1

I have a MySQL RDS instance that's got 100+ queries all stuck waiting for a table metadata lock.

The good news

I have permission to kill all of the sessions, and have 0 concern for data loss as this database is completely reloaded as needed.

The bad news

It's a MySQL instance running as a hosted RDS instance in AWS AND while I have access to the "root" MySQL account, I don't have access to the AWS console that this RDS instance is hosted in.

What I've tried

When this has happened previously, I look through the connections and run CALL mysql.rds_kill(_each_and_every_process_id_one_at_a_time). However, this time, when I try run one of these kill statements, it simply hangs. (I'm doing this through MySQL workbench)

What options do I have? Is there someway to "escalate" my mysql.rds_kill call so that it's handled right away? Do I try executing this outside of workbench somehow?

  • 2
    In SHOW PROCESSLIST, which query has the highest "Time"? Can you kill it? – Rick James Mar 16 at 3:26
  • Check the free storage space for the RDS instance, in the console. Lack of free disk could cause a hang. Also check the error log in the console for unusual entries. – Michael - sqlbot Mar 17 at 1:33
1

Despite Workbench freezing on me, I was able to connect to the MySQL instance via the command line and run CALL mysql.rds_kill(_each_and_every_process_id_one_at_a_time) successfully.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.