Monitoring Cloud Infrastructure
In this lab, using AWS Systems Manager, I ran a CloudWatch agent command.
CloudWatch is an AWS service that can set alarms and through SNS(Simple Notification Service) can set notifications on parameters you set.
I then used Parameter Store to create a monitor to observe web server metrics through CloudWatch logs.
I then ran a command linked it to the configuration document prefix of AmazonCloudWatch-ManageAgent
The next step was interesting. Attempting to access a test website that doesn’t exist to generate log data by adding /start to the end of the test website.
Navigating to the newly named Log Management(formerly Log groups), I found the log activity in HttpAccessLog
I then created a metric filter on HttpAccessLog to uniquely identify 404 errors
Get Mr. Raph’s stories in your inbox
Join Medium for free to get updates from this writer.
I also created an alarm for when an excess of 404 errors(when a website is not found)
I then created an SNS subscription, shared my email, confirmed the subscription and went to the test server to create some entries for the log.
I did TestIP/start2 TestIP/start3 TestIP/start4 TestIP/start5 TestIP/start6 as I made the threshold 5 or more times for the metric.
After waiting, I did indeed get an email alert regarding the repeated 404 errors:
I then went to the test server instance, then the Monitoring tab to take note of the metrics:
I was unable to find Rules under Events to proceed and there is unclear information on if Amazon EventBridge replaced Events under CloudWatch so the exercise was halted.
