c-kruse

The other week while working away from home and using my phone with a poor connection as a mobile hotspot for my intel based macbook, I ran into a need for an aarch64 based system. I thought I would share what I came up with, not because I think it is especially useful, but because it was a lot of fun working through this without a stable internet connection. It has been a long time since I have been forced to work this way....

In my previous post Sensu Go Needs a Queue I talked about one of Sensu’s use cases for a queue. In this post I am looking at NSQ, a promising distributed messaging platform that is easy to operate and scales horizontally. After analyzing NSQ’s functionality and its operation under load similar to what I expect a large sensu-go cluster may put on it, I have found that NSQ is not a suitable candidate for sensu-go’s round robin message queue solution....

Over the past two years I’ve been working with the supremely talented sensu team developing and maintaining the open source monitoring solution, sensu-go. Throughout this journey a significant challenge has emerged - persistent reliability and stability problems in large deployments pushing the limits of sensu-go and its database, etcd. With the next major version of sensu-go, 7.0, we aim to apply our learnings and alleviate these stability and reliability issues....

I have a theory that there are many choreographed AWS Lambda functions out there in the world doing asynchronous background work that would be better served by a managed central orchestration layer, and that a better developer experience on top of AWS Step Functions would make writing these workflows faster, cheaper and more sustainable than the choreographed version I frequently see. The Choreographed Lambda Workflow First, what do I mean when I describe choreographed AWS Lambda functions, where they might be a good fit, and where there’s likely dragons....