Raindogtoo Mar 5 7:35AM 2018 CLI
I kicked off a full restore of my media server repository on Saturday as a test. This is a 1.1TB repository and the Duplicacy storage is on a local hard drive attached to a Linux server. It took several hours to complete the initial full backup.
The using the -stats option, I could see that the estimated time for completion of the restore was over 24 hours.
Is that to be expected? Would it help to play around with the -threads option to improve the performance?
I had to kill the restore job and when I attempted to restart it, I received an error message stating that some files already existed and that I needed to use the -overwrite option.
Does that mean that if I restart an incomplete restore job, that it must replace all the files that it's already restored? Please tell me that's not the case.
Thanks
gchen Mar 5 10:24PM 2018
The -threads option should improve the restore performance.
If a file to be restored already exists in the repository, Duplicacy will compare the file hash to see if it needs to be restored. If the file hash is different, it will try to break the existing file into chunks and only download chunks that are different and then run an in-place update.