Continuous Monitoring: How Kambr Keeps Clients Happy 24/7

Raphael De Lio
Kambr
Published in
4 min readApr 21, 2022

I will start this story by sharing a bit about our culture and how we are encouraged to show appreciation for our colleagues and their work within Kambr.

Chief Commercial Officer Mike Buccialia praising the excellent work of the customer support team

Ana Barceló is the head of the support team at Kambr. She joined Kambr a year ago, in April 2021 as our first Kambrian whose sole job was specifically supporting our clients and their operations.

Since then, Ana has been continuously working on building a support team, standardizing the support processes and protocols, organizing on-calls, and thinking of how we can build or start using better tools to help us give our clients the support they need to keep their operations working smoothly.

I’m not going to lie, supporting our applications is not the easiest job. And as Kambr started transitioning to Microservices, it started becoming even more difficult and complex.

Where should I check the logs? This given process failed, which service is responsible for it? Which team should I forward this bug to? I’m receiving these generic error messages…. Should I keep caring or just start ignoring them?

Yes, among all these questions and trouble, Ana and her team were able to keep our customers happy and their processes running as expected. Thank you very much!

But how can we keep providing Ana and her team the best tools to support our clients? Continuous monitoring is the answer.

What is Continuous Monitoring

A lot can go very wrong with an application, and when we’re talking about an application that manages millions of dollars of revenue, things can go dark very quickly.

“The sooner you can get the bad news, the sooner you can start working on the fix and the sooner you can stop bleeding.” — David Clinton

Continuous monitoring is using a stack of tools and technologies to keep a close and unblinking eye into our processes and applications. Allowing us to be notified and start working on solutions the quickest.

But not only, a well implemented continuous monitoring solution should also be able to track the quality of the service we are providing, such as:

  • Website latency
  • Database performance
  • Data accuracy
  • Data privacy
  • Vulnerabilities
  • Compliance standards

If you’re not watching, how can you know you’re not on the right track?

The steps to a Continuous Monitoring Solution

David Clinton, who has a course on The Big Picture of Continuous Monitoring, separates the steps to implementing a Continuous Monitoring solution into four steps:

  • Research
  • Selection
  • Implementation
  • Consumption

Let’s dig into each one of them.

Research

First of all you need to understand how the software is supposed to work and how it’s currently working. Otherwise, you cannot measure if it’s not working correctly.

In this research, you will make sure you understand:

  • What makes up the technology stack of the application.
  • What are the current security controls in place.
  • How data is supposed to be collected, processed, shared and stored.
  • The business goals and limitations.
  • What has gone wrong with your application on the business or infrastructure side that you are afraid of happening again?

Selection

With this information in hand, you will be much more capable of selecting the right tools to build your monitoring solution. However, this will require much more research, a lot of demoing, trials and errors until you find the correct one.

Implementation

This is the step where you will configure and tune the tools that you’ve selected in a way that best fit the needs of your company’s business.

In this step, you will also test the whole solution to make sure it works as expected.

Consumption

Finally, you will need to define what’s going to happen with the information your solution is providing you. You will need to give meaning to all this data. And also:

  • Who should be receiving these reports?
  • Defining a process of how alerts should be assessed and handled.
  • How can we digest all the data our monitoring applications are generating?
  • Who can help us give meaning to this data?

Conclusion

The success of our clients is critical for our operations. We have previously discussed how Kambr is using the ELK Stack to continuously monitor its integrations and be able to provide the support our clients need.

And although Ana, her team and many other Kambrians are the answer to how we keep our customers happy 24/7, we still have a lot room to get even better. There’s a lot more we want to do and we’re working on continuing to scale our monitoring solutions and making sure we are able to keep providing the support our clients need in the most efficient way.

I hope you have enjoyed this story. If you still haven’t read our previous story on this matter, I invite you to do so by clicking on the link below:

--

--

Raphael De Lio
Kambr
Editor for

Software Consultant @ Xebia - Dutch Kotlin User Group Organizer: https://kotlin.nl