Newbie needs help with snapshots/version

Hi!

I don’t know if I just searched for the wrong topic but I was unable to find an answer for my basic problem…

I would like to set some snapshots/versions for the backups.
Something like 1 x last year, 12 x last months, 4 x last weeks, 7 x last days.
I found that it is possible to set this while starting Duplicacy for the first time - I don’t remember… As I use Duplicacy within a docker-container I searched the gui for some options to get this done but haven’t found where to set this up.

If there is someone being so kind to give me a helping hand I would deeply appreciate it.

I’m on a personal licence, so no trial anymore…

Kind regards,

OK… It seems to be the ‘prune’ option…

Is it that I can just add this option to the Global Options?

like: ‘prune -keep 1:7 -keep 7:30 -keep 30:180 -keep 180:360 -keep 360:720 - keep 0:900’ ??

Not the Global Options, just Command options…

Omit the prune command from the option (you already added the job as a prune) - you just want flags here, cap it off with -a, and put them in the correct order:

-keep 0:900 -keep 180:360 -keep 30:180 -keep 7:30 -keep 1:7 -a

(I dunno know why you’d choose that schedule though, especially for snapshots >1 year… backups a year apart kinda make it an archive not a backup. Personally, I use -keep 30:365.)

I’m just ‘playing around’ as a first step. By the time I’m running out of storage I will have to adjust it…

But I changed it to: -threads 1 -keep 180:360 -keep 30:180 -keep 7:30 -keep 1:7 -a

Thank you for your help!

Starting the backup-job I do get the error ‘invalid options’…?

Anything wrong with the syntax?

Show your backup-job’s options.

Hi @saspus!

It was my fault trying to get Duplicacy running for the first time.

But I’m having problems with the backup-jobs/schedules… Right now the ‘check’-job never stops as it takes more than 24h to finish and I do have my ‘test’-job setup to run daily as I would like to get daily backups. How do I change this to a daily backup, a weekly prune and a check every month?

This is my setup at the moment:

check -chunks -a downloads and checks every chunk in the storage the first time it’s ran; on subsequent invocations it will only check chunks not previously checked. It can take significant amount of time. You can add “-v” or “-d” global flags to the command, and see in the logs what’s going on.

Is there any reason you are checking chunks? What is the target storage?

Right now the target is an sftp-server - thinking about adding minio/s3 to it as well. Source and target are located at zfs-pools.

A little bit of history… I lost a lot of data etc. while I was using btrfs and a single backup. Unfortunately btrfs just had all files being only 2/3 of the original size… Movies just played 2/3, pictures could not be opened etc. I wasn’t too concerned about it as I had my backup - I thought… but the backup had the same corrupted files on it… TBs of data gone, my wife nearly killed me because of all the files of our marriage were lost… Took some months to get at least some of them back.

I just want to make sure that the files of the backup are 1,000,000,000 % usable!!! Therefore I would like to run a check and have lots of snapshots…

Btw - is there a way to have different schedules as I would like to have daily backups of some files and weekly for other files? As much as I do get it the gui is just letting you setup one schedule with just one setting of the time it runs.

As long as zfs pool is scrubbed periodically — it’s as good as it gets. Running check right after backup will ensure that no shenanigans perspired during transfer -but duplicacy already protects against this by uploading to a tmp files and renaming only after transfer succeeds. But there is no harm is running check -chunks either — you don’t pay for egress, so why not.

Side comment — why not just do snapshot replication (zfs send/zfs receive)? What benefits are you getting with duplicacy in this scenario?

This is very strange. Was it Synology (with their lvm extension) or naked btrfs? Striped configurations of btrfs were not stable until recently (that why Synology would not use them)

There is tiny + at the bottom of the schedules page that adds a new schedule.

Will have a look into it - at first I would like to have a good gui backup solution up and running… :innocent:

No - btrfs raid5 was labled as usable and I trusted it… I was using an ubuntu server as my backup location at that time…

Thank you for your help!