Toolset to dump postgresql data into a pgdata tarball
- you need a service account with permissions to create a clone postgres SQL
- copy the yaml file from samples/ and adjust parameters to your infrastructure
- apply the yaml to your cluster to:
- create an SQL clone
- dump the whole database locally (watch out, you will need a PVC on disk if that DB is larger than a few GBs)
- pg_restore that whole database into a local (in-pod) postgresql instance and stop it
- tarball the PGDATA folder
- Upload that tarball to google storage
Then, you are ready to use that tarball in an integration test, with https://github.com/streamingfast/graphnode-operator