Error AtlasError Documentation and Resolution
Back to search

GKE

GKE image pull failed with i/o timeout

i/o timeout

A Google Kubernetes Engine node timed out while connecting to the image registry, usually because of network reachability, firewall, NAT, DNS, or internet access problems.

i/o timeoutnetworking

Observed message

i/o timeout

Why it happens

  • The cluster cannot reach the image registry endpoint over the network.
  • Firewall, DNS, Private Google Access, or NAT configuration is incomplete.
  • A network-isolated cluster lacks the required egress path to the registry.

How to fix it

  1. Verify registry connectivity, DNS resolution, and firewall behavior for the cluster nodes.
  2. Enable or confirm Private Google Access or NAT where required.
  3. Check whether an external registry endpoint has working internet connectivity.

Official reference

Google Cloud: Troubleshoot image pulls in GKE

Related errors

0/1 nodes are available: 1 node(s) didn't have free ports for the requested pod ports.GKE could not schedule the pod because nodes did not have free requested pod ports401 UnauthorizedGKE image pull failed with 401 UnauthorizedNotReadyGKE node entered NotReady statusImagePullBackOffGKE pod failed to pull its container image

Explore more

networkingBrowse more networking errors
GKE image pull i/o timeout: causes and fixes | Error Atlas