Docs » Get started with the Splunk Distribution of the OpenTelemetry Collector » Discover and configure metrics sources automatically

Caution

Preview: Discovery mode in Splunk Distribution of OpenTelemetry Collector

Preview features described in this document are provided by Splunk to you โ€œas isโ€ without any warranties, maintenance and support, or service-level commitments. Splunk makes this preview feature available in its sole discretion and may discontinue it at any time. Use of preview features is subject to the Splunk General Terms .

Discover and configure metrics sources automatically ๐Ÿ”—

Use the discovery mode of the Splunk Distribution of OpenTelemetry Collector to detect metric sources and create a configuration based on the results.

Discovery mode can detect several types of metric sources on the host, such as databases and servers. With this information, the Collector generates configuration you can modify and adopt, or incorporate into your exiting configuration automatically by default.

The main advantage of using discovery mode is that you donโ€™t need to manually update the OpenTelemetry Collector configuration for supported metric sources. This is helpful in environments when you deploy and activate host services dynamically or when adding a new supported target database to your infrastructure.

Note

Discovery mode is available starting from version 0.72.0 and higher of the Splunk Distribution of the OpenTelemetry Collector.

How discovery mode works ๐Ÿ”—

When you run the Collector in discovery mode, it tests built-in configurations for supported metric receivers against endpoints discovered on your platform by observer extensions. This happens before starting the Collector service.

For any dynamically instantiated receiver that retrieves metrics matching the success criteria, the Collector translates the discovery configuration to a receiver creator instance with the known working rules, as well as the required observer extension. See Receiver creator receiver for more information. At the same time, the Collector adds the configuration to the metrics pipeline at runtime.

For any receiver that can establish a connection with a service, but not receive the expected metrics, discovery mode suggests which properties to set, or what extensions or settings to configure on the service to successfully retrieve telemetry. You can define any target-specific configuration values that are required, for example authentication information, using discovery properties to tune the discovery process.

Supported host services and applications ๐Ÿ”—

Discovery mode supports the following host services and applications:

Service

Receiver

MySQL

Smart Agent with collectd/mysql monitor type. See MySQL.

PostgreSQL

Smart Agent with postgresql monitor type. See PostgreSQL.

OracleDB

Oracle DB receiver. See Oracle Database receiver.

NGINX

Smart Agent with collectd/nginx monitor type. See NGINX.

Redis

Redis receiver.

Discover active metric sources ๐Ÿ”—

To discover any active and supported metric sources, run the following command on the desired monitoring host:

bin/otelcol --discovery --dry-run

The --dry-run option ensures that the resulting configuration isnโ€™t applied to the Collector at runtime. The sample configuration appears in the console as YAML instead. For example:

$ Discovering for next 10s...
Partially discovered "smartagent/postgresql" using "docker_observer"
endpoint "5c9c80ba4319395c26255b6374f048ca973d3618fdd4b92a9ed601c7dddbff6a:5432":
Please ensure your user credentials are correctly specified with
`--set splunk.discovery.receivers.smartagent/postgresql.config.params::username="<username>"`
and `--set splunk.discovery.receivers.smartagent/postgresql.config.params::password="<password>"`
or `SPLUNK_DISCOVERY_RECEIVERS_smartagent_x2f_postgresql_CONFIG_params_x3a__x3a_username="<username>"`
and `SPLUNK_DISCOVERY_RECEIVERS_smartagent_x2f_postgresql_CONFIG_params_x3a__x3a_password="<password>"`
environment variables.

When discovery mode canโ€™t access a discovered service to extract metric data, it provides instructions and the original log error message. In the example, discovery mode canโ€™t authenticate to the discovered PostgreSQL server due to missing or incorrect credentials, which you can provide through custom discovery properties. See Configure or fix discovery properties.

Note

The Linux installation script of the Collector supports the --discovery option. When turning on discovery mode through the installation script, the resulting configuration is applied directly to the metrics pipeline. For example:

curl -sSL https://dl.signalfx.com/splunk-otel-collector.sh > /tmp/splunk-otel-collector.sh && \
sudo sh /tmp/splunk-otel-collector.sh --realm <realm> โ€“ <token> --mode agent --discovery

Configure or fix discovery properties ๐Ÿ”—

To fix most of the issues identified by discovery mode, add or edit the configuration settings suggested in the status messages. You can define the required settings in the following ways:

  • Use the --set option to specify settings to be used by discovery mode at runtime. For example:

    --set splunk.discovery.receivers.smartagent/postgresql.config.params::username='${PG_USERNAME_ENVVAR}'
    
  • Set the environment variable for the setting. Each discovery property has an equivalent environment variable form using _x<hex pair>_ encoded delimiters for non-word characters [^a-zA-Z0-9_]:

    For example:

    export SPLUNK_DISCOVERY_RECEIVERS_smartagent_x2f_postgresql_CONFIG_params_x3a__x3a_username='${PG_USERNAME_ENVVAR}'
    
  • Define the properties in the config.d/properties.discovery.yaml file. See Define properties through the properties file.

When issues are detected, discovery mode suggests which parameters and environment variables youโ€™ve to use to complete the missing configuration settings.

Note

By default, the duration of the discovery process is 10 seconds, which you can increase by setting the SPLUNK_DISCOVERY_DURATION environment variable. For example: export SPLUNK_DISCOVERY_DURATION = 20s.

Customize discovery settings ๐Ÿ”—

By default, discovery mode reads the built-in configuration provided by the Collector executable. You can provide your own configuration to modify settings or adjust the existing configuration in case of a partial discovery status.

