ShowMerit articleshttps://showmerit.com/abhimanyu/articles/Articles by Abhimanyu BajajOpenTelemetry Java Agent in Splunk APM: One Env Var, Zero Code, Full Service Maphttps://showmerit.com/abhimanyu/articles/opentelemetry-java-agent-in-splunk-apm-one-env-var-zero-code/https://showmerit.com/abhimanyu/articles/opentelemetry-java-agent-in-splunk-apm-one-env-var-zero-code/Thu, 05 Nov 2026 12:00:00 GMTOne environment variable put two Spring Boot services on a Splunk APM service map, with traces and JVM metrics, no code change and p95 about level.October: Kubernetes memory semantics and Collector edge fixeshttps://showmerit.com/abhimanyu/articles/open-source-summary-2026-10/https://showmerit.com/abhimanyu/articles/open-source-summary-2026-10/Sun, 01 Nov 2026 12:00:00 GMTIn October I had 10 PRs merged across four projects, mostly around Kubernetes memory semantics, Helm chart behavior, Collector receiver correctness, and active component ownership.EKS App Slow, CPU Fine? Scale on Splunk Latency with KEDAhttps://showmerit.com/abhimanyu/articles/eks-app-slow-cpu-fine-scale-on-splunk-latency-with-keda/https://showmerit.com/abhimanyu/articles/eks-app-slow-cpu-fine-scale-on-splunk-latency-with-keda/Tue, 27 Oct 2026 12:00:00 GMTBuild an EKS lab where KEDA scales a Python service on request latency from Splunk Observability Cloud instead of CPU. In my recorded run, the app scaled from 1 to 5 ready replicas while CPU stayed low.From Broken Ingress to Healthy ALB: An EKS and Splunk Hands-On Labhttps://showmerit.com/abhimanyu/articles/from-broken-ingress-to-healthy-alb-an-eks-and-splunk-hands/https://showmerit.com/abhimanyu/articles/from-broken-ingress-to-healthy-alb-an-eks-and-splunk-hands/Thu, 22 Oct 2026 12:00:00 GMTBuild a small EKS lab where an Ingress starts with an empty ADDRESS, then install the AWS Load Balancer Controller with Pod Identity. Prove the ALB works with curl, target health, controller logs, and Splunk metrics.Trace a Java service on EKS with one annotation, zero code changeshttps://showmerit.com/abhimanyu/articles/trace-a-java-service-on-eks-with-one-annotation-zero-code/https://showmerit.com/abhimanyu/articles/trace-a-java-service-on-eks-with-one-annotation-zero-code/Wed, 21 Oct 2026 12:00:00 GMTBuild an EKS lab where a plain Java service starts emitting OpenTelemetry traces after one Deployment annotation. No source edits, no image rebuild, and the Collector logs prove service.name=orders-java.Roll back an EKS release in 16 seconds without restarting a podhttps://showmerit.com/abhimanyu/articles/eks-canary-releases-alb/https://showmerit.com/abhimanyu/articles/eks-canary-releases-alb/Mon, 19 Oct 2026 12:00:00 GMTA real traffic dial for EKS releases: send 10% to v2 with one Ingress annotation, and roll back in 16 seconds without restarting a pod. Built and tested in a real AWS lab, every step included.Scale a Fargate SQS worker from zero with FILL(running, 0)https://showmerit.com/abhimanyu/articles/scale-to-zero-sqs-workers-on-ecs-fargate/https://showmerit.com/abhimanyu/articles/scale-to-zero-sqs-workers-on-ecs-fargate/Sun, 18 Oct 2026 12:00:00 GMTI reproduced an ECS Fargate scale-to-zero SQS worker and got it scaling from zero. The fix was to wrap the missing RunningTaskCount metric with FILL(running, 0).