Kubernetes Image Volumes with PullPolicy: Always Cause Excessive Pulls During CrashLoops
Is it fixed? Not yet.
Ongoing since Nov 5, 00:29 UTC. First reported Nov 5, 00:29 UTC.
What happened
When a container crashloops, Kubernetes repeatedly pulls every image volume with PullPolicy: Always on every syncPod() call, even if the volumes are not mounted by the crashing container. This introduces unnecessary churn for the container runtime and consumes slots in the node's image pull queue.
What caused it?
Under active investigation. No report has identified a root cause yet.
No verified workaround yet
None of the linked reports describe a mitigation. This page updates when one appears.
Timeline
- Investigating(current)
- Workaround available(upcoming)
- Resolved(upcoming)
- First report on GitHub issue
Image Volumes with `PullPolicy: Always` are pulled on any syncPod() call when a container is crashlooping
Source
Evidence
1 report · 1 GitHubShow
- [1]GitHub First reportImage Volumes with `PullPolicy: Always` are pulled on any syncPod() call when a container is crashlooping(opens GitHub issue in a new tab)
What happened? When a container crashloops, every image volume with pullpolicy: always (even those not mounted by the crashing container) is pulled every time SyncPod is called (which is at least on every backoff step and at most on any event on the pod). Each image pull takes a spot in the node's image pull queue and introduces unnecessary churn for the container runtime. What did you expect to happen? Only the image volumes mounted by the crashing container should be pulled at the time of…
by adamwathieu 0upvotes 5comments
FAQ
What happened in the Kubernetes issue?
When a container crashloops, Kubernetes repeatedly pulls every image volume with PullPolicy: Always on every syncPod() call, even if the volumes are not mounted by the crashing container. This introduces unnecessary churn for the container runtime and consumes slots in the node's image pull queue.
What is affected by the Kubernetes issue?
Kubernetes.
When did the Kubernetes issue start and end?
It started Nov 5, 00:29 UTC and is still ongoing (14 d 4 h so far). This page updates as it develops.
How do I get alerts for Kubernetes?
Subscribe to the Kubernetes RSS feed for every new Kubernetes incident, or connect an AI assistant to WhatBrokeToday over MCP and ask whether Kubernetes is down. Kubernetes feed
Where does this information come from?
WhatBrokeToday collects public status pages, developer community reports and security advisories, groups the ones about the same problem, and links every report under Evidence on this page. This one has 1 report from 1 independent source.