mondane / timebackup Goto Github PK

Currently, this script operates in a "push" mode: the backup client runs the script and connects to the backup server.
I'm proposing a "pull" mode, where the backup server can be configured with a cron job to run the script to connect to the client's SSH server, and then "pull" the new data to backup from the client.
I have a use-case for this, but if you think it's too esoteric, then I wont go through the pull-request process, and only add it to my local version.

Based on http://unix.stackexchange.com/questions/92346/why-does-find-mtime-1-only-return-files-older-than-2-days

I'm thinking of changing the mtime parameter in the find commands to mmin to have a better selection of files. This would hopefully improve the hourly and daily rotate so not several backups from the same day are saved.

@rossica , what do you think about this?

OferLapid commented wrote 13 april 2017:

Is this package ready for use by non technical users?
I would love a short explanation on setup and usage
Thanks!

@OferLapid a bit of a technical background is easiest, do you grasp https://github.com/Mondane/timebackup/blob/develop/README.md ?

I'm considering a feature which would allow multiple instances of RSync to run, speeding up the entire backup. I have many folders backing up, and some of them are larger/have more churn, than the others. Since Rsync is pretty low bandwidth (about 13 MB/s on my network), and my NAS has much more disk capacity, allowing multiple instances to work on different directories could save time.
I was looking at this answer as a way because it doesn't require extra software:
https://stackoverflow.com/a/46611168

Would you accept this feature if it were coded?