We need to look into what is already there in deployment/choreo/development/docker/mongodb/Dockerfile. The idea is that we must use the backed up data already in https://github.com/LDFLK/data-backups/ to this image. The idea is to build a container image which already has data baked into it. The reason is when we use the cloud environment, this image must come up as quickly as possible with all data and serve. If we chose the loading the container without data and then do the back up, it won't scale as it will take a long time to backup and release.
We already did this for the neo4j and it is here: deployment/choreo/development/docker/neo4j/Dockerfile. There are certain things like the Choreo user requirement when deploying in Choreo. Other than that the rest of the stack is pretty much normal.
There is another example for Postgres but it is work in progress here: #447
You can study the databack script here deployment/development/init.sh, it has the way to restore the back from GitHub for Mongodb as well.
We need to look into what is already there in
deployment/choreo/development/docker/mongodb/Dockerfile. The idea is that we must use the backed up data already in https://github.com/LDFLK/data-backups/ to this image. The idea is to build a container image which already has data baked into it. The reason is when we use the cloud environment, this image must come up as quickly as possible with all data and serve. If we chose the loading the container without data and then do the back up, it won't scale as it will take a long time to backup and release.We already did this for the
neo4jand it is here:deployment/choreo/development/docker/neo4j/Dockerfile. There are certain things like the Choreo user requirement when deploying in Choreo. Other than that the rest of the stack is pretty much normal.There is another example for Postgres but it is work in progress here: #447
You can study the databack script here
deployment/development/init.sh, it has the way to restore the back from GitHub for Mongodb as well.