Automating Default Exception Queue Cleanup in Cisco ECE

In Cisco ECE, email activities that cannot be processed or routed correctly may end up in the Default Exception Queue. Over time, this queue can collect activities that do not require any manual action from an agent.

A common example is technical email activity that only needs to be completed, not reprocessed or assigned to an agent.

This post shows how to automatically clean up the Default Exception Queue by using an Alarm Workflow.

Initial Situation

In Cisco Finesse, an agent or administrator can open Pick Activities, select Default exception queue, and see the activities currently waiting in that queue.

In this example, the queue contains several email activities:

Activity ID: 1058
Activity ID: 1112
Activity ID: 1234

Closing these activities manually is not practical if they appear regularly.

Solution Overview

Instead of manually closing each activity, we can create an Alarm Workflow that runs on a schedule, checks the Default Exception Queue, and automatically completes matching activities.

The basic workflow is:

Start node
   ↓
Alarm node
   ↓
Completion node

This does not delete emails from the system. It completes the selected activities.

Creating the Alarm Workflow

In the Administration Console, go to the required department and create a new Alarm Workflow.

Workflow Diagram

The workflow contains three nodes:

Start → Alarm → Completion

The purpose of each node is:

Start node      — selects the queue and defines the schedule
Alarm node      — selects activities that match the condition
Completion node — completes the selected activities

Configuring the Start Node

In the Start node, select the queue that the workflow will monitor → Default exception queue

On the Schedule tab of the Start node, define when the workflow should run. Our workflow is configured to run multiple times:

This schedule is useful for a lab demonstration. In production, running the workflow every minute is usually unnecessary. A safer interval is every 15–30 minutes or once per hour.

Configuring the Alarm Node

The Alarm node defines which activities should continue through the workflow. In this example, the condition is:

Activity → Minutes after due >= 1

This means the workflow selects activities that are at least one minute past their due time. This detail is important: an activity without a Due on value will not match this condition and will not be completed by this workflow. In production you have to configure your criteria which emails should be completed.

The Alarm node can also send a notification. In our demo, the workflow sends an email:

To: Customer98@cvp8.com
Subject: Some Activities were completed in DEQ

Message body:

Hello,

Some Activities were automatically completed in the Default Exception Queue

This is useful for testing and validation. After the workflow runs, the notification shows which activities were processed.

Result

After the workflow runs, only one activity remains in the Default Exception Queue:

Activity ID: 1058

Activities 1112 and 1234 were automatically processed by the workflow.

The notification comes and confirms that the workflow selected the matching activities and passed them to the Completion node.

What Should Be Completed Automatically

Only activities that do not require agent action should be completed automatically. Practical examples include:

Undeliverable
Delivery Status Notification
Mail delivery failed
Mailbox unavailable
Mailbox full
Auto-Reply
Out of office
mail-daemon
postmaster

These are usually technical responses from mail systems or automatic replies. An agent should not have to process each of them manually.

The key rule is simple: automatically complete only those activities that clearly do not require agent action. For all other cases, the Default Exception Queue should remain a control point, not a trash bin.