Skip to main content
Unlisted page
This page is unlisted. Search engines will not index it, and only users having a direct link can access it.

🐳 On-prem

Fused offers an on-prem version of our application in a Docker container. The container runs in your computing environment (such as AWS, GCP, or Azure) and your data stays under your control.

Our container is currently distributed via a private release. Email info@fused.io for access.

Commands​

run-config​

run-config runs the user's jobs. The job configuration can be specified either on the command line, as a local file path, or as an S3/GCS path. In all cases the job configuration is loaded as JSON.

Options:
  --config-from-gcs FILE_NAME   Job step configuration, as a GCS path
  --config-from-s3 FILE_NAME    Job step configuration, as a S3 path
  --config-from-file FILE_NAME  Job step configuration, as a file name the
                                application can load (i.e. mounted within the
                                container)
  -c, --config JSON             Job configuration to run, as JSON
  --help                        Show this message and exit.

version​

Prints the container version and exits.

Environment Variables​

The on-prem container can be configured with the followin environment variables.

  • FUSED_AUTH_TOKEN: Fused token for the licensed user or team. When using the FusedDockerAPI, this token is automatically retrieved.
  • FUSED_DATA_DIRECTORY: The path to an existing directory to be used for storing temporary files. This can be the location a larger volume is mounted inside the container. Defaults to Python's temporary directory.
  • FUSED_GCP: If "true", enable GCP specific features. Defaults to false.
  • FUSED_AWS: If "true", enable AWS specific features. Defaults to false.
  • FUSED_AWS_REGION: The current AWS region.
  • FUSED_LOG_MIN_LEVEL: Only logs with this level of severity or higher will be emitted. Defaults to "DEBUG".
  • FUSED_LOG_SERIALIZE: If "true", logs will be written in serialized, JSON form. Defaults to false.
  • FUSED_LOG_AWS_LOG_GROUP_NAME: The CloudWatch Log Group to emit logs to. Defaults to not using CloudWatch Logs.
  • FUSED_LOG_AWS_LOG_STREAM_NAME: The CloudWatch Log Stream to create and emit logs to. Defaults to not using CloudWatch Logs.
  • FUSED_PROCESS_CONCURRENCY: The level of process concurrency to use. Defaults to the number of CPU cores.
  • FUSED_CREDENTIAL_PROVIDER: Where to obtain AWS credentials from. One of "default" (default to ec2 on AWS, or none otherwise), "none", "ec2" (use the EC2 instance metadata), or "earthdata" (use EarthData credentials in FUSED_EARTHDATALOGIN_USERNAME and FUSED_EARTHDATALOGIN_PASSWORD).
  • FUSED_EARTHDATALOGIN_USERNAME: Username when using earthdata credential provider, above.
  • FUSED_EARTHDATALOGIN_PASSWORD: Password when using earthdata credential provider, above.
  • FUSED_IGNORE_ERRORS: If "true", continue processing even if some computations throw errors. Defaults to false.
  • FUSED_DISK_SPACE_GB: Maximum disk space available to the job, e.g. for temporary files on disk, in gigabytes.