I understand that if you want to only transfer changed blocks across the network that the remote machine must be running rsync server and the local machine must be running rsync client. What if I don't care what gets transferred over the network, I just care what gets written to disk (i.e. only changed blocks)? Is there an option to do that in rsync without needing an rsync server on the other end?
I have an OpenSolaris server that I'm using for backups. It's running ZFS, so if only part of a file changes, I can take a snapshot and be able to have access to both copies of the file but only store the differences. However, if the entire file gets copied back down to disk, ZFS isn't smart enough to realize that the new blocks copied to disk are the same as the ones that just got overwritten - so it marks all blocks of the file as changed and doesn't provide any space-savings. If rsync had a mode that made sure only to write to disk the blocks that changed, this would get me around this problem. Also, since many of the computers I'll be backing up are Windows machines, I don't want to hog their CPU, and I don't care how much data gets transferred over the network, I'd rather avoid installing the rsync server on those machines. -- View this message in context: http://www.nabble.com/Write-only-changed-blocks-to-disk-using-rsync-client-only-tp20662653p20662653.html Sent from the Samba - rsync mailing list archive at Nabble.com. -- Please use reply-all for most replies to avoid omitting the mailing list. To unsubscribe or change options: https://lists.samba.org/mailman/listinfo/rsync Before posting, read: http://www.catb.org/~esr/faqs/smart-questions.html
