Thank you all for your responses.
To clarify, I haven’t made any manual or explicit changes to the system configuration, including the fstab entry I referred to. I do use NFS shares, those were intentionally set up for my specific needs.
For now, I’ll keep my home share as it is, since the system configured setup works fine for my current use. Once I gain a better understanding of how to modify it safely, I’ll making adjustments.
Thanks again for your help!
How did you "log in"? I assume ssh. If yes, then what you are seeing is just the directory of your home directory on the filesystem. It is not smb or nfs since those are network sharing protocols. This is no different than clicking on your Documents folder on windows.
Hi,
It’s been a while and I’ve tried a lot of things since then, but here’s what I remember:
I haven’t used Windows in a long time. On my Linux machine, I browsed the network using Nemo and found the "home" share there. I was most likely asked for credentials, and after logging in, I mounted it and added an entry to my fstab.
The fstab entry uses cifs, which led me to believe it's a Samba (SMB) share. But since you mentioned it's not Samba, does that mean my fstab entry is incorrect?
I also created a second user on another computer and repeated the process. Once I found the share and entered credentials, I was in—and again, I added a fstab entry using cifs.
No SSH involved—just network browsing and mounting via the file manager.
So now I’m wondering: is cifs the right option to use in fstab for this setup, or should I be using something else?
Thanks in advance!
Just to be fair to my own question earlier—I wasn’t entirely sure how Home folders actually work in OMV or Debian in general (and I still don’t know if it's the same behavior in both).
It turns out that I'm not supposed to see a shared folder at the top level called Homes where I would then pick my personal folder.
Instead, once I log in, I’m taken directly into my Home folder, and I only see its contents.
It just drops me right into my personal Home.
So, this behavior is simpler than I expected—but also different from what I had imagined.
One thing I still don’t know is: what kind of share is this? SMB? NFS? Something else?
I hope this helps others who are using Home folders in OMV for the first time and wondering about how it behaves.
Is there a way to have docker compose using the new appuser ID?
Is it correct to think that when creating a container and not specifying the ID, it will use a default one? probably not the correct one.
appuser needs to be created. It also doesn't have to be named appuser. You can create a group in the OMV web interface and then create a user in the OMV web interface and make that user part of the group.
If you use the same user for all containers, someone will have access to all containers if they get the user's credentials. Up to you. It is just like using the same password for every website. Unique passwords for every site is a better idea but requires effort (which is worth it in my opinion).
I must have wrongly understood the appuser. Has to be created.
So, ideally we should create a user for each containers for maximum security.
Effectively, I could end up having a dozen of users.
It is an interesting decision to make.
Considering is a home server lab.
Thank for the suggestions.
Where did you get that idea?
I guess I misunderstood the wiki!
Appuser have to be created first (or any other user)
Alles anzeigenThe command is just id
But looking at your system, it looks as though you're using root as your user.
I generally create another user, and make sure it's a member of the 'users' group.
Codeid uid=1000(newuser) gid=100(users) groups=100(users)...
So with the above, this is what I would put in my compose file
Codeenvironment: - PUID=1000 - PGID=100
Hi,
So, I ssh to my system as root.
But looked for ID for appuser that from the wiki it seems a default user created. however, I cant find it.
Just id gives a root ID 0 and a group 0. I will not use that!
I was also thinking of creating a user. And since appuser does not seem to exist, I could create that to use for the containers.
But I read I should also be careful not to use group 100.. or I though I understand that.
From the wiki, the best practice would be to create a user for each container but I think is unpractical.
What are the security implications when creating a user (just a user ) and using it for docker?
Would you suggest to create it using CLI? Or OMV GUI would be good enough?
That part is also not so clear to me (I probably missing enough knowledge to understand that per well).
PS
Is not a raspberry pi. I use a nas hardware (ugreen 2 bays. + a NVMe for compose files, backup and docker. the actual media is on a raid0 as is less important - few HDs that will server movies and TVshows)
Thank you
Have you looked at these guide?
https://wiki.omv-extras.org/doku.php?id=omv7:docker_in_omv
Hi.
i did. Not at first but eventually I read it.
