Monitoring is not just Dashboards

A well-functioning Serverless Team ensures that, on a Friday afternoon, releases are seamlessly integrated into the main branch, rigorously tested, and deployed to various environments prior to reaching production. 

Serverless Team requires monitoring

However, you are no longer in the office - instead, you are enjoying time with family and friends or pursuing your hobbies. Your trust lies in the robust test suite and automation, which effectively identifies and resolves all issues. Health checks are conducted meticulously, and automation is ready to initiate a rollback if problems arise during the deployment of the new code in various environments. In the rare event of an anomaly, there is a robust notification system that notifies you and the team immediately, even if the problem has already triggered a rollback, allowing you to investigate it on Monday.

Automation and testing culture

Of course you wouldn't do this every Friday, however it is important to address it when necessary. This is when monitoring and health checks are not dashboards, but an integral part of the development and deployment cycle.

It is a common occurrence that monitoring and dashboard functionalities are not automated or that features provided by cloud providers, which can be highly powerful, are not set up at all. Even fundamental aspects like proper logging and machine-readable logs are missing, which can be utilized for further automation. You gain many more opportunities for insights, troubleshooting, and state checks when you ensure that crucial actions throughout the system are logged in a standardized manner and the logs are stored in a cloud provider's service, such as CloudWatch in AWS. You can even transform log messages into metrics that impact the deployment of your applications. All of this is highly cost-effective, simple, and swift.

Dashboards are less suitable for emergencies 

Dashboards are, of course, crucial. You should have automation in place that generates dashboards for your services, allowing you to identify trends, investigate issues with combined root causes, and observe the behavior of a specific setup. However, they are less suitable for emergencies, as these should be addressed by automation, automated health checks, and systems to restore your infrastructure to a healthy state.

„The assurance that my modifications are swiftly and securely released enables me to excel in my role and amplifies my job satisfaction. “

Florian Motlik
Florian Motlik, Senior Cloud Solutions Architekt

Athena or Cloudwatch Insights query for log search    

Otherwise, your systems remain distant from the ideal Serverless cloud of services, which effortlessly runs countless instances as needed. This is the destination you aspire to reach, not only for its productivity benefits but also because it significantly enhances the contentment of your skilled developers. There is no greater expense than losing valuable team members with extensive institutional wisdom due to them believing they can achieve more elsewhere. 

Any questions?

If you have any questions about your monitoring strategy or how to best set it up in AWS, especially with all the services from CloudWatch to Logs Insights, Metrics, Athena, etc., we're happy to help.

Start your individual journey with ORBIT.