics/cables and shut/unshut the port. LC/0/4/CPU0:2022 Dec 7 08:38:50.059 UTC: npu_drvr[130]: %PKT_INFRA-ERRDIS-6-ERROR_DISABLE : TenGigE0/4/0/35/0: Error disabling due to reason: bad-link-error LC/0/4/CPU0:2022 Dec 7 08:38:50.060 UTC: ifmgr[261]: %PKT_INFRA-LINK-3-UPDOWN : Interface TenGigE0/4/0/35/0, changed state to Error Disabled

Solution

The steps in order to recover the interface from the err-disable status in such situation are:

1. Fix the source of the error which can be fiber, optics, and layer 1 transport.
2. Clear NPU counters by running command: clear controller npu stats asic-counters instance all location 0/x/CPU0.
3. Manually bounce the interface by running command: shutdown and commit and then no shutdown and commit.