Graphs with Unrealistic Spike Values

[Szollosy]

Hello,

We have approximately 42 Cisco Nexus 9000 switches (N9K-C93180YC-FX), but only one of them is experiencing issues with the graphs. There are some spikes with unrealistic value.

This happens on several ports, not just on the Overall Traffic.

It’s interesting that this switch is in a vPC with another one, and this one doesn’t have the same graph issues

Does anyone know how to fix this?
Thanks!

$ ./validate.php

Component	Version
LibreNMS	23.10.0-61-g7c006e962 (2023-11-07T19:25:28-03:00)
DB Schema	2023_11_04_125846_packages_increase_name_column_length (273)
PHP	8.1.24
Python	3.6.8
Database	MariaDB 10.3.39-MariaDB
RRDTool	1.7.0
SNMP	5.8
===========================================

[OK] Composer Version: 2.6.5
[OK] Dependencies up-to-date.
[OK] Database connection successful
[OK] Database Schema is current
[OK] SQL Server meets minimum requirements
[OK] lower_case_table_names is enabled
[OK] MySQL engine is optimal
[OK] Database and column collations are correct
[OK] Database schema correct
[OK] MySQl and PHP time match
[OK] Distributed Polling setting is enabled globally
[OK] Connected to rrdcached
[OK] Active pollers found
[OK] Dispatcher Service is enabled
[OK] Locks are functional
[OK] Python wrapper cron entry is not present
[OK] Redis is functional
[OK] rrdtool version ok
[OK] Connected to rrdcached

[laf]

Their’s an FAQ on this: FAQ - LibreNMS Docs

[Szollosy]

@laf Yes, I read the documentation, but it didn’t help… as I mentioned, all the other devices are functioning normally, and only this one is having issues with the graph. Furthermore, the throughput of this device is quite low, averaging 1Gb

[laf]

Have you ran what it says in the FAQ?

You could also use one of the removespikes.* scripts in the script folder, that can flatten the graphs but you have to run it manually when you spot the issues.