The DevOpsBeast Blog
Field notes on Kubernetes, GPUs, Linux, and the rest of the production stack, from engineers who run real infrastructure.
The scheduler used to tell you which taint blocked your pod. Now it says 'node(s) had untolerated taint(s)' and stops. The detail still exists, just not where you are looking.
Read postLiveness probes that fire before your app is ready. Readiness probes that check the database. Exec probes leaking zombie processes by the thousands. The six mistakes that turn health checks into the cause of the outage they were supposed to prevent.
Read postA node runs out of memory. The kernel and the kubelet both pick which pod to kill. Neither of them picks the leaky one. They pick the well-behaved BestEffort pod next door. The QoS, oom_score_adj, and eviction-priority story most engineers never learn.
Read postKubernetes certificates expire silently. No warning, no alert, no graceful degradation, just a dead cluster. Here is how to fix it in five minutes and how to make sure it never happens again.
Read postThree completely different problems hide behind exit code 137. Most engineers fix the wrong one and the pod keeps crashing.
Read post