TimTheSettler

Basically everyone's suggestion here is correct in one way or another. As Aran points out, syncthing is a synchronization tool. It's sort of like a backup in the sense that you have another copy somewhere else and it's even possible to snapshot the data but it's not a true backup. The reason why all the suggestions are right is because you need to get the data off the phone and syncthing is the easiest way to do that. Set it up so that your phone is a device and your unRAID server is a device and then link the pictures on your phone to a folder on your unRAID server. That will mirror the pictures from the phone to the unRAID server. Next, you should set up duplicacy or Vorta and then back up the pictures folder that's on the unRAID server. That's the true backup and those products will create efficient snapshots for you (using deduplication). So, whenever you take a picture on the phone it's get synchronized to the unRAID server and then backed up. If you accidentally delete a picture on the phone then that deletion is synchronized to the unRAID server and the file is deleted but Duplicacy/Vorta will have a snapshot of that picture from a previous backup (assuming the backups are frequent enough). You can then restore the picture from the backup.

I'm not a command-line guy so I would go with #1.

Parity is redundancy which implies duplication. So parity is a form of backup. Parity also helps in the ease of recovery. But other things can happen to that server or the data so a separate backup server serves a few more purposes: Separate hardware (complete server backup) Offsite location A snapshot in time (history) Many people say, "I want to back up my data", and then there are lots of ideas about backup software, parity, multiple machines, etc. Backing up your data is about protecting your data and backup is only part of that protection (encryption and access to the data are others). A good backup strategy involves multiple parts and parity is a nice, easy, and effective way to get started. But I digress because we've already established that people here are using parity in their main servers. Do you need it in a backup server? Only if your backup data contains important info not held elsewhere (like the snapshot history for me).

I'm glad it works for you guys. As you can guess, I'm paranoid. Burned too many times and don't have time to deal with massive restores. On one hand I love to putter around and try new things but that's my hobby time. When something fails and I rely on it then it cuts into my real life and that's too valuable. In the end I feel that a few hundred bucks is cheaper than me tearing my hair out at 3am after six hours of trying to fix/recover from a problem that came up.

Point noted, good to know. I was mainly comparing unRAID to TrueNAS.

No parity? On a backup system? Interesting...

His 5 Pros are: Resource consumption - Agree ZFS Support - Who cares? I wouldn't list this as a plus (or minus). It's just a feature. Hardware requirements - Agree Parity - Agree GUI - Agree His 5 Cons are: Transfer speed - Maybe Learning curve - Disagree (Thanks to the GUI) although a wizard to get you started would be a nice addition. ARM support - Really? Single Admin - Seems like a good point but I'm on the fence. Not free - Free is nice but it also means that it might not be supported very well (or at all) down the road.

Two things to consider with pictures on phones: You'll want them backed up. At some point the phone will run out of space. The easiest way to solve the "running out of space" problem is to buy a microSD card and add more space. This might not be possible or it might be too expensive. The other thing that solves both problems but requires a bit of work is to install syncthing on the phone and your unRAID server. Link the folder on the phone to a folder on unRAID (full bidirectional sync). When the phones are on the same network as the unRAID server (in your house - easiest/safest solution) then the photos are synchronized between the phone and unRAID. Once in a while you or mom or the wife will need to move photos from that folder on to another folder on the unRAID server. This way the photos are still archived and sitting on the server but moving them out of the synchronized folder means that they are also removed from the phone thereby freeing up space. So, just to clarify a bit here. Each phone has a corresponding folder on the unRAID server. Then, once in a while you move the photos from each of those folders to an archive folder somewhere else. The easiest way to clean up the photos is to find a family event like a past birthday and move all those into a folder just for that event (like an album). The drawback is that they are off the phone but if you're running out of space then so what. Your "photo album" (folder) is still on unRAID and can be accessed. Personally I have ALL photos synchronized to ALL my devices (phones, tablets). We're talking about 77,000 photos (250GB). I just make sure that any new phone I buy has a microSD slot and then I buy a 256GB (or 512GB) card. Now I'll never run out, I have all photos available to me, and everything is backed up. (Make sure to encrypt the card when you first put it in.) Actually my setup is more sophisticated so read on if you're interested. There's the camera folder on the phone and a separate pictures folder. The camera folder is bidirectional sync which means that changes on either side are reflected back to the other device/server. The pictures folder is one-way sync from unRAID to the phone. All common pictures are put into the pictures folder and these are pushed to the phone. This means that the phone has all family photos. When you take a picture on the phone it's saved to the camera folder. The camera folder is specific to the device but is synchronized to unRAID as a backup. At some point you manage the camera folder (on unRAID) by moving photos from the camera folder to the pictures folder (like all photos for a birthday). This moves the photos out of the camera folder on the phone but pushes them back up to the phone and into the pictures folder. The benefit here is that each person takes their own photos but by moving the photos from the camera to the pictures you are sharing those photos because the pictures can be replicated to all devices. If you're not sure if this makes sense but you're curious about how it works then I can explain it better with a diagram but if no one cares then I won't bother with diagrams. In the pictures above note that camera folder is two-way sync between the camera folder on the phone to unRAID (left pic). The right pic shows a one-way sync from unRAID back up to the phone into the pictures folder on the phone. Your photos app (Google Photos or Samsung Gallery) will show all pics in both folders.

