Runs a Directory sync service at a biobank site.
This service keeps the BBMRI Directory up to date with the number of samples, etc. kept in the biobank. It also updates the local FHIR store with the latest contact details etc. from the Directory.
It is assumed that you have access to a FHIR store containing information about patients and samples, as well as the details of your biobank. The data in this store should conform to the GBA profile.
If you are running a Bridgehead, then you will already have a suitable store, running under Blaze.
The service is able to use Quartz to start synchronization at regular intervals. These are specified using cron syntax. If no cron information is supplied, Directory sync will only be run once.
You need to provide URL, user and password for the Directory.
Additionally, you will need to supply a URL for the local FHIR store.
See below for ways to get these parameters to the application.
It is generally advised that you assign collections to all Specimens. This is best achieved using an extension to the Specimen resource, which contains a reference to a collection, see the FHIR profile. The IDs of these collections will be passed on to the Directory during synchronization.
However, for some biobanks it may not be feasible to assign individual specimens to a collection. A fallback solution is available in this case: the default collection ID. This is a single ID that will be assigned to all specimens without collection references. There are two possible ways to specify this ID: either as environment variable (see DS_DIRECTORY_DEFAULT_COLLECTION_ID in the table below) or as an orphan collection in FHIR that has no references from any specimens but whose identifier is a collection ID.
We recommend using Docker for running Directory sync.
First, you will need to set up the environment variables for this:
| Variable | Purpose | Default if not specified |
|---|---|---|
| DS_DIRECTORY_URL | Base URL of the Directory | https://directory-backend.molgenis.net |
| DS_DIRECTORY_USER_NAME | User name for logging in to Directory | |
| DS_DIRECTORY_USER_PASS | Password for logging in to Directory | |
| DS_DIRECTORY_DEFAULT_COLLECTION_ID | ID of collection to be used if not in samples | |
| DS_DIRECTORY_MIN_DONORS | Minimum number of donors per star model hypercube | 10 |
| DS_DIRECTORY_MAX_FACTS | Max number of star model hypercubes to be generated | |
| DS_DIRECTORY_ALLOW_STAR_MODEL | Set to 'True' to send star model info to Directory | False |
| DS_DIRECTORY_MOCK | Set to 'True' mock a Directory. In this mode, directory-sync will not contact the Directory. All Directory-related methods will simply return plausible fake values. | False |
| DS_DIRECTORY_ONLY_LOGIN | Set to 'True' to log in to the Directory, without carrying out any further actions. | False |
| DS_FHIR_STORE_URL | URL for FHIR store | http://bridgehead-bbmri-blaze:8080 |
| DS_TIMER_CRON | Execution interval for Directory sync, cron format | |
| DS_RETRY_MAX | Maximum number of retries when sync fails | 10 |
| DS_RETRY_INTERVAL | Interval between retries (seconds) | 20 seconds |
DS_DIRECTORY_USER_NAME and DS_DIRECTORY_USER_PASS are mandatory. If you do not specify these, Directory sync will not run. Contact Directory admin to get login credentials.
DS_DIRECTORY_DEFAULT_COLLECTION_ID is not mandatory, but you will need to specify it if your ETL does not assign Directory collection IDs to specimens.
If you set DS_DIRECTORY_ALLOW_STAR_MODEL to 'True', then the star model summary information for your data will be generated and sent to the Directory. You are advised to talk to your local data protection group before doing this.
If DS_TIMER_CRON is not specified, Directory sync will be executed once, and then the process will terminate.
The DS_RETRY_ variables specify how Directory sync reacts to failure. RETRY_MAX should be greater than zero. A typical case where retries may be necessary is when you simultaneously start Directroy sync and FHIR store. It takes some time for a FHIR store to start, whereas Directory sync starts immediately, so Directory sync needs to have a way to poll the store until is ready.
If DS_DIRECTORY_MOCK is set to 'True', the Directory will not be contacted, but you will still need to specify login credentials (which will then be ignored).
For your convenience, we recommend that you store these variables in a .env file. The file could look like this:
DS_DIRECTORY_URL=https://bbmritestnn.gcc.rug.nl
[email protected]
DS_DIRECTORY_USER_PASS=qwelmqwldmqwklmdLKJNJKNKJN
DS_FHIR_STORE_URL=http://store:8080/fhir
With this configuration, the synchronization will be done just once, with the test Directory server. The login details are fake, you will need to replace them with something sensible.
To run from Docker:
docker run --env-file .env samply/directory_sync_service
Alternatively, you can use docker-compose. An example docker-compose.yml file has been included in this repository. To start:
docker-compose up
You can also build and run this service on your own computer, if Docker is not an option.
Please run from within Linux.
You will need to install git, Java 19 and Maven. Clone this repository and cd into it.
From the command line, run:
mvn clean install
You need to provide the necessary parameters via environment variables, as documented above. These should be set in the same shell as the one that you use to run java.
Once you have done this, you can start the service directly from the command line:
java -jar target/directory_sync_service\*.jar
Copyright 2024 The Samply Community
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.