Alarm creation flawed

[msoosgds]

The two alarms for EFS are being created concurrently, which can lead to a race condition and a fail:

  | 16:48:03 UTC+0100 | CREATE_COMPLETE | AWS::CloudWatch::Alarm | CriticalAlarm |  
  | 16:48:03 UTC+0100 | CREATE_FAILED | AWS::CloudWatch::Alarm | WarningAlarm | A separate request to update this alarm is in progress.
  | 16:48:03 UTC+0100 | CREATE_IN_PROGRESS | AWS::CloudWatch::Alarm | CriticalAlarm | Resource creation Initiated

As you can see, the create for one hasn't terminated when the other is started, leading to a failure of "A separate request to update this alarm is in progress.". Then it completes the alarm creation, but from then on, it's all rolled back of course because of the failure due to the concurrent creation. Please fix :)

[darrylsosborne]

I'm looking into this. This is the first time hearing of this condition. I'll try to reproduce the issue. If I'm unable to reproduce it (I've run this 100's of times without seeing this condition), I could possibly add an alarm dependency on one to force them to be created serially. Let me investigate.

[msoosgds]

Please use google to search to find similar issues:

https://www.google.co.uk/search?q=CREATE_FAILED+A+separate+request+to+update+this+alarm+is+in+progress