A unified data governance solution that helps manage, protect, and discover data across your organization
Hi Mofei Zhuang Thanks for the detailed report. A consistent HTTP 408 from the Purview /entity/bulk API typically points to a timeout caused by request size, backend latency, or network issues.
Here are troubleshooting steps to try:
Break Large Requests into Smaller Batches: You're querying 9 GUIDs in one call try splitting into smaller batches (2–3 GUIDs) to reduce processing load and isolate any problematic entity.
Run with Diagnostic Tools: Use curl or Postman to measure latency:
curl -v --trace-time --max-time 30 "<your URL>"
Capture traffic via Fiddler/Wireshark to check DNS, TLS, or timeout phases.
Enable Diagnostic Logs: In Azure Portal → Purview > Diagnostic Settings, enable:
-
AtlasAPIRequests -
RequestLatency -
ThrottlingEvents
Send logs to Log Analytics for insights into delays or throttling.
Retry with Backoff: Use exponential backoff (e.g., 2s → 4s → 8s) and set a max wait time around 30–60s.
Check Network & Azure Health:
- Use Network Watcher to test latency to your Purview endpoint.
- Visit Azure Service Health for any incidents in your region.
Hope this helps. If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.