Transitioning from Synology DSM to TrueNAS - Help requested on some aspects

TrueNAS General
1

I’m at a crossroads and after some advice to decide if I stay on Truenas and persevere or go back to DSM on Synology

If you can help in any are I will be very grateful :slight_smile:

I was using synology DSM suited my needs. The cost of their hardware in Australia is getting too high for what I need.

I changed to TrueNAS Scale (current build) for greater reliability on my own choice of hardware. After a few months I have learned a lot and still need to learn a LOT more :).

I run an online NAS/Server 24/7 and have a second that boots once a week or month depending on my workload to do a backup of my online unit.
My Main NAS was made the “offline” one when I built my TrueNAS machine.

The activities and preferred methods I describe below I would appreciate some advice on achieving with Truenas.

I will be grateful for help in any area.
Please note I’m very hardware savvy, and windows/DOS savvy. I’m not good with linux - it’s been over 20 years since I did my training with it and I’m not at ease yet editing in command line or such activities.

1. Backup my laptop and desktop PC’s to NAS weekly
I was using Syno Backup for Business. I have Macrium Reflect and think it will be a suitable alternative?

2. My media collection is large and most is not replaceable.
Photos, VHS movies I digitised, my BR and UHD collection I rip etc. About 9TB currently.

To protect from accidental deletion, ransomware etc, I used a read/write accessible folder “Temp Media” to drop new media into from my PC, while permanent media was read only to external users.

I would log into DSM and use the file browser to shift what I wanted to keep into my permanent folders.

I don’t know how to access such a file browser with Truenas since it lacks the windows type environment of DSM.
I think I found a “file manager” app but have not yet got it working since apps are very different to DSM too! I was flat out getting plex to work and learning how its container folders needed to point to real folders outside.

Note: I prefer not to use different accounts through my PC to login as read or write to manage media if I can avoid it, because that can get unwieldy for me.

What file explorer options exist or alternative methods to achieve my outcome?

3. Backup of one NAS to another.
I was using Hyperbackup. I believe RSYNC will be my best tool now?

I’m trying to work out the best approach.

Hyperbackup produces a backup file containing versions. I don’t know how to create this type of outcome in RSYNC - I thought it was aimed at copying files or creating a synced state where my target NAS would add new items and remove if I allow it to any deleted ones from the source?

Is RSYNC my only choice?
Where is a really simple idiots guide to RSYNC? I’ve looked at a lot of guides, videos and such and often think they are written for experts :stuck_out_tongue:

I think that’s all for now - thanks if you read this far, and thanks again if you are able to help walk me through some of this or set me on the best path.

2

First of all, relax. I fully understand your excited + worried situation here. Linux has come a long way since last 20 years, and does not bite that often now a days , especially the selected ones coming from windows background :grinning:

  1. Yes, Macrium Reflect is a perfect choice for this. I am using it in the same scenario and works perfectly fine.

  2. What you can do is keep 2 datasets here and take a manual snapshot at first. Also you could lock that snapshot to prevent from accidental deletion. The second dataset can be used for temporary media storage and you can review them over just a smb (samba share) from windows explorer and move required file whenever you find appropriate. TrueNAS should handle this as a server side file copy and your network will not be involved much in data transfer. Ensure to set automatic snapshots for both of these datasets as you feel necessary. The main media dataset, setup a local backup to another pool.

  3. If you can have the 2nd nas also to have zfs, you should look at zfs replication than using RSync. If you choose to go that way I would recommend, you either use TrueNAS-TrueNAS replciation, or (as you mentioned you are experienced with systems already) consider looking at syncoid for managing replication. That will give you better control on the replication lifecycle.

Hope this helps

1 Like
3

Thanks :slight_smile:
I shall relax and continue to take my time. I’m a bit worried not having a running backup regime yet still ok because my offline backup has all non-replaceable media to date.

I even have a 3rd NAS which was my backup until it was dropped by a removalist. Catastrophic failure is possible but that one is currently held together by airframe tape :P.

I did think making my “offline” NAS a Truenas setup too would be the better choice. I did not wish to until I learned enough about Truenas to know I could do what I want and manage it.

Probably the 2 areas that give me frustration at the moment are my understanding of permissions, ACL and the meaning of the different variables I can use there. It’s far more involved than DSM which has tick boxes for read/write/full control… that’s it. I think I have working permissions but I can’t say if they are safe or correct. Does that make sense?

