I've been looking through the docs, and I wanted to see if there's a way to print hashes from the rclone check command. Seems like check is already getting hashes from the remote, or downloading and hashing when using the --download flag, but there isn't a way to print those hashes from the check command.
The use case is that my org needs to provide content hashes for each file we manage. Currently I'm using a Jenkins pipeline to copy files from one remote to another, then run a check to verify integrity, and I could add a step to that to run hashsum on each file, but that would add a lot more time and if there's a way to cut that down some, I'd like to investigate it.
Yes, just didn't want to go through each file on the remote and re-hash it if I didn't have to, especially when I'm already running check --download
I'm usually copying genomic data, so these can be up to 50 GB per file, couple hundred files at a time. Anything I can do to cut down on time is preferred.
That is good to know, it'd be a lot quicker to diff that output. Thanks for the tip! I guess I need to be a little more diligent about looking at log levels before assuming something isn't happening.
Rclone will log any hashes it uses for the rclone check but it will only use one (if multiple are available).
Generally speaking which hashes are available depends on the backend in use - you can see which backends support which hashes in the overview docs.
You can of course calculate any hash with rclone hashsum --download but rclone check ---download will default to MD5 as that is fast and very good at detecting bit rot. There isn't away of forcing rclone check to use any particular checksum, but there probably could be!