Issue 114
Thank you for all of the well-wishes last week–whatever I came down with knocked me on my arse and I ended up spending several days just in bed. I’m slowly recovering now, thankfully. <3 you folks
This issue is sponsored by:
Serverless has grown in popularity with application developers because it abstracts away the pesky operations pieces of development. Download SignalFx’s Definitive Guide to Serverless Monitoring and Observability and learn from dozens of SignalFx customers using serverless in production.
Articles & News on monitoring.love
Observability & Monitoring Community Slack
Come hang out with all your fellow Monitoring Weekly readers. I mean, I’m also there, but I’m sure everyone else is way cooler.
Observability & Robots with Ian Sherman
I linked to their article on robots and observability in here a while back and I had the opportunity to interview Ian Sherman, their Head of Engineering, about what they’re up to and how it all works. I love the unique application of standard observability tooling to their space.
From The Community
From Observability to Significance in Distributed Information Systems
Wherein Mark Burgess of CFEngine and promise theory fame discusses observability. Very cool stuff.
Your Load Generator is Probably Lying to You
“The number one indicator you should never get rid of is the maximum value. That’s not noise, it’s the signal, the rest is noise.” and many other goodies in this article+video.
Why are software developers confused by Kafka and immutable logs?
This article starts off with some musings about overly-complex software architecture, but it starts to get really good about halfway in when the stuff about Kafka shows up. Stick with it; I promise it’s worth the read.
For you folks holding off from using TimescaleDB because there wasn’t any hosted option, the day has come. I think their focus on “first to run on AWS, GCP, and Azure” is a bit odd marketing-wise, but I suspect it’s a shot over the bow at InfluxData…
Some Observations On the Messy Realities of Incident Reviews
When Dr. Richard Cook talks incidents, you want to listen.
.Net Logging Performance Comparison
Some blessed soul spent what appears to be quite a lot of time analyzing the performance of all of the most common .NET logging frameworks and their calls. It’s pretty great.
Grafana Tutorial: Simple Synthetic Monitoring for Applications
I like this article, but one caveat not mentioned here is that it ignores any frontend/UI problems. For example, you might see a drastic drop in ecommerce checkouts or user logins, not see any issues with backend systems, and start blaming your monitoring for false negatives–only to realize a UI bug removed the checkout or login button. Using DOM element selection in your synthetics is important for catching these sorts of issues.
Observing Software Delivery Performance
Kind of a cool little tool to surface the key metrics identified in the DORA State of DevOps reports to everyone with a REST API, Postgres, and Grafana.
Listen to a Recorded Incident Response Call
The folks at PagerDuty recorded a (simulated) incident to illustrate incident response procedures in action. Neat.
Lessons learned from running Kafka at Datadog
Because everyone loves right Kafka, right?
Splunk has always been the dominant player in log analytics-as-SIEM, so I’m pretty excited to see Elastic formally enter the space with a competitor. This should be interesting to see where things go from here.
BGP Leak Highlights the Fragility of the Internet with Real Consequences
In case you needed a reminder that the internet is built on toothpicks, bubblegum, and a weird level of trust unbefitting of the modern nature of the internet.
This issue is sponsored by:
LightStep is one of the new breed of tools out there I’m excited about. Designed with modern, high-scale, high-traffic architectures in mind, LightStep makes it easy to spot, diagnose, and solve performance issues. Check it out here.
See you next week!
– Mike (@mike_julian) Monitoring Weekly Editor