backup & index scripts

Dockerfile

@@ -1,33 +1,37 @@
 FROM php:7.4-apache
 LABEL maintainer="Niels Lippke<nlippke@gmx.de>"
 ENV VER 5.1.13
+ENV SEEDDMS_BASE=/var/www/seeddms
+ENV SEEDDMS_HOME=/var/www/seeddms/seeddms
 
 # Update and install necessary packages
 RUN apt-get update && apt-get install --no-install-recommends gnumeric libpng-dev catdoc poppler-utils \
-    id3 docx2txt tesseract-ocr tesseract-ocr-deu ocrmypdf imagemagick vim parallel dos2unix cron -y
+    id3 docx2txt tesseract-ocr tesseract-ocr-deu ocrmypdf imagemagick vim parallel dos2unix cron rsync -y
 RUN docker-php-ext-install gd mysqli pdo pdo_mysql && \
     pear channel-update pear.php.net && pear install Log
 
 # Get seeddms
 RUN curl -fsSL https://downloads.sourceforge.net/project/seeddms/seeddms-${VER}/seeddms-quickstart-${VER}.tar.gz | tar -xzC /var/www
-RUN mv /var/www/seeddms51x /var/www/seeddms && touch /var/www/seeddms/data/conf/ENABLE_INSTALL_TOOL
+RUN mv /var/www/seeddms51x /var/www/seeddms && mkdir /var/www/seeddms/backup && mkdir -p /var/www/seeddms/import/admin && \
+    rm -rf /var/www/seeddms/conf && ln -s /var/www/seeddms/data/conf /var/www/seeddms/conf && touch /var/www/seeddms/conf/ENABLE_INSTALL_TOOL
 
 # Copy settings-files
 COPY sources/php.ini /usr/local/etc/php/
 COPY sources/000-default.conf /etc/apache2/sites-available/
 COPY sources/settings.xml /var/www/seeddms/data/conf/settings.xml
-COPY sources/ocrmypdf.sh /usr/local/bin
 COPY sources/seeddms-entrypoint /usr/local/bin
+COPY sources/*.sh /usr/local/bin/
 
 RUN chown -R www-data:www-data /var/www/seeddms/ && \
-    dos2unix /usr/local/bin/ocrmypdf.sh && chmod a+rx /usr/local/bin/ocrmypdf.sh && \
+    dos2unix /usr/local/bin/*.sh && chmod a+rx /usr/local/bin/*.sh && \
     dos2unix /usr/local/bin/seeddms-entrypoint && chmod a+rx /usr/local/bin/seeddms-entrypoint && \
-    a2enmod rewrite
+    a2enmod rewrite && \
+    echo "export SEEDDMS_BASE=$SEEDDMS_BASE" >> /usr/local/bin/seeddms-settings.sh && \
+    echo "export SEEDDMS_HOME=$SEEDDMS_HOME" >> /usr/local/bin/seeddms-settings.sh
 
-RUN cp -a /var/www/seeddms/data /var/www/seeddms/data.bak
 
 # Volumes to mount
-VOLUME [ "/var/www/seeddms/data", "/var/www/seeddms/www/ext" ]
+VOLUME [ "/var/www/seeddms/backup", "/var/www/seeddms/import", "/var/www/seeddms/www/ext" ]
 
 ENTRYPOINT [ "/usr/local/bin/seeddms-entrypoint"]
 CMD ["apache2-foreground"]

README.md

@@ -6,10 +6,56 @@ This image supports OCR processing for images and PDFs.
 
