fix(release): share sibling paths through the workspace (#1)
quality-gate / quality (push) Successful in 2m42s
quality-gate / container (push) Successful in 2m9s
immutable-release / release (push) Failing after 6m7s

The release job stopped at the bind-source gate. RUNNER_TEMP is
container-internal on this runner and no runner mount exposes it on the Docker
host, so no sibling container could ever share it; only the workspace is
host-mounted.

Move every path shared between the runner's shell steps and its sibling
containers under $GITHUB_WORKSPACE/.release-work: the release directory, the
release builder's HOME and NuGet cache, both candidate image tars, and the
container SPDX inventory. Resolution now maps the workspace alone to its host
path, and each sibling binds that source at $GITHUB_WORKSPACE and works from
there instead of /source, so a shared path is the same string on both sides of
the boundary. publish-release.sh follows with a single bind and requires the
release directory to sit inside the workspace.

Ignore .release-work in Git so the tag gate's cleanliness check stays true while
artifacts accumulate, and in Docker so artifacts written between the two
candidate builds cannot alter the build context the byte-comparison gate
depends on. Skip it in the dependency inventory as well, keeping the restored
package cache out of the license policy scan.
This commit is contained in:
KyuubiYoru
2026-08-22 20:31:18 +02:00
parent c5bc9651e5
commit 4233f12368
5 changed files with 61 additions and 51 deletions
+16 -7
View File
@@ -11,11 +11,21 @@ token="${RENDEZVOUS_RELEASE_TOKEN:?RENDEZVOUS_RELEASE_TOKEN is required}"
username="${RENDEZVOUS_RELEASE_USERNAME:?RENDEZVOUS_RELEASE_USERNAME is required}"
release_builder="${RENDEZVOUS_RELEASE_BUILDER:?RENDEZVOUS_RELEASE_BUILDER is required}"
# Sibling-container runners execute this script inside a container while
# `docker run` starts containers on the host daemon, so bind sources must be
# host paths. The workflow resolves them once and exports them; a direct host
# run keeps the local paths.
# `docker run` starts containers on the host daemon, so the bind source must be
# a host path. The workflow resolves the workspace once and exports it; a direct
# host run keeps the local path. Only the workspace is bound, so the release
# directory has to live inside it, and binding it back onto its own path keeps
# every shared path identical on both sides of the boundary.
workspace_source="${RENDEZVOUS_WORKSPACE_SOURCE:-$root}"
release_dir_source="${RENDEZVOUS_RELEASE_DIR_SOURCE:-$release_dir}"
[[ -d "$release_dir" ]] || {
echo "Release directory does not exist: $release_dir" >&2
exit 1
}
release_dir="$(cd "$release_dir" && pwd)"
if [[ "$release_dir" != "$root"/* ]]; then
echo "Release directory must live inside the workspace: $release_dir" >&2
exit 1
fi
docker_config="$(mktemp -d)"
curl_config="$(mktemp)"
release_request=""
@@ -52,9 +62,8 @@ cosign public-key --key env://COSIGN_PRIVATE_KEY >/dev/null || {
run_release_builder() {
docker run --rm \
--user "$(id -u):$(id -g)" \
--volume "$workspace_source:/source:ro" \
--volume "$release_dir_source:$release_dir" \
--workdir /source \
--volume "$workspace_source:$root" \
--workdir "$root" \
"$release_builder" "$@"
}