Monitoring Redis with Prometheus, Node exporter and Grafana

Prometheus is a server designed to collect and store metrics received from processes (exporters). The accumulated data can be viewed both through the Prometheus web interface and through separate visualization tools, for example, Grafana.

This instruction will help:

• deploy Prometheus 2.45.1 server in CentOS 8.4 operating system in VK Cloud;
• install the Node exporter extension in the Redis 5 Single configuration DBMS;
• customize the visualization of data from Redis in Prometheus and Grafana 10.

Prometheus server, DBMS and Grafana will be deployed on separate VMs.

1. Create a virtual network, for example, monitoring-net.

2. Create VM for Prometheus server:

   • name: Centos_8_5_Prometheus;
   • operating system: CentOS 8.4;
   • network: monitoring-net;
   • public IP address: assign it, as an example is given 87.239.239.239;
   • security groups: default, all.

   The internal IP address of the created instance: 10.0.3.7.

3. Create a DB instance:

   • name: Redis-5;
   • DBMS: Redis 5;
   • configuration type: Single;
   • network: monitoring-net;

   The internal IP address of the created instance: 10.0.3.13.

4. Deploy Grafana 10 is on the monitoring-net network.

1. Install the Node exporter extension for the Redis-5 DB instance. When installing, specify the parameter listen_port = 9100.

2. Connect to the Centos_8_5_Prometheus VM.

3. Download Prometheus and unzip the downloaded archive:

   Copy
   sudo suexport VERSION="2.45.1"wget https://github.com/prometheus/prometheus/releases/download/v$VERSION/prometheus-$VERSION.linux-amd64.tar.gz -O - | tar -xzv -C /tmp

4. Copy the contents of the repository prometheus-2.45.1.linux-amd64:

   Copy
   mkdir /etc/prometheusmkdir /var/lib/prometheuscp /tmp/prometheus-$VERSION.linux-amd64/prometheus /usr/local/bincp /tmp/prometheus-$VERSION.linux-amd64/promtool /usr/local/bincp -r /tmp/prometheus-$VERSION.linux-amd64/consoles /etc/prometheuscp -r /tmp/prometheus-$VERSION.linux-amd64/console_libraries /etc/prometheuscp /tmp/prometheus-$VERSION.linux-amd64/prometheus.yml /etc/prometheus/

5. (Optional) Delete files from the temporary folder:

   Copy
   rm -rf /tmp/prometheus-$VERSION.linux-amd64

6. Create a group and a user prometheus, assign him rights to the associated repositories:

   Copy
   groupadd --system prometheususeradd --system -g prometheus -s /bin/false prometheuschown -R prometheus:prometheus /var/lib/prometheus /etc/prometheuschown prometheus:prometheus /usr/local/bin/prometheus /usr/local/bin/promtool

7. In the file /etc/prometheus/prometheus.yml add content to the scrape_configs block:

   Copy
   - job_name: "node"  scrape_interval: 10s  static_configs:    - targets: ["10.0.3.13:9100"]

8. Create a Prometheus startup script /etc/systemd/system/prometheus.service with the contents:

   Copy
   [Unit]Description=PrometheusWants=network-online.targetAfter=network-online.target[Service]User=prometheusGroup=prometheusExecStart=/usr/local/bin/prometheus \    --config.file /etc/prometheus/prometheus.yml \    --storage.tsdb.path /var/lib/prometheus \    --web.console.templates=/etc/prometheus/consoles \    --web.console.libraries=/etc/prometheus/console_librariesExecReload=/bin/kill -HUP $MAINPID[Install]WantedBy=default.target

9. Run Prometheus:

   Copy
   systemctl daemon-reloadsystemctl start prometheus.servicesystemctl enable prometheus.service

10. Make sure that Prometheus is working correctly:

    Copy
    systemctl status prometheus.service

    The output of the command must contain the active status.

    An example of the expected output

11. Wait a few minutes for the data to accumulate.

12. Go to the address http://87.239.239.239:9090.

    The Prometheus server web interface opens.

13. (Optional) Review the available data:

    • Create a search query to view the data in tabular form.
    • Open the template graph sets to http://87.239.239.239:9090/consoles/index.html.example.

14. (Optional) Unbind floating address from VM Centos_8_5_Prometheus.

1. Go to the Grafana web interface.
2. Add new data source: to Prometheus server URL specify http://10.0.3.7:9090.
3. Install sets of graphs for visualizing the received data, for example, by importing a ready-made version — Node Exporter Full.

1. Create a test load on the Redis-5 VM in a way that is convenient for you.

   Example with the sysbench utility

2. Make sure to change the indicators of the Grafana charts.

The created resources are charged and quotas are spent. If you don't need them anymore:

1. Delete deployed virtual machines.
2. Delete a virtual machine with Grafana.
3. Remove the floating IP address assigned to the Centos_8_5_Prometheus VM.
4. Delete the port to which the virtual IP address is assigned.
5. Delete a monitoring-net network.