Best Practices for DataKit Self-Log Collection¶

DataKit does not collect its own logs by default, but provides a configuration entry to manually enable it. The following describes two ways to collect DataKit self-logs in different deployment environments.

Overview¶

DataKit self-logs are divided into two categories:

Main logs: Runtime logs of DataKit, recording collector running status, error information, etc.
GIN logs: Access logs of DataKit HTTP service, recording API request information

By default, the DataKit main log path is /var/log/datakit/log, and the GIN log path is /var/log/datakit/gin.log.

Do not enable debug log mode

Never enable DataKit debug logs while configuring self-log collection. Enabling debug mode will generate a large amount of logs, which may cause log loop collection and disk space exhaustion.

Configuration Methods¶

Host DeploymentKubernetes

In host deployment environments, you need to manually create a logging collector configuration file to collect DataKit self-logs.

Navigate to the conf.d/log directory under the DataKit installation directory, create a logging-dk-self.conf file (the filename can be customized), and configure it as follows:

[[inputs.logging]]
  logfiles = [
    "/var/log/datakit/log",
  ]

  # ========== Log Processing Configuration ==========
  source = "datakit"
  service = "datakit"

  # Optional: custom tags
  # [inputs.logging.tags]
  #   env = "production"

[[inputs.logging]]
  logfiles = [
    "/var/log/datakit/gin.log",
  ]

  # ========== Log Processing Configuration ==========
  source = "datakit-gin"
  service = "datakit"

After configuration, restart the DataKit service to make the configuration take effect.

Verify Configuration

After restarting, you can verify whether the configuration is effective through the datakit monitor command:

datakit monitor -V

In Kubernetes environments, DataKit configures self-log collection through Pod Annotation datakit/logs.

Edit the datakit.yaml file, find the annotations section in the Pod Template, and modify the datakit/logs Annotation to change "disable": true to "disable": false to enable log collection:

apiVersion: apps/v1
kind: DaemonSet
metadata:
  name: datakit
  namespace: datakit
spec:
  template:
    metadata:
      annotations:
        datakit/logs: |
          [
            {
              "disable":       false,
              "type":          "file",
              "path":          "/var/log/datakit/log",
              "source":        "datakit",
              "service":       "datakit",
              "storage_index": ""
            },
            {
              "disable":       false,
              "type":          "file",
              "path":          "/var/log/datakit/gin.log",
              "source":        "datakit-gin",
              "service":       "datakit",
              "storage_index": ""
            }
          ]

After modification, apply the configuration:

kubectl apply -f datakit.yaml

DataKit Pods will automatically restart to apply the new configuration.

Configuration Details¶

For detailed configuration field descriptions and advanced configuration options, see the Log Collection Configuration Documentation.

Notes¶

Log Loop Collection Risk¶

Avoid Log Loop Collection

Improper configuration may cause log loop collection:

DataKit collects its own logs → writes to log files → DataKit collects again → infinite loop

To avoid this problem:

Do not enable debug mode: Debug mode will generate a large amount of logs
Configure log level appropriately: Set [logging] level = "info" in datakit.conf
Monitor log volume: Regularly check log collection volume, and investigate promptly if abnormal growth is found