ci: restore image publishing with a persistent Docker config
CI / test (push) Successful in 1m38s
CI / chart (push) Successful in 2s
CI / image (push) Failing after 3m29s

A multi-stage kaniko build replaces / with the last stage's root filesystem,
which deletes $HOME/.docker/config.json. Kaniko then resolves credentials at
push time without them, asks the registry for an anonymous token and Gitea
rejects the upload with 401 reqPackageAccess.

Keep the Docker credential store in /kaniko/.docker for the whole image job via
DOCKER_CONFIG and let docker login write there directly. Restore the image job
and document that CI publishes images again.
This commit is contained in:
TapTap committed 2026-11-13 15:51:01 +00:00
1 parent 60e8dadce6
commit a6cc46b747
3 files changed
+61 -11

No files matched your search

+3 -3
View File
@@ -3,9 +3,9 @@
# Builds and pushes the application image from the local machine, and can
# optionally deploy the Helm chart to the current kube context.
#
# Image pushes intentionally do not run in Gitea Actions: kaniko inside the job
# container was rejected by the registry with 401 reqPackageAccess, while the
# same credentials work from here.
# CI also builds and pushes the image (see .gitea/workflows/ci.yaml). This
# script is the local alternative, mainly for releases that also deploy with
# --deploy.
set -euo pipefail
ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"