Router Hardware Management Guide for Cisco 8000 Series Routers, Cisco IOS XR Releases

PDF

Router Hardware Management Guide for Cisco 8000 Series Routers, Cisco IOS XR Releases

Fabric link management for uncorrectable errors

Updated: July 1, 2026

Want to summarize with AI?

Describes how the router uses Forward Error Correction (FEC) to detect uncorrectable errors on fabric links, retune noisy links, and permanently shuts down a link after repeated failures.

Forward Error Correction (FEC) is a error control method in data transmission that

allows the transmitter to send redundant data so the receiver can recognize only the portion of the data that contains no apparent errors, and
enables the receiver to detect and correct a limited number of errors.

The Cisco IOS XR router will not bring the link to the data plane if the link is noisy at inception or during bring up. If the link becomes noisy after it is brought up, the fabric link will be reset and re-tuned. If this event occurs five times within an hour, the fabric link shuts down permanently. After the link is up, the polling interval for link error is 10 minutes.

Fabric link management feature uses FEC as the criteria to determine if a link is good. The router receives a notification for every bad FEC on each fabric port. FEC can correct up to 15 bit errors; beyond this threshold, the error is considered uncorrectable. This feature allows fabric links to operate without errors.

In Cisco IOS XR Release 24.2.11, the FEC feature is enabled only on Q200-based line cards and fabric cards.

Table 1. Feature history table
Feature Name	Release Information	Feature Description
Fabric Link Management for Uncorrectable Errors	Release 25.4.1	Introduced in this release on: Fixed Systems (8010 [ASIC: A100])(select variants only) This feature is supported on: 8011-32Y8L2H2FH 8011-12G12X4Y-A/D
Fabric Link Management for Uncorrectable Errors	Release 25.1.1	Introduced in this release on: Fixed Systems (8700 [ASIC: K100], 8010 [ASIC: A100])(select variants only) This feature is supported on: 8011-4G24Y4H-I 8712-MOD-M
Fabric Link Management for Uncorrectable Errors	Release 24.4.1	Introduced in this release on: Fixed Systems (8200 [ASIC: P100], 8700 [ASIC: P100])(select variants only); Modular Systems (8800 [LC ASIC: P100])(select variants only) *This feature is supported on: 8212-48FH-M 8711-32FH-M 88-LC1-36EH 88-LC1-12TH24FH-E 88-LC1-52Y8H-EM
Fabric Link Management for Uncorrectable Errors	Release 24.2.11	You can now run your fabric links error-free using the forward error correction (FEC) technique. The feature allows you to determine the link quality by monitoring the noisy fabric links during and post bring-up. This feature introduces the hw-module fabric-fec-monitor disable command.

FEC bin index

The FEC bin index indicates the number of bit errors.

For example, FEC bin index of a good link is {12642341946,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,} and for a bad link it is {221568554,532162,414888,101277,144171,3317,61681,4065,741,1977586,138,62,17,6,1,}.

If the FEC bin index is equal to or greater than ten non-zero bits (bin ≥ 10 non-zero) before the link is up, the FEC will indicate that the link is bad. When the link is up and the FEC bin index is equal to or greater than 13 non-zero bits (bin ≥ 13 non-zero), the FEC will indicate that the link is bad.

Link states for a noisy fabric link

When a fabric link is noisy, any of these link states is possible:

Link does not come up at all.
Link comes up, but fluctuates.
Link comes up, but generates uncorrectable errors.

The network traffic flow is not affected if the link never comes up; however, packet drops may occur in the other two link states.

How the router monitors FEC fabric links

Summary

The key components involved in the fabric link monitoring process are:

Router: A network device that monitors fabric links by processing FEC reports, performing retuning, logging system messages, and managing link states based on failure frequency.
Fabric link: A physical connection that undergoes quality monitoring, retuning, and snapshot collection to ensure stable data plane operation, with a permanent shutdown occurring after the sixth failure within one hour.
MAC port: A network interface that stops and reactivates between the second and fifth link failure to facilitate recovery.

Workflow

For every FEC report, the router performs this process to monitor the fabric links:

If the fabric link is faulty before the link is up, the router retunes and checks again for the FEC improvement.
If the link quality does not improve after retuning, the router displays the syslog message after tuning for 100 times and does not bring the link to the data plane.
Post link up, when the fabric link becomes noisy, the router collects a snapshot and retunes after the first failure.
From second failure to fifth failure, the MAC port stops and re-activates. Retune is done as part of this process.
If the fabric link fails for the sixth time within an hour, the router permanently shuts down the link.

FEC fabric link system log messages

This topic helps you to identify the syslog messages the router displays after retuning a FEC fabric link.