I am confused by the app user.
from what I read I take the follwoning:
1) there is a default use appdata.` But needs to be taken some permission away to avoid accessing folders we dont want.
2) never use root user (of course not a good idea)
When I check for the default app user, I cant find it:
root@ugreen-nas:~# id appuser
id: ‘appuser’: no such user
Do I misunderstanding about this default user?
If I create a simple user for docker, I dont understand how can I define compose files to use the newly created user.
Could you please explain me that or point me where to find this in the wiki?
Thank you
Hi,
I would like to install my docker files/containers correctly.
I read that is best NOT to run the containers under any admin user or users with access to Homes.
Using OMV7 docker compose, what PUID and PGUID is used by default?
For future containers, the compose files uses the ID. So far I have only admin and a user with admin rights and access to Homes.
Would it be best to create a user for docker files?
Greetings, Today April 2025 I had the exact same issue. No DB modification. followed the instruction in the wiki and had same as Winkimania. sudo apt-get install --reinstall openmediavault-compose
As suggested by Ryecoaaron did the trick.
the solution is still valid today.
Thank you
The worst-case scenario is a fire or burglary at your home, and the server has disappeared along with all the data. Or malware encrypts all your files, and you receive an email requesting a €2,000 payment for the decryption key.
A RAID doesn't solve these scenarios. Consider setting up a full backup (or several), preferably remote (at a friend's or relative's house) and preferably versioned. https://www.raidisnotabackup.com/
Regarding recovering from a degraded RAID, you can search the forum and find many threads with detailed explanations.
Hi Chente,
Thank you for your earlier suggestion—I’ve done some research on handling degraded RAID arrays, and I wanted to clarify my question since I’m using Btrfs RAID 1 (not MDADM).
First, I want to emphasize that OMV is my favorite open-source NAS solution, and I fully understand that RAID 1 is not a backup—it’s purely for hardware redundancy.
My original question was about how to distinguish between a recoverable disk issue (replace disk) and a scenario requiring a backup restore. While OMV’s GUI offers basic checks (Storage > File Systems), I’ve found that Btrfs-specific CLI commands (btrfs filesystem show, btrfs scrub, btrfs device stats, etc.) provide much deeper insight into array health.
If time allows, I'd like to create a beginner-friendly tutorial on monitoring Btrfs RAID 1 in OMV. This would cover how to:
As someone who's just gone through this learning process myself, I think it could help other OMV users maintain their NAS more confidently and safely. Would this be useful to the community?
Hi.
The situation I have is that I dont seem to fully understand how User home directory works.
What I did:
I followed the doc link: Users and it seems I am doing all the right steps.
created a Shared Folder called Users
Under User Management I Enabled User home directory:
under Services | SMB/CIFS | Settings I checked Enabled for the Enable user home directories option
Therefore, under Storage|Shared Folders I now have a Users referenced share.
When I browse the network on my linux laptop, I see the folder called Homes (I assume that even if I created Users shared folder, OMV will show it as a Homes under smb when searching in network).
Problem:
when I browse the network and see Homes, I only see 3 files .bash_logout, .bashrc and .profile
I dont see the User folder name.
I created 2 Users for testing.
Wouldn't I see 2 folders under Homes with the users names?
Also, I dont get asked for credentials when accessing Homes.
What am I doing wrong?
When I SSH the NAS, I also am able to see the created users.
Yet, under smb://my-nas.local/homes/ there are no users to choose.
if I manually add an existing user_name I simply have an error "Could not find "smb://my-nas.local/home/testuser""
I also tested by adding a samba share and reference Users.
Of course I can than see Users and all the users in it. Despite I set it not public, when I access the User folder using the password of the testuser, I can access any other user too.
Thank you
If you want to repeat the questions on this thread that you asked here: RE: Raid Setup Settings you might get an answer ( it's not possible to "quote" from another thread ).
Not sure I understand what you mean here.
