From f8c45fc28964a69d51fe41f967c9fe0734b68a72 Mon Sep 17 00:00:00 2001
From: "Danilo M." <danix@danix.xyz>
Date: Mon, 7 Sep 2026 13:09:39 +0200
Subject: image-builder: abort early when the docker daemon is unreachable

A dead dockerd is a global failure, not a per-variant one, but every stage
treated it as the latter: the digest probes read an unreachable daemon as
"no digest, rebuilding to be safe", the build then failed, and the variant
loop logged "WARNING: variant X failed; continuing" and moved on. Cron kept
exiting nonzero into a log nobody read, so a daemon that died in August went
unnoticed for 25 days while no image was ever rebuilt.

Add require_docker() to lib.sh, following the existing require_mount()
precedent, and call it before the variant loop in all three stages. One loud
error, exit 1, instead of a nightly pile of warnings.

In build-full-image.sh the check goes before the --force cache prune, since
that prune also talks to the daemon.

Self-check stubs `docker` so it never contacts a real daemon.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AD4jP4wFMpBgEvd9ZP9Xhu
---
 image-builder/bootstrap.sh           | 1 +
 image-builder/build-full-image.sh    | 2 ++
 image-builder/build-sbo-testbuild.sh | 1 +
 image-builder/lib.sh                 | 9 +++++++++
 image-builder/test-image-builder.sh  | 8 ++++++++
 5 files changed, 21 insertions(+)

diff --git a/image-builder/bootstrap.sh b/image-builder/bootstrap.sh
index 250da44..7316116 100755
--- a/image-builder/bootstrap.sh
+++ b/image-builder/bootstrap.sh
@@ -556,6 +556,7 @@ DOCKERFILE
 # Main
 # ============================================================================
 
+require_docker
 rc=0
 for VERSION in "${BUILD_VARIANTS[@]}"; do
     if changelog_changed "${VERSION}"; then
diff --git a/image-builder/build-full-image.sh b/image-builder/build-full-image.sh
index f6814bd..3706186 100755
--- a/image-builder/build-full-image.sh
+++ b/image-builder/build-full-image.sh
@@ -228,6 +228,8 @@ DOCKERFILE
 # Main
 # ============================================================================
 
+require_docker
+
 # A --force build passes --no-cache, so the stale build cache is dead weight:
 # it still loads the containerd snapshotter's lease/lock bookkeeping and has
 # raced the layer export ("failed to open writer: ref ... locked ... unavailable").
diff --git a/image-builder/build-sbo-testbuild.sh b/image-builder/build-sbo-testbuild.sh
index 8778897..f982a91 100755
--- a/image-builder/build-sbo-testbuild.sh
+++ b/image-builder/build-sbo-testbuild.sh
@@ -134,6 +134,7 @@ DOCKERFILE
     _log "=== Done: ${TB_TAG} ==="
 }
 
+require_docker
 require_pkgs
 rc=0
 for VERSION in "${BUILD_VARIANTS[@]}"; do
diff --git a/image-builder/lib.sh b/image-builder/lib.sh
index dfb076b..8a22828 100644
--- a/image-builder/lib.sh
+++ b/image-builder/lib.sh
@@ -72,6 +72,15 @@ txz_hash() {
     done <<< "$files" | sha256sum | cut -d' ' -f1
 }
 
+# require_docker
+# Assert the docker daemon answers before any variant is attempted. A dead
+# daemon is global, not per-variant: without this each variant "fails and
+# continues", the digest probes read as "no digest, rebuild to be safe", and
+# cron logs warnings nightly while nothing is ever built. Fail loudly once.
+require_docker() {
+    docker info >/dev/null 2>&1 || _err "docker daemon not reachable; is dockerd running?"
+}
+
 # require_mount VERSION
 # Assert the NAS tree for VERSION is mounted (file:// mirror only).
 # x86_64 tree dir is slackware64-${VERSION}.
diff --git a/image-builder/test-image-builder.sh b/image-builder/test-image-builder.sh
index 525ea8f..5781b22 100755
--- a/image-builder/test-image-builder.sh
+++ b/image-builder/test-image-builder.sh
@@ -75,6 +75,14 @@ MIRROR="file://$tmp/mnt"
 MIRROR="https://example/x"
 ( require_mount current ) 2>/dev/null; check_rc "require_mount http skips" 0 $?
 
+# --- require_docker: stub `docker` so the check never touches a real daemon ---
+docker() { return "${DOCKER_RC:-0}"; }
+DOCKER_RC=1
+( require_docker ) 2>/dev/null; check_rc "require_docker daemon down" 1 $?
+DOCKER_RC=0
+( require_docker ) 2>/dev/null; check_rc "require_docker daemon up" 0 $?
+unset -f docker
+
 echo "----"
 echo "PASS: $pass  FAIL: $fail"
 [[ "$fail" -eq 0 ]]
-- 
cgit v1.2.3