resume function

[yveskretzschmar]

this is a really quick and dirty resume function by writing current filepos of every worker to file
needed this because i wanted to sync really big files and when ssh disconnected the sync process began from the first block

i use this in a bash script like this:

until python /opt/remotebackup/blocksync.py [ ... ]
do
sleep 30
done
rm syncpos*$(basename $image)*

i hope this helps and someone can rework this in a better way

[theraser]

Sorry for the late answer, for some reason there was no notification about the PR...

This is a interesting feature, thanks! I'm just a little bit worried about what happens when you don't want this feature to be enabled. I think there should be an additional command line option for enabling/disabling this feature because most people would expect a similar behavior to rsync, which always starts at the beginning. Could you integrate this, too?

And would it be useful to remove the syncpos-files automatically after a complete sync of a file is finished? In case of a lost connection as you described it it is good to leave the files as they are, but in case of a finished sync those files are not required anymore until you start a new sync.

[klemens-u]

+1 for this feature!