I've been using Unraid since 2008, and for the past year or so I've been seriously considering replacing my 8 year old server.  But a limitation on funds, and the new emerging Unraid capabilities had me holding off on making my move, until the most opportune time.  Well that time is now, as I'm pretty sure my trusty el cheapo Intel Motherboard has died.

 

Before the board died, I originally planned on setting up a pretty powerful new system, capable of hosting a Windows desktop replacement vm, and an Unraid backup server vm.  But because of available funds at the moment, I'm going to have to pare down these ideas a bit. 

 

My Windows Desktop, even though 5 years old, still works fine.  And it was pretty decent hardware at the time of purchase, so unless I have a hardware failure I'm good on that front, hopefully for another year or 2. 

 

I don't currently have a backup server, I know some folks on here do, and some consider it overkill.  My thought process was that if I had a multi drive failure, or had a file succumb to bit rot, that a backup server would be a good idea.  And having it run on the same hardware as my main unraid system, even seemed like it might be a cost effective way of implementing it.  But I'm wondering how necessary it is, considering I'd have to buy a case that was capable of housing 24+ drives, a processor and mobo capable of vt-x, potentially extra controller cards, and etc...  Unraid 6.2 is going to have dual parity capabilities, which means that the likelihood of losing data from a multi-drive failure is reduced.  So maybe that part isn't as big a concern. 

 

I'm still concerned about bit-rot though, I'm pretty sure I've had files get randomly corrupted in the past, but I don't know if it's a big enough concern to validate the extra cost associated with a backup server.  I have a vague notion that there might be ways to protect against this without backups.  Checksums and par files???  But how to implement something like that simply, and in a way that was easily manageable isn't quite clear to me.  Anyone who could explain it, like they were talking to5 year old Linux noob would be much appreciated.  Is there a software gui frontend that could control this?  A docker maybe?

 

Are there any other things that could be implemented to protect files against corruption or make remediation thereof easy?

 

Thanks in advance for any and all advice...

What is your current backup plan? unRAID or indeed any RAID system is not a backup. The only thing that counts as a backup is an extra copy of your files.

 

A separate server is better than having 2 systems in one server since it is possible for some hardware issue to make both inaccessible or even some data permanently lost in both. And separate is also easier to maintain and to troubleshoot.

 

Having something offsite is better still. Until I built my backup server, I was only backing up critical files to externals and then storing them in 2 offsite locations. Now that I have a backup server I still store backups of critical files offsite as before, and my backup server only has the less critical files.

Point taken. 

 

Right now the motherboard in my server is dead.  I'm assuming all 36TB of files on there are fine and just inaccessible without a functioning motherboard.  But I guess I don't know that for a fact do I?  Hmmmm.  Scary thought.