Posts by justtim

In the end I solved it by moving the folder /var/lib/docker to the new location and simply creating a symlink. Maybe a little dirty but it does the job for now.

Quote from chente

In that case, here is a procedure.

Tried this, caused Docker to start empty, none of my containers/volumes etc. are visible. Also tried to add --data-root /srv/dev-disk-by-uuid....... to /lib/systemd/system/docker.service but that resulted in an error when I tried to start the service.

Quote from olduser

daemon.json maybe? I don't use the plugin so I don't know what steps it does or doesn't.

This file is populated by the plugin. I will check if /lib/systemd/system/docker.service is modified when the Docker storage location is changed in the plugin.

Quote from chente

I would do the same but without copying the data from that folder. Unless you have some special configuration, what is in that folder can be downloaded automatically when you start the containers.

I use Docker volumes so I do want to preserve the data in there.

Hi, I would like to move the Docker storage location to another disk to make it easier to expand the disk when it is filling up.

What I've tried so far:

1. Add new disk to OMV (VM in Proxmox)

2. Mount disk in OMV

3. Copy all data from /var/lib/docker to /srv/dev-disk-by-uuid....... using rsync -aP

4. Change Docker storage location under Services -> Compose -> Settings to /srv/dev-disk-by-uuid......

5. Restart Docker from Services -> Compose -> Settings (even tried reinstall Docker)

After perfoming these steps Docker is started, the location has been changed but all the volumes, containers and images are not visible.

Am I missing something here?

Thanks, just so you know, all seems well with onedrive 2.4.25-1 so maybe it is an option to include that version? It is just officially not stable yet.

I was able to work around it for now by first installing the onedrive 2.4.25-1 (Trixie) package: https://packages.debian.org/source/trixie/onedrive. Afterwards, the plugin installed succesfully but I had to do chown onedrive:users /var/cache/onedrive/ before I was able to authorize successfully.

I was unable to install the OneDrive plugin on a fresh OMV7 install:

The following packages have unmet dependencies:

openmediavault-onedrive : Depends: onedrive (>= 2.4.25) but it is not going to be installed

I tried to manually install onedrive but that didn't help:

The following packages have unmet dependencies:

openmediavault-onedrive : Depends: onedrive (>= 2.4.25) but 2.4.23-1 is to be installed

Quote from macom

You can boot a linux live distro on your laptop/PC and do this.

So, add the new disk to a Linux sytem, format in ext4, change the label to volume2, and add it back in my Odroid? I guess that is worth a shot!

Quote from macom

Sorry. I meant label, not ID

On OMV (currently) drives are mounted as "dev-disk-by-label-xxx".

Previously they were mounted as "dev-disk-by-UUID-xxxxx"

Ok, that is indeed the case, the broken disk is still visible as dev/disk/by-label/volume2.

So how would I go about setting the same label for the new disk? That cannot be done if the old disk is still present, and I cannot remove the old disk because the shares (and probably other stuff) still has references to the old disk.

I will try adding the old UUID to the new disk first, to see if that does the trick. If it doesn't, it looks like the only way is to go with @Adoby's suggestion, cleanup all references, remove the old disk and add the new disk.

Quote from macom

If the disk is mounted "by-id" should it not be enough to just adjust the id? I would try if it is working.

Where would I adjust the ID? Can you elaborate a on this?

Quote from macom

It seems to be also possible to change the UUID if needed
askubuntu.com/questions/132079…a-disk-to-whatever-i-want

This is an interesting option. I guess I could give this a shot as soon as my new disk arrives, since I have nothing to lose (I'll make a backup of my SD boot card first)

Thanks Adoby, ok, so there is no easy way to replace the existing volume with a new one and keep the volumename?

Hi, I have a SBC (Odroid HC2), with just one 3TB data disk and an SD-card containing the boot partition.
I make regular backups of said data disk with Duplicati.
Last weekend, the data disk crashed, and I am planning to replace it with a new one (probably SSD while I’m at it).
What would be the best way to go about this? Can I simply replace the now broken disk (shows “missing” in OMV) with the new one, give it the same volume name, and restore my folders? It looks like I cannot delete the “missing” volume from the GUI...

I'm starting to think the start options and environment variables are kicked off in the Docker run command when the container is started and are not actually stored anywhere, could that be true?
Stil doesn't explain how Watchtower knows about all of them though...

Let me clarify: The options I am using to start a container with (see screenshot) are saved in appdata outside Docker?
Where would I be able to find them on a Debian OMV install? Can you give me an example of such a config file on the file system?

Hi, I am kind of new to Docker but have quite a lot of containers running nonetheless and love it!

There is something I can't get my head around though, and I hope someone can explain this.
When I run a container from an image, and set a lot of start options/environment variables, these are visible in the details of the running container. My question is, where are these saved? Are they saved to a Dockerfile somewhere?
When Watchtower starts to update images and re-create containers, it knows what run options and environment variables to set, so it must get them from somewhere, right?

This worked for me too!

I noticed /opt/EasyRSA-3.0.3/openssl-1.0.cnf holds the following variables:

default_days = $ENV::EASYRSA_CERT_EXPIRE # how long to certify for
default_crl_days= $ENV::EASYRSA_CRL_DAYS # how long before next CRL

Does anyone know where to access/change these variables?

Ok, I figured this out. There where still entries to the removed disk in /etc/monit/conf.d/openmediavault-filesystem.conf. Manually removed the entries there, rebooted, and the error is gone.

Hi,

After having some trouble removing shared folders and filesystems for a crashed external backup disk, I was able to clean everything up (partly manually from config.xml).
The only issue I have left is monit complaining when booting the system:

Status failed Service mountpoint_srv_dev-disk-by-label-backup 


Date:        Wed, 26 Jun 2019 10:23:01
Action:      alert
Host:        odroidxu4
Description: status failed (1) -- /srv/dev-disk-by-label-backup is not a mountpoint


Your faithful employee,
Monit

Any idea where to clean up the remaining mountpoint apparently still listed somewhere?
There are no more references in config.xml or in fstab.

Quote from ryecoaaron

You have users that are allowed to use the system but you don't want to vpn?

Yes. I want to limit my VPN access as much as possible to reduce the attack surface.