The priority order for discovery configuration values from lowest to highest is:

  • Default bundle.d component configuration files, built into the Collector executable

  • config.d/<receivers or extensions>/*.discovery.yaml component configuration files

  • config.d/properties.discovery.yaml properties file content in mapped form

  • config.d/properties.discovery.yaml properties file content using --set form

  • SPLUNK_DISCOVERY_<xyz> property environment variables available to the Collector process

  • --set splunk.discovery.<xyz> property command line options

Define properties through the properties file ๐Ÿ”—

You can override or add properties by creating the etc/otel/collector/config.d/properties.discovery.yaml file. Each mapped property in the file overrides existing discovery settings. For example:

splunk.discovery:
  receivers:
     smartagent/postgresql:
      config:
        params:
          username: "${PG_USERNAME_ENVVAR}"
          password: "${PG_PASSWORD_ENVVAR}"

You can use the --discovery-properties=<filepath.yaml> argument to load discovery mode properties that you donโ€™t want to share with other Collectors. If you specify discovery properties using this argument, properties contained in config.d/properties.discovery.yaml are ignored.

Create custom configurations ๐Ÿ”—

To create custom discovery configurations, follow these steps:

  1. Navigate to the config.d folder in /etc/otel/collector/config.d on Linux.

  2. Create a <name>.discovery.yaml file and place it inside a subdirectory of config.d, for example extensions or receivers where <name> is the name of the component you want to use.

  3. Edit the <name>.discovery.yaml files to add the desired configuration. For example, if youโ€™re adding a receiver, discovery mode loads the content inside the receivers object of the Collector configuration.

Custom configurations consist of the fields you want to override in the default configuration. For example:

# <some-receiver-type-with-optional-name.discovery.yaml>
  <receiver_type>(/<receiver_name>):
     enabled: <true | false> # true by default
     rule:
        <observer_type>(/<observer_name>): <receiver creator rule for this observer>
     config:
        default:
           <default embedded receiver config>
        <observer_type>(/<observer_name>):
           <observer-specific config items, merged with `default`>
     status:
        metrics:
           <discovery receiver metric status entries>
        statements:
           <discovery receiver statement status entries>

Use the --dry-run option to check the resulting discovery configuration before using it with the Collector.

See the Discovery receiver README file for more information.

Define a custom configuration directory ๐Ÿ”—

To define a custom directory for discovery settings, use the --config-dir option as in the example:

otelcol --discovery --config-dir <custom_path>

Usage example ๐Ÿ”—

The following example shows how to install the Collector on Linux using discovery mode to find a MySQL database and retrieve metrics.

  1. Install the Collector on the host where MySQL is running. Include the --discovery flag:

    curl -sSL https://dl.signalfx.com/splunk-otel-collector.sh > /tmp/splunk-otel-collector.sh && \
    sudo sh /tmp/splunk-otel-collector.sh --realm <realm> โ€“ <token> --mode agent --discovery
    
  2. Retrieve the Collector logs with the following command and review the output of the discovery process:

    journalctl -u splunk-otel-collector -f
    

    In the following sample logs, the MySQL database has been partially discovered. The error message indicates the problem, which in this case is bad credentials:

    Partially discovered "smartagent/collectd/mysql" using "host_observer" endpoint "(host_observer)[::]-3306-TCP-1757": Make sure y
    our user credentials are correctly specified using the `--set splunk.discovery.receivers.smartagent/collectd/mysql.config.username="<username>"` and `--set splunk.discovery.receivers.smartagent/collect
    d/mysql.config.password="<password>"` command or the `SPLUNK_DISCOVERY_RECEIVERS_smartagent_x2f_collectd_x2f_mysql_CONFIG_username="<username>"` and `SPLUNK_DISCOVERY_RECEIVERS_smartagent_x2f_collectd_
    x2f_mysql_CONFIG_password="<password>"` environment variables. (evaluated "{\"collectdInstance\":\"monitor-smartagentcollectdmysqlreceiver_creatorhost_observerendpoint3306host_observer3306TCP1757\",\"k
    ind\":\"receiver\",\"message\":\"mysql plugin: Failed to connect to database splunk.discovery.default at server ::: Access denied for user 'splunk.discovery.default'@'localhost' (using password: YES)\"
    ,\"monitorType\":\"collectd/mysql\"}")
    
  3. Provide the necessary credentials by creating the properties.discovery.yaml file in the /etc/otel/collector/config.d directory with the following content:

    splunk.discovery:
      receivers:
        smartagent/collectd/mysql:
          enabled: true
          config:
            username: "<username>"
            password: "<password>"
            databases: "[{name: '<database name>'}]"
    
  4. Restart the Collector with the following command:

    sudo systemctl restart splunk-otel-collector
    
  5. Tail the Collector logs again to confirm that it has discovered the MySQL database successfully:

    journalctl -u splunk-otel-collector -f
    
  6. When successful, the logs include a line similar to the following:

    Successfully discovered "smartagent/collectd/mysql" using "host_observer" endpoint "(host_observer)[::]-3306-TCP-1757".
    

Troubleshooting ๐Ÿ”—

If you are a Splunk Observability Cloud customer and are not able to see your data in Splunk Observability Cloud, you can get help in the following ways.

Available to Splunk Observability Cloud customers

Available to prospective customers and free trial users

  • Ask a question and get answers through community support at Splunk Answers .

  • Join the Splunk #observability user group Slack channel to communicate with customers, partners, and Splunk employees worldwide. To join, see Chat groups in the Get Started with Splunk Community manual.

To learn about even more support options, see Splunk Customer Success .