---
slug: "Kubernetes-The-node-was-low-on-resource-ephemeral-storage"
title: "What to Do When Your Kubernetes Node Shows \"The node was low on resource: ephemeral-storage\""
description: "Detect that an element does not exist with Pyppeteer — catch the `waitForSelector` timeout and return `None`."
url: "https://www.ytyng.com/en/blog/Kubernetes-The-node-was-low-on-resource-ephemeral-storage"
publish_date: "2021-01-15T14:38:47Z"
created: "2021-01-15T14:38:47Z"
updated: "2026-05-11T13:21:38.912Z"
categories: ["kubernetes"]
keywords: ""
featured_image_url: "https://media.ytyng.com/resize/20230812/a50fe6eec6454db6a3f07b49568be64f.png.webp?width=768"
has_video: false
has_music: false
video_urls: []
music_urls: []
lang: "en"
---

# What to Do When Your Kubernetes Node Shows "The node was low on resource: ephemeral-storage"

<ul>
<li class="p1">Pod exits with the error: "The node was low on resource: ephemeral-storage."</li>
<li class="p1">Storage usage of the node is at 85%</li>
<li class="p1">There is an unusually heavy process running in Kubernetes, causing the load average to be abnormally high (like 100)</li>
</ul>
<p class="p1">In this case, it is a storage shortage.</p>
<p class="p2"></p>
<p class="p1">The 85% figure is the default value for Kubernetes' image-gc-high-threshold. When storage usage exceeds this, image garbage collection starts.</p>
<p class="p1">However, because the cause of the storage bloat was not directly related to Kubernetes, it could not be handled and resulted in issues.</p>
<p class="p2"></p>
<p class="p1">This time, there was leftover junk from Docker containers, so I removed it using docker prune.</p>
<p class="p2"></p>
<pre class="p1">docker container prune</pre>
<p class="p1">Or</p>
<pre class="p1">docker system prune</pre>
<p class="p1"></p>