Grafana Admin study plan

Inventory every Grafana administration responsibility area (server configuration, authentication, data sources, dashboards and folders, provisioning, alerting, users and permissions, backups, upgrades, plugins) and write a one-page map of what you already know versus what is new

Install Grafana locally from the official binary or container image and start it with a custom configuration file rather than defaults

Locate and annotate the key sections of grafana.ini (server, database, security, auth, users, analytics, log, paths, plugins) in your own notes

Configure the instance to use an external PostgreSQL or MySQL database instead of the embedded SQLite store and confirm it starts cleanly

Configure environment-variable overrides for at least five settings and verify each takes effect over the config file

Set up provisioning directories for data sources and dashboards so that a fresh container starts fully configured with no manual clicks

Create an organization, a team, and at least three users, assigning Viewer, Editor, and Admin roles and testing what each can actually do

Enable anonymous access in a scratch instance, observe the effect, then disable it and document the risk

Configure LDAP or OAuth (generic OAuth or GitHub) authentication against a test directory or identity provider and map an external group to a Grafana role

Set up at least two data sources of different types and restrict one of them to a specific team via permissions

Create a service account with a scoped token and use it to call the HTTP API for a dashboard lookup

Explain and test the difference between basic role assignment and dashboard/folder-level permissions

Build a dashboard-as-code workflow: export a dashboard's JSON, store it in version control, and re-provision it after deletion

Create alert rules with notification policies and a contact point, then trigger and resolve a real alert using a test data source

Configure silences and mute timings and confirm each suppresses notifications as expected

Document the alerting evaluation path from rule to contact point in your own words

Perform a full backup: dump the Grafana database and copy the provisioning, plugins, and dashboard JSON directories to a separate location

Restore the backup into a clean instance and verify dashboards, users, and data sources survive

Configure and test SMTP so Grafana can send an invitation or alert email to a local test mailbox

Set up a reverse proxy with TLS in front of Grafana and confirm the root_url and cookie settings are correct

Upgrade Grafana across at least one minor version and record the pre-upgrade checks, database migration output, and rollback path

Install, enable, and then disable a plugin, noting how plugin signatures and the allow_loading_unsigned_plugins setting behave

Tune and interpret the metrics and logging settings: enable slow query logging, review the log output, and identify the health and metrics endpoints

Review the security settings (admin password, secret key, cookie settings, disable gravatar and update checks) and harden your instance

Diagnose three deliberately injected faults: a wrong database connection, an expired service account token, and a data source with bad credentials

Use Grafana's own logs, the health endpoint, and the API to identify each fault's root cause before fixing it

Write troubleshooting decision trees for authentication failures, data source errors, and alert delivery failures

Practice explaining each fix concisely as you would to a colleague in a ticket response

Re-read your week 1 knowledge map and close every remaining gap with hands-on practice

Complete a timed self-assessment of 40 or more questions you write yourself from your notes, targeting 80 percent or better

Rebuild the entire environment from your provisioning repository and backup procedure without consulting notes

Review the areas you scored lowest on and schedule a focused follow-up session for each