Interpreting interface errors graph


I am having little difficulties understanding the interface errors graphs:

  1. The average value seems to be higher than than maximum value - is this a rounding / conversion issue?
  2. How should these values be interpreted?
  3. Is the graphed value just the delta count from the previous counter divided by the graph’s interval?

See this example figure from a 100 Gbit/s interface from a Cisco Nexus C9364C device (Cisco NX-OS)


Component Version
LibreNMS 1.69-69-g0a7586b9f
DB Schema 2020_10_12_095504_mempools_add_oids (191)
PHP 7.4.3
Python 3.8.5
MySQL 10.3.25-MariaDB-0ubuntu0.20.04.1
RRDTool 1.7.2


[OK] Composer Version: 2.0.8
[OK] Dependencies up-to-date.
[OK] Database connection successful
[OK] Database schema correct
[INFO] Detected Python Wrapper
[OK] Connection to memcached is ok
[WARN] Global lnms shortcut not installed. lnms command must be run with full path
sudo ln -s /opt/librenms/lnms /usr/local/bin/lnms
[WARN] Bash completion not installed. lnms command tab completion unavailable.
sudo cp /opt/librenms/misc/lnms-completion.bash /etc/bash_completion.d/
[WARN] Your install is over 24 hours out of date, last update: Tue, 01 Dec 2020 07:45:53 +0000
Make sure your cron is running and run ./ by hand to see if there are any errors.

My interpretation is that this is millibps, while your maximum is in bps. Not many errors at all, but they are errors, and depending on the load, they will increase.

Are you running ACI or NXOS? On NXOS you should verify if you are running FEC or not. If not, you can enable FEC and it will probably “fix” the link. NXOS has as far as I know limited commands to verify FEC “show interface fec” is the only command I know.

Is this a optical link? Based on my experience 100G is really sensitive to optical values. Adding attenuators to make the link “perfect” really does solve issues like this.

ACI and yes, it’s optical.