How to monitor your Database Instance with Scaleway Cockpit

Reviewed on 19 June 2024 • Published on 19 June 2024

You can view your Managed Database for PostgreSQL or MySQL metrics and logs using Scaleway Cockpit. Managed Databases are fully integrated into Cockpit.

Monitoring is available for:

Primary Database Instance nodes
Secondary nodes (if High Availability is activated)
Any Read Replicas

This page explains how to get started with Scaleway Cockpit for viewing your Managed Database metrics and logs.

Before you start

To complete the actions presented below, you must have:

A Scaleway account logged into the console
Owner status or IAM permissions allowing you to perform actions in the intended Organization
A PostgreSQL or MySQL Database Instance

How to activate your Cockpit

Scaleway Cockpit is activated by default if you have created any Scaleway resources integrated with Cockpit.

How to access the Grafana dashboard

To view your Managed Database’s metrics and logs, use the Grafana dashboard which is accessible from the Scaleway console:

First, you must create a user and their associated credentials (a username and password) for Grafana
Then, you can use these credentials to access the Grafana dashboard for all your Scaleway products, via the Scaleway console.

Create a Grafana user and retrieve their username and password by following this procedure.
From the Scaleway console Cockpit overview page, click Open dashboards in the top right corner.
Enter your Grafana username and password to log in.

You are directed to the Scaleway Cockpit dashboards homepage.

How to view and understand your Managed Database metrics

Click Open dashboards on the top right from the Scaleway Cockpit dashboards homepage.
Click either RDB MySQL Overview or RDB PostgreSQL Overview, depending on your engine of choice, on the left side menu.

You are directed to the RDB Overview dashboard.

Metrics are refreshed every minute, and report on the following data:

CPU usage
Memory usage
Total connections
Disk usage (in % and in Bytes)
Replication lag in seconds (when High Availability mode is activated or when Read Replicas are created).

Choosing metric parameters

At the top of the dashboard, you can configure the following parameters, that control which metrics are displayed:

Resource name - Select a specific Database Instance to monitor.
Node - Choose a specific node (main, replica, or Read Replica). In a High Availability (HA) cluster, both nodes might switch from main to replica and vice versa. So you might see two versions of the same node, as main and as replica. If you select main for a specific node, you will only see the node’s activity when it was the main, and vice-versa.
Timeframe - Get metrics from the last five minutes up to 31 days (free plan) and 365 days (premium plan).

Viewing logs

The log section is displayed below the metrics section.

Click Logs to collapse the tab and view your logs.

Log configuration is done by Scaleway. Note that:

The log_error_verbosity parameter is set to VERBOSE for PostgreSQL and 2 for MySQL by default.
On PostgreSQL, all queries longer than 5 seconds are logged. This parameter cannot be edited.
On MySQL, slow queries are not activated by default, but can be activated in the advanced settings of the Database Instance. The setting is called slow_query_log and the time can be edited with long_query_time.

How to activate managed alerts

You can activate managed alerts for your Managed Databases with Scaleway Cockpit.

Note

Refer to the How to activate managed alerts to learn how to do so.

By default, three managed alerts are configured for Managed Databases:

MySQLHighStorageUsage or PostgreSQLHighStorageUsage- You are alerted if disk usage is higher than 80% for 10 consecutive minutes.
MySQLHighCPULoad or PostgreSQLHighCPULoad- You are alerted if CPU usage is higher than 80% for 10 consecutive minutes.
MySQLTooManyConnections or PostgreSQLTooManyConnections- You are alerted if your connection usage is higher than 80% for 10 consecutive minutes.

You can also configure your own managed alerts and set values for each of the parameters below to define when you want to be alerted:

MySQL

Alert	Description
`rdb_instance_mysql_mysql_global_status_threads_connected`	Total active connections on MySQL
`rdb_instance_mysql_mysql_global_variables_max_connections`	Maximum number of connections configured
`rdb_instance_mysql_mysql_slave_status_seconds_behind_master`	Replica lag in seconds
`rdb_instance_mysql_node_cpu_seconds_total`	Total amount of CPU time consumed by the Database Instance, measured in seconds
`rdb_instance_mysql_node_filesystem_avail_bytes`	Total amount of remaining available space, in bytes
`rdb_instance_mysql_node_filesystem_size_bytes`	Total size of the filesystem, in bytes
`b_instance_mysql_node_memory_MemAvailable_bytes`	Amount of memory available
`rdb_instance_mysql_node_memory_MemTotal_bytes`	Total installed physical memory (RAM)

PostgreSQL

Alert	Description
`rdb_instance_postgresql_node_cpu_seconds_total`	Total amount of CPU time consumed by the Database Instance, measured in seconds
`db_instance_postgresql_node_filesystem_avail_byte`	Total amount of remaining available space, in bytes
`rdb_instance_postgresql_node_filesystem_size_bytes`	Total size of the filesystem, in bytes
`db_instance_postgresql_node_memory_MemAvailable_bytes`	Amount of memory available
`rdb_instance_postgresql_node_memory_MemTotal_bytes`	Total installed physical memory (RAM)
`rdb_instance_postgresql_pg_replication_lag`	Replica lag in seconds
`rdb_instance_postgresql_pg_settings_max_connection`	Maximum number of concurrent Database Instance connections
`rdb_instance_postgresql_pg_stat_database_numbackends`	Number of active Database Instance connections.