fix

fix: tika image name

.env.example

@@ -56,8 +56,11 @@ TRAEFIK_SERVICES_TLS_CONFIG="tls.certresolver=letsencrypt"
 #   certificates:
 #     - certFile: /certs/opencloud.test.crt
 #       keyFile: /certs/opencloud.test.key
-#       stores:
-#         - default
+#   stores:
+#     default:
+#       defaultCertificate:
+#         certFile: /certs/opencloud.test.crt
+#         keyFile: /certs/opencloud.test.key
 #
 # The certificates need to be copied into ./certs/, the absolute path inside the container is /certs/.
 # You can also use TRAEFIK_CERTS_DIR=/path/on/host to set the path to the certificates directory.
@@ -175,7 +178,11 @@ START_ADDITIONAL_SERVICES=""
 # search/tika.yml or by using the following command:
 # docker compose -f docker-compose.yml -f search/tika.yml up -d
 # Set the desired docker image tag or digest.
-# Defaults to "apache/tika:latest-full"
+# Defaults to "apache/tika:latest"
+# The slim variant is recommended for most use cases as it provides core text extraction
+# functionality with a smaller image size and faster startup time.
+# Only use the full variant (apache/tika:latest-full) if you need specialized features
+# like advanced OCR or specific image processing capabilities.
 TIKA_IMAGE=
 
 ### IMPORTANT Note for Online Office Apps ###
@@ -214,7 +221,8 @@ COLLABORA_HOME_MODE=
 ### Virusscanner Settings ###
 # IMPORTANT: If you enable antivirus, you also MUST configure the START_ADDITIONAL_SERVICES
 # envvar in the OpenCloud Settings above by adding 'antivirus' to the list.
-# The maximum scan size the virus scanner can handle, needs adjustment in the scanner config as well.
+# The maximum scan size the virus scanner can handle, needs adjustment in the scanner config as well:
+# For ClamAV, set CLAMD_CONF_StreamMaxLength in antivirus/clamav.yml to the same or a higher value.
 # Usable common abbreviations: [KB, KiB, MB, MiB, GB, GiB, TB, TiB, PB, PiB, EB, EiB], example: 2GB.
 # Defaults to "100MB"
 #ANTIVIRUS_MAX_SCAN_SIZE=

README.md

@@ -16,6 +16,7 @@ OpenCloud Compose offers a modular approach to deploying OpenCloud with several
 - **Full text search** with Apache Tika for content extraction and metadata analysis
 - **Monitoring** with metrics endpoints for observability and performance monitoring
 - **Radicale** integration for Calendar and Contacts
+- **ClamAV** antivirus scanning with ClamAV
 
 ## Quick Start Guide
 
@@ -147,6 +148,14 @@ This setup includes:
 - Full text search functionality in the OpenCloud interface
 - Support for documents, PDFs, images, and other file types
 
+**Tika Image Variant:**
+By default, OpenCloud Compose uses `apache/tika:latest` which provides:
+- Smaller image size (~300MB vs ~1.2GB for the full variant)
+- Faster container startup and deployment
+- Core text extraction functionality for common document formats (PDF, Office docs, text files, etc.)
+
+The base variant is recommended for most use cases. If you need advanced features like specialized OCR processing or specific image format support, you can override the image by setting `TIKA_IMAGE=apache/tika:latest-full` in your `.env` file.
+
 ### With Radicale
 
 Enable CalDAV (calendars, to-do lists) and CardDAV (contacts) server.
@@ -223,6 +232,25 @@ This exposes the necessary ports:
 If you're using **Nginx Proxy Manager (NPM)**, you **should NOT** activate **"Block Common Exploits"** for the Proxy Host.
 Otherwise, the desktop app authentication will return **error 403 Forbidden**.
 
