Start the Native Backend

The Java microservices provide CEDAR's artifact, repository, user, terminology, submission, and supporting APIs. They all use the infrastructure configured on the previous page and run from the JARs produced by cedarcli build java.

Connect CEDAR to BioPortal

CEDAR uses BioPortal to search ontologies and controlled terms. Create a BioPortal account, obtain its API key, and set it in $CEDAR_HOME/set-env-external.sh:

export CEDAR_BIOPORTAL_REST_BASE="https://data.bioontology.org/"
export CEDAR_BIOPORTAL_API_KEY="your-api-key"

The terminology service reads these values when it starts. Restart that service after changing them.

Build and Start the Microservices

If the Java source has not been built yet, build it in dependency order:

cedarcli build java

Start the complete microservice tier and inspect the result:

cedarcli native start microservices
cedarcli native status

Every microservice should report a healthy application port and a current binary. A STALE binary means the service is healthy but still running an older JAR and must be restarted. Frontends will still appear stopped until the next page.

If a service does not become healthy, follow its log by process name. For example:

cedarcli native logs resource

Initialize an Empty Installation

A new installation needs its initial users, folders, permissions, and search state. Run the system reset only after the infrastructure and microservices are healthy:

CEDAR_PROFILE=develop source "$CEDAR_HOME/cedar-development/bin/templates/cedar-profile-native.sh"
cedarat system-reset

Confirm the reset when prompted. This command deletes existing CEDAR content, so do not use it to repair a development installation whose data you intend to keep.

For later backend work, individual services can be restarted without disturbing the rest of the stack:

cedarcli native restart terminology
cedarcli native stop microservice terminology
cedarcli native start microservice terminology