 ## How to run
 
-`docker run --name seeddms -d -v <local>:/var/www/seeddms/data -p 8080:80 seeddms`
+`docker run --name seeddms -d -v dms-data:/var/www/seeddms/data -p 8080:80 nlippke/seeddms:5.1.13`
+
+or as compose file
+
+```yaml
+version: '2'
+
+services:
+  dms:
+    image: nlippke/seeddms:5.1.13
+    ports:
+      - "8080:80"
+    environment:
+      - TZ=Europe/Berlin
+      - 'CRON_INDEX=0 0 * * *'
+      - 'CRON_BACKUP=0 23 * * *'
+    mem_limit: 2g
+    volumes:
+      - dms-data:/var/www/seeddms/data
+      - /share/Container/container-data/seeddms/extensions:/var/www/seeddms/seeddms/ext
+      - /share/Container/container-data/seeddms/backup:/var/www/seeddms/backup
+      - /share/Container/container-data/seeddms/import:/var/www/seeddms/import
+    logging:
+      options:
+        max-size: "10m"
+        max-file: "1"
+volumes:
+  dms-data:
+```
+
+## Default configuration
+
+The image is preconfigured. Nevertheless you're guided through the installation steps upon first start for a review.
+
+1. `/var/www/seeddms/data` is the central data directory. It is not intended to be bound to a host directory. Instead use a docker volume.
+2. `/var/www/seeddms/backup` is where backups are being stored. Bind it to a host directory.
+3. `var/www/seeddms/import` is being used as drop folder. Bind it to a host directory.
+4. Optionally mount `/var/www/seeddms/seeddms/ext` to allow upload of extensions.
+
+Backup and import directories should be readable/writeable by uid 33!
+
+## Backup
+
+Backup is done by syncing the `data` folder (partially) to the backup folder. Use environment variable `CRON_BACKUP` for automatic scheduling.
+
+## Full text search
+
+Indexing documents can take some time (especially on low powered NAS). Therefore indexing is done asynchronously by a job. Use `CRON_INDEX` for scheduling this job. 
 
 ## Additional information
 
 The image is base on https://github.com/ludwigprager/docker-seeddms.
 
-

sources/000-default.conf

@@ -11,12 +11,3 @@
 	ErrorLog ${APACHE_LOG_DIR}/error.log
 	CustomLog ${APACHE_LOG_DIR}/access.log combined
 </VirtualHost>
-
-
-<Directory "/var/www/seeddms/www/">
-    DirectoryIndex index.php
-    AllowOverride All
-    Order deny,allow
-    Deny from all
-    Allow from all
-</Directory>

sources/ocrmypdf.sh

@@ -1,5 +1,7 @@
 #!/bin/bash
 
+set -e
+
 inputpdf=$1
 tmpdir=/tmp/seed
 lockfile=$tmpdir/`basename $0`

sources/seeddms-entrypoint

@@ -1,17 +1,23 @@
 #!/bin/sh
-set -e
 
-echo "args  $@"
+rm -rf /tmp/seed
 
-# first run?
-if [ ! -f /var/www/seeddms/data/content.db ]; then
-    mkdir -p /var/www/seeddms/data
-    cp -a /var/www/seeddms/data.bak/* /var/www/seeddms/data
+if [ -n "$TZ" ]; then
+  echo $TZ > /etc/timezone
 fi
 
-if [ ! -L "/var/www/seeddms/conf" ]; then
-	rm -rf /var/www/seeddms/conf
-    ln -s /var/www/seeddms/data/conf /var/www/seeddms/conf    
+rm -f /etc/cron.d/seeddms
+
+if [ -n "$CRON_INDEX" ]; then
+   echo "$CRON_INDEX su -s /bin/bash -c /usr/local/bin/seeddms-generate-index.sh www-data" >> /etc/cron.d/seeddms
+fi
+
+if [ -n "$CRON_BACKUP" ]; then
+   echo "$CRON_BACKUP su -s /bin/bash -c /usr/local/bin/seeddms-generate-backup.sh www-data" >> /etc/cron.d/seeddms
+fi
+
+if [ -e /etc/cron.d/seeddms ]; then
+	crontab /etc/cron.d/seeddms
 fi
 
