ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
DevOps & observability · #kubernetes
Learn how to mount a ConfigMap of dashboard JSON files to Grafana's provisioning directory and keep dashboards in sync with version control.
Sample loss in Prometheus remote write is usually backpressure. Sizing queue shards, batch size and retry backoff lets a growing cluster absorb bursts without unbounded memory growth.
k3OS turns the operating system into a Kubernetes resource, letting the cluster manage OS upgrades in an immutable, zero‑downtime fashion. Learn how the two‑phase upgrade works, see a concrete example, and weigh the trade‑offs before deploying.
Managing Kubernetes scale with Gardener Shoot Clusters: The multi‑cluster management bottleneck occurs when scaling to dozens of clusters across cloud providers. Gardener turns cluster lifecycle into a declarative object, simplifying provisioning and scaling.
Learn how to implement a minimal OpenTelemetry Collector sidecar in Kubernetes to handle OTLP trace ingestion, probabilistic sampling, and batch exporting to Jaeger.
Compare GKE Autopilot and Standard mode across node control, pricing, and workload fit, then verify your cluster mode with kubectl.
Update a Kubernetes Deployment image with readiness probes, rolling-update limits, and rollback for zero-downtime releases.
YAML anchors and aliases let you deduplicate configuration inside a single file, but the merge key << is a non‑standard extension with uneven parser support. Use anchors for small, local reuse; switch to Helm, Kustomize, or Jsonnet when sharing across files or teams.
A concise architecture note for running Kubeflow Pipelines v2 on Kubernetes with the smallest viable footprint, covering trust boundaries, operational signals, failure modes, and when to redesign.
grafana
prometheus
k3os
gardener
opentelemetry
grpc
google
yaml
kubeflow