We have been evaluating different products for moving data from Win to S3 bucket. I am using basic commands given in the command line documentation. We do not want to use the GUI.
We need to backup from different systems to the same bucket. When we tried from another system, getting “try repair”.
We don’t want to have any restrictions in pushing files from different locations/systems. Another internal system is maintaining the files to be backed up and not sure if --no-local-db = false is the solution.
Would like to make a rule in AWS to push files to Glacier a some point. So Duplicati should just backup and restore if available.
Would like to compress it at the time of backup. Could not restore the entire folder. However, I am able to restore each files under a folder separately. Do I have to use any special command to restore the entire folder.
I should be able to see the files in S3 through AWS and download the compressed file directly if needed as well.
Is this Duplicati is the right product for me? Appreciate any help in this regards.
Welcome to the forum!
1 - it is important that each computer (and even each backup set on each computer) back up to a unique location. You can back up to the same S3 bucket, but you should use a different folder for each computer and backup set.
2 - Duplicati needs to maintain its own local database for tracking files, backup versions, deduplication chunks, etc. It needs one database for each backup set on each machine.
3 - People have had mixed results with Glacier. Try searching the forum for Glacier for more threads on that topic.
4 - Duplicati uses compression (and deduplication) by default. You can use the command line to restore an entire folder, or just individual files.
5 - Files will not be in native format. You must use Duplicati to restore any data as it needs to be decrypted, uncompressed, and rehydrated.
If you want a synchronization tool where files stay in native format on the S3 storage, then Duplicati is definitely not the right product. But if you want a backup tool that will efficiently store multiple versions on the back end, then maybe it will work. You have some interesting requirements though (not wanting Duplicati to have a local database, etc).