Configuration
Torch is configured using environment variables that control its behavior and integration with FHIR servers. This allows for flexible deployment and customization based on the specific requirements of your environment.
Configuration Overview
This document provides an overview of the environment variables used to configure TORCH, including their default values and descriptions. These variables can be set in the application.yml file or as system environment variables when running the application.
When running TORCH as a container, it is recommended to set these variables or pass them as environment variables during container startup.
Backend
Environment Variables
SERVER_PORT
Since 1.0.0-alpha
The port on which the server listens for incoming connections.
Default: 8080
TORCH_PROFILE_DIR
Since 1.0.0-alpha
The directory path where profile definitions are stored.
Default: structureDefinitions
TORCH_MAPPING_CONSENT
Since 1.0.0-alpha
Path to the JSON file containing consent mappings formatted for FHIR.
Default: mappings/consent-mappings_fhir.json
TORCH_MAPPING_TYPE_TO_CONSENT
Since 1.0.0-alpha
File that maps FHIR Resource Types to their associated time fields used to check consent validity.
Default: mappings/type_to_consent.json
TORCH_FHIR_USER
Since 1.0.0-alpha
The username used for authentication with the FHIR server.
Default: – (none)
TORCH_FHIR_PASSWORD
Since 1.0.0-alpha
The password used for authentication with the FHIR server.
Default: – (none)
TORCH_FHIR_OAUTH_ISSUER_URI
Since 1.0.0-alpha
The URI of the OAuth issuer used for OAuth authentication.
Default: – (none)
TORCH_FHIR_OAUTH_CLIENT_ID
Since 1.0.0-alpha
Client ID used for OAuth authentication.
Default: – (none)
TORCH_FHIR_OAUTH_CLIENT_SECRET
Since 1.0.0-alpha
Client secret used for OAuth authentication.
Default: – (none)
TORCH_FHIR_URL
Since 1.0.0-alpha
Base URL of the FHIR server that TORCH connects to.
Default: – (none)
TORCH_FHIR_MAX_CONNECTIONS
Since 1.0.0-alpha
Maximum number of concurrent connections allowed to the FHIR server. Must be equal to TORCH_MAXCONCURRENCY + 1
.
Default: 5
TORCH_FHIR_PAGE_COUNT
Since 1.0.0-alpha
Number of entries per page in FHIR search responses.
Default: 500
TORCH_FHIR_DISABLE_ASYNC
Since 1.0.0-alpha
Set to true
to disable the use of the Asynchronous Interaction Request Pattern for FHIR operations.
Default: false
TORCH_FLARE_URL
Since 1.0.0-alpha
Base URL of the FLARE server used in the pipeline.
Default: – (none)
TORCH_RESULTS_DIR
Since 1.0.0-alpha
Directory path where results are stored.
Default: output/
TORCH_RESULTS_PERSISTENCE
Since 1.0.0-alpha
ISO 8601 duration indicating how long result files are persisted (e.g., PT2160H
means 90 days).
Default: PT2160H
TORCH_BATCHSIZE
Since 1.0.0-alpha
Size of data batches processed at once.
Default: 500
TORCH_MAXCONCURRENCY
Since 1.0.0-alpha
Maximum level of concurrency for data processing operations.
Default: 4
TORCH_MAPPINGS_FILE
Since 1.0.0-alpha
Path to the file containing ontology mappings defined using Clinical Quality Language (CQL).
Default: ontology/mapping_cql.json
TORCH_BUFFERSIZE
Since 1.0.0-alpha
Size (in MB) of the buffer used by the web client interacting with the FHIR server.
Default: 100
TORCH_CONCEPT_TREE_FILE
Since 1.0.0-alpha
File containing the concept tree mapping used for resource classification.
Default: ontology/mapping_tree.json
TORCH_DSE_MAPPING_TREE_FILE
Since 1.0.0-alpha
File containing the concept tree mapping specifically for DSE (Data Set Extensions).
Default: ontology/dse_mapping_tree.json
TORCH_USE_CQL
Since 1.0.0-alpha
Flag indicating whether to enable Clinical Quality Language (CQL) for processing.
Default: true
TORCH_BASE_URL
Since 1.0.0-alpha
Base server URL before any proxy, used for accessing TORCH directly.
Default: – (none)
TORCH_OUTPUT_FILE_SERVER_URL
Since 1.0.0-alpha
URL used to access the result files in TORCH_RESULTS_DIR
via a proxy or file server.
Default: – (none)
LOG_LEVEL_DE_MEDIZININFORMATIKINITIATIVE_TORCH
Since 1.0.0-alpha
Logging level for core TORCH functionality.
Default: info
LOG_LEVEL_CA_UHN_FHIR
Since 1.0.0-alpha
Logging level for the HAPI FHIR library.
Default: error
SPRING_PROFILES_ACTIVE
Since 1.0.0-alpha
Active Spring profile used to configure the application context.
Default: active
SPRING_CODEC_MAX_IN_MEMORY_SIZE
Since 1.0.0-alpha
Maximum allowed in-memory size for Spring codecs when processing data.
Default: 100MB