Skip to content

Commit ea2e855

Browse files
authored
Merge pull request #180434 from lobrien/patch-mir-limits
Added section per Zhiyong
2 parents 7ec3b32 + 3b9ee8f commit ea2e855

File tree

1 file changed

+9
-0
lines changed

1 file changed

+9
-0
lines changed

articles/machine-learning/how-to-troubleshoot-online-endpoints.md

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -236,6 +236,15 @@ Although we do our best to provide a stable and reliable service, sometimes thin
236236

237237
If you are having trouble with autoscaling, see [Troubleshooting Azure autoscale](../azure-monitor/autoscale/autoscale-troubleshoot.md).
238238

239+
## Bandwidth limit issues
240+
241+
Managed online endpoints have bandwidth limits for each endpoints. You find the limit configuration in [Manage and increase quotas for resources with Azure Machine Learning](how-to-manage-quotas.md#azure-machine-learning-managed-online-endpoints-preview) here. If your bandwidth usage exceeds the limit, your request will be delayed. To monitor the bandwidth delay:
242+
243+
- Use metric “Network bytes” to understand the current bandwidth usage. For more information, see [Monitor managed online endpoints](how-to-monitor-online-endpoints.md).
244+
- There are two response trailers will be returned if the bandwidth limit enforced:
245+
- `ms-azureml-bandwidth-request-delay-ms`: delay time in milliseconds it took for the request stream transfer.
246+
- `ms-azureml-bandwidth-response-delay-ms`: delay time in milliseconds it took for the response stream transfer.
247+
239248
## HTTP status codes
240249

241250
When you access online endpoints with REST requests, the returned status codes adhere to the standards for [HTTP status codes](https://aka.ms/http-status-codes). Below are details about how endpoint invocation and prediction errors map to HTTP status codes.

0 commit comments

Comments
 (0)