From the course: Docker: Continuous Delivery

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Monitoring and reverting

Monitoring and reverting - Docker Tutorial

From the course: Docker: Continuous Delivery

Start my 1-month free trial

Monitoring and reverting

“

- [Instructor] And now a little aside to take a finer look at monitoring. When building deployment systems, I try to distinguish between monitoring and forensics, also called logging. Monitoring tends to be about right now. Is the system up? Did I just break that? And forensics tends to be about what just happened. What was that request? Where's this bug? Monitoring tends to be aggregated. I like systems that will tell me, "What is the error rate right now?" What is the distribution of latencies? What is the slowest request in the last 10 minutes? Things like that. Whereas forensics tends to be detailed. Do you want to store each individual thing so that when you discover that there was a bug, you can go back and say, "Hmm, what was the request "that caused this bug? "Ah, it was this request." And then you look up the request and the response and maybe if you're lucky, you've got a system where you can feed the request back in as an input and watch it flow through the system and…

Contents