Fix pod_kill to use pod-failure action to prevent immediate pod recreation

[Copilot]

The pod_kill chaos experiment was using the pod-kill action, which deletes pods and triggers immediate Kubernetes controller recreation, defeating the fault injection.

Changes

• inject_symp.py: Changed inject_pod_kill() to use pod-failure action instead of pod-kill

  • pod-kill forcefully deletes pods → controller recreates immediately
  • pod-failure makes pods unavailable for duration without deletion → prevents recreation

• chaos-yaml/pod-kill.yaml: Updated template to reflect correct action

The fault now persists for the specified duration (e.g., 100s) as intended.

chaos_experiment = {
    "spec": {
-       "action": "pod-kill",
+       "action": "pod-failure",
        "mode": "one",
        "duration": duration,
        # ...
    }
}

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Pod killing does not work</issue_title>
<issue_description>Problem id: pod_kill_hotel_res-localization-1

test-hotel-reservation   search-5c64c8f5bc-cz8hr                                    1/1     Running      0              91s
test-hotel-reservation   user-66d5d6f874-9gh86                                      1/1     Running      0              40s

The user service should get killed by chaos-mesh but k8s recreates the pod immediately. Is this expected?</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Pod killing does not work #111

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.