flake: Build Linux Docker images - GHCR manifest timeout

[flake-investigator]

CI Run: https://github.com/coder/coder/actions/runs/25671604621
Failed Job: https://github.com/coder/coder/actions/runs/25671604621/job/75358104546 (build)
Failure time: 2026-05-11T13:07:48Z

Commit:

• SHA: abe9f44e036d3a9af41c84d9afe6290e0c9f72f9
• Author: Michael Suchacz
• coder/coder@abe9f44

Error excerpt:

--- Creating multi-arch Docker image (ghcr.io/coder/coder-preview:2.34.0-rc.0-devel-abe9f44e0)
Get "https://ghcr.io/v2/coder/coder-preview/manifests/v2.34.0-rc.0-devel-abe9f44e0-arm64": Get "https://ghcr.io/token?account=ibetitsmike&scope=repository%3Acoder%2Fcoder-preview%3Apull&service=ghcr.io": net/http: request canceled (Client.Timeout exceeded while awaiting headers)

Root cause classification:

• Infrastructure / registry timeout while fetching GHCR manifest during multi-arch build.

Data race / panic / OOM checks:

• Searched logs for data race warnings, panics, and OOM indicators — none found.

Assignment analysis:

• Intended blame command: git blame -L <multiarch manifest section> scripts/build_docker_multiarch.sh
• History (proxy): git log --oneline -10 --follow scripts/build_docker_multiarch.sh
  • 8196221 chore: parallel makefile attempt 3 (Dean Sheather)
• Assigning to @deansheather as the most recent modifier of the build script; please reassign if infra ownership differs.

Related issues searched (no matches):

• "Client.Timeout exceeded" + ghcr.io
• "ghcr.io/token"
• "build_docker_multiarch.sh"
• "Build Linux Docker images"

Reproduction (best effort):

• Re-run the build job; failure appears network/registry dependent.

[flake-investigator]

New occurrence in https://github.com/coder/coder/actions/runs/25672426916
Job: https://github.com/coder/coder/actions/runs/25672426916/job/75361046050 (build)
Commit: 67aa62557984a2cbb33307ffc76fb475e2f49300 (Michael Suchacz) coder/coder@67aa625

Failure during GHCR Login:

Error response from daemon: Get "https://ghcr.io/v2/": Get "https://ghcr.io/token?account=ibetitsmike&client_id=docker&offline_token=true&service=ghcr.io": net/http: request canceled (Client.Timeout exceeded while awaiting headers) (Client.Timeout exceeded while awaiting headers)

Looks like another GHCR network timeout, this time at the login step before build/push.

[flake-investigator]

New occurrence in main CI:

• CI Run: https://github.com/coder/coder/actions/runs/25791859937
• Failed Job: https://github.com/coder/coder/actions/runs/25791859937/job/75759237856 (build)
• Failure time: 2026-05-13T10:07:15Z

Error excerpt:

failed to configure transport: error pinging v2 registry: Get "https://ghcr.io/v2/": net/http: request canceled (Client.Timeout exceeded while awaiting headers)

Looks like the same GHCR registry timeout while pushing the multi-arch manifest.