Hi ,
Thanks for reaching out to Microsoft Q&A.
A 408 Request Timeout
error in ADF’s Web Activity, particularly when triggering Synapse, can be tricky to troubleshoot.
Here’s a structured approach to diagnose and address this:
- Verify if the Synapse endpoint itself is responsive by testing it outside ADF, e.g., using Postman or a simple script. This can confirm if the issue is specifically with ADF or the Synapse service itself. Look for any service-level or firewall rule changes that may have impacted the endpoint's accessibility.
- Although you used AutoResolveIntegrationRuntime (default Azure IR), connectivity issues can sometimes arise if the underlying node experiences a network delay or issue. However, if a self-hosted IR temporarily fixed the issue, it suggests an intermittent connectivity problem on the Azure IR network side.
- Ensure that your timeout settings in the Web Activity are set to 5 minutes or longer, as you did, to allow for extended processing times. Consider adding a retry policy to the Web Activity if you haven’t already. ADF’s Web Activity supports up to three retries with an exponential backoff mechanism, which could be helpful if this is an intermittent issue.
- DNS Resolution: Verify if the endpoint URL resolves correctly by conducting a DNS lookup. Intermittent DNS resolution issues can cause the request to fail, so confirm that the Synapse endpoint’s DNS is accessible from the IR location.
- Certificates and SSL Validation: Any recent changes to SSL/TLS policies on the endpoint could affect SSL handshakes, so ensure that certificate chains are intact and meet Azure’s validation requirements.
- If other services are making requests to Synapse, you may encounter throttling. Check Synapse’s usage metrics and logs to identify if you are hitting any concurrency or resource limits.
- Enable diagnostic logging in both ADF and Synapse to get detailed logs on the failure. These logs might provide more context, particularly about networking, connectivity, or other service-related errors. Use Log Analytics or a similar monitoring tool to track any patterns in latency, especially around the 11 AM CET timeframe when the issue started. I have listed out most of the options. This structured approach should help pinpoint the root cause, whether it’s on the ADF, Synapse, or network side.
Please feel free to click the 'Upvote' (Thumbs-up) button and 'Accept as Answer'. This helps the community by allowing others with similar queries to easily find the solution.