Discontinuous graph in LibreNMS

[Sompong_Manthong]

Hi,

I have got the problem about the graph discontinuous and gateway timeout , Could someone guild me to start the investigation?

Below is my captured information.

./daily.sh
Could not connect to database, check logs/librenms.log.
Could not connect to database, check logs/librenms.log.
Updating to latest codebase OK
Updating Composer packages OK
Updated from cfb70f73d to 473a7d360 OK
Updating SQL-Schema OK
Updating submodules OK
Cleaning up DB OK
Fetching notifications OK
Caching PeeringDB data OK
Caching Mac OUI data OK
You have mail in /var/mail/librenms

++++++++++++++++++++++

./validate.php

Component	Version
LibreNMS	23.2.0-41-g473a7d360 (2023-03-19T10:07:52+07:00)
DB Schema	2023_03_14_130653_migrate_empty_user_funcs_to_null (249)
PHP	8.1.2-1ubuntu2.11
Python	3.10.6
Database	MariaDB 10.6.12-MariaDB-0ubuntu0.22.04.1
RRDTool	1.7.2
SNMP	5.9.1
===========================================

[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 and column collations are correct
[OK] Database schema correct
[OK] MySQl and PHP time match
[OK] Active pollers found
[OK] Dispatcher Service not detected
[OK] Locks are functional
[OK] Python poller wrapper is polling
[OK] Redis is unavailable
[OK] rrdtool version ok
[OK] Connected to rrdcached
You have mail in /var/mail/librenms

Example of the graph:

CPU/Memory Utilization:

Gateway timeout error:

[RR1]

You have to start looking at system resources first
how many devices ?
what are the processor and memory specifications
look at the poller logs, to see which devices is causing high polling time
how many threads per core

is this a new setup or it was working fine before

logs to check
librenms log
webserver logs
php logs

[hamrad]

I see a wrong (in my opinion) number of pollers for your system:
python3 /opt/librenms/poller-wrapper.py 100
I see that your system have only 12 cores and all of them has 100% utilization. Try to reduce the number of pollers to 50, for example.
Also, try to increase a Timeout value for SNMP: Global Settings - Poller - SNMP - Timeout (also, try changing the amount of “Max OIDs” and/or “Retries” here).