The latest articles on DEV Community by Aron Fyodor Asor (@aronasorman). https://dev.to/aronasorman https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F277861%2F05a501a3-4877-416a-8f54-c4eebba40211.jpg https://dev.to/aronasorman en Aron Fyodor Asor Sat, 23 Nov 2019 03:05:04 +0000 https://dev.to/aronasorman/my-top-4-takeaways-from-kubecon-2019-4g7g https://dev.to/aronasorman/my-top-4-takeaways-from-kubecon-2019-4g7g <ul> <li><p>When your pod is terminating, it takes a while for requests to stop hitting your pod. In your pod definitions, always have a <code>preStop</code> with <code>sleep 20</code> to allow Endpoint changes to propagate. After <code>preStop</code> is when Kubernetes hits your containers with a <code>SIGTERM</code>.</p></li> <li><p>Always have CPU requests. Be very wary of CPU limits -- don't add them unless you really need to. This allows your container to use more CPU when it's available. But when CPU pressure comes in, your app is guaranteed the amount of CPU cycles it requested.</p></li> <li><p>A corollary to the above: there is always tension between your cluster's utilization and stability. You can increase your utilization rate i.e. by increasing the difference between requests and limits. But that ups the chance of eviction. Reduce utilization, and you'll see less evictions. Find the balance for you.</p></li> <li><p>Automation is not the end-all-be-all. When your automation doesn't work, someone has to fix it after all.</p></li> </ul> kubernetes devops Aron Fyodor Asor Sat, 23 Nov 2019 02:56:33 +0000 https://dev.to/aronasorman/testing-testing-29bi https://dev.to/aronasorman/testing-testing-29bi <p>Hello world! How is this different from medium or reddit?</p>