Appdata once a month (automated). Flash drive every three months or so (manually). All data (documents, pictures, media, etc.) (~28TB) synchronized real-time (syncthing) between four servers in three different locations with daily backup (Vorta/Borg).

That's cool. This is probably a stupid question but how do you do that? Not to take anything away from @Hoopster and rsync because I like that solution, especially the logging/email part, but I really like the backup software I see out there. Since Vorta/Borg (and others) use dedupe it's pretty efficient and there's a history of file locations if files are moved around. I used to have files in six different places and couldn't remember why I had them there and they were taking up lots of space because of that. My backup runs each night and I have a detailed prune cycle so I keep daily info for a couple weeks then weekly info for a few months then monthly info for a couple years, etc. My largest archive is 25TB so it runs pretty good at that size (although if you add a TB of data it can take a day to process all that). Here's an example of a couple movie folders that I renamed. In the end they don't take up any extra space and I know what happened if I wonder about it later. Also, thanks to @Aran and @Hoopster. As @ConnerVT points out, your suggestions are gold. I love the remote management.

Do people actually say "WD"? It sounds weird.

Most backup software uses deduplication which is really nice.

I'm using Vorta (Borg). You can use it on your regular server and point the Vorta container to your backup server as a target. However, I do something a bit weird. I syncthing my data from my regular server to the backup server and then back up that data using Vorta. So the backup server has two copies of everything, one that is synchronized and the other is the backup. The drawback is that you need double the space. The good thing is that it's easy to restore (because syncthing repropogates everything back to all the other servers and all the work (Vorta backup process) happens on my most powerful server (the backup server). In your case I would just use the backup server as a target. Bring it up, run Vorta on your main file server so that the backup server gets a new copy, and then shut down the backup server until next time.

The answer might be somewhere in the forum topic quoted above but I find that some of these topics have lots of info and suggestions and in the end it's hard to know what needs to done. People just post a link and say "It's in there." but where in there is it? Maybe @Squid can help to point out what exactly needs to be done for you in 6.9.2.

I'm assuming that the xbox is using DHCP to get an IP address from your internal network. In this case the DHCP server is going to send back a gateway IP address which is usually the IP address of your router. The xbox then sends all external traffic (outside of the local network) to the gateway. In this case you need to set up the VPN connection on your gateway device (the router). If the router supports the VPN service then you're laughing and you just need to set it up. If the router does not support this then you can go through the unRAID server but now you need to tell the xbox that unRAID is your gateway and you need to set up the unRAID server to talk to the VPN service. No one can really tell you the exact steps because no one is privy to the exact setup that someone might have at home. The best you can do is to understand the basics here and then use the many, many guides that exist out on the internet to figure out how to make it all work for you.

