Abnormal high values on all graphs of some devices

Good day.

Did anyone faced with issue where graphs on network device show abnormal high values, this values are beyond port speed and device capacity. For example Port speed is 1G, but graph show 46G and higher. The same applies to all other ports on this device.

Graph image - 1 hosted at ImgBB — ImgBB

My validate.sh


Component Version
LibreNMS 22.3.0-9-g51a864682
DB Schema 2021_12_15_233760_create_bill_tags_table (237)
PHP 7.3.20
Python 3.6.8
MySQL 10.3.28-MariaDB
RRDTool 1.7.0
SNMP 5.8

[OK] Composer Version: 2.5.4
[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 schema correct
[OK] MySQl and PHP time match
[FAIL] No active polling method detected
[OK] Dispatcher Service not detected
[OK] Locks are functional
[FAIL] No active python wrapper pollers found
[OK] Redis is unavailable
[OK] rrd_dir is writable
[OK] rrdtool version ok

Is there any commonality between the devices doing this? Either vendor, model, or the network path to them?

I have several posts around about spikes caused by high latency and lossy links where the polling is interrupted. There and some SNMP tweaks which may help (max repeaters)

There are also RRD tuning options in the docs which can cap spikes to the interface speed to help reduce the impact.

Yes, this issue applies only to some Cisco Nexus 5548 devices, but some are not afftected by this problem. NX-OS is the same on all boxes.

Can you please point me what SNMP tweaks and RRD tuning can be done in order to avoid such issues, i’ll try and may be in will help

Thank you.

Max Repeaters is mentioned here: Performance - LibreNMS Docs

I have a big sob story which mentions some of my testing and successes with it here: Debugging graph spikes from high latency links - #6 by rhinoau

The RRD tuning is referred to here: RRDTune - LibreNMS Docs

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.