Beiträge von bolehill

Thanks BernH. I'll post how I get on, once the hardware arrives.

A couple of days ago, my server died on me. It started misbehaving, so I opted for a re-boot. The re-boot became a shut down and I couldn't power up the machine at all. I've checked the usual suspects - wall socket, PSU, on/off switch etc to no avail.

The mobo was pretty old (Gigabyte GA-M68MT-D3P rev 1.0), with an Phenom 905e processor and 16GB ram. It didn't have to work too hard, just serve music and films via a Jellyfin docker.

Despite being a bit despondent initially, I'm coming round to the thought that this might not be a bad thing. The hardware is pretty long in the tooth, so my thoughts move on to new kit and re-installing OMV.

From previous reinstallations, I know that I can pick up the directories on the data drives fairly easily, the the big question however is what's the best way to boot OMV? Can I use the existing OS drive or is a full re-install the better option? I'm guessing the latter.

I'm planning on a new mobo, processor and ram - probably Ryzen 3/4 (with graphics), and re-using the three hard drives that contain data. The boot drive is a Samsung SSD which I'd clear for reinstallation if that's be better option.

Advice appreciated!

TIA

Chris

My OMV installation is running on a Gigabyte GA-M68MT-SP2 (revision 1.0), circa 2010...

Originally with an Athlon II, but latterly upgraded to a Phenom X4.

It works perfectly, doing exactly what was planned - nothing particularly special! Just a simple media server (Jellyfin)/file server.

I've thought about upgrading the system, but why spend money on replacing something that works, and works well?

If I do spend money it will be on a bigger hard drive...

chente - I've made my peace with this. Suppressing my 'completist/perfectionist' tendencies after having discovered the dangers of a misplaced space in the command line instruction to delete a folder, I'm good.

The drive will get formatted at some stage in the future and the problem will be gone. In the meanwhile, if I don't look, it's not really there .

Appreciate your help.

The folders and Samba shares were created in OMV6. I physically disconnected the drives before I installed OMV7 (to make sure the drives didn't get accidently formatted), then physically reconnected the drives after OMV7 was up and running, then re-established the shares and mapped the folder locations.

I'll take a look at /srv

Thanks for the help/advice.

Thanks chente. To confirm - the drives are already mounted?

I have done a clean install of OMV 7. Very straightforward and everything works as well as it did previously.

One thing I noticed as I was reconnecting all my shares, was that I had quite a few 'orphan' folders. It looks like formally deleting unwanted shares in OMV doesn't actually delete the folder on the drive. (Previous posts here had said the same, but on older versions of OMV).

So my question is how to delete these safely and without upsetting OMV and Debian?

Do I need to Mount the drive, then delete the folders, and finally unmount the drive?

The OP's system is running primarily off storage batteries, with the grid as backup...

I replaced my Athlon II with a Phenom 905e recently and used Arctic MX4. It works just fine.

The key thing is to make sure all the old paste has been removed. I used Isopropyl alcohol to clean it off the heatsink. In your case, you will need to clean the cpu die and the heatsink.

Not worth paying over the odds for 'special' paste imho.

Thanks for the reply. Sadly the nomodeset edit of grub didn't work, so I slept on it.

The nights sleep must have helped, as I got there in the end through trial and error:

1 - reversed the video cable just in case there was a bad pin connection (didn't work)

2 - set the monitor to 1024x768 (didn't work either, but then...)

3 - incrementally changed some settings in the bios:

- unganged the memory

- disabled just about everything that wasn't needed in a running system

- tweaked the on-board graphics settings to something minimal

- loaded 'fail-safe' defaults

and voila, something clicked and I'm up and running.

Not sure about the root cause, but clearly something in the bios wasn't meshing with the monitor properly.

Thanks again for the help ryecoaaron!

I found the grub.cfg file, but it's difficult to see where to add the nomodeset line.

The research I've done suggest amending the line ending with "quiet splash" to "quiet splash nomodeset", but as you can see there's no such line in the boot/grub/grub.cfg file in the OMV installation media.

Thanks for your help this far.

set default=0

loadfont $prefix/dejavu-bold-16.pf2
loadfont $prefix/dejavu-bold-14.pf2
loadfont $prefix/unicode.pf2
set gfxmode=auto
insmod all_video
insmod gfxterm
insmod png

set color_normal=light-gray/black
set color_highlight=white/black

if [ -e /isolinux/splash.png ]; then
    # binary_syslinux modifies the theme file to point to the correct
    # background picture
    set theme=/boot/grub/live-theme/theme.txt
elif [ -e /boot/grub/splash.png ]; then
    set theme=/boot/grub/live-theme/theme.txt
else
    set menu_color_normal=cyan/blue
    set menu_color_highlight=white/blue
fi

terminal_output gfxterm

insmod play
play 960 440 1 0 4 440 1

menuentry "Install" {
    linux    /install/vmlinuz vga=normal quiet file=/cdrom/install/preseed.cfg file=/cdrom/install/preseed.cfg file=/cdrom/install/preseed.cfg
    initrd    /install/initrd.gz
}

submenu 'Advanced options...' {

# Memtest (if any)
menuentry "memtest86+" {
    linux16 /live/memtest
}

}

I've been using OMV for about a couple of years, pretty much trouble free. The O/S drive started to get a bit full and having a decent sized spare SSD I thought I'd perform a re-install from scratch. Previous installs had been a breeze, so I prepped my USB drives and got started.

Step 1 - disconnected all the drives (1x ssd, 3x mechanical hd)

Step 2 - swap sda1 out and put the Samsung 840EVO in (250GB) to become the new sda1, Connected power and sata to this drive only.

Step 3 - Plug the bootable USB into a port and hit the power button. USB drive created with Rufus and OMV iso download.

The system boots and I can see GRUB load and then there's an AntiX screen with three options. I left it to its own devices and after a couple of seconds the monitor loses input and a message "input not supported" pops up. I'm using the same monitor that had worked on previous installs.

My system is self built

Gigabyte GA-M68MT-D3P motherboard (integrated Nvidia graphics)

16GB RAM

AMD Phenom 905e proc

plus the Samsung 840 SSD

Iiyama monitor connecting via D-SUB (there are no other possible connections from the motherboard)

I also tried a bare Debian install which yielded the same result.

Previous posts in the forum on this subject were related up upgrades, but I'm down at bare metal level so can't try the solution.

Any help would be much appreciated.

Zitat von deej78

Can someone write some easy to follow guides for this please and stick to the main page so the beginners can sort this out?

The guides that are out there already are sufficient. I'm by no means a proficient Linux user, and I can get by with Windows.

My first thoughts on the upgrade were "No, no, no. How do I work around this so I can keep Portainer"? Then I left it be for a couple of days as I thought it was a horrible mistake that would go away eventually. Of course it didn't.

My next step was the sensible one that I ought to have taken first. I read the user guides and made some notes. Simple notes in the do this, then do that style, with references to the published guides. Make sure you get some peace and quiet while you do this - no distractions!

This worked for me. It didn't go totally smoothly, but I learned a lot and I got there in the end (it took about 2 hours, including reading the guides and making notes). By the time I'd finished, I liked what I saw and ditched Portainer!

Docker every time. I use Portainer (via OMV) to create and store my various installation stacks. In the event of a need to re-install Jellyfin (or other docker image) for any reason, its an absolute doddle.

I replaced my processor about a year ago, swapping a very old Athlon II for a slightly less old Phenom 905e.

OMV booted without a hitch and correctly identified the new proc in the dashboard.

According to Docker, the change only affects a few accounts. The quote that follows is from the link you posted:

"For those of you catching up, we recently emailed accounts that are members of Free Team organizations, to let them know that they will lose features unless they move to one of our supported free or paid offerings. This impacted less than 2% of our users. Note that this change does not affect Docker Personal, Docker Pro, Docker Team, or Docker Business accounts, Docker-Sponsored Open Source members, Docker Verified Publishers, or Docker Official Images".

I have a free personal account and have not received any e-mail from Docker regarding changes to that account, so I'm guessing that I'm unaffected.

Thanks macom. That cleared the error!

So far I've successfully deployed Jellyfin from Portainer, creating the various settings from the GUI.

After adding additional disk space, I thought it might be worth trying to deploy via a Custom Template (still within Portainer). Composing the file was pretty easy as there's lots of help available.

Portainer helpfully shows errors in the file, and its proving difficult to resolve the error in the final line (restart:unless-stopped)

YAML file as follows -

services:

jellyfin:

image: lscr.io/linuxserver/jellyfin:latest

container_name: jellyfin

environment:

- PUID=1000

- PGID=1000

- TZ=Europe/London

volumes:

/srv/dev-disk-by-uuid-33cf22b6-9ac9-4430-87c3-4b83defeace7/config:/config

/srv/dev-disk-by-uuid-33cf22b6-9ac9-4430-87c3-4b83defeace7/tv:/data/tvshows

/srv/dev-disk-by-uuid-c7f127de-5b40-40e1-b2a4-9d8ebb7bae1f/films:/data/movies

/srv/dev-disk-by-uuid-33cf22b6-9ac9-4430-87c3-4b83defeace7/photos:/data/photos

/srv/dev-disk-by-uuid-33cf22b6-9ac9-4430-87c3-4b83defeace7/music:/data/music

ports:8096:8096

restart:unless-stopped

I've tried deploying the stack, and it fails to deploy with the error:

failed to deploy a stack: yaml: line 16: could not find expected ':'

I can't see where the missing colon should be - can anyone shed a little light please?

I did now!

Thanks!

I check whether there are updates to apply each morning, and today the login page wont load, just a light blue screen with 'Loading...' in the middle and the mouse pointer as a stopwatch. The server itself is running and I can access it through PuTTY.

The server is used as a media server only (Jellyfin running in a Docker). I've checked all the drives via du - none are close to full: The root drive is showing 8%, the media drive is 78% and the drive I use for backing up data from various PCs is 2%.

After a quick trawl through this forum, I've tried all the usual suspects:

- Clearing the browser cache

- Followed the [How to] Fix full OS filesystem - GUI login loop fixes

- System reboot

Can anyone offer advice please?