Rdiff-image-get

NAME

rdiff-image-get - restore an rdiff-image backup

SYNOPSIS

rdiff-image-get [options] image-dir

DESCRIPTION

rdiff-image-get downloads backups created by rdiff-image-cron(1) to a cache directory, unpacks (ie restore) backups in the cache directory and cleans up the cache. The cache directories stores backups download in previous runs and in particular the latest full backup. Since the full backups are produced infrequently this usually means only the much smaller difference file needs to be downloaded on a second run, saving a lot of time. rdiff-image-get always does a full restores of a backup. It doesn't support partial restores - eg restoring one file.

OPTIONS

-b backup-name, --backup=backup-name: Instead of downloading and unpacking the latest backups found use backup backup-name. backup-name has the format yyyymmdd-hhmmss, which is the date-time the backup was made. The last digits may be omitted if the resulting name is unambiguous.
-c cache-dir, --cache=cache-dir: Use cache directory cache-dir instead of the current directory.
-d, --dry-run: Go through the motions, but don't alter anything. Usually combined with --echo or --progress.
-e, --echo: Print shell commands describing each action. The S3 commands assume you have s3-bash installed.
-i, --index: Instead of downloading images, print an index of images available.
-l, --cleanup: Remove all backups that were not used from the cache. In other words, remove all bar the latest backups if --backup wasn't used, or all bar the backup specified by --backup if it was supplied.
-n, --no-secret: Don't download / unpack / cleanup secret backups.
-o, --overwrite: When doing an unpack overwrite image-dir if it exists. If this option isn't given the program will refused to unpack over an existing directory.
-p, --progress: Print a description of each step as it is about to be done.
-r root-cmd, --root-cmd=root-cmd: Use the command root-cmd to unpack the archive as root. Root is usually needed to unpack archives as they contain device files. sudo(8) and fakeroot(1) are suitable root-cmd's.
-s bucket[:key:secret], --s3=bucket[:key:secret]: Download the backups into the Amazon S3 bucket whose name is bucket. Key and secret are the credentials needed to log into S3. They can alternatively be supplied using the RDIFF_IMAGE_S3_CREDENTIALS environment variable. An alternative to using this option is to download the backups to the cache directory using say Firefox's s3fox plugin.
-t tarfile, --tar=tarfile: Write a copy of reconstructed backup to tarfile. Tarfile has the string {{backup}} in it replaced with the date-time (yyyymmdd-hhmmss) the backup was made, and the string {{kind}} replaced with "base" or "secret".
-u status-url[,image-url], --url=status-url[,image-url]: Download the images using HTTP. Status-url is an index page containing a list of the currently available backup files. Those files must be located in the directory image-url. If image-url isn't given the backup files must lie in the same directory as status-url. An alternative to using this option is to download the backups and their .sha1 files to the cache directory using say wget(1), and manually checking the result with sha1sum(1).

ARGUMENTS

image-dir: If supplied a backup will be unpacked to this directory. If it contains the string {{backup}} all occurrences will be replaced with the date-time the backup was made. The date-time has the format yyyymmmdd-hhmmss.

ENVIRONMENT VARIABLES

RDIFF_IMAGE_S3_CREDENTIALS: If the S3 credentials aren't supplied to the --s3 option they can be supplied in this environment variable. The format is key:secret.
http_proxy: If present, the --url option will do its HTTP downloads via this proxy. The format of the variable is: http://home.name:port

EXAMPLES

To list the backups available on S3:

: rdiff-image-get \
--index \
--s3=bucket:key:secret

Download the latest backup from via HTTP, unpack it to a directory named after the backup, and after all that is done remove all bar the backups unpacked from the cache:

: rdiff-image-get \
--url=http://host.name/backups/ \
--cleanup backup-{{backup}}

AUTHOR

Russell Stuart <russell-rdiffimage@stuart.id.au>