Skip to main content
Pentaho Documentation

Administer Pentaho Worker Nodes

Parent article

The Pentaho Server can be configured for worker nodes, after completing the installation and setup process, so that transformations and jobs can be run as work items. You can take advantage of worker nodes, for example when using Kafka, to scale out across clusters when you have workload increases. Results from the transformations and jobs occur normally within the Pentaho platform, and the Administration Application provides monitoring for the health and performance of the worker node instances and services.

Monitor the Worker Nodes

The monitoring reporting functions for worker instances and processes are performed by PWN. Master instances run an essential set of services, which includes the administration application, cluster-coordination, synchronization, and service deployment. Worker instances can run any services besides those previously listed. More information on a work-item by work-item basis is available on Mesos and Chronos, which are also services available on PWN.

Access the PWN Administration Application

You need your user name and password to access the Pentaho Worker Nodes Administration Application (Admin App).

Perform the following actions to open the Admin App.

Procedure

  1. Open a web browser to go to the PWN Administration Application:

    https://<PWN-instance-IP-address>:8000
  2. In the Security Realm field, select the location where your user account is defined.

    To log in using the local admin account, select Local.
  3. Enter your user name and password, then click Log In. The Home page appears.

  4. Click Admin App to open the tool.

View the Dashboard

You can view the status of the PWN system, resource allocations, instances, and services.

Perform the following actions to monitor the system and services.

Procedure

  1. Access the PWN Administration Application and log in.

    By default, the application opens to the Dashboard page.
  2. Use the left-side menu to navigate the application.

    You can view the Pentaho Worker Nodes services and instances, including the state and status of the service and its resource allocations.

View Instances

You can view all master and worker instances, which are identified by IP address. Single-instance systems have one master instance while multi-instance systems have either one or three master instances.

Perform the following actions to monitor instances.

Procedure

  1. Access the PWN Administration Application and log in.

  2. Click Instances.

    The Master and Worker instances are displayed, including their state, service units, and resource allocations.
  3. Click on an instance to view details.

Monitor Services

You can access details, status, and results about the files running on worker nodes. Services perform functions essential to the health or functionality of the system.

Perform the following actions to monitor services.

Procedure

  1. Access the PWN Administration Application and log in.

  2. Click Services.

    Pentaho Worker Nodes Services appear, including the number of instances, state, and resource allocations.
  3. Click on any service to view more details, including volumes, network, configuration, scale, and events.

View Jobs

Jobs are units of work that are triggered and live only until their execution finishes. You can view details, status, and configuration of the Jobs running on worker nodes.

Perform the following actions to manage jobs:

Procedure

  1. Access the PWN Administration Application and log in.

  2. In the left-side panel, click Jobs. Pentaho Worker Nodes Job Types are displayed, including its resource allocation.

  3. Under Job Types, select a specific job to see triggered jobs status and resource allocation. You can also access its configuration.

Worker Node Queue

The jobs and transformations that are sent to worker nodes are queued in Chronos, the scheduling service. Based on the availability of resources, Chronos sends the request to a specific node for execution.

Worker Node Logs

You can configure the worker node logs to be sent to ELK stack via Kafka. The ELK can be either internal or external to workerNode.