In this blog post, we will show you how to add external exporters to Percona Monitoring and Management (PMM) and monitor various services. Our focus will be on Valkey/Redis, but the principles apply to other technologies as well.
PMM uses VictoriaMetrics, which is compatible with Prometheus. Prometheus has numerous exporters that you can use.
When you install PMM Client on a server, the following exporters are also installed:
- node_exporter collects general system metrics.
- mysqld_exporter collects MySQL server metrics.
- mongodb_exporter collects MongoDB server metrics.
- postgres_exporter collects PostgreSQL performance metrics.
- proxysql_exporter collects ProxySQL performance metrics.
- rds_exporter collects Amazon RDS performance metrics.
- azure_database_exporter collects Azure database performance metrics.
In addition to these, you can add any additional exporters, known as external exporters. The PMM Client will connect to these exporters and send the metrics to the PMM Server. Let’s see how we can add the Valkey/Redis exporter.
Valkey/Redis exporter
The Prometheus exporter for Valkey, which is also Redis-compatible, is available here. Follow the installation instructions for the exporter. For simplicity, we will download and set it up using systemd. Execute the following commands on your Valkey/Redis server:
|
1 2 3 4 5 |
> wget https://github.com/oliver006/redis_exporter/releases/download/v1.61.0/redis_exporter-v1.61.0.linux-amd64.tar.gz > tar xfv redis_exporter-v1.61.0.linux-amd64.tar.gz > sudo groupadd --system prometheus > sudo useradd -s /sbin/nologin --system -g prometheus prometheus > sudo cp redis_exporter-v1.61.0.linux-amd64/redis_exporter /usr/local/bin/ |
For security reasons, you can add the following user to your Redis/Valkey configuration file:
|
1 |
user monitor +client +ping +info +config|get +cluster|info +slowlog +latency +memory +select +get +scan +xinfo +type +pfcount +strlen +llen +scard +zcard +hlen +xlen +eval allkeys on >{{PASSWORD}} |
Of course, you have to change the password.
Create the /etc/systemd/system/valkey_exporter.service systemd file with the following content:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 |
[Unit] Description=Valkey/Redis exporter Documentation=https://github.com/oliver006/redis_exporter Wants=network-online.target After=network-online.target [Service] Type=simple User=prometheus Group=prometheus ExecReload=/bin/kill -HUP $MAINPID ExecStart=/usr/local/bin/redis_exporter --log-format=txt --namespace=redis --web.listen-address=:9121 --web.telemetry-path=/metrics --redis.user=monitor --redis.password={{PASSWORD}} SyslogIdentifier=valkey_exporter Restart=always [Install] WantedBy=multi-user.target |
Then enable the daemon:
|
1 2 3 |
> sudo systemctl daemon-reload > sudo systemctl enable valkey_exporter > sudo systemctl start valkey_exporter |
At this point, the Valkey/Redis exporter is already running and exposing the metrics on port 9121.
Now, we have to tell the PMM Client to collect the metrics from that port.
|
1 |
> pmm-admin add external --service-name=valkey --listen-port=9121 |
With this simple command, PMM Client starts collecting metrics and sends them to PMM Server. If you do not have a PMM Server or Client configured yet please follow this link.
Dashboards
Now, let’s visualize the data we collect. PMM is based on Grafana, and Grafana has many publicly available dashboard templates. To find a Redis dashboard, visit the Grafana website and search for “Redis dashboard.” Choose your favorite template and load it into PMM. To do this, navigate to Dashboards -> Import and enter the Dashboard ID that you chose.
Now, you have a Valkey/Redis dashboard in PMM. You can customize this dashboard as you like.
Alerting
We also want an alert if the Valkey/Redis service is unavailable. We can create an alert rule based on the redis_up metric. Follow these steps to set up the alert:
- Go to Alerting -> Alert Rules and create a new alert rule.
- Choose Grafana managed alerts.
- Use the following expression for the alert condition:
sum by (service_name, node_name) (avg_over_time(redis_up[1m])) == 0 - Change the Type from Range to Instant.
- Set an evaluation period that works for you. In this example, it is 30s, and the alert will be triggered if the condition is true for 1 minute.
- Alert if no data or all values are null (exporter is down)
This configuration ensures that you will be alerted if the Valkey instance is down for more than one minute.
If you want more alert rules, you can find a nice collection here.
Conclusion
By following these steps, you can monitor Valkey/Redis with PMM. The process is straightforward and can be adapted for other exporters and services. PMM’s flexibility in integrating with various exporters makes it a powerful tool for comprehensive monitoring. Additionally, with Grafana dashboards and alerting, you can visualize and be alerted to the health of your services in real time.
Percona Monitoring and Management is a best-of-breed open source database monitoring solution tool for MySQL, PostgreSQL, MongoDB, and the servers on which they run. It allows you to monitor, manage, and improve the performance of your databases, no matter where they are located or deployed.
Download Percona Monitoring and Management Today
About the Author
Tibor KoroczTibi joined Percona in 2015 as a Consultant. Before joining Percona, among many other things, he worked at the world’s largest car hire booking service as a Senior Database Engineer. He enjoys trying and working with the latest technologies and applications which can help or work with MySQL together. In his spare time he likes to spend time with his friends, travel around the world and play ultimate frisbee.
