Introduction
As a SaaS customer, you might occasionally notice an unexpected spike in your data transfer usage. Understanding exactly what is driving this consumption—whether it is a highly active IP address, a specific automated user, or a frequently downloaded artifact path—is crucial for managing your platform limits and optimizing your CI/CD workflows.
This article will walk you through the steps to independently investigate data transfer spikes using the Usage Explorer in your MyJFrog portal.
Resolution
The MyJFrog portal provides built-in analytics that allow you to drill down into your usage metrics. By exporting your data transfer logs with specific "Group by" filters, you can easily identify the root cause of any usage anomalies.
Step-by-Step Instructions
- Access the Portal: Log in to your MyJFrog Portal using your administrator credentials.
- Navigate to Usage Explorer: From the main dashboard, locate and click on the Usage Explorer tab.
- Set the Timeframe: Adjust the calendar filters to select the specific Date Range where you observed the spike in data transfer.
- Initiate Export: Click the Export CSV button located on the Usage Explorer dashboard.
- Select "Group By" Parameters: A prompt will appear asking how you want to structure your exported data. To find the source of the spike, select the Group by dropdown. You can group the data by various dimensions to isolate the heavy consumers:
- Repository: To see if a specific local, remote, or virtual repository is causing the high traffic.
- User: To identify if a specific developer or service account is pulling large amounts of data.
- IP Address: To check for external systems or specific build nodes driving the usage.
- Artifact Path: To pinpoint exact files or images (like a massive Docker layer) being downloaded repeatedly.
- Action: To understand what types of requests are consuming bandwidth.
- Analyze and Adjust: Download the generated CSV file. Sort the data by size/transfer volume to identify the top contributors.
Once you have identified the source (e.g., a script repeatedly pulling a large artifact without caching), you can adjust your configurations, implement better caching strategies, or restrict user access to prevent future spikes.
Conclusion
By leveraging the granular export options within the MyJFrog Usage Explorer, you can quickly diagnose data transfer spikes and take actionable steps to optimize your usage.