Re: correct-i386-2 proved useless.
- To: Luca Santarelli <LucaChan@Tin.It>
- Cc: email@example.com
- Subject: Re: correct-i386-2 proved useless.
- From: Philip Hands <firstname.lastname@example.org>
- Date: 06 Oct 2000 11:59:20 +0100
- Message-id: <[🔎] email@example.com>
- In-reply-to: <firstname.lastname@example.org> (Luca Santarelli's message of "Fri, 22 Sep 2000 21:27:15 +0200")
- References: <email@example.com> <firstname.lastname@example.org>
Luca Santarelli <LucaChan@Tin.It> writes:
> >Correct, that's the way it's supposed to work. (At least at this moment;
> >I'd love to see some changes in the rsync protocol to enable resuming.)
> Me too :-) I'm sure the people behind it will make it! :-)
I don't know if this is covered in the psuedo-image-kit readme, but if
not, then maybe it should be.
Firstly, it should be emphasised that it is almost never a good idea to
delete an image and start again, because even if the image is a bit
broken, rsync can make good use of it to make the real image on a
Secondly, there's the thing about rsync discarding partial images. If
you use --partial and --compare-dest then you can have rsync keep
partial images, but you may end up needing more disk space.
A useful tip for resuming partial transfers is that rsync can pick the
matching bits out of a file, so if you stick all the files that might
contain any part of the eventual image together, you will get the best
out of rsync.
To illustrate, if you have two files, image.psuedo and image.partial
after a first attempt at grabbing an image, then if you stick
image.psuedo onto the end of image.partial:
cat image.psuedo >> image.partial
rsync ..../source_image image.partial
rsync will whiz past the first bit, and then carry on from where it
left off. This does end up using even more temporary disk space