Cloudwatch alarms strange behavior

Question

We're implementing some new alarms recently that needs to monitor CPU for 1 min period. Alarms are acting in very strange way. They show Good status for a min and for the min after they show Insuffucient data sign

Reason: Insufficient Data: 1 datapoint was unknown.

• Does this mean any problems in the configuration?
• If the alarms condition meet, will fire in all cases or not?

Solution

Do you have detailed monitoring enabled on your instance?

If you create CloudWatch Alarms for period 1 minute without detailed monitoring enabled, it will stuck at Insufficient Data for 4 minutes after showing OK for every 5th minute . This is because with normal monitoring the reporting data comes with the interval of 5 minutes whereas in detailed monitoring, this interval is 1 minute.

If you do not want to enable detailed monitoring which is paid, change all alarms to 5 minutes period and it will be all fine.

I think Amazon should not let users create alarms for 1 minute period when detailed monitoring is not enabled. This is not your fault.

Comments

[Bassam Gamal]

This isn't related at all to the question. I didn't mention AutoScaling at all. It's CloudWatch only issue