Instance gone without any notification and nothing recover after half hour.

[Han-Jen Cheng]

I found my service on GAE gone (replying 502) accidently.

https://console.cloud.google.com/appengine/instances shows no instance running during the period.

No any warning or alert message in dashboard or stackdriver or event panel. No instance start for over half hour (longer than monthly SLA allowance already)

What can I do to debug the issue or how can I get any help to prevent it happen?

Seems I'm not alone and it had been seen for a while.

https://groups.google.com/d/topic/google-appengine/6snQVOQdYwE/discussion

[Carl Smith]

I don't have a problem with AppEngine (at the moment), but if its any help to support, then Google Cloud Build is also reporting 502 errors

error pulling image configuration: received unexpected HTTP status: 502 Bad Gateway

ERROR

ERROR: build step 0 "gcr.io/cloud-builders/docker" failed: exit status 1

[Will Shepherdson]

We are experiencing the same issue. Our service is running, but no instances are running either. However, no incident is being tracked on GCP status:

https://status.cloud.google.com/

[Julie (cloud platform support)]

Thanks for reporting this. There was a known issue which occurred the same time that may be related to your issue which has since been resolved. Are you still experiencing the issue as it may be possible that there are no instances if your app.yaml is configured to automatic scaling and no minimum number of instances defined. If this is not the case please file an issue tracker for private App Engine issues with the time frame the issue began, your project ID and the affected service/version so we can determine the cause of the 502 errors. 

[Han-Jen Cheng]

Hello Julie:

Thanks for your information! Actually I didn't ran into such problem recently. But beside the single occurrence of such issue I'm also wondering how the SLA work in such case and how can we find/track the issue once it happen?

Julie (cloud platform support)於 2019年7月5日星期五 UTC+8上午5時00分44秒寫道：

[Julie (cloud platform support)]

This is the App Engine SLA which defines the allowable downtime and the specific percentages when downtime is exceeded. You can fill out this form to request a review  if it seems SLA was not met, this should be done no more than 30 days after the close of the month for which you are requesting an adjustment. If you suspect you are being affected by a known issue, check the dashboard and issue tracker where known issues and their progress for resolution may be posted. If you have access to the Google Cloud Support, they will be able to confirm it. 

[Han-Jen Cheng]

Thanks for your answering, appreciate!

Julie (cloud platform support)於 2019年7月12日星期五 UTC+8上午3時52分56秒寫道：