From 7739b827878a7dd125ec934911b50e1a3cd9cc47 Mon Sep 17 00:00:00 2001 From: fogelito Date: Tue, 30 Jun 2026 14:28:49 +0300 Subject: [PATCH] ci: retry docker compose pull to survive transient registry timeouts The "Load and Start Appwrite" step ran `docker compose pull` as a plain step with no retry, so a single transient Docker Hub timeout (e.g. a traefik manifest fetch timing out) failed the whole job. Wrap the pull in a 3-attempt retry loop across all test-gating jobs. `up --wait` stays single-shot so genuine container health failures still surface immediately rather than being masked by retries. Co-Authored-By: Claude Opus 4.8 (1M context) --- .github/workflows/ci.yml | 55 ++++++++++++++++++++++++++++++++++++---- 1 file changed, 50 insertions(+), 5 deletions(-) diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml index e3883e1838..219270d06e 100644 --- a/.github/workflows/ci.yml +++ b/.github/workflows/ci.yml @@ -306,7 +306,16 @@ jobs: - name: Load and Start Appwrite timeout-minutes: 5 run: | - docker compose pull --quiet --ignore-buildable + # Docker Hub registry timeouts are a common transient CI failure + # (e.g. a traefik manifest fetch timing out). Retry the image pull; + # keep `up` single-shot so genuine container health failures still surface. + for attempt in 1 2 3; do + if docker compose pull --quiet --ignore-buildable; then + break + fi + echo "::warning::docker compose pull failed (attempt ${attempt}/3); retrying in 15s" + sleep 15 + done docker compose up -d --quiet-pull --wait - name: Environment Variables @@ -353,7 +362,16 @@ jobs: - name: Load and Start Appwrite timeout-minutes: 5 run: | - docker compose pull --quiet --ignore-buildable + # Docker Hub registry timeouts are a common transient CI failure + # (e.g. a traefik manifest fetch timing out). Retry the image pull; + # keep `up` single-shot so genuine container health failures still surface. + for attempt in 1 2 3; do + if docker compose pull --quiet --ignore-buildable; then + break + fi + echo "::warning::docker compose pull failed (attempt ${attempt}/3); retrying in 15s" + sleep 15 + done docker compose up -d --quiet-pull --wait - name: Wait for Open Runtimes @@ -443,7 +461,16 @@ jobs: _APP_DATABASE_DOCUMENTSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'documentsdb_db_main' || '' }} _APP_DATABASE_VECTORSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'vectorsdb_db_main' || '' }} run: | - docker compose pull --quiet --ignore-buildable + # Docker Hub registry timeouts are a common transient CI failure + # (e.g. a traefik manifest fetch timing out). Retry the image pull; + # keep `up` single-shot so genuine container health failures still surface. + for attempt in 1 2 3; do + if docker compose pull --quiet --ignore-buildable; then + break + fi + echo "::warning::docker compose pull failed (attempt ${attempt}/3); retrying in 15s" + sleep 15 + done docker compose up -d --quiet-pull --wait - name: Wait for Open Runtimes @@ -526,7 +553,16 @@ jobs: _APP_DATABASE_DOCUMENTSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'documentsdb_db_main' || '' }} _APP_DATABASE_VECTORSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'vectorsdb_db_main' || '' }} run: | - docker compose pull --quiet --ignore-buildable + # Docker Hub registry timeouts are a common transient CI failure + # (e.g. a traefik manifest fetch timing out). Retry the image pull; + # keep `up` single-shot so genuine container health failures still surface. + for attempt in 1 2 3; do + if docker compose pull --quiet --ignore-buildable; then + break + fi + echo "::warning::docker compose pull failed (attempt ${attempt}/3); retrying in 15s" + sleep 15 + done docker compose up -d --quiet-pull --wait - name: Run tests @@ -584,7 +620,16 @@ jobs: _APP_DATABASE_DOCUMENTSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'documentsdb_db_main' || '' }} _APP_DATABASE_VECTORSDB_SHARED_TABLES: ${{ matrix.mode != 'dedicated' && 'vectorsdb_db_main' || '' }} run: | - docker compose pull --quiet --ignore-buildable + # Docker Hub registry timeouts are a common transient CI failure + # (e.g. a traefik manifest fetch timing out). Retry the image pull; + # keep `up` single-shot so genuine container health failures still surface. + for attempt in 1 2 3; do + if docker compose pull --quiet --ignore-buildable; then + break + fi + echo "::warning::docker compose pull failed (attempt ${attempt}/3); retrying in 15s" + sleep 15 + done docker compose up -d --quiet-pull --wait - name: Wait for Open Runtimes