Assigned
Status Update
No update yet.
Comments
su...@google.com
su...@google.com
Hello,
Thank you for reaching out to us with your request.
We have duly noted your feedback and will thoroughly validate it. While we cannot provide an estimated time of implementation or guarantee the fulfillment of the issue, please be assured that your input is highly valued. Your feedback enables us to enhance our products and services.
We appreciate your continued trust and support in improving our Google Cloud Platform products. In case you want to report a new issue, Please do not hesitate to create a new issue on the
Once again, we sincerely appreciate your valuable feedback; Thank you for your understanding and collaboration.
Description
The issue at hand is related to the behavior of App Engine with warmup request enabled during new version deployments with steady or existing QPS. The expected behavior was that non-warming requests would continue to be routed to the existing version and only switch to the new version after the associated warming request had completed. However, it was observed that non-warming requests were getting queued on the clone for the new version while the warming request was still in progress. This resulted in potential latency spikes and impacted user experience.
What you expected to happen:
The expectation was that non-warming requests would be routed to the new version only after the warming request had completed, ensuring a seamless transition with no increase in user latency.
Steps to reproduce:
Deploy a new version of an application on App Engine with warmup request enabled.
Maintain steady or existing QPS while performing the deployment.
Observe the routing of non-warming requests to the new version even before the associated warming request has finished.
Other information (workarounds you have tried, documentation consulted, etc):
In this case, the user reported this behavior to Google Cloud Support, and an internal feature request has been initiated to address and align the behavior of App Engine more closely with the expected outcome. However, there is no specified ETA for when this feature request will be released or published. The recommendation for now is for the user to consider migrating to Cloud Run, where the desired behavior is already in place, ensuring that the feature request would be immediately satisfied.