olares-cluster

Inspect and manage Kubernetes pods, workloads, jobs, and logs on Olares via olares-cli.

5.2k|321|Updated Apr 29, 2024
One-click install
npx skills add https://github.com/beclab/Olares --skill olares-cluster
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: olares-cluster
Source: https://github.com/beclab/Olares/tree/main/cli/skills/olares-cluster
Command: npx skills add https://github.com/beclab/Olares --skill olares-cluster

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires olares-cli, and includes references (resource) components.

What problem does it solve?

Operators of an Olares personal cloud need a raw Kubernetes runtime view to inspect pods, containers, workloads, jobs, cronjobs, namespaces, nodes, and middleware, and to perform controlled mutations like scaling, restarting, or deleting objects without leaving the CLI.

Core Features & Use Cases

  • Runtime inspection: List and get pods, workloads (Deployment/StatefulSet/DaemonSet), jobs, cronjobs, namespaces, nodes, and middleware, with YAML, events, and aggregated application status views.
  • Logs and exec: Tail container logs with poll-based follow, and run one-shot commands inside containers over the K8s exec WebSocket with a namespace permission gate.
  • Controlled mutations: Scale, restart, stop, start, or delete workloads, suspend/resume cronjobs, and rerun jobs, all behind confirmation prompts and server-side authorization.
  • Use Case: A user notices an app is misbehaving and asks the agent to check pod status, pull recent logs, and restart the deployment, with each destructive step confirmed before execution.

Quick Start

Ask the agent to list all pods in your Olares cluster and show the recent logs of a specific pod using olares-cli cluster commands.

Frequently Asked Questions about olares-cluster

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I view pod logs on an Olares cluster from the CLI?

Use olares-cli cluster pod logs <namespace>/<pod> to stream plain-text logs to stdout. Add -f to poll-follow, --tail and --since to bound the initial fetch, and -c to pick a container in multi-container pods.

How do I restart a deployment on Olares with olares-cli?

Run olares-cli cluster workload restart <ns/name> --kind deploy, which deletes the workload's pods in parallel-bounded batches so the controller recreates them. The verb is destructive and asks for confirmation unless --yes is passed.

Can I exec into a container on Olares like kubectl exec?

Yes, olares-cli cluster pod exec and container exec run commands over the K8s exec WebSocket, requiring Olares 1.12.7 or later. One-shot mode is for agents; interactive -it requires a real terminal, and a namespace permission gate restricts which namespaces each role can exec into.

What is the difference between cluster workload stop and market stop on Olares?

cluster workload stop scales the Kubernetes controller to zero replicas and does not touch the app-store lifecycle row. market stop/resume updates the Olares Market application lifecycle, so use the market skill for app-level stop and resume.

Why does olares-cli cluster workload get fail with --kind is required?

Every workload verb except list requires --kind to disambiguate Deployment, StatefulSet, or DaemonSet. Add --kind deploy, sts, or ds; for list you can omit it or use all to get a merged cross-kind view.

When should I use olares-doctor instead of the cluster skill?

Use olares-doctor when diagnosing why an app is broken, such as stuck installs, crash loops, or image pull failures, since it orchestrates cluster commands into symptom-to-root-cause routing. The cluster skill stays the raw runtime view and the place that mutates K8s objects.