Load data into CrateDB using collectd

This usage guide shows how to configure and start collectd and CrateDB so that collectd sends system metrics and CrateDB stores them.

Prerequisites

Use Docker or Podman to run all components. This approach works consistently across Linux, macOS, and Windows.

Files

First, download and save all required files to your machine.

• compose.yaml

• Dockerfile

• collectd-cratedb.conf

Services

Start services using Docker Compose or Podman Compose. If you use Podman, replace docker with podman (or enable the podman‑docker compatibility shim) and run podman compose up.

docker compose up

To send the collected data to CrateDB, collectd is configured to load its postgresql plugin.

collectd configuration collectd-cratedb.conf

# collectd configuration for storing metrics into CrateDB.
# https://cratedb.com/docs/guide/integrate/collectd/

# https://collectd.org/documentation/manpages/collectd.conf.html#plugin-postgresql

LoadPlugin postgresql

<Plugin postgresql>
        <Writer sqlstore>
                Statement "INSERT INTO doc.collectd_data (p_time, p_host, p_plugin, p_plugin_instance, p_type, p_type_instance, p_value_names, p_type_names, p_values) VALUES ($1, $2, $3, $4, $5, $6, $7, $8, $9);"
                StoreRates true
        </Writer>
        <Database cratedb>
                Writer sqlstore
                # Commit every 30 seconds (adjust based on write volume)
                CommitInterval 30
                Host cratedb
                User crate
                Password crate
        </Database>
</Plugin>

Provision database

Create a database table that stores collected metrics.

docker compose run --rm --no-TTY postgresql psql "postgresql://crate:crate@cratedb:5432/" <<SQL
CREATE TABLE doc.collectd_data (
   p_time timestamp with time zone,
   p_host TEXT,
   p_plugin TEXT,
   p_plugin_instance TEXT,
   p_type TEXT,
   p_type_instance TEXT,
   p_value_names TEXT,
   p_type_names TEXT,
   p_values TEXT,
   month GENERATED ALWAYS AS date_trunc('month',p_time)
) PARTITIONED BY (month);
SQL

Explore data

After the first scraping interval, metrics will show up in the designated table in CrateDB, ready to be inspected.

docker compose run --rm --no-TTY postgresql psql "postgresql://crate:crate@cratedb:5432/" -c "SELECT * FROM doc.collectd_data ORDER BY p_time LIMIT 5;"

           p_time           |    p_host    | p_plugin  | p_plugin_instance |   p_type   | p_type_instance | p_value_names |   p_type_names    |   p_values   |           month
----------------------------+--------------+-----------+-------------------+------------+-----------------+---------------+-------------------+--------------+----------------------------
 2025-09-20 13:57:12.822+00 | 9cde293016c2 | interface | gre0              | if_errors  |                 | {'rx','tx'}   | {'gauge','gauge'} | {nan,nan}    | 2025-09-01 00:00:00.000+00
 2025-09-20 13:57:12.822+00 | 9cde293016c2 | memory    |                   | memory     | cached          | {'value'}     | {'gauge'}         | {4600500224} | 2025-09-01 00:00:00.000+00
 2025-09-20 13:57:12.822+00 | 9cde293016c2 | interface | gre0              | if_dropped |                 | {'rx','tx'}   | {'gauge','gauge'} | {nan,nan}    | 2025-09-01 00:00:00.000+00
 2025-09-20 13:57:12.822+00 | 9cde293016c2 | interface | erspan0           | if_octets  |                 | {'rx','tx'}   | {'gauge','gauge'} | {nan,nan}    | 2025-09-01 00:00:00.000+00
 2025-09-20 13:57:12.822+00 | 9cde293016c2 | interface | ip_vti0           | if_dropped |                 | {'rx','tx'}   | {'gauge','gauge'} | {nan,nan}    | 2025-09-01 00:00:00.000+00
(5 rows)