Configure Monitoring with Datadog

This topic describes how to integrate SingleStore with Datadog and collect metrics used to monitor a cluster.

Note

The SingleStore integration does not include any Datadog events.

Prerequisites

Install the Datadog Agent. The SingleStore check is a part of the Datadog Agent package, and no additional installation is required.

For containerized environments, enable Autodiscovery.

Configure Datadog

You can configure Datadog to collect metrics and logs. The configuration steps differ based on whether SingleStore is deployed on the host or inside a container.

Note

To verify if the Datadog Agent is collecting metrics/logs from SingleStore, run the Agent’s status subcommand, and find singlestore under the Checks section.

Host Environment

Configure Metric Collection

To collect SingleStore performance metrics:

  1. Configure the conf.d/singlestore.d/conf.yaml file at the root of the Agent’s configuration directory. Here’s a sample conf.yaml file.

  2. Restart the Agent.

Note

By default, the SingleStore integration collects metrics from the following tables: MV_GLOBAL_STATUS, AGGREGATORS, and LEAVES. To collect system level metrics (for example, CPU, disk, network IO, and memory), set collect_system_metrics as true in singlestore.d/conf.yaml file.

Configure Log Collection

Log collection is disabled by default in the Datadog Agent. To enable log collection:

  1. Set logs_enabled as true in the datadog.yaml file.

  2. Add the log files in singlestore.d/conf.yaml to start collecting SingleStore logs. For example,

     logs:
        - type: file
          path: /var/lib/memsql/<NODE_ID>/tracelogs/memsql.log
          source: singlestore
          service: "<SERVICE_NAME>"
  3. Restart the Agent.

Containerized Environment

Configure Metric Collection

To collect SingleStore performance metrics in a containerized environment, update the following parameters in the configuration file:

Parameter

Value

<INTEGRATION_NAME>

singlestore

<INIT_CONFIG>

blank or {}

<INSTANCE_CONFIG>

{"host": "%%host%%", "port": "%%port%%", "username": "singlestore_user", "password": "user_password"}

See Autodiscovery Integration Templates for information on updating these parameters.

Note

By default, the SingleStore integration collects metrics from the following tables: MV_GLOBAL_STATUS, AGGREGATORS, and LEAVES. To collect system level metrics (for example, CPU, disk, network IO, and memory), set collect_system_metrics as true in singlestore.d/conf.yaml file.

Configure Log Collection

Log collection is disabled by default in the Datadog Agent. To enable log collection, update the following parameter:

Parameter

Value

<LOG_CONFIG>

{"source": "singlestore", "service": "<name_of_the_service>"}

See Kubernetes Log collection for more information.

Service Checks

The SingleStore integration supports the following service checks:

Service Check

Description

singlestore.can_connect

Returns CRITICAL if the Agent is unable to connect to the SingleStore instance or collect metrics. Otherwise, returns OK.

List of Metrics

See SingleStore Datadog Integration docs for a complete list of all the metrics that the Datadog Agent collects from the connected SingleStore cluster.