 rm -f /var/run/cron*

sources/seeddms-generate-backup.sh

@@ -0,0 +1,36 @@
+#!/bin/bash
+
+. /usr/local/bin/seeddms-settings.sh
+
+lockfile=$tmpdir/`basename $0`
+mkdir -p $tmpdir
+
+if [ -e "$lockfile" ]; then
+   log warn "indexing skipped because other backup is already running"
+   exit 1
+fi
+
+if ( set -o noclobber; echo "locked" > "$lockfile"); then
+  trap 'rm -f "$lockfile"; exit $?'  INT TERM KILL EXIT
+else
+  exit 1
+fi
+
+backupdir=$SEEDDMS_BASE/backup
+
+mkdir -p $backupdir/data
+
+rsync -avu --delete $SEEDDMS_BASE/data/1048576 $backupdir/data
+if [ $? != 0 ]; then
+	log error "Backup (rsync) failed"
+fi
+
+cp -auf $SEEDDMS_BASE/data/content.db $backupdir/data
+if [ $? != 0 ]; then
+	log error "Backup (database) failed"
+fi
+
+cp -auf $SEEDDMS_BASE/data/conf $backupdir/data
+if [ $? != 0 ]; then
+	log error "Backup (config) failed"
+fi

sources/seeddms-generate-index.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+
+set -e
+
+. /usr/local/bin/seeddms-settings.sh
+
+lockfile=$tmpdir/`basename $0`
+mkdir -p $tmpdir
+
+if [ -e "$lockfile" ]; then
+   log warn "indexing skipped because other indexer is already running"
+   exit 1
+fi
+
+if ( set -o noclobber; echo "locked" > "$lockfile"); then
+  trap 'rm -f "$lockfile"; exit $?'  INT TERM KILL EXIT
+else
+  exit 1
+fi
+
+pushd $SEEDDMS_HOME/utils > /dev/null
+
+/usr/local/bin/php $SEEDDMS_HOME/utils/indexer.php --config $SEEDDMS_BASE/conf/settings.xml > /dev/null
+
+popd > /dev/null

sources/seeddms-settings.sh

@@ -0,0 +1,10 @@
+#!/bin/bash
+
+tmpdir=/tmp/seed
+logfile=$SEEDDMS_BASE/data/log/`date +%Y%m%d`.log
+
+function log() {
+   printf "%s  [%s] -- (localhost) %s\n" "`date +"%b %d %H:%M:%S"`" $1 "$2" >> $logfile
+}
+
+export PATH=$PATH:/usr/local/bin

sources/settings.xml

@@ -52,7 +52,7 @@
        - cacheDir: where the preview images are saved
        - backupDir: where the backups are saved
     -->
-    <server rootDir="/var/www/seeddms/seeddms-5.1.13/" httpRoot="/" contentDir="/var/www/seeddms/data/" stagingDir="/var/www/seeddms/data/staging/" luceneDir="/var/www/seeddms/data/lucene/" logFileEnable="true" logFileRotation="d" enableLargeFileUpload="false" partitionSize="2000000" cacheDir="/var/www/seeddms/data/cache/" dropFolderDir="" backupDir="/var/www/seeddms/data/backup/" checkOutDir="" createCheckOutDir="false" repositoryUrl="https://repository.seeddms.org/dd278323b03fd2aa5510b8fa9addf5ea/" maxUploadSize="" enableXsendfile="false">
+    <server rootDir="/var/www/seeddms/seeddms/" httpRoot="/" contentDir="/var/www/seeddms/data/" stagingDir="/var/www/seeddms/data/staging/" luceneDir="/var/www/seeddms/data/lucene/" logFileEnable="true" logFileRotation="d" enableLargeFileUpload="false" partitionSize="2000000" cacheDir="/var/www/seeddms/data/cache/" dropFolderDir="/var/www/seeddms/import/" backupDir="/var/www/seeddms/backup/" checkOutDir="" createCheckOutDir="false" repositoryUrl="https://repository.seeddms.org/dd278323b03fd2aa5510b8fa9addf5ea/" maxUploadSize="" enableXsendfile="false">
     </server>
     
     <!-- enableGuestLogin: If you want anybody to login as guest, set the following line to true