I'm writing some backup software that needs functionality like partclone
(8), and having some trouble with the man pages. In particular, I don't see any difference between the pages for partclone.dd
and partclone.imager
, and the two binaries, while different, are the exact same size. Moreover, I found a question somewhere with an answer that made a distinction between them, but it was not clear why.
So I'm really wanting to know what is different between them. I had supposed they were just dd
disguise, but I'm ready to be proven wrong.
I ran 4 copies of a fairly large EXT4 partition to see what the differences could be. That, together with some further inside info, seems to settle the matter for me.
The 4 copies I made were by the following programs:
partclone.ext4
partclone.dd
partclone.imager
dd
(the real one)I discovered that the first was by far the smallest, which makes sense because it does not have to back up everything -- just the stuff that is occupied.
I discovered that
partclone.dd
produces exactly the same image that the realdd
produces. Just with a different command interface.I discovered that
partclone.imager
produces a considerably larger image thandd
, which is consistent with the clone format I found athttps://github.com/Thomas-Tsai/partclone/blob/master/IMAGE_FORMATS.md
. This includes checksums and bitmaps.It makes sense to me that in cases where dd could make sense, I'll use
partclone.imager
rather thandd
because of the added robustness and checking in the event of damage to the backup.The added parameters
--clone
,--restore
and--disk-to-disk
are needed bypartclone.imager
because the image is not identical to the source partition.