Danylo·Just nowI broke my GitOps control plane on purpose. Here’s what survived.Seven failure-injection tests on four real Ubuntu hosts — and the gaps the test still does not cover.
Vortex 404·6h agoGrafana’s AI Assistant Can Now Find an Incident, Investigate It, and Open the Fix as a Pull RequestWatchers detect what you never wrote an alert rule for, Investigations explain why, and Managed Sandboxes propose the actual code change —…
Jon Handler·15h agoFrom Log Lines to Agent Investigations: An Observability Maturity Model for Amazon OpenSearch…Every on-call team climbs the same ladder. The agent at the top is only as good as the rungs underneath it.
Alexey Horey·18h agoDynamic adaptation of Monitoring.When wearing my SRE hat I am driven by two ideas:
Sabbagh·22h agoOne Zone Lost Power and Banks Went Down. Is Your Multi-Zone Real?I cut the power to one zone of a test Kubernetes cluster. Traffic recovered in 50 seconds. Three things never came back on their own.
Richard King·23h agoHardware Tiering & Specification: Repurposed Silicon, 30W Baseline, and Repatriating $19,300 in…Author: Richard King / Ubinity AI
Malathi B Shetty·1d agoSLA vs SLO vs SLI: Understanding the Difference Without Getting ConfusedWhile learning DevOps and SRE, I came across three terms quite often:
Jiten I·1d agoZero errors, wrong answers: catching stale data behind an AI agent with CloudWatch OmniI broke the service behind my AI agent on purpose. Every dashboard stayed green. This is how I caught it with Amazon CloudWatch Omni.
InChaos Engineering Field Guide — Experiments, Learnings, and Best Practices.byAbhishek Datta·1d agoThe Outage That Outlives Its TriggerChaos Engineering and SRE recovery testing for retry storms, cold caches and metastable failures
