Hello, Community Members: I have tested the behaviors of Backup API and Snapshot API, which are written in the URL below. https://lucene.apache.org/solr/guide/7_4/making-and-restoring-backups.html#making-and-restoring-backups ()
As far as I observed the behavior of Backup API, I now know the followings: - Solr's back up simply means to make a copy of a full-sized index. - Solr's restore means to make another copy of a full-sized index from a backup directory and refer the copy as the index. - Backup / Restore APIs belong to Replication Handler. Also, I know the following for Snapshot API. - Solr can make a snapshot at any time (i.e. does not matter if it is after commit/backup/ restore..) - snapshot_N contains a name of the snapshot(commitName) and the current Index path. - N in snapshot_N is the identical number to segments_N. I believe, by observing the Snapshot API behavior, it is impossible to "backup" or "restore" Solr's Index. So, my questions are" - What is the difference between Backup API and Snapshot API? The above Solr's Guide says "The snapshot functionality is different from the backup functionality as the index files aren’t copied anywhere." But, then how Snapshot API help me to make a backup? - Or, more basically, when should I use Snapshot API? - What is the rule of thumb operation for Solr's backup system? Sincerely, Kaya Ota
