rsync
zsync
Is there any other better sync/backup tool available on linux. I have a local installation of my site, a wiki. I wish to make changes to my local site, and then just wish to press a button :D or run the command over a folder, it should check for the diff for my local files vs live-site files, and does the needful in the direction i wish.
That is why filezilla/ cuteFTP will not work for as they donot perform any Diff, will very inefficiently just replace the full folder. It will be awesome if there is any frontend available to rsync/zsync. Preferably zsync as the site, it is tweaked for http transfers.
|
|
|||||||||
|
|
SiteCopy might be a solution. It has worked for me with a similar situation. It unfortunately does not have a GUI, but it would be simple enough to create a shortcut to run it. |
|||
|
|
|
Sounds like you want to use a versioning tool like Take a look at the Subersion Homepage and/or git HOWTO. |
|||||||||
|
|
wiki's tend to have revision control built-in, and many are file-based (vs stored in a database), so rsync should work perfectly fine. I know folks who do this for 'TWiki' for replicating their installations to mulitple servers. Perhaps you only have 'ftp' access to your wiki files? you might consider 'wget' to pull from ftp (rather than the http interface) with the recursive (-r) and timestamping (-N) flags set so that it only transfers file that are 'newer' (which isn't exactly a diff). Once you have a 'copy' of what is out on the ftp server, you'd mark the update time somehow (often with just a 'touch' of a specific marker file). You would then edit normally via your local installation of the same wiki, then use 'find $dir --newer touchmarkerfile' to identify the updates for ftp and transfer them over via a script around an ftp delivery tool. I have used such a solution before (though I had the advantage of sucking the changes back to the main server via 'wget', so just used the recursive timestamping approach again. In hindsight, if I had 'ssh' access (I didn't), I would have simply used 'rsync -globtru[n]cv' to simply pull (or push) the files in each direction. |
|||
|
|
|
What I don't see is what kind of data format your wiki is in. If you are using a database backend, you can probably get away with some kind of replication. Both MySQL and PostgreSQL support this out-of-box, as well as others. |
|||
|