sd-imager

This package contains 4 shell scripts. All shell scripts must be run using bash and are not expected to work with other shells.

The purpose of these scripts is to allow disk image copies to and from SD cards so that copies of the SD cards may be saved lest the original card becomes corrupt, or to allow downloaded images (e.g. for installable operating systems) to be written to SD cards.

Although this is intended for use with SD cards, it should work as well with any removable block device (e.g. a DVD writer).

The 4 scripts are:

sd-save

This will copy an image stored on the SD card to a file.

The command format is:

sd-save options

Options

The options are described in long form, but each of the options can be used with a single letter, thus the -input= option can also be used as -i=.

• -help

  Will display a help message.

• -input=source_device

  source_device is the device name where the SD card is attached (e.g. sdc).

  Note that this has no path (e.g. /dev/sdc would be wrong, it must be sdc,and no individual partitions are allowed, so sdc1 would also be incorrect).

• -output=target_file

  target_file is the name of the file to which the image is to be written.

  If the file is written uncompressed then the file must have a .img suffix. If the file is to be saved in compressed form then, depending on the type of compression, the file suffix must me either .img.gz or .img.bz2 or .img.xz.

• -sparse

  If this switch is included and the file is being saved in uncompressed form then the output will attempt to skip writing null data blocks and so improve its performance. This switch is ignored if the output file is a compressed file.

• -trim

  If this switch is included then data beyond the the end of last partition in the partition table will not be copied.

  You may not use this switch for a source_device that does not contain partitions.

sd-write

This will copy an image stored in a file to an SD card.

The command format is:

sd-write options

Options

The options are described in long form, but each of the options can be used with a single letter, thus the -input= option can also be used as -i=.

• -input=source_file

source_file is the name of the file to from which the image is to be read.

The file must have a suffix of .img or .iso or .img.gz or .iso.gz or .img.bz2 or .iso.bz2 or .img.xz or .iso.xz.

• -output=target_device

  target_device is the device name where the SD card is attached (e.g. sdc). This is where the image is to be written.

  Note that this has no path (e.g. /dev/sdc would be wrong, it must be sdc, nor are separate partitions allowed, so sdc1 would also be incorrect).

  Optionally, the output may be directed to the null device, which might be useful for testing purposes where one does not actually wish to write to a physical device.

• -sparse

  If this switch is included and the file is being saved in uncompressed form then the output will attempt to skip writing null data blocks and so improve its performance.

img-mount

This will mount a stored disk image (only uncompressed images may be used) on a mount point.

The command format is:

img-mount image_file mount_point

image_file is the name of the file where the disk image is stored.

Optionally, if the image contains partitions, then image_file should be of the form image_file:partition_number (e.g. test.img:1). The partition_number starts with 1 as the first partition.

You may use the command fdisk -l image_file to determine which partitions exist on the image.

mount_point is the directory where the image is to be mounted.

img-umount

This will dismount a stored disk image from a mount point.

The command format is:

img-umount mount_point

mount_point is the directory where the image was mounted.