SOLR¶

SOLR is a full-text search engine built on top of Apache Lucene. We use it to search the manuscripts and the Wordpress content.

A nightly cron of server/import_solr.py harvests following sources and indexes them into SOLR:

the TEI XML files <front> (as category:front)
the TEI XML files <body> (as category:chapter)
the Wordpress pages (as category:post)

The wordpress cap-meta-search-plugin queries the SOLR database through the python application server and presents the results to the user.

Installation¶

mkdir -p ~/solr
cd ~/solr
wget https://dlcdn.apache.org/solr/solr/9.6.1/solr-9.6.1.tgz
tar -xzf *.tgz
ln -s solr-9.6.1 solr

sudo systemctl enable solr
sudo systemctl start solr

cd ~/prj/capitularia/capitularia
make solr-init
make solr-import

Service¶

SOLR runs as systemd service /etc/systemd/system/solr.service

Start with systemctl start solr

Stop with systemctl stop solr

Icinga¶

SOLR can be monitored through /usr/local/lib/nagios/plugins/check_solr.py -P -H localhost