PostgreSQL monitoring with Netdata

Collects database health and performance metrics.

Requirements

• python-psycopg2 package. You have to install it manually and make sure that it is available to the netdata user, either using pip, the package manager of your Linux distribution, or any other method you prefer.

• PostgreSQL v9.4+

Following charts are drawn:

1. Database size MB

   • size

2. Current Backend Processes processes

   • active

3. Current Backend Process Usage percentage

   • used
   • available

4. Write-Ahead Logging Statistics files/s

   • total
   • ready
   • done

5. Checkpoints writes/s

   • scheduled
   • requested

6. Current connections to db count

   • connections

7. Tuples returned from db tuples/s

   • sequential
   • bitmap

8. Tuple reads from db reads/s

   • disk
   • cache

9. Transactions on db transactions/s

   • committed
   • rolled back

10. Tuples written to db writes/s

-   inserted
-   updated
-   deleted
-   conflicts

11. Locks on db count per type

    • locks

12. Standby delta KB

    • sent delta
    • write delta
    • flush delta
    • replay delta

13. Standby lag seconds

    • write lag
    • flush lag
    • replay lag

14. Average number of blocking transactions in db processes

    • blocking

Configuration

Edit the python.d/postgres.conf configuration file using edit-config from the Netdata config directory, which is typically at /etc/netdata.

cd /etc/netdata   # Replace this path with your Netdata config directory, if different
sudo ./edit-config python.d/postgres.conf

When no configuration file is found, the module tries to connect to TCP/IP socket: localhost:5432.

socket:
  name         : 'socket'
  user         : 'postgres'
  database     : 'postgres'

tcp:
  name         : 'tcp'
  user         : 'postgres'
  database     : 'postgres'
  host         : 'localhost'
  port         : 5432

---