Inserting a Custom Pipeline

This section describes how to modify an existing pipeline or create a new pipeline that takes advantage of the Deployment Template structure.

About the sample pipelines
For testing purposes, the Deployment Template includes two sets of pipelines and configuration files.
Sample pipeline overview
This section describes the high-level steps that are necessary to integrate a new/existing pipeline with a deployment template.
Location of pipeline configuration files
A deployment template's pipeline configuration files should be located on the primary server (i.e., the server on which the deployment template is installed, and usually the server hosting the EAC Central Server).
Creating a new project
Once the reference configuration files have been deleted, a new pipeline configuration project can be created.
Modifying an existing project
Modifying an existing Developer Studio project to match a new deployment template application is a somewhat tedious task. In fact, it is often easier to simply create a new deployment template application instead.
Configuring a record specifier
The deployment includes support for both baseline and partial index updates. In order to support partial updates, an application must include a record specifier, which is a property marked as the unique identifier of records in the index.
Forge flags
In order to reduce the amount of configuration required to integrate a pipeline into a deployment template, a standard deployment template application runs the primary and partial update Forge processes with an abbreviated set of flags.
Input record adapters
The record adapters load the source data.
Dimension adapters
The --inputDir flag specified to forge overrides the input URL for dimension adapters.
Indexer adapters
Since --outputPrefix and --outputDir flags are both included, the deployment template will override any values specified for the Indexer Adapter "URL" and "Output prefix" properties.
Output record adapters
Output record adapters are often used to generate debug or state information. By default, the location to which this data is written will be overridden by the --outputDir flag.
Dimension servers
The --stateDir flag will override the URL value for all Dimension Server components, and place any autogen state files in the [appdir]/data/state/ directory.
Common errors
This section provides troubleshooting information for commonly received errors.