I'm definitely not an expert in this but a VPN is an encrypted tunnel that you set up between two machines. Usually it's from one network to another network (LAN over WAN, like the internet, connecting to another LAN). Typical uses back in the day were to connect your home computer to your work's network or for a work site to connect to another work site. The benefit, if the routing is set up properly, is that each device appears to be connected directly to each other even though there might be a vast distance between the devices. The secondary benefit (although it's become the primary now) is the encrypted communication between the devices. Nowadays people use a VPN like IPVanish to hide their IP or to encrypt all traffic across the internet. I personally don't bother with this for the following reasons: Encrypted connections are slower than normal connections and require extra processing on either end to encrypt and decrypt the the data. A long time ago the processing needed to encrypt/decrypt was a problem but nowadays it's a moot point because processors are more powerful or encryption logic is added to the processor (https://en.wikipedia.org/wiki/AES_instruction_set). But the connection is still technically slower because encrypted data is usually larger than the data you're intending to send and encrypted data does not compress well or at all (https://docs.actian.com/vectorhadoop/5.1/index.html#page/Security/How_to_Compute_the_Width_of_Encrypted_Data.htm) although this also tends to be a moot point nowadays. If I truly need an encrypted connection then I ensure that I'm using an SSL connection (HTTPS). If you're connected to a web site using SSL and you're using VPN then you're double-encrypting your data (which is ok but unnecessary). The end-point of your VPN tunnel (IPVanish) decrypts your data and therefore knows what you're doing. You need to ensure that they respect your privacy. Some VPN companies will log your activity while others won't on purpose. Note here too that your data is still going to the destination website as cleartext (unless using SSL) but it now appears that you're a member of the VPN's local network. In other words your IP has changed. There's an extra hop to the VPN for all connections. Instead of being connected directly to your destination you first go through the VPN provider's servers. So what's the point of the Wireguard VPN that's integrated into unRAID? The idea here is twofold, you can access your home network but also your home server becomes like the IPVanish server. In other words, you connect to your home machine and all data sent there is encrypted and you appear to be a member of that local network. Usually it ends here and the goal is to simply access your home LAN from a remote location (like the coffee shop) but with the added benefit that the connection is encrypted and secure. The other benefit is that you can access your home devices, media, and data remotely so it appears that you're at home even though you're somewhere else (in another country perhaps). Now your home VPN connection appears like the IPVanish connection. The biggest drawback to using it this way is that all data traffic is going to your home network first and then out into the world. This means that your home bandwidth is used a lot (lots of data being routed into and out of your home internet connection). If you're trying to set up a VPN connection from the unRAID server to IPVanish (which is what you're trying to do) then you need to search for "IPVanish Linux client" (https://www.ipvanish.com/vpn-setup/linux/) and install the client on the server and then you need to set up the routing so that all connections go through there. If it's done right then all data sent out from the unRAID server goes through IPVanish. In this case the end server that you're connecting to will see you as a device on the IPVanish network. So if the IPVanish server is in the US then it appears that you are also in the US. Keep in mind that all connections will now go through this IPVanish server which means that you have an added hop to this server.

I might be missing something here but I have Plex running as a docker on my unRAID server and I do NOT have a Plex account at all. I'm able to view all my local content using the Plex app on my Roku device. Once in a blue moon (very rare) I get a window showing up that tries to get me to sign up for an account but I just bypass it. One note is that I can only access the server locally. I have VPN access between all my servers and I'm unable to get to the GUI remotely but that all makes sense since I don't have an account.

Interesting that the new NT version is very similar to the existing Exos drives. Seagate Datasheet comparing the NE and NT versions (20TB and 18TB): https://www.seagate.com/content/dam/seagate/migrated-assets/www-content/datasheets/pdfs/ironwolf-pro-20tb-DS1914-21-2206US-en_US.pdf Seagate Datasheet for Exos drives (20TB and 18TB): https://www.seagate.com/content/dam/seagate/migrated-assets/www-content/datasheets/pdfs/exos-x20-channel-DS2080-2111US-en_CA.pdf Product Manual for Seagate NT drives (22TB and 20TB): https://www.seagate.com/content/dam/seagate/migrated-assets/www-content/product-content/ironwolf/en-us/docs/204482900b.pdf

Sounds like syncthing might be the way to go for you. Install that on all the Windows machines and your unRAID server (I assume you have unRAID since you're here although it's not explicitly mentioned). Hard drive space gets to be the issue here. You can connect all the devices together so that all data is shared on all devices and you can even encrypt data on certain devices. For example, you could have it do a four-way sync between all devices where the docs and pics show up as encrypted on the child's device. The other option is to sync all devices to the unRAID server. This is what I do. All phones and devices see and sync to the unRAID server. They don't see or sync among themselves since they don't have the space for all that data. In the example below I have a number of folders (27) shared across seven devices. The "Audio" folder is encrypted and the "Pictures (Phone)" folder is a send-only folder. The syncthing01/02/03/04 devices are my main file/backup servers so all folders are synced among those machines. Only certain folders are synced to syncthing05 (limited space) and only a couple folders (camera/pics) are shared with the phone/tablet devices. When you set all this up you need to think about where you want the data to be on each device and will there be space for that data on those devices. The only other option is to use regular SMB shares.

