bodyshop/_reference/New Hasura Deployment.md

Postman Method: POST https://db.imex.online/v1alpha1/pg_dump Set x-hasura-admin-secret header. Body is RAW JSON:

{
  "opts": ["-O", "-x", "--schema-only", "--schema", "public"],
  "clean_output": true
}

Save output. Manually export hasura metadata on production.

Go to new instance. Run SQL

CREATE EXTENSION pg_trgm

Run SQL from PG Dump Import hasura metadata.

//Done before IO BETA Release Step 1: Nuke local migrations Delete all the contents of your local migrations directory.

$ rm migrations/* Step 2: Reset the migration history on server On the SQL tab of console, execute the following statement:

TRUNCATE hdb_catalog.schema_migrations; Step 3: Pull the schema and metadata from server Setup fresh migrations by taking the schema and metadata from the server:

(available after version alpha45)

create migration files (note that this will only export public schema from postgres)

$ hasura migrate create "init" --from-server

note down the version

mark the migration as applied on this server

$ hasura migrate apply --version "" --skip-execution If you are using schemas other than public, use --schema "schema_name" flag to indicate each one of them in the create command. This flag can be used multiple times. See more details about the usage in the docs.

Step 4: Verify the status Execute the following command to verify status of migration:

$ hasura migrate status You have brand new migrations now!

This can also be used to combine (kind of squash) all of your migration files into a single one. You're snapshotting the state of a server and adding it as a new migration.