The second aspect is apps. I’m not skilled yet with docker apps and so forth. I’ve needed to get my head around the way the system works. It’s been interesting learning since I know the theory of how VM’s function I just don’t have practical experience.

I do think as I learn more about permissions, credentials and such I will feel more at ease with using and setting up my devices using the NAS.

I did not know about zfs replication. That sounds like a simple approach I should examine and another reason to change my xpenology NAS across to Truenas.

Answer 2 above I’m unsure of what you mean exactly.
I currently use SMB on my laptop to move files and yes it does do it direct. What I’m unsure of is how to not have write access to all of the media from my laptop, yet still be able to move files between the 2 data sets.

Are you using the snapshot to replace the protection I achieved by only having read access? Since the media doesn’t change, the snapshot should never really take up any space right? In this context is it my protection of data and I recover the permanent media using the snapshot if I were to make a mistake and delete the incorrect files?

I have a rudimentary understanding of snapshots and little practical experience using them. I’d need to make some, play with files and learn recovery etc to be sure I’m ok with it in case of emergency.

4

Yes, your snapshot will serve the purpose of a read-only copy.

I know its a bit difficult to get the head around this concept initially. Let me try to explain it in a simpler way as I perceive it. Think of the first snapshot as a Full backup and all the following snapshots as diff backups. Now if the full backup sitting at the starting end is deleted, the diff backup after that automatically becomes a full backup. Again this is just an analogy I made up to simplify things and exact pieces behind it providing this result may vary - its just for understanding purpose, and from what you mentioned about your Windows background I hope that a full/diff will be one thing you are already familiar with.

But again, do not solely rely on snapshots, but make a separate backup as you need. They are just your first line of defense against deletion and not a replacement for backups.

From my experience, TrueNAS is solid as a NAS. But my personal preference for VMs is other solutions. So is the case for containers, but I do understand some may prefer to have a Swiss knife, and others like specialized tool for specific activities. Again just my opinion, and rest I leave it to your choice.

Finally, start with a sample dataset (create a miniature model of your most important data and copy a subset of actual data to it. - like only a few documents, some mp3s etc.) . Try making first snapshot. Then add some files, take next snapshot. then delete some file, take next snapshot… finally create clones of each snapshot and do a compare (like with beyond compare) and see for yourself to understand with your actual sample data. This will bring confidence.

Once you are happy with the snapshots concept, next create a backup dataset and start replicating your sample dataset. again make changes to folders/files in sample and be comfortable going through the backup dataset and restore it to a 3rd dataset to do the comparison. This will give the confidence and better understanding of the concepts for you to start trusting your data with zfs and TrueNAS.

If you want to go one more step ahead, install a Debian VM on hyper-v, install zfs, connect your disk with zfs sample dataset to it. import the pool and ensure you can access data from it. If you reach at this stage, rest you will be able to manage on your own with confidence.

Wish you all the best with your new journey in zfs and TrueNAS. I am sure you will learn a lot and enjoy from it. Good luck and see you on the other side. :handshake:

5

Here’s the simplest and clearest explanation I’ve seen of snapshots:

2 Likes
6

That is a good summary thank you. I thought of a snapshot as a map of the data that prevents deleted blocks being overwritten which allows them to be recovered until I delete the snapshot. Is that assessment correct?

In my scenario of media I can take a snapshot and each time I add media take a new one; If I keep an original and occasionally update that one to remove extra snapshots I should be protected against mass corruption like ransomware or deletion (even say mistakenly deleting the main folder of all movies).

Is that correct?

7

Quick question.
For the learning process are you advising a dataset on my backup NAS or one on my online NAS? I’m unsure if using the online NAS with another dataset can still let me learn the same skills but it would be more convenient to begin learning each element of the required steps for my needs.

8

Pretty much. Typically you’d take snapshots on a schedule, for a specified lifetime.

9

By online nas you meant a “local nas that is always online”. yes then create a new learn-dataset and learn snapshots with that. Then use your backup nas as a target for snapshot replication.

If online nas means a VPS on the internet (not in your local lan) - do the opposite. Create your learn-dataset on backupnas (which I believe is locally available) and learn snapshots with that. Then use the VPS-nas as target for snapshot replication (or even better create a local backup-dataset on the local nas and learn replication first.)

The end goal is to familiarize with snapshots, rolling back with cloned snapshot, replication, reverse replication etc. As long as you get hold of the concepts and the replication delays (in case of slow upload to internet) are managable - everything is fine.