+1-208-473-2904 (USA - Sales)
0-800-051-8984 (UK - Sales)
0-800-181-0665 (GER - Sales)
When doing the backup of InnoDB tables it’s possible to omit the secondary index pages. This will make the backups more compact and this way they will take less space on disk. The downside of this is that the backup prepare process takes longer as those secondary indexes need to be recreated. Difference in backup size depends on the size of the secondary indexes.
For example full backup taken without and with the --compact option:
#backup size without --compact 2.0G xb_backup #backup size taken with --compact option 1.4G xb_compact_backup
Compact backups are not supported for system table space, so in order to work correctly innodb-file-per-table option should be enabled.
This feature was introduced in Percona XtraBackup 2.1.
To make a compact backup innobackupex needs to be started with the --compact option:
$ xtrabackup --backup --compact --target-dir=/data/backups
This will create a compact backup in the /data/backups.
If you check at the xtrabackup-checkpoints file in the target-dir folder, you should see something like:
backup_type = full-backuped from_lsn = 0 to_lsn = 2888984349 last_lsn = 2888984349 compact = 1
When --compact wasn’t used compact value will be 0. This way it’s easy to check if the backup contains the secondary index pages or not.
Preparing the compact require rebuilding the indexes as well. In order to prepare the backup a new option --rebuild-indexes should be used with --apply-logs:
$ xtrabackup --prepare --rebuild-indexes /data/backups/
Output, beside the standard innobackupex output, should contain the information about indexes being rebuilt, like:
 Checking if there are indexes to rebuild in table sakila/city (space id: 9)  Found index idx_fk_country_id  Rebuilding 1 index(es).  Checking if there are indexes to rebuild in table sakila/country (space id: 10)  Checking if there are indexes to rebuild in table sakila/customer (space id: 11)  Found index idx_fk_store_id  Found index idx_fk_address_id  Found index idx_last_name  Rebuilding 3 index(es).
Additionally, you can use the --rebuild-threads option to process tables in multiple threads when rebuilding indexes, e.g.:
$ xtrabackup --prepare --rebuild-indexes --rebuild-threads=16 /data/backups/
In this case Percona XtraBackup will create 16 worker threads with each thread rebuilding indexes for one table at a time. It will also show thread IDs for each message
Starting 16 threads to rebuild indexes.  Checking if there are indexes to rebuild in table sakila/city (space id: 9)  Found index idx_fk_country_id  Checking if there are indexes to rebuild in table sakila/country (space id: 10)  Checking if there are indexes to rebuild in table sakila/customer (space id: 11)  Found index idx_fk_store_id  Found index idx_fk_address_id  Found index idx_last_name  Rebuilding 3 index(es).
Since Percona XtraBackup has no information when applying an incremental backup to a compact full one, on whether there will be more incremental backups applied to it later or not, rebuilding indexes needs to be explicitly requested by a user whenever a full backup with some incremental backups merged is ready to be restored. Rebuilding indexes unconditionally on every incremental backup merge is not an option, since it is an expensive operation.
The xtrabackup binary does not have any functionality for restoring a backup. That is up to the user to do. You might use rsync or cp to restore the files. You should check that the restored files have the correct ownership and permissions.