README.md

Setting up postgres

By default, Scouter-Server uses postgres and is configured to run with both pg_partman and pg_cron extensions in order to manage partitions and schedule tasks. For an example of setting this up with a local postgres instance, see the docker-compose.yml and Dockerfile files.

User Roles and DB Config

Upon application startup, Scouter-Server will attempt to run an initial migration to setup database tables, partitioning and cron configurations. Prior to application startup, you will need to create the scouter schema, as well as the pg_partman and pg_cron extensions and create a user and database for Scouter-Server to use (unless you are using a superuser). If using a non-superuser when running the application, the user will need to have a variety of permissions on the database and the scouter and cron schemas.

This must be run by a superuser when the database is created

CREATE SCHEMA if not exists scouter;
CREATE EXTENSION if not exists pg_partman SCHEMA scouter;
CREATE EXTENSION if not exists pg_cron;

Example non-superuser permissions

• This was tested on cloudsql with postgres 15

-- create user
CREATE USER my_user WITH PASSWORD 'my_pass';

-- scouter schema
GRANT ALL ON SCHEMA scouter TO my_user;
GRANT ALL ON ALL TABLES IN SCHEMA scouter TO my_user;
GRANT EXECUTE ON ALL FUNCTIONS IN SCHEMA scouter to my_user;
GRANT EXECUTE ON ALL PROCEDURES IN SCHEMA scouter TO my_user;
GRANT ALL PRIVILEGES ON TABLE scouter.part_config_sub TO my_user;
GRANT ALL PRIVILEGES ON TABLE scouter.part_config TO my_user;

-- pg cron
GRANT USAGE ON SCHEMA cron TO my_user;
GRANT ALL ON ALL TABLES IN SCHEMA cron TO my_user;