BestHub
Sign in
Tagged articles

preStop Hook

6 articles · Page 1 of 1
Linux Cloud-Native Ops Stack
Linux Cloud-Native Ops Stack
Apr 23, 2026 · Cloud Native

Kubernetes Interview: What Exactly Happens When You Delete a Pod?

When a pod is deleted in Kubernetes, the API server timestamps the request, services stop routing traffic, kubelet runs any preStop hook, sends SIGTERM followed by a configurable grace period, then SIGKILL if needed, cleans up resources, and finally removes the pod metadata, with controllers optionally recreating a replacement pod.

DeploymentGraceful TerminationKubernetes
0 likes · 7 min read
Kubernetes Interview: What Exactly Happens When You Delete a Pod?
360 Quality & Efficiency
360 Quality & Efficiency
Apr 14, 2023 · Cloud Native

Ensuring Zero‑Downtime Rolling Updates in Kubernetes: Causes and Solutions

This article analyzes why Kubernetes rolling updates can still cause service interruptions during pod startup and termination, explains the underlying mechanisms of Kubelet and Endpoint Controller, and provides practical steps such as readiness probes and preStop hooks to achieve smoother, near‑zero‑downtime deployments.

KubernetesReadiness ProbeZero Downtime
0 likes · 7 min read
Ensuring Zero‑Downtime Rolling Updates in Kubernetes: Causes and Solutions
System Architect Go
System Architect Go
Mar 27, 2023 · Cloud Native

Understanding Kubernetes Endpoint Propagation and Graceful Pod Deletion

Deleting a pod triggers endpoint removal, but various components like kube-proxy, CoreDNS, and ingress controllers may still route traffic until the endpoint fully propagates, so you must wait or use preStop hooks to delay deletion and handle SIGTERM gracefully within the configurable shutdown period.

Endpoint PropagationGraceful ShutdownKubernetes
0 likes · 5 min read
Understanding Kubernetes Endpoint Propagation and Graceful Pod Deletion
Alibaba Cloud Native
Alibaba Cloud Native
Jun 2, 2020 · Cloud Native

Why Kubernetes Deployments Cause Service Outages and How to Prevent Them

This article explains why a typical Deployment + LoadBalancer setup can experience downtime during updates, analyzes the pod lifecycle, endpoint synchronization, iptables/ipvs and SLB interactions, and provides concrete configuration steps—including readiness probes, preStop hooks, and Service traffic policies—to achieve zero‑downtime deployments.

Cluster ModeDeploymentENI
0 likes · 12 min read
Why Kubernetes Deployments Cause Service Outages and How to Prevent Them