Azure API Management

Inside scoop from the API Management team

Release Notes – March 17, 2017

On Friday, March 17, 2017, we started deploying a regular service update. We upgrade production service instances in batches, and it usually takes about a week for the rollout to complete. Customers in the Central US EAUP region should have access shortly.

Below is the list of the new features, improvements and bug fixes in this release.

New functionality

  • Now you can use management API to get API Management service instance in a VNET to pick up configuration updates, e.g. DNS server change [1]
  • Having issues with a VNET? No problem! Use the API to check if your API Management instance has connectivity to Azure the services it depends on [1]
  • Now you can easily import API Apps as APIs – we use the Swagger/OAS if one is available and create a “wildcard” API otherwise
  • We added mock-response policy which is handy in many situations, including making freshly designed APIs instantly callable. UI support for it is on the way, so keep an eye out!

[1] Documentation is being updated and should appear here shortly.

Updates and fixes

  • We significantly improved the scalability and throughput of the Log to Event Hub policy to minimize the loss of events sometimes occurring in high request volume situations (note that Event Hubs should be scaled independently and appropriately)
  •  Now changing values of properties referenced from quota and rate limit policies works as it should and doesn’t result in an obscure error
  • We made some smart tweaks to the Git backend and expect heavy users of the feature to see dramatic performance improvement of all Git operations
  • Saving policies containing expressions is much faster now (like 10 times faster in some cases)
  • Now you can use context.Request.Certificate in expressions within the send-request and send-one-way-request policies
  • Now literally any string can be used as a value of the reason attribute in the set-status policy