From 6d7469e9b81ff91b9848bc7dd7091cbaa89362a4 Mon Sep 17 00:00:00 2001
From: Marcel <marcel@familienarchiv>
Date: Tue, 14 Apr 2026 15:29:28 +0200
Subject: [PATCH] fix(deploy): increase OCR healthcheck start_period, comment
 ocr_cache volume, add token hint
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- start_period 60s → 120s: Zenodo download on cold start can exceed 60s on slow connections
- ocr_cache volume comment: documents what the cache stores for future operators
- .env.example: add token generation command to prevent weak placeholder in production

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
---
 .env.example       | 1 +
 docker-compose.yml | 4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/.env.example b/.env.example
index 6ba5dcf9..5b928337 100644
--- a/.env.example
+++ b/.env.example
@@ -23,6 +23,7 @@ PORT_MAILPIT_SMTP=1025
 
 # OCR Training — secret token required to call /train and /segtrain on the OCR service.
 # Also set in the backend so it can pass the token through. Must not be empty in production.
+# Generate with: python3 -c "import secrets; print(secrets.token_hex(32))"
 OCR_TRAINING_TOKEN=change-me-in-production
 
 # Production SMTP — uncomment and fill in to send real emails instead of catching them
diff --git a/docker-compose.yml b/docker-compose.yml
index 35660e0f..0637d417 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -87,7 +87,7 @@ services:
     memswap_limit: 12g
     volumes:
       - ocr_models:/app/models
-      - ocr_cache:/root/.cache
+      - ocr_cache:/root/.cache  # Hugging Face / ketos model download cache — prevents re-downloads on container recreate
     environment:
       KRAKEN_MODEL_PATH: /app/models/german_kurrent.mlmodel
       TRAINING_TOKEN: "${OCR_TRAINING_TOKEN:-}"
@@ -102,7 +102,7 @@ services:
       interval: 10s
       timeout: 5s
       retries: 12
-      start_period: 60s
+      start_period: 120s
 
   # --- Backend: Spring Boot ---
   backend: