Beiträge von Sean

Thanks for your reply. I think I found the reason for the error in the OMV-extras documentation: Don't use a mergerfs file system. (Maybe the guide could be extended by one sentence?)

Now I created the shares on another drive, and docker is running. But, unlike to what it says in the guide, my containers are not running. They also don't show if I enter "docker image ls" or "docker images" at the command line...

Hi,

I tried to migrate from the "old" omv-extras docker to docker-compose following KM0201's guide https://forum.openmediavault.org/index.php?thread/48003-guide-using-the-new-docker-plugin/ , but in step 11, after entering the new paths and trying to apply the changes, I got this error:

Failed to execute command 'export PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin; export LANG=C.UTF-8; export LANGUAGE=; omv-salt deploy run --no-color compose 2>&1' with exit code '1': orion.local:
----------
          ID: docker_install_packages
    Function: pkg.installed
      Result: False
     Comment: Problem encountered installing package(s). Additional info follows:

              errors:
                  - Running scope as unit: run-rf92e55227d6b4ca390754b6cc4f3c819.scope
                    E: Sub-process /usr/bin/dpkg returned an error code (1)
     Started: 15:27:45.906497
    Duration: 46967.524 ms
     Changes:
              ----------
              containerd:
                  ----------
                  new:
                  old:
                      1.4.13~ds1-1~deb11u4
              containerd.io:
                  ----------
                  new:
                      1.6.21-1
                  old:
              docker-ce-cli:
                  ----------
                  new:
                      5:24.0.4-1~debian.11~bullseye
                  old:
              docker.io:
                  ----------
                  new:
                  old:
                      20.10.5+dfsg1-1+deb11u2
              runc:
                  ----------
                  new:
                  old:
                      1.0.0~rc93+ds1-5+deb11u2
----------
          ID: docker_compose_install_packages
    Function: pkg.installed
      Result: True
     Comment: All specified packages are already installed
     Started: 15:28:32.887698
    Duration: 1598.188 ms
     Changes:
----------
          ID: docker_purged_packages
    Function: pkg.purged
      Result: False
     Comment: Problem encountered removing package(s). Additional info follows:

              errors:
                  - Running scope as unit: run-r2282fc5c8e204c8e8fa5f9787da0341f.scope
                    E: Sub-process /usr/bin/dpkg returned an error code (1)
     Started: 15:28:34.486162
    Duration: 12448.088 ms
     Changes:
              ----------
              installed:
                  ----------
                  docker-compose:
                      ----------
                      new:
                      old:
                          1.25.0-1
              removed:
                  ----------
----------
          ID: configure_compose_global_env_file
    Function: file.managed
        Name: /srv/mergerfs/Data/compose/global.env
      Result: True
     Comment: File /srv/mergerfs/Data/compose/global.env is in the correct state
     Started: 15:28:46.946963
    Duration: 265.331 ms
     Changes:
----------
          ID: /etc/systemd/system/docker.service.d/waitAllMounts.conf
    Function: file.managed
      Result: True
     Comment: File /etc/systemd/system/docker.service.d/waitAllMounts.conf is in the correct state
     Started: 15:28:47.212528
    Duration: 3.11 ms
     Changes:
----------
          ID: systemd_daemon_reload_docker
    Function: cmd.run
        Name: systemctl daemon-reload
      Result: True
     Comment: Command "systemctl daemon-reload" run
     Started: 15:28:47.217266
    Duration: 461.093 ms
     Changes:
              ----------
              pid:
                  94412
              retcode:
                  0
              stderr:
              stdout:
----------
          ID: configure_etc_docker_dir
    Function: file.directory
        Name: /etc/docker
      Result: True
     Comment: The directory /etc/docker is in the correct state
     Started: 15:28:47.678987
    Duration: 3.258 ms
     Changes:
----------
          ID: /etc/docker/daemon.json
    Function: file.serialize
      Result: True
     Comment: File /etc/docker/daemon.json is in the correct state
     Started: 15:28:47.682494
    Duration: 6.626 ms
     Changes:
----------
          ID: docker
    Function: service.running
      Result: False
     Comment: Job for docker.service failed because the control process exited with error code.
              See "systemctl status docker.service" and "journalctl -xe" for details.
     Started: 15:28:47.719754
    Duration: 411.986 ms
     Changes:
----------
          ID: create_usr_local_bin_dir
    Function: file.directory
        Name: /usr/local/bin
      Result: True
     Comment: The directory /usr/local/bin is in the correct state
     Started: 15:28:48.132922
    Duration: 6.682 ms
     Changes:
----------
          ID: /usr/local/bin/docker-compose
    Function: file.symlink
      Result: True
     Comment: Symlink /usr/local/bin/docker-compose is present and owned by root:root
     Started: 15:28:48.140212
    Duration: 30.104 ms
     Changes:
----------
          ID: configure_compose_scheduled_jobs
    Function: file.managed
        Name: /etc/cron.d/omv-compose-backup
      Result: True
     Comment: File /etc/cron.d/omv-compose-backup is in the correct state
     Started: 15:28:48.170926
    Duration: 49.667 ms
     Changes:

Docker shows as installed, but not running in the dashboard (it is activated in the omv-extras setting, though).

Maybe it's a little thing. I looked through some parts of the pinned thread regarding docker-compose, but I couldn't find a solution to this problem. Any ideas?

Zitat von ryecoaaron

Nope. I've used three with 8 ports each.

Wow, 24 disks... That's a bit out of my league!

Zitat

Clone the old to the new with Clonezilla.

Thanks for the pointer! Looks fairly straigtforward... So I boot the NAS from the Clonezilla USB stick, clone the system disk, and reboot, right?

Hi,

I am currently running OMV on a system with a system SSD and a 5-HDD Snapraid array. Since the disks are filling up, I want to extend the storage with another 5-disk pool of Western Digital Ultrastar DC HC550 18TB disks. I could squeeze all 10 HDDs into the current case, but then there would be almost no space in between for ventilation. So I decided to buy a new case, and my favourite is the Fractal Design Meshify 2. In addition to gaps between the individual disks, the HDD mounts have some vibration dampening, which should also improve disk longevity. (A cheaper alternative would be the Aerocool Cipher, but here the disks are closer to each other and there is no vibration dampening.)

Are there any experiences regarding one of these cases or alternative cases capable of holding 10 HDDs? One restriction: Since the Ultrastar doesn't have any screw holes in the middle position, I can't use cases which rely on these screw positions, like the Fractal Node 804.

Using two SATA controllers with 4 ports each shouldn't pose a problem, right?

Secondly, once I'm at it, I would also like to move the OS (OMV) to a larger and newer SSD. What is the easiest way to do this?

Any help is appreciated!

OK, while trying to reproduce the behaviour, I believe I finally managed to find the cause: The "News" directory, where all the new files go, actually exists an all file systems, but the "temp" directory, which is used for incoplete downloads, does *not* exist on "Data1". So the download starts on one of the other volumes, and after completion of the download, the file stays on the same volume, even when it's moved to another directory. So I guess you were right after all.

Thanks again for your help!

Sorry if I sounded irritated - I didn't mean to yell, I only meant to emphasize, and I do appreciate your help.

I will try to give a better example tomorrow.

But the path *is existing* on both volumes! If you look at the last screenshot: The directory written to is "Media/News". A new directory for the new file is generated on Data1, but no file is written to it. Instead, the same directory is generated on volume "Data4", where the files end up being stored.

Hi trapexit,

I finally managed to dig a bit deeper and found the following:

When new files are created in the pool, it never happens on Data1. Occasionally, a new directory is created on Data1, but without the containing files; instead, the directory is also created on another drive in the pool (with the files). So it is possible to write to the file system.

I added some screenshots of the mounted files, the file systems, and the mergerfs configuration.

Edit: I also added a screenshot of a newly created directory: The directory is created on both Data1 and Data4, but the files are written to Data4 (although it has less free space).

Does this behaviour make any sense to you?

OK, after trying some more functions, I found that "reset permissions" solved the problem; now everything is owned by "root users", previously it was "docker 1000".

Could it be that a recent docker update screwed up the group of the docker user somehow?

I guess files newly created by docker will have the same problem again. How can I change this? In the docker submenu of the workbench, there is no option to change the group of the docker user.

Edit: ... and in the "User" dialog, docker belongs to the groups "users, docker", and in "etc/group", docker has id 995. How can I assign the files created by docker to the "users" group? Also: Shouldn't the ACL permissions override the unix user/group permissions?

Thanks for your reply. However, I have disabled the preview pane, and the problem even occurs when only one explorer window is opened. So I think this is a different problem.

I recently have had a similar problem. It appears that files that were created by the docker user can't be modified by other users, even if they have read/write permission for the folder in question.

Another strange thing is that sometimes the error message says that the file is in use (even though it isn't; I even rebooted to make sure of that), sometimes it says that the user doesn't have permission.

Any Ideas?

Thanks for clearing that up.

Zitat

Not sure why you were uninstalling sharerootfs

No particular reason other than removing everything that's not needed because it can use resources and compromise security.

Hi,

I recently uninstalled some unused plugins, among them sharerootfs. At the next scheduled "snapraid sync", I received the error message "snapraid: command not found", and when I checked the installed plugins, snapraid really was gone.

I thought that I maybe had selected it for uninstallation by accident, but after I reinstalled snapraid, sharerootfs was back as well. The logs also show that the plugins were (un)installed back-to-back.

So my question is: Does snapraid require sharerootfs somehow? In that case, I would have liked a warning before uninstalling sharerootfs, because during uninstallation of snapraid, my configuration was also lost.

OK, I created the issue. Let's see what happens...

OK, sorry if I didn't express myself clearly. The username is "admin". But I entered (by autofill or copy and paste, or maybe I hit space accidentally) "admin ", and the additional space is not recognisable in the login dialog unless the cursor is behind it.

So I tried to log in many times and kept getting "Error 400 - Bad request", and I couldn't find the reason because login and password looked correct. And my suggestion was to modify the login dialog of the web UI so that it removes white space from the user name (which, as you said, is illegal anyway).

Hi,

after the update to OMV 6, I couldn't log in to the web ui from my phone, even though I verified a dozen times that user name and password were correct. I suspected the browser, my ad blocker, android...

But now I found out it was a trailing space in the user name: "admin " rather than "admin", which is (almost) impossible to see.

Maybe the username should be cleaned from whitespace before verifying the login...

Zitat von ryecoaaron

Guessing you are using usb drives?

No, they are internal drives. And according to SMART, they are also healthy.

Zitat

[ 2.429682] ata7: SATA link up 6.0 Gbps (SStatus 133 SControl 300)

Zitat von ryecoaaron

The upgrade "converts" the pool but it seems like there are too many problems when using the old mount point /srv/UUID.

OK, I followed your advice, pointed the shares to another drive, deleted and recreated the pool. Let's see if it works reliably now. I hated the UUID as the mount name anyway.

Zitat

Is this a pool that was upgraded from omv 5 to 6?

Yes. Do I have to do something to "convert" an OMV 5 pool?

OK, now I'm really confused because after another reboot everything is working again - and everything I changed was stopping the docker containers before the reboot.

Thanks to both of you for helping! If it turns out that something is still broken, I'll be back.

Zitat von ryecoaaron

If the pool can't be restarted after your change, it usually fails. Not sure how the plugin can fix that. That is why rebooting is usually easier and when you apply after the reboot, it is restarting the same pool.

Was this pool created from an OMV 5 to 6 upgrade? I expect the changes to still be pending. And as I said above, restarting the same pool after the reboot should have less issues.

I tried to restart the pool after a reboot. Same error.

You mean delete all the shares referencing the merged file system? That delete button is also greyed out...

I stopped the docker containers using the merged FS, but that didn't do it.