For anyone interested in what @positronicP and @tjb_altf4 are talking about you can read about it here --> https://blog.westerndigital.com/host-managed-smr-dropbox/ As @tjb_altf4 points out, Dropbox uses HM-SMR (Host-managed SMR) but the host (OS) needs to be aware of this technology for it to be of use. So for most of us we're talking about DM-SMR (Device-managed SMR - as mentioned by @tjb_altf4 above). Maybe at some point in the future HM-SHR will be supported by Linux and unRAID. I personally think that if you have cache drives then DM-SMR is ok as a regular drive.

The ultimate solution would be an app where you specify a folder and tell it to create a disc archive. It then reads the directory structure into memory and splits up the files so that they fit optimally on a bunch of discs. You then insert the discs and it burns them and records a serial number which can be used to link back to the disc's contents. Since this process is a manual one you would need to run the archival process manually. Each time it runs it collects the directory structure and compares it to the last one and creates new discs for changed and new files. Similar to a backup app but it would need to know which disc the proper version of the file is on. It would also need a feature where a new disc can be created which can replace an existing but lost or destroyed disc (same content as previous disc but with new serial number recorded in the app). A simpler version is that you use Brasero to burn a disc and then separately have a script that reads the disc, records the files on it, links them to an ID, and puts all the data into a database (or Excel). Let me know when you have either of these built.

The Microsoft site shows no such restriction. In fact you can create a Windows VM with only one pinned CPU (although a paired CPU/thread is recommended). Also note that CPU pinning simply limits the app/VM to using only that CPU. unRAID will use that CPU depending on the load of the server. To make the CPU pin exclusive you need to use CPU isolation. In general, @Gragorg's advice is good. I don't mean to slag what he's saying. It's just that you should use what you have right now because it's what you've got and might be all you need. If you find that you need more power then change the VM to have more CPUs or RAM and if you're maxing out the resources of your unRAID server then upgrade the hardware. This is easy to do and unRAID will simply see a more powerful engine under the hood. The beauty of the VM concept is that you can fiddle with the CPU/RAM assignments to get the desired performance you want. If anyone feels that this advice is wrong then please correct me. My goal here is to get people using unRAID without them having to spend money on a system they may not need. It's very easy to upgrade the motherboard, CPU, and RAM without impacting your server. I've already done it.

Shares are controlled by user access, not by machine access. You can't restrict access to a share based on the machine that's accessing it. Although you don't log in, Windows will have created a user for you. It's just that this is the only user for that computer. The drawback here is that this user does not have a password so if you want to restrict access through a share you'll need to give this user a password or create another user with a password. I recommend that you create another user and give this user a password (first create the user without a password and then give this user a password - doing this bypasses the security questions that Microsoft adds when giving the user a password). This user will be a special user that unRAID uses to access that computer. You can continue to use your main user that you do today without a password (nice and easy to log in to Windows). Once you do that you can create your share and then change the permissions by removing "Everyone" and replacing this with the user that you just created. When you create the share in unRAID it will ask you for a user name and password at which point you provide this new user and its password.

There was a time when I saw these used by companies (as replacements to tape backups) but no one does that anymore. It was always expensive technology and is still so. However, I loved the idea that seldom used stuff (based on modify date or access date) would be archived off but can be brought back automatically without human interaction. The OS or software would know where that file is. I think this is what you're referring to but as I mentioned, almost no one uses it and it's expensive. I use a CD suitcase and burn things like pictures to DVDs (or now Blu-Ray) from time to time. It's a manual process but pretty easy. Keep an Excel file with the disc number, it's contents, and date it was burned. If the disc gets old (10 years?) then burn a new copy of the files that were on it and update the speadsheet.

I researched this myself and in the end I chose Vorta (Borg) because of a recommendation from a friend who is a Linux administrator. He uses it in a production setting so how can you beat that? Anyway, Borg is free and it's pretty powerful but it's a command line utility. I prefer a GUI interface which is where Vorta comes in. Vorta is a docker app you can download in unRAID and it's a GUI interface to Borg. Although Vorta is a bit kludgy it does the trick. You can create one or more archives and each archive is encrypted. One other feature I like is the ability to mount an archive point. So if you need a file from a week ago then you mount that archive date and pull the file out. The decision for you is, do you create one archive or many? Each archive benefits from deduplication but the more you put into an archive means that it's harder to manage that archive. I have multiple archives so that I can control the schedule and docker. Each archive has it's own docker app because I found that trying to create multiple archives in a single Vorta docker didn't work as I expected. Within each archive I keep many archive points (dates/snapshots) because there's no reason not to with deduplication. Can you clarify what you mean here? In Vorta you can see the progress and check the archive. In your case you just need one archive.