Run Powerpipe

Powerpipe is simple to install and manage and does not require any special expertise to get started. It's distributed as a single binary file - just download and run it!

Mod Location

Powerpipe always runs in the context of a mod, which is a collection of Powerpipe benchmarks, controls, dashboards, and queries. You can import and use resources from other mods so you can get started without even writing any code! You can explore the available mods on the Powerpipe Hub.

Powerpipe loads the mod from the current directory by default, but you can pass the --mod-location flag or set the POWERPIPE_MOD_LOCATION to set it to a different path.

Configuration Files

Powerpipe will load configuration files (*.ppc) according to the configuration search path. You can change this path with the --config-path argument or the POWERPIPE_CONFIG_PATH environment variable, but it defaults to .:$POWERPIPE_INSTALL_DIR/config (.:~/.powerpipe/config). This allows you to manage your workspaces centrally in the ~/.powerpipe/config directory, but override them in the mod location if desired.

Operating Modes

Powerpipe can operate in 2 modes.

By default, Powerpipe runs in Client-only Mode. Powerpipe loads the mod, runs the command, and exits. Interactive dashboards are not enabled in Client-Only Mode.

If you run Powerpipe in Server Mode mode, Powerpipe will run a dashboard server (on port 9033 by default). In this mode, you can browse Interactive dashboards by navigating to http://localhost:9033/ in your web browser.

Selecting a database

Most public mods are written without specifying a database on each query. As a result, all queries in the mod run against the 'active' database. For Steampipe, the active database is postgres://steampipe@localhost:9193/steampipe, thus Powerpipe will run against a local Steampipe instance. For Powerpipe, it's duckdb:~/.tailpipe/data/default/tailpipe.db. ` Some mods allow you to set the database via a variable. In this case, you can pass the connection to the variable when you run a powerpipe command:

or start the dashboard server:

For the database connection types (connection.steampipe, connection.postgres, connection.mysql, connection.duckdb, connection.sqlite, connection.tailpipe) Powerpipe even supports passing them as connection strings:

or pipes workspaces (you will need to log in first)!

Targeting specific Postgres/Steampipe schemas/connections

A PostgreSQL database contains one or more schemas. A schema is a namespaced collection of named objects, like tables, functions, and views. When writing a query, you may qualify the table name with the schema name (e.g. select * from schema_name.table_name) or use an unqualified name (select * from table_name). When using unqualified names, the system determines which table is meant by following a search path; the first matching table in the search path is taken to be the one wanted.

Usually, Powerpipe mods use unqualified queries to "target" whichever connection is first in the search path, but you can specify a different path if you want:

The --search-path argument will replace the entire search path. Often you just want to move a single connection to the front of the path:

Steampipe creates a schema for each connection and aggregator, and understanding how the search path works is important when using Steampipe and Powerpipe. See the Using search_path to target connections and aggregators guide for more information.