How to Manage Workflow Executions in n8n

Click the Executions icon in the left sidebar of the n8n editor. This opens a list of all workflow executions across your instance. Each row shows the workflow name, execution status (Success, Error, or Running), start time, and duration. Click any row to open the execution details, where you can see the data flowing through each node. Use the filter controls at the top to narrow results by status, workflow name, or date range. This is your primary tool for monitoring what your workflows are doing.

By default, n8n only saves the final result of each execution. If a workflow fails midway, you cannot see which nodes completed successfully. To fix this, click the gear icon on any workflow canvas to open Workflow Settings. Toggle Save Execution Progress to on. Now n8n saves the output of each node as it completes, so if the workflow errors on step 5, you still have the data from steps 1 through 4. This is invaluable for debugging complex workflows but increases database storage usage.

On a self-hosted n8n instance, execution records grow indefinitely unless you configure pruning. Set the following environment variables to automatically delete old executions. EXECUTIONS_DATA_PRUNE sets pruning to active. EXECUTIONS_DATA_MAX_AGE controls how many hours to keep records. EXECUTIONS_DATA_PRUNE_MAX_COUNT sets the maximum number of records to keep. Add these to your Docker Compose file, systemd environment file, or .env file depending on your deployment method.

1# Docker Compose example
2environment:
3  - EXECUTIONS_DATA_PRUNE=true
4  - EXECUTIONS_DATA_MAX_AGE=168        # 7 days in hours
5  - EXECUTIONS_DATA_PRUNE_MAX_COUNT=5000
6  - EXECUTIONS_DATA_SAVE_ON_ERROR=all
7  - EXECUTIONS_DATA_SAVE_ON_SUCCESS=none  # Don't save successful runs
8  - EXECUTIONS_DATA_SAVE_MANUAL_EXECUTIONS=true

To delete executions manually, go to the Executions tab in the left sidebar. Use the filters to find the executions you want to remove. Select individual executions using the checkboxes on the left, or use the Select All checkbox at the top to select all visible executions. Then click the Delete button. For bulk deletion, filter by a date range or status first, then select all and delete. This is useful for one-time cleanup or removing test executions before going to production.

Different workflows have different logging needs. A critical payment workflow should save all executions, while a test workflow should save none. Click the gear icon on a workflow to open Workflow Settings. Under Save Successful Production Executions, choose Default (follows global setting), Yes, or No. Do the same for Save Failed Production Executions. Set manual execution saving separately. This lets you keep detailed logs for important workflows while reducing storage for trivial ones.