The router displays these syslog messages after retuning:

If the link..	then the router displays this syslog message
is noisy at inception or during bring up	`LC/0/2/CPU0:Jan 13 00:56:03.939 UTC: npu_drvr[128]: %FABRIC-NPU_DRVR-3-NPU_CPA_GEN_ERR_INFO : Link 0/254 has tuned 100 times and failed to come up. FEC bin is filled to 11` The router displays this syslog message after tuning for 100 times
is noisy post bring up, the router permanently shuts down the link	`LC/0/2/CPU0:Jan 13 00:20:16.251 UTC: npu_drvr[128]: %FABRIC-NPU_DRVR-3-NPU_CPA_GEN_ERR_INFO : FEC check failures on link 0/254. FEC bin is filled to 14`

If the link..

then the router displays this syslog message

is noisy at inception or during bring up

LC/0/2/CPU0:Jan 13 00:56:03.939 UTC: npu_drvr[128]: %FABRIC-NPU_DRVR-3-NPU_CPA_GEN_ERR_INFO : 
Link 0/254 has tuned 100 times and failed to come up. FEC bin is filled to 11

The router displays this syslog message after tuning for 100 times

is noisy post bring up, the router permanently shuts down the link

LC/0/2/CPU0:Jan 13 00:20:16.251 UTC: npu_drvr[128]: %FABRIC-NPU_DRVR-3-NPU_CPA_GEN_ERR_INFO :
 FEC check failures on link 0/254. FEC bin is filled to 14

Verify the FEC links

Procedure

Verify the FEC link information using show controllers npu link-info command.

Example:

Router#show controllers npu link-info rx 254 255 fsm instance 0 location 0/2/CPU0 detail
Sat Jan 13 00:39:49.448 UTC
 
Node ID: 0/2/CPU0
Link ID: 0/2/0/254    255      Oper State: DOWN   0/FC1/2/158
Event                      State                 Timestamp
+--------------------------------------------------------------------------------+
LINK_UP_INTR               UP                    Sat Jan 13 00:18:44 2018
BAD_FEC_BELOW_THR          PORT_ACTIVATE_DELAYED Sat Jan 13 00:19:16 2018
LINK_MON                   FAB_PORT_CREATED      Sat Jan 13 00:19:17 2018
LINK_MON                   ACTIVATED             Sat Jan 13 00:19:19 2018
LINK_UP_INTR               MAC_UP                Sat Jan 13 00:19:24 2018
LINK_UP_INTR               PEER_DISCOVERY        Sat Jan 13 00:19:24 2018
LINK_UP_INTR               PEER_DETECTED         Sat Jan 13 00:19:24 2018
LINK_UP_INTR               TOPOLOGY_CHECK        Sat Jan 13 00:19:24 2018
LINK_UP_INTR               SYNC_WAIT             Sat Jan 13 00:19:24 2018
LINK_UP_INTR               KEEPALIVE_START       Sat Jan 13 00:19:24 2018
LINK_UP_INTR               CHECK_REACH           Sat Jan 13 00:19:24 2018
LINK_UP_INTR               UP                    Sat Jan 13 00:19:24 2018
BAD_FEC                    UP                    Sat Jan 13 00:20:16 2018
DIS_PERM_SHUT              MAC_UP                Sat Jan 13 00:20:16 2018
DIS_PERM_SHUT              STOPPED               Sat Jan 13 00:20:16 2018
+--------------------------------------------------------------------------------+

This table describes the significant fields shown in this example.

Table 2. show controllers npu link-info field descriptions
Field	Description
`BAD_FEC_BELOW_THR`	There are FEC failures, but the number of failures has not exceeded the predefined threshold (in this case, 5 per hour). The router retunes and checks for FEC improvement.
`BAD_FEC`	This part of the log entry indicates that FEC detected failures, and the number of these failures surpassed a predefined threshold. As a result, the decision was made to permanently shut down the affected interface or port as a protective measure.
`DIS_PERM_SHUT`	The link or port has been intentionally disabled and is in a shutdown state after FEC fails for the threshold limit (After fifth failure).

Disable fabric link management for uncorrectable errors

Fabric link management for uncorrectable errors is enabled by default. To disable this feature, use the hw-module fabric-fec-monitor disable command in XR Config mode mode.

Procedure

	1.	Enter the global configuration mode. Example: `Router# configure`
	2.	Disable the fabric FEC monitor. Example: `Router(config)# hw-module fabric-fec-monitor disable`
	3.	Save the changes. Example: `Router(config)# commit`

Need help?

Open a support case

(Requires a Cisco Service Contract)

The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.