I got to the conclusion that BTRFS RAID1 is best over BTRFS RAID1 MD from the article I mention.
That article explained very well what is the difference among the 2 options.
I got to understand that MD is best for Ext4 but not so much for BTRFS!
In case I got that the wrong way, Please let me know.
Ok, thank you.
Macom posted a really good link: https://www.dont-panic.cc/capi…rfs-raid1-vs-mdadm-raid1/
Things are getting clear what to use.
As a since note, as soon as I installed MD, my system became unresponsive and I actually lost the SMB service (is there but cant be founds from any of the computers I have).
Sounds illogical that MD did anything on the SAMBA service but it just happen at the very end of MD installation. Nor I can access it with domain name anymore!
Btrfs raid1 vs. mdadm raid1
Really good article.
Things are getting much clear now.
From this article I take it is better to use BTRFS RAID1 rather than MD BTRFS RAID1.
Effectively is best the native OMV RAID1 system and not installing MD. (MD seem to be more effective for ext4)
Ok,, now.. back to OMV...
if one disk has a "Btrfs RAID Pitfalls", how OMV handle that?
How it gives the info of a "degraded arrays"?
Since that is the wake-up call for using the back up, would I consider the second disk my backup (BTRFS substantially make a copy of one of the 2 disks in RAID1).
Is there a way to know how to recover a degraded array using OMV?
Basically I want to be prepare for the worst case scenario and know what to do or how to recognize that.
And really a big Thanks for that link.
I've just answered your question in your previous thread, RE: wipe 2 disks coming from an other NAS
Starting a new thread for each question that belongs to a previous thread will just cause confusion on the forum
I agree. It is redundand in some way.
When I wrote this post was to have a separate more dedicated post not only for me but for those that need to search for RAI1 vs MD.
Also, I write this before the answer you game the other post.
I will continue to the other original post
Sorry for the confusion
Hi
I need to understand the difference between OMV7 different RAID creation procedures;
The File System options gives me the option to choose FS and type of RAID (my preffered choice is BTRFS and RAID1 as I have 2 bays)
There is also a Plugin called MD that seem to do the same
traditionally, MD will mirror the files and the procedure will take over 30 hours (I have x2 16TB drives). I had this long mirroring time when using the original OS of my new UGREEN nas (2 bays)
The File System raid1 options I chosen, it was instantly created.
Is there a way to understand that? Mind I dont have great experience in RAID.
Thank you
Can someone explain this to me?
Did you figure the difference between the RAID option and the MD plugin?
It is not clear to me either.
maybe I dont understand how RAID works or how it works in OMV.
My goal is to have RAID1 formatted in BTRFS
It seems I can do that by simply wipe the disk and create and mount a file system. Choosing the file system I want (BTRFS) and the disks to be used after selecting RAID1.
Would that give me a raid1? (the creation process seem to me too quick to mirror 16TB )
Label: none uuid: e32f6490-c13a-434b-8039-1d94d100e3a3
Total devices 2 FS bytes used 160.00KiB
devid 1 size 14.55TiB used 2.01GiB path /dev/sda
devid 2 size 14.55TiB used 2.01GiB path /dev/sdb
Data, RAID1: total=1.00GiB, used=0.00B
System, RAID1: total=8.00MiB, used=16.00KiB
Metadata, RAID1: total=1.00GiB, used=144.00KiB
GlobalReserve, single: total=5.50MiB, used=0.00BWhat do you mean with "If you want a traditional Linux software raid mirror"?
sound like if I dont use MD, there is a non traditional form of Raid mirroring?
PS
I have installed MD
when the installation finished, I got kicked out the webUI with a message"software failure... could not read the all thing.
could not reboot not even SSH the nas. I had to pull the plug.
Now I am back but it does not connect with domain name, only IP address.
I dont know what the MD has done but made things different!
Have you installed the MD plugin on OMV7? Because it looks like you've created BTRFS RAID1 profile.
My goal is to create a raid1
I did not install any plugin yet.
You mean this?
penmediavault Linux MD (Multiple Device) plugin