Server Monitoring Client

Track system metrics from Platform

The Linux Monitoring Client is a software module based on the Linux Client that keeps track of different system metrics, and publishes them to the Platform in order to keep track of a server or an embedded Linux device or a fleet of them.

Resource Tracking

The client tracks different metrics for System Information as well as for six different types of resources: general system information, CPU, system memory, network interfaces, filesystem mount points, and I/O drives.

Apart from uploading the resource information to Platform, the resources can be queried directly to the client or from the local server running the client at http://localhost:7890/monitor

General System Information

There is some system information that can be relevant from a server administration standpoint.

Currently, the metrics the client is retrieving are:

  • Hostname

  • Uptime

  • OS version

  • Quantity of normal and security updates (Ubuntu server)

  • If a reboot is required to apply updates (Ubuntu server)


To evaluate the performance of the CPU, the following metrics are tracked:

  • Number of cores

  • CPU usage

  • CPU load for 1m, 5m and 15m

System Memory (RAM)

As having reliable information about the system memory used is vital for capacity planning and maintaining the integrity of your servers, the following metrics are collected:

  • Total, used and available memory capacity

  • Memory usage

  • Total, used and available swap memory capacity

  • Swap memory usage

Network Interfaces

For each network interface the following is tracked:

  • Internal/Private IP

  • Incoming and Outgoing network speed

  • Incoming and Outgoing total network transfer

It will also retrieve the public IP of the server.

Filesystem Mount Points

Each filesystem corresponds to one partition of a drive, from which it will track:

  • Capacity

  • Capacity free

  • Capacity used

  • Capacity usage (percentage)

I/O Drives

To provide insight on how a server is performing, the following I/O drive metrics are monitored:

  • Input and Output operation speed

  • Disk usage


The basic device configuration is tracked over a JSON file located in the server. Additional configurations are tracked trough the device properties to allow live configuration changes through the Platform.

Basic device configuration

The configuration will be auto created on first launch when it is auto provisioned.

The file structure is as follows:

  "device": {
    "credentials": "<device_credentials>",
    "id": "<device_id>",
    "name": "<device_name>"
  "server": {
    "ssl": true,
    "url": "<server_to_connect_url",
    "user": "<thinger.io_account>"

If launched with root, the default path for the configuration file is: /etc/thinger_io/monitor/app.json

If launched with non-root, the default path for the configuration file is: /home/<user>/.config/thinger_io

Configuration properties

As mentioned before, the configuration for the resources will be in the device properties.

It contains five sections:

  • defaults: boolean value indicating whether the name of the first element of each resource will have a default name instead of the resource name. Useful when tracking different devices with one dashboard.

  • interfaces: JSON array with the name of the network interfaces to track.

  • filesystems: JSON array with the mount point of the partitions to track.

  • drives: JSON array with the Linux devices to track.

  • server: Object containing the host IP to listen in and port to launch the local server.

Here is an example resources property value:

  "defaults": true,
    "drives": [
    "filesystems": [
    "interfaces": [
    "server": {
      "host": "",
      "port": 7890

First Launch

This module has the feature to auto provision the device in the Platform, but also allows to input a configuration if it's already provisioned.

It is also capable of creating a system service so the monitor is always running regarding if the server has been rebooted.

With auto provision

Download the installation script from the last software release, and run the software as:

./ -t '<create_device token>'

For update and reboot abilities it needs to be run with root user

Without auto provision

If the device is already provisioned, we will need to set the user id, device id and credentials in the configuration file and launch the software as:

./thinger_monitor [-c <config file path>]

If the operating system is not Ubuntu or OpenSSL is installed in a different folder than default, it is neccesary to indicate the certificates directory by declaring the variable SSL_CERT_DIR before calling the installer or the binary:

Ex: SSL_CERT_DIR=/etc/ssl/certs ./

Last updated