Home Assistant uses database to store events and parameters for history and tracking. The default database used is SQLite and the database file is stored in your configuration directory (e.g., <path to config dir>/home-assistant_v2.db); however, other databases can be used. If you prefer to run a database server (e.g., PostgreSQL), use the recorder integration.

To work with SQLite database manually from the command-line, you will need an installation of sqlite3. Alternatively DB Browser for SQLite provides a viewer for exploring the database data and an editor for executing SQL commands. First load your database with sqlite3:

$ sqlite3 home-assistant_v2.db
SQLite version 3.13.0 2016-05-18 10:57:30
Enter ".help" for usage hints.

It helps to set some options to make the output more readable:

sqlite> .header on
sqlite> .mode column

You could also start sqlite3 and attach the database later. Not sure what database you are working with? Check it, especially if you are going to delete data.

sqlite> .databases
seq  name             file
---  ---------------  ----------------------------------------------------------
0    main             /home/fab/.homeassistant/home-assistant_v2.db


Get all available tables from your current Home Assistant database:

sqlite> SELECT sql FROM sqlite_master;

	event_id INTEGER NOT NULL, 
	event_type VARCHAR(32), 
	event_data TEXT, 
	origin VARCHAR(32), 
	time_fired DATETIME, 
	created DATETIME, 
	context_id VARCHAR(36), 
	context_user_id VARCHAR(36), context_parent_id CHARACTER(36), 
	PRIMARY KEY (event_id)
CREATE TABLE recorder_runs (
	start DATETIME, 
	"end" DATETIME, 
	closed_incorrect BOOLEAN, 
	created DATETIME, 
	PRIMARY KEY (run_id), 
	CHECK (closed_incorrect IN (0, 1))
CREATE TABLE schema_changes (
	change_id INTEGER NOT NULL, 
	schema_version INTEGER, 
	changed DATETIME, 
	PRIMARY KEY (change_id)
	state_id INTEGER NOT NULL, 
	domain VARCHAR(64), 
	entity_id VARCHAR(255), 
	state VARCHAR(255), 
	attributes TEXT, 
	event_id INTEGER, 
	last_changed DATETIME, 
	last_updated DATETIME, 
	created DATETIME, 
	context_id VARCHAR(36), 
	context_user_id VARCHAR(36), context_parent_id CHARACTER(36), old_state_id INTEGER, 
	PRIMARY KEY (state_id), 
	FOREIGN KEY(event_id) REFERENCES events (event_id)
CREATE TABLE sqlite_stat1(tbl,idx,stat)
CREATE INDEX ix_events_context_user_id ON events (context_user_id)
CREATE INDEX ix_events_event_type ON events (event_type)
CREATE INDEX ix_events_context_id ON events (context_id)
CREATE INDEX ix_events_time_fired ON events (time_fired)
CREATE INDEX ix_recorder_runs_start_end ON recorder_runs (start, "end")
CREATE INDEX ix_states_entity_id ON states (entity_id)
CREATE INDEX ix_states_context_user_id ON states (context_user_id)
CREATE INDEX ix_states_last_updated ON states (last_updated)
CREATE INDEX ix_states_event_id ON states (event_id)
CREATE INDEX ix_states_entity_id_last_updated ON states (entity_id, last_updated)
CREATE INDEX ix_states_context_id ON states (context_id)
CREATE INDEX ix_states_context_parent_id ON states (context_parent_id)
CREATE INDEX ix_events_context_parent_id ON events (context_parent_id)

To only show the details about the states table (since we are using that one in the next examples):

sqlite> SELECT sql FROM sqlite_master WHERE type = 'table' AND tbl_name = 'states';


The identification of the available columns in the table is done and we are now able to create a query. Let’s list your Top 10 entities:

sqlite> .width 30, 10,
sqlite> SELECT entity_id, COUNT(*) as count FROM states GROUP BY entity_id ORDER BY count DESC LIMIT 10;
entity_id                       count
------------------------------  ----------
sensor.cpu                      28874
sun.sun                         21238
sensor.time                     18415
sensor.new_york                 18393
cover.kitchen_cover             17811
switch.mystrom_switch           14101
sensor.internet_time            12963
sensor.solar_angle1             11397
sensor.solar_angle              10440
group.all_switches              8018


If you don’t want to keep certain entities, you can delete them permanently by using the services provided by the recorder.

For a more interactive way of working with the database, check the Data Science Portal.