Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Troubleshooting CDN Connection
If you received an email alert regarding a CDN traffic drop in your Microsoft Clarity project, the logging pipeline detected a sudden decrease or complete stoppage of edge log data flowing into your AI Visibility dashboard. The following steps guide you to diagnose and resolve CDN connection issues.
Step 1: Verify the integration status
Before modifying your infrastructure, check whether the integration is still active within the Clarity platform.
- Sign in to your Microsoft Clarity account.
- Navigate to your specific project and select Settings -> AI Visibility.
- Check the status of your connected CDN provider. If the status shows an error or appears disconnected, you might need to reconnect your CDN.
Step 2: Check for infrastructure or configuration changes
Most sudden traffic drops are caused by accidental configuration overrides during routine server deployments, architecture updates, or token expiry. Verify the following on your CDN platform:
Cloudflare Enterprise
- LogPush job deletion: Cloudflare pushes HTTP request logs via LogPush jobs. Ensure that the specific LogPush job created for Microsoft Clarity hasn't been paused, modified, or accidentally deleted.
- Plan downgrades: Verify your Cloudflare account tier, as LogPush availability can vary by plan level.
Cloudflare Non-Enterprise
- API token validity: Confirm the API token you shared with Clarity still exists and hasn't been revoked or regenerated in your Cloudflare account.
- Token permissions: Ensure the token still grants read access to Analytics for the connected zone. If permissions were reduced or the zone was removed from the token's scope, recreate the token and update it inside Clarity Settings -> AI Visibility.
Amazon CloudFront
- Kinesis data firehose streams: CloudFront utilizes an AWS Kinesis Data Firehose destination to stream request logs to Clarity. Log into the AWS Management Console, navigate to CloudFront -> Logging, and ensure the pipeline destination is intact, active, and unthrottled.
Fastly
- HTTPS logging endpoints: Fastly forwards edge logs via an HTTPS logging endpoint. Ensure that a recent version deployment in your Fastly Service Configuration didn't clone an older active version that lacked the Clarity logging endpoint.
Azure Front Door
- Diagnostic settings: Check your Azure Front Door profile under Monitoring -> Diagnostic settings. Ensure the forwarding mechanism (such as the event hub rule) is properly active and has not been deleted.
Akamai
- DataStream deactivation: Check your Akamai Control Center under Common Services -> DataStream. Ensure that the telemetry stream responsible for forwarding edge request logs to Clarity hasn't been set to Deactivated.
WordPress
Clarity AI Visibility for WordPress is included in the Microsoft Clarity plugin. In your WordPress admin dashboard:
- Go to Plugins.
- Locate Microsoft Clarity and verify it shows as Active.
- If it was deactivated, click Activate to re-enable it.
- Confirm the plugin is connected to your Clarity project and avoid disabling it during maintenance or bulk plugin updates.
Step 3: Check permissions and token expiration
Security policies or updated API tokens can cause immediate authentication blocks between your CDN and Clarity.
- Rotated credentials: If your organization recently rotated API keys, secrets, or IAM roles associated with your AWS, Azure, or Cloudflare platforms, the logging pipeline fails.
- Re-linking: You may need to update the credentials inside Clarity Settings -> AI Visibility to restore the authorization handshake.
Step 4: Rule out site-wide traffic anomalies
Sometimes a drop in AI Visibility traffic indicates a broader network issue rather than a specific AI Visibility problem.
- Global CDN downtime: Check your CDN provider's official status page to ensure they aren't experiencing an active outage or edge log delays.
- Drastic traffic changes: Review your normal human traffic metrics in Clarity. If overall site traffic has also dropped to zero, it suggests a site-wide DNS or hosting outage rather than a localized AI Visibility breakdown.
Step 5: After applying the fix
Once you've fixed the configuration at your server or CDN level, data ingestion resumes automatically.
Note
Once logs begin streaming again, it can take up to a few hours for the pipeline to process the metrics and resume updating your AI Bot Activity dashboard.
If you confirmed that your CDN is sending logs but your dashboard remains blank, reach out to Clarity Support with your Project ID for further assistance.