Below is a list of DevOps articles/sites that I discovered over the past month and found interesting.
DevOps Roadmap
First is a link to a DevOps roadmap. This is a more technological than philosophical view of what kinds of skills can be acquired in order to perform DevOps activities when on a team that does delivery via DevOps practices.
Diagramming and Documentation
Next is a programmatic diagramming tool that allows you to set up diagrams for your implementations via Python. One of the most difficult parts of any progressive technology implementation is ensuring that your materials remain up to date and available to new and existing team members. By putting this information into the code repository, and by allowing those artifacts to become part of the build pipeline, you can get much closer to that idealistic view of well-developed and well-run technology by closing this particular
https://diagrams.mingrammer.com/
DoorDash Outage
Next is an article that describes a postmortem from a Kubernetes outage at DoorDash. This article helps illustrate the importance of not only reading the docs but also expressing the importance of testing your various failure scenarios to ensure that your entire team understands the various failure mechanisms and their impacts on a normal operating system.
https://doordash.engineering/2022/08/09/how-to-handle-kubernetes-health-checks/
Kubernetes Storage
Sticking with Kubernetes (and any flavor of Kubernetes you may be interested in); dealing with storage as part of your implementation OnPrem is a challenge. This article does a great job of illustrating that challenge and what I means to anyone who has not yet made it to the cloud. What I hope you take from this is, it is not impossible to accomplish and there are many solutions to this problem, but the downstream operational impacts of doing this are far broader than the initial implementation.
https://refaktory.net/blog/posts/self-hosted-kubernetes-solving-the-storage-problem
Traffic Visualization
Visualizing what is happening in your cluster can be a challenge. I have taken the strategy of implementing Linkerd into my lab cluster to support both mTLS along with traffic visualization, but that is not the only way to accomplish this. Lots of teams are familiar with Grafana as their source of information, especially with the proliferation of something like Prometheus. k8spacket is a tool that helps bring a version of that visualization directly into Grafana dashboards.
