Backing Up

Backing up the database

You can backup the database using mysqldump using the following syntax. The database name, username and password follows that which was used in the install guide - however the username and password should be different for you if you’d changed them as suggested.

mysqldump -umysqluser -pmysqlpassword seek_production > seek_production.sql

Backing up files

The is also no harm in backing up the entire SEEK directory, however the essential directory to backup it the filestore/ directory. This is the directory that contains all asset files, and other potentially important files.


Should you ever need to restore the database you can do so by logging into mysql

mysql -umysqluser -pmysqlpassword

and creating the database (assuming it no longer exists), connecting to it, and importing the dump:

>create database seek_production;
>\r seek_production
>\. ./seek_production.sql

The files can be restored simply by copying the filestore/ directory back into the SEEK root directory.

Incremental Backups

Incremental backups allow you to recover a particular version of the backup from a given point in time, depending upon the frequency of the backup and when the back ups started.

For our own SEEK deployments, we use rdiff-backup within a simple script that is run nightly using crontab. e.g.


logger "Local Rdiff backup started"


/usr/bin/rdiff-backup --terminal-verbosity 9 --print-statistics --terminal-verbosity 2  --exclude-globbing-filelist $EXCLUDE  --include-globbing-filelist $INCLUDE  --exclude / / $BACKUPDIR

logger "Local Rdiff backup complete"

where backup.include is a file that includes a list of paths (files or directories) to be backed up, and backup.exclude is a list of paths that should be excluded.

BACKUPDIR can also be an ssh url, e.g