Skip to content

[BUG] kagent 100% CPU and not responding well to queries #911

@LutzLange

Description

@LutzLange

📋 Prerequisites

  • I have searched the existing issues to avoid creating a duplicate
  • By submitting this issue, you agree to follow our Code of Conduct
  • I am using the latest version of the software
  • I have tried to clear cache/cookies or used incognito mode (if ui-related)
  • I can consistently reproduce this issue

🎯 Affected Service(s)

Not Sure

🚦 Impact/Severity

Blocker

🐛 Bug Description

I'm trying out kagent with subagents.

I've created a new agent and added k8s-agent and helm-agent as resources.
I did remove the doc-query tool from booth agents before doing this. I've asked my new agent to list pods running on my kubernetes cluster.

I can see the call to the to k8s_agent and a response, but I don't get a text response.
The k8s-agent pod goes to 100% CPU.

Image Image

Here is the UI:

Image

🔄 Steps To Reproduce

I'm running on a kind cluster on my Ubuntu 24.04.

I've used this guide for setup.

🤔 Expected Behavior

I should get a text response with the running pods.
k8s-agent pod should not eat away CPU cycles at 100% after one call.

📱 Actual Behavior

No response

💻 Environment

Ubuntu 24.04 LTS
Kind Cluster 1.34.0
Kagent 0.6.12

Kagent Debug:

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

Projects

Status

Backlog

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions