Slug: document-retry Version: 1.0.0 Infer: No Event Aware: Yes

Overview

Document Retry Mechanism

This model provides an automated retry system for documents that may have failed processing. It uses scheduled events to identify documents needing retry and applies a configurable retry strategy.

How It Works

The document retry model periodically checks for documents matching specific filter criteria, tracks retry attempts using document labels, and performs reprocessing until either:

The document processes successfully
The maximum number of retry attempts is reached (at which point the document is marked with a failure status)

Core Functionality

Runs on a scheduled basis
Identifies documents that match the retry filter but are not already in the failed status
Tracks retry attempts using incrementing numeric labels
Reprocesses documents that haven’t exceeded the retry limit
Marks documents as failed when they exceed the retry limit

Process Flow

Retry Count Tracking

The system tracks retry attempts using document labels with a configurable prefix followed by a number (e.g., retry-1, retry-2).

Configuration Parameters

document_store_ref: Reference to the document store to query
retry_filter: Query filter to identify documents for retry
failed_status_id: Status ID to assign when max retries are exceeded
number_of_retries: Maximum retry attempts allowed
label_prefix: Prefix for retry count labels
project: Project endpoint for status definitions
assistant: Assistant endpoint for document reprocessing

This model helps ensure that temporary processing failures don’t result in permanently failed documents, improving overall processing reliability.

Inference Options

The following options can be configured when using this model for inference:

Name	Label	Type	Description	Default	Required
`database_host`	Database Host	string	N/A	-	No
`database_port`	Database Port	integer	N/A	-	No
`database_username`	Database Username	string	N/A	-	No
`database_password`	Database Password	password	N/A	-	No
`database_name`	Database Name	string	N/A	-	No
`document_store_ref`	Document Store	documentStore	N/A	-	No
`retry_filter`	Retry Filter	string	N/A	exists(statistics.recentExecutions.execution.status: ‘FAILED’)	No
`failed_status_id`	Failed Status	documentStatus	N/A	-	No
`number_of_retries`	Number of Retries	integer	N/A	3	No
`label_prefix`	Label Prefix	string	N/A	retry-	No
`assistant`	Assistant	assistant	N/A	-	No

Model Details

Provider: Kodexa AI

Introduction

​Overview

​Document Retry Mechanism

​How It Works

​Core Functionality

​Process Flow

​Retry Count Tracking

​Configuration Parameters

​Inference Options

​Model Details