Application Troubleshooting Using ZDX Features

This KB demonstrates how to troubleshoot an application issue using various ZDX parameters. In this knowledge base, Zoom is used as the example application. Below, we outline the process of troubleshooting an application.

Troubleshooting Zoom application issues when using Zscaler ZDX involves analysing performance and connectivity data. Zscaler ZDX (Digital Experience) provides insights into network and application performance across the path from the user's device to the application. Here's an example approach to troubleshoot the issue:

1. ZDX Dashboard

Scenario: Users report issues with Zoom calls having audio delays and poor video quality.

Steps:

  1. Access the Application Health Tab:

    • Navigate to the ZDX dashboard and select "Zoom" from the list of monitored SaaS applications.

    • Check the "Overall Health" score for Zoom to identify if it's a global issue or isolated to specific users/locations.

  2. Key Metrics to Analyze:

    • Latency: If latency is above 150ms, calls may experience delays.

      • Example: A user’s latency shows 220ms to zoom.us servers.

    • Packet Loss: Check for packet loss above 1%. Frequent losses result in degraded audio/video quality.

      • Example: ZDX shows 5% packet loss in the path from the user to Zoom's data center.

Action:

  • Use ZDX’s Path Trace to identify where latency or packet loss occurs:

    • A congested router at the ISP level might be the issue.

    • Suggest using a different ISP or escalate to the current provider.

  • If local network quality is poor, optimize Wi-Fi or move to a wired connection.


2. ZDX Dashboard: User Metrics

Scenario: A user reports that Zoom keeps disconnecting during meetings.

Steps:

  1. View User Experience Score:

    • Go to the “Users” tab in ZDX and select the affected user.

    • Review the User Experience Score for Zoom.

    • Identify if their score is low (below 70%).

  2. Device Health:

    • Check the CPU and Memory Utilization for the device.

      • Example: The dashboard indicates CPU utilization at 95% during Zoom calls, causing performance degradation.

    • Look for issues with Wi-Fi Signal Strength or Connection Speed:

      • Example: Wi-Fi signal strength is below 50%, leading to poor connectivity.

  3. Network Path:

    • Analyze the path between the user and Zoom.

    • Example: The ZDX path trace shows intermittent DNS resolution failures for zoom.us.

Action:

  • Suggest closing other applications consuming CPU.

  • Advise upgrading the user's Wi-Fi equipment or moving closer to the router.

  • For DNS issues, recommend switching to a reliable DNS provider (e.g., Google DNS or OpenDNS or Zscaler trusted resolver).


3. ZDX Dashboard: Application Reachability

Scenario: Zoom fails to connect, displaying an error like “Cannot reach Zoom servers.”

Steps:

  1. Zoom Server Accessibility:

    • Check Application Reachability in ZDX.

4. ZDX Dashboard: Network Path Analysis

Scenario: A group of users in a specific location reports poor Zoom performance.

Steps:

  1. Network Path Trace in Web or Cloud Path Probes

    • Check for bottlenecks in the route between the users and Zoom servers.

  2. Analyze Hop Performance:

    • Example: A specific ISP node shows 35% packet loss, causing delays.

Action:

  • Contact the ISP for node-level troubleshooting.

  • Temporarily route office traffic via an alternative ISP link.

5. Use Case: Global vs. Local Issue Isolation

Scenario: Zoom services are down for several users in different regions.

Steps:

  1. Global Health Analysis:

    • In ZDX, view the health of Zoom services globally.

    • Check if Zoom servers are having widespread outages.

  2. Isolated Region Check:

    • Filter data by location to identify affected regions.

    • Example: Users in Asia are affected, but users in North America are not.

Action:

  • Escalate to Zoom support if it’s a global outage.

  • Work with local ISPs or IT teams to resolve regional issues.

By leveraging ZDX dashboards and correlating its insights, you can pinpoint and resolve issues effectively. For troubleshooting any other application, the above parameters can be followed.