In order to create a backup of the CoreMedia Search Engine you have to do two things in the following order:
Back up the state of the Feeders
Back up the Solr index
Back up the state of the Feeders
For the Content Feeder this step can be skipped, as it stores its state in the Solr index.
The CAE Feeder in contrast stores its state in a dedicated SQL database. This database has to be backed up and it is important to do so before taking the backup of the Solr index.
The reason for this is that if the Solr index is fresher than the CAE Feeder database, the CAE Feeder will possibly redundantly refeed some documents which is OK, but if the Solr index is older than the CAE Feeder database the commits between the time of the CAE Feeder backup and the Solr backup would be lost.
If your database / tools provide the feature of hot backup, you do not have to stop the CAE Feeder for taking backups.
Back up of the Solr index
To take a hot back up of the Solr index you can use Solr's ReplicationHandler.
Once configured, a backup can be taken with the following HTTP request to the Solr Master server.
Replace <core> with the name of the Solr core you want to back up.
http://<host>:<port>/solr/<core>/replication?command=backup
You can find the snapshot files under the Solr core's data directory afterwards.
For details see
https://cwiki.apache.org/confluence/display/solr/Index+Replication.
 |  |  |
 |