Cloud SQL First Generation hang this morning?

[Allen Firstenberg]

Our automated processes on GCE went to execute a DROP TABLE command on one of our Cloud SQL tables this morning around 3am ET. This is part of a regular set of commands that are issued this time every week. The Cloud SQL instance is a First Generation D4.

Most weeks it takes about 3 seconds.

This week it took over 2.5 hours - until 5:41:35 am ET.

The CPU usage graph on our cloud console also indicates no CPU activity on the Cloud SQL instance until this time as well.

The GCE status page reports no issues during this time.

Is there any way we can figure out what happened and how to prevent it from happening again?

Thanks!

[Faizan]

Hello Allen,

Do you have query log enabled on your CloudSQL instance ? If yes, you can query the log table which might provide you with more insight. You can refer to this link for more information on slow query log. If its not enabled, you can visit the link for steps to enable it to debug the issue in future.

I hope that helps.

Faizan