Description

This article provides steps to troubleshoot the "Input Errors" or "Resource Errors" incrementing on the interfaces of a QFX device.
Also it provides the suggestions to reduce/eliminate these errors.


User@Switch> show interfaces et-1/0/96 extensive | match errors
Input errors:
Errors: 0, Drops: 54402857, Framing errors: 0, Runts: 0, Bucket drops: 0, Policed discards: 0, L3 incompletes: 0, L2 channel errors: 0, L2 mismatch timeouts: 0, FIFO errors: 0,
Resource errors: 54402857
Output errors:
Carrier transitions: 3, Errors: 0, Drops: 0, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0, Bucket drops: 0

Symptoms

Mostly the "Input Errors" / Resource Errors" are seen to be incrementing if 
a) The interface is receiving more traffic than it can handle or it is oversubscribed.
b) If the traffic coming in on the interface is less than the bandwidth of the interface but the traffic is having micro burst.  


User@Switch> show interfaces et-1/0/96 extensive | match errors
Input errors:
Errors: 0, Drops: 54402857, Framing errors: 0, Runts: 0, Bucket drops: 0, Policed discards: 0, L3 incompletes: 0, L2 channel errors: 0, L2 mismatch timeouts: 0, FIFO errors: 0,
Resource errors: 54402857
Output errors:
Carrier transitions: 3, Errors: 0, Drops: 0, Collisions: 0, Aged packets: 0, FIFO errors: 0, HS link CRC errors: 0, MTU errors: 0, Resource errors: 0, Bucket drops: 0

 

Solution

1. We need to identify what is the rate of increment on the interface.
2. Check what kind of traffic is incoming on the interface for example if it is "lossy" , "lossless" , or "lossless-headroom" traffic. Lossy traffic is the best effort traffic.
3. Based on the above information we need to adjust the shared buffer settings on the QFX device on the ingress buffer.

Setting-1
set class-of-service shared-buffer ingress percent 100
set class-of-service shared-buffer ingress buffer-partition lossless percent 20
set class-of-service shared-buffer ingress buffer-partition lossless-headroom percent 10
set class-of-service shared-buffer ingress buffer-partition lossy percent 70
set class-of-service shared-buffer ingress buffer-partition lossy dynamic-threshold 10

If the above configuration reduces the drops but if the are still seen we can try to increase the lossy buffer more and reduce the lossless-headroom percent.

Setting-2
set class-of-service shared-buffer ingress percent 100
set class-of-service shared-buffer ingress buffer-partition lossless percent 10
set class-of-service shared-buffer ingress buffer-partition lossless-headroom percent 10
set class-of-service shared-buffer ingress buffer-partition lossy percent 80
set class-of-service shared-buffer ingress buffer-partition lossy dynamic-threshold 10

We can try to increase the shared buffer to lossy maximum.

Setting-3
set class-of-service shared-buffer ingress percent 100
set class-of-service shared-buffer ingress buffer-partition lossless percent 5
set class-of-service shared-buffer ingress buffer-partition lossless-headroom percent 5
set class-of-service shared-buffer ingress buffer-partition lossy percent 90
set class-of-service shared-buffer ingress buffer-partition lossy dynamic-threshold 10


We can refer to the following documents for curating the buffer settings

https://www.juniper.net/documentation/us/en/software/junos/traffic-mgmt-qfx/topics/example/cos-shared-buffer-allocation-lossless-qfx-series-configuring.html



 

Modification History

2023-12-03 : Article Created
Removed the buffer cli enable as it is for the egress queue drops.

21 May 2024: Added the PR associated