Logging is the process of collecting metrics from a server and logging them. For managing a large number of server deployments, these logs should be sent to a database server. The database can be queried by a monitoring program to help track down issues and optionally visualize the data.
[Log collector] --> [Database] --> [Monitoring dashboard with API and/or UI]
collectd gathers and distributes metrics of a running system. A full list of metric collection plugins that collectd supports can be found here. 
collectd is available in most major repositories. 
CentOS and RHEL:
$ sudo yum install epel-release $ sudo yum install collectd
Debian and Ubuntu:
$ sudo apt-get install collectd
By default, collectd gathers information about the CPU utilization (
interface bandwidth, and
memory usage. The hostname and desired plugins should be defined in the
/etc/collectd.conf configuration file at a minimum.
Hostname <FQDN> LoadPlugin <PLUGIN>
Define a database server that the network plugin should send the logs to. This can be InfluxDB, Logstash, or any other logging database. Alternatively, the
dbi plugin can be used to connect to many different database back-ends directly.
<Plugin network> Server "<DB_ADDRESS>" "<DB_PORT>" </Plugin>
Start the service.
$ sudo systemctl start collectd
“collectd – The system statistics collection daemon.” collectd.org. Accessed July 9, 2020. https://collectd.org/
“[collectd] Download.” collectd.org. Accessed July 9, 2020. https://collectd.org/download.shtml
“Monitoring Linux performance with Grafana.” August 28, 2017. Accessed July 9, 2020. https://opensource.com/article/17/8/linux-grafana