Using rsync for data transfer and synchronization
Rsync is a data transfer tool that can be used much like the scp
command.
When transferring data, rsync
checks the difference between the source and
target files and only transfers the parts that have changed. This makes rsync
suitable for:
- Synchronizing folders. Using
scp
orcp
would copy and transfer everything, whilersync
will only copy and transfer the modifications. - Transferring large files.
rsync
can be set to save progress, so if the transfer is interrupted, it can be resumed at the same point.
The basic command syntax of rsync
is:
If the data source or target location is a remote site, it is defined with the syntax:
However, both the target and source can also be located on the same machine. In that case you can just give directory paths to source and target sites.
The table below lists the most commonly used options:
Option | Argument | Description |
---|---|---|
-r |
Recurse into directories | |
-a |
Use archive mode: copy files and directories recursively and preserve access permissions and timestamps | |
-v |
Verbose mode | |
-z |
Compress | |
-e |
ssh |
Specify the remote shell to use |
-n |
Show what files would be transferred | |
--partial |
Keep partially transferred files | |
--progress |
Show progress during transfer | |
-P |
Same as --partial --progress |
So the command for transferring a local folder to Puhti, while showing the progress and keeping partially transferred files, would for example be:
This would either:
- Create a folder on Puhti at
/path/to/target/folder
if the folder was not present before. In this case, everything in the local folder will be transferred. - Synchronize the source and target folders if the folder already exists on Puhti. In this case, only changes we have made will be transferred.
And the same thing in reverse:
Warning
rsync
will always overwrite any changes made to the target, even if they
are newer than the source!