How to Fix AWS CloudWatch Alarm Incidents

Here are the steps to fix incidents that trigger from AWS CloudWatch Alarms.

AWS CloudWatch Incidents

  • You can create alarms in the AWS CloudWatch on a specific metric. A metric can be failed login attempts, CPU usage, Server is down etc.
  • When an alarm triggers ( after it crosses the defined threshold limit, as configured), a Remedy ticket automatically triggers the support team.
  • The support (operations) team should analyze the root cause of the issue.

How to resolve AWS incidents: 7 steps

AWS remedy incidents

Step#1

Check the ticket description to know the AWS account (It may get from Production, Development, or Integration accounts). Then login into that account using your IAM role credentials.

Step#2

After you log in, in the AWS console search for CloudWatch.

Step#3

Next, click on the alarms list, and go to the alarm in question. It could be in the red, which means in alarm state.

Step#4

Now, you can verify the graphical display and timestamp when the alarm triggers (timestamp).

Step#5

Go to the log groups list, and select the log-group in which the alarm metric is defined. Below, you will find the relation between the log group and the alarm.

Creating an alarm from a log group in the AWS.
Image is adapted from AWS blogs

Step#6

Select the relevant log group, and go to the latest log stream. The other way, you can find the log stream by filtering the time.

Step#7

Once you get the event details, you can check the error and find the root cause of the issue.

Keep reading

Related

Author: Srini

Experienced software developer. Skills in Development, Coding, Testing and Debugging. Good Data analytic skills (Data Warehousing and BI). Also skills in Mainframe.