Running in Production
To run xpanse in production, we must ensure that all necessary dependent systems in the stack are also deployed together as a stack.
All required profiles and configurations must be known beforehand.
Run using jar
Download the released runtime jar from GitHub releases. You can get the latest from here.
After downloading, we can start the jar file using the Java command.
java -jar xpanse-runtime-*.jar
Run using Docker image
You can start the runtime using our released docker image, and this is the preferred way. This image contains all necessary tools preinstalled. Below is an example of running a single instance of xpanse.
$ docker pull ghcr.io/eclipse-xpanse/xpanse:${release-version}
$ docker run -d -p 8080:8080 --name xpanse -e "SPRING_PROFILES_ACTIVE=oauth,zitadel,mysql" ghcr.io/eclipse-xpanse/xpanse:${release-version}
$ docker logs xpanse
It's safe to provide all properties as environment variables rather than passing them directly in the command line.
In case of this, the same property name must be set in UPPERCASE for all four variables.
For running, using docker image, we can use the --env-file option of the docker run command to store all sensitive data.
Again, here the property names must be in UPPERCASE.
Running API behind a proxy
For running the runtime application behind a proxy, we must ensure that the proxy forwards the
correct X-Forwarded-*headers to the API.
This is necessary since the API has some features where the
links to HTML pages are returned and this link will have the correct protocol and host only when
these headers are set.
In the case of NGINX, the configuration will look like this
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto https;
proxy_set_header Host $host;