+### ClamAV anti-virus
+
+Enable anti-virus scans for uploaded files.
+
+Using `-f` flags:
+```bash
+docker compose -f docker-compose.yml -f antivirus/clamav.yml -f traefik/opencloud.yml up -d
+```
+
+Or by setting in `.env`:
+```
+COMPOSE_FILE=docker-compose.yml:antivirus/clamav.yml:traefik/opencloud.yml
+```
+
+**Important:** adjust the variable in `.env` to start the antivirus service. Add additional services separated by comma, e.g. `notifications,antivirus`:
+```
+START_ADDITIONAL_SERVICES="antivirus"
+```
+
 
 ## SSL Certificate Support
 
@@ -318,7 +346,7 @@ Key variables:
 | `INSECURE`                    | Skip certificate validation                           | true                         |
 | `COLLABORA_DOMAIN`            | Collabora domain                                      | collabora.opencloud.test     |
 | `WOPISERVER_DOMAIN`           | WOPI server domain                                    | wopiserver.opencloud.test    |
-| `TIKA_IMAGE`                  | Apache Tika image tag                                 | apache/tika:latest-full      |
+| `TIKA_IMAGE`                  | Apache Tika image tag                                 | apache/tika:slim             |
 | `KEYCLOAK_DOMAIN`             | Keycloak domain                                       | keycloak.opencloud.test      |
 | `KEYCLOAK_ADMIN`              | Keycloak admin username                               | kcadmin                      |
 | `KEYCLOAK_ADMIN_PASSWORD`     | Keycloak admin password                               | admin                        |

antivirus/clamav.yml

@@ -0,0 +1,31 @@
+---
+services:
+  opencloud:
+    environment:
+      POSTPROCESSING_STEPS: "virusscan"
+      STORAGE_USERS_DATA_GATEWAY_URL: "http://opencloud:9200/data"
+      ANTIVIRUS_MAX_SCAN_SIZE: ${ANTIVIRUS_MAX_SCAN_SIZE:-100MB}
+      ANTIVIRUS_INFECTED_FILE_HANDLING: abort
+      ANTIVIRUS_MAX_SCAN_SIZE_MODE: ${ANTIVIRUS_MAX_SCAN_SIZE_MODE:-partial}
+      ANTIVIRUS_WORKERS: 1
+      ANTIVIRUS_CLAMAV_SOCKET: /var/run/clamav/clamd.sock
+      ANTIVIRUS_SCANNER_TYPE: clamav
+    volumes:
+      - clamav-socket:/var/run/clamav
+  clamav:
+    image: clamav/clamav:${CLAMAV_DOCKER_TAG:-latest}
+    environment:
+      # Accepts a number with optional K, M or G suffix. Must be greater or equal to ANTIVIRUS_MAX_SCAN_SIZE above.
+      # K = KiB (1024), M = MiB (1024 * 1024), G = GiB (1024 * 1024 * 1024)
+      CLAMD_CONF_StreamMaxLength: 100M
+    networks:
+      opencloud-net:
+    volumes:
+      - clamav-socket:/tmp
+      - clamav-db:/var/lib/clamav
+    logging:
+      driver: ${LOG_DRIVER:-local}
+    restart: always
+volumes:
+  clamav-db:
+  clamav-socket:

search/tika.yml

@@ -1,7 +1,10 @@
 ---
 services:
   tika:
-    image: ${TIKA_IMAGE:-apache/tika:latest-full}
+    image: ${TIKA_IMAGE:-apache/tika:latest}
+    # Using the base variant for smaller image size and faster startup
+    # The base variant includes core functionality for text extraction
+    # Full variant is only needed for specialized OCR/image processing
     # release notes: https://tika.apache.org
     networks:
       opencloud-net:

weboffice/collabora.yml

@@ -46,7 +46,7 @@ services:
     restart: always
 
   collabora:
-    image: collabora/code:25.04.4.2.1
+    image: collabora/code:25.04.7.1.1
     # release notes: https://www.collaboraonline.com/release-notes/
     networks:
       opencloud-net: