Why do we use circleci/openjdk:11-jdk for the deploy job? (Done , changed to 'cimg/node:19.8.)
Ideally, the work of configuring the server (installing software) should have been handled in avni-infra, and the CI should just hand over the responsibility to Ansible. See here on how we did this for avni-etl. By doing it this way, we can be sure that the scripts on Ansible are being used. Right now we are missing this functionality on the avni-infra repository, and have another version on the avni-media repository which will need to be removed later
Since this is working, we can keep it while I create a card to fix this.
Important note:
Config for reference
Client-config:
NEXT_PUBLIC_IMAGE_LIST_URL=https://localhost:6013/etl/media NEXT_PUBLIC_DOWNLOAD_REQUEST_URL=http://localhost:3010/media-viewer/requestDownload NEXT_PUBLIC_DOWNLOAD_LIST_URL=http://localhost:3010/media-viewer/allData
http://localhost:3010 === NESTjs
Server-config:
AWS_ACCESS_KEY_ID={{AWS_ACCESS_KEY}} AWS_SECRET_ACCESS_KEY={{AWS_SECRET_KEY}} S3_BUCKET_NAME={{STAGING/PROD_S3_BUCKET_NAME}} S3_ZIP_UPLOAD_BUCKET_NAME={{STAGING/PROD_S3_BUCKET_NAME}} AWS_REGION={{AWS_DEFAULT_REGION}}
DB_HOST={{STAGING/PROD_DB_HOST}} DB_PORT={{DB_PORT}} DB_USERNAME={{DB_USER}} DB_PASSWORD={{DB_PWD}} DB_DATABASE={{DB_DB}}