Skip to main content

Puppet monitoring with Netdata

Monitor status of Puppet Server and Puppet DB.

Following charts are drawn:

  1. JVM Heap

    • committed (allocated from OS)
    • used (actual use)
  2. JVM Non-Heap

    • committed (allocated from OS)
    • used (actual use)
  3. CPU Usage

    • execution
    • GC (taken by garbage collection)
  4. File Descriptors

    • max
    • used

Configuration

Edit the python.d/puppet.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/puppet.conf
puppetdb:
url: 'https://fqdn.example.com:8081'
tls_cert_file: /path/to/client.crt
tls_key_file: /path/to/client.key
autodetection_retry: 1

puppetserver:
url: 'https://fqdn.example.com:8140'
autodetection_retry: 1

When no configuration is given, module uses https://fqdn.example.com:8140.

notes

  • Exact Fully Qualified Domain Name of the node should be used.
  • Usually Puppet Server/DB startup time is VERY long. So, there should be quite reasonable retry count.
  • Secure PuppetDB config may require client certificate. Not applies to default PuppetDB configuration though.

Was this page helpful?

Need further help?

Search for an answer in our community forum.

Contribute