CI "context deadline exceeded"

By Daniel Zoghalchali·Latchkey

A client aborted an operation because its context deadline (timeout) passed before the work finished. In CI this is common from Go tooling, Docker, and kubectl when a remote call is slow.

What this error means

An operation fails with context deadline exceeded, often pulling an image, calling an API, or waiting on a resource. A re-run usually succeeds when the slowness was transient.

Error: context deadline exceeded
failed to pull image "registry/app:latest": context deadline exceeded

Common causes

A slow remote call exceeded the deadline

A transiently slow registry, API, or network made the call run past the client timeout.

The deadline is too tight

A short configured timeout cannot cover a legitimately slow operation.

How to fix it

Retry, and raise the deadline if needed

Retry the operation and give it a longer timeout where the client allows.

for i in 1 2 3; do
  kubectl --request-timeout=60s get pods && break
  sleep 5
done

Reduce the slow path

1. Use a closer mirror/cache for pulls.
2. Reduce concurrent calls hitting the same endpoint.
3. Authenticate to avoid throttling-induced slowness.

How to prevent it

• Set deadlines that cover realistic latency.
• Retry deadline-prone operations in CI.
• When a transient infra slowdown trips the deadline, managed runners detect the mechanical blip and automatically retry, so a one-off failure does not break the build.

Frequently asked questions

Related guides

References

• GitHub Actions documentation