The only machine I have ever worried about too much tmpfs on had 512mb of ram.
Hope it is decent quality with wear leveling. Otherwise it won't last long replacing tmpfs.
Maybe you use /tmp differently than I do.
The SSD has worked for over 4 years - I'll stick with the turkey fallacy for now.
By the way: Do you know why I started using OMV (see nickname)? Because I hoped for a NAS that didn't try to outsmart its users.
It has worked very well so far. Would be a pity if that changed.
OMV makes sure to use tmpfs for /tmp. I don't see any way to disable that. You could delete /srv/salt/omv/deploy/systemd/20tmp.sls but it would come back with every omv update. /tmp isn't a symlink on my systems. Both /tmp and /var/tmp are using tmpfs though. How did you install OMV? Why do you do want /tmp on a regular device?
I installed OMV around version 3 or so and updated incrementally.
Usually I use tmpfs too if the machine has enough memory.
As for the why:
1) the machine has only 8GB and I didn't want to cut off any more of that than necessary
2) I have a separate 128GB SSD for /var so why not use that?
I'll try to hack /srv/salt/omv/deploy/systemd/20tmp.sls in a way it doesn't keep coming back. 
Nov 05 12:30:57 omv systemd[1]: tmp.mount: Failed to run 'mount' task: Too many levels of symbolic links
Nov 05 12:30:57 omv systemd[1]: tmp.mount: Failed with result 'resources'.
and
The unit tmp.mount has entered the 'failed' state with result 'resources'.
Nov 05 12:39:02 omv systemd[1]: Failed to mount Mount /tmp as tmpfs to reduce write access to the root device.Yeah - I know.
/tmp -> /var/tmp
and /var is mounted on its own device
/dev/sda 122548168 6651820 115879964 6% /varSo I believe I know better than OMV and would like it to accept/swallow my design choice without constantly falling on its face when applying configuration.
HowTo?
Well - I found the Discord channel, with a message once every 10 months or so. Will try IRC.
I find myself in the bizarre situation of reading signatures with invitation links to a OMV discord channel, all of which are expired.
Google cache hits of reddit discussions mentioning OMV has a Discord channel, which aren't archived when I visit them.
Pasting some invitation links from Google cache that bring me on a "strange place in Discord with no text channels at all" (quote Discord).
So ... do we or don't we have some Discord Chatroom(s)?
I see these errors in /var/log for postfix
Nov 11 21:17:01 OMV-1 postfix/postsuper[12977]: fatal: scan_dir_push: open directory hold: No such file or directory
Nov 11 22:17:02 OMV-1 postfix/postsuper[14362]: fatal: scan_dir_push: open directory hold: No such file or directory
Nov 11 23:17:01 OMV-1 postfix/postsuper[16130]: fatal: scan_dir_push: open directory hold: No such file or directoryWhat am I supposed to do to rectify the situation?
OMV version is important here.
# cat /etc/issue
openmediavault 3.0.86 (Erasmus) \n \l
The issue is resolved.
Solution:
The omv boot USB stick was the ONLY ONE that somehow set up correctly the VGA mode that worked with my KVM console monitor
(this thing https://noteboox.de/media/imag…-1501ge-kvm-konsole_1.jpg)
Unfortunately the installed OMV does not and therefore the integrated graphic somehow blocks the boot.
First time I encounter this. I found out after boot sticks for Ubuntu, vanilla Jessie (8.9) didn't show a picture
on the monitor.
Attached a different monitor and voila! All works as smooth as expected.
The one successful install was also on a different monitor.
Because BIOS settings, startup picture and of course the OMV Grub menu worked fine, it took me a long time to figure out the monitor (well and obviously the VGA settings of the software) being the culprit. All looked really 1:1 like a grub failure after install.
Hi,
let me explain the title. I installed OMV on my designated mainboard
https://www.aliexpress.com/ite…VGA-HDMI/32574624109.html
which a friend brought from China. So I though I'd replace my aging Qnap with some DIY NAS.
I install from USB to a 1TB mSATA (Samsung EVO 850) on the board
The 1st install went fine, I could boot the mSATA, configure the web frontend, conect via ssh.. all well.
Then I got bold and did apt update && apt upgrade
The machine continued to work for a while, then I lost contact. Since then it never booted from mSATA again.
I installed OMV to it - now for the 7th time. WIthout any problems, the installation goes through, no errors,
the mSATA is there, fdisk sees it, tells me there is a dos boot sector.
But all I get when booting is a screen with a blinking cursor after a while the machine reboots.
When I boot some rescue disk, and then try to boot existing OS, that fails too.
When I boot into a rescue system, mount the /dev/sda1 partition (with OMV on it), that works, FS seems to not have any errors whatsoever, everything looks right.
chrooting into that is impossible, suddenly binaries are missing, but not the corrupted kind of missing, it just looks as if I was on a totally different disk.
I've been doing Linux now since 1995 but I never ever experienced something like this.
And honestly I'm out of ideas. Maybe you got some for me.
pre-edit:
What I did and didn't write up above (because 2 days aren't written in a few lines)
Because I thought I might have done something to the BIOS settings, I popped out the 2032 battery for a couple of minutes - which by the way didn't seem to have any effect on the BIOS settings. Battery back.
Of course I revised BIOS settings like 50 times now. I almost feel at home in the American Megatrends BIOS there...
edit:
Installing the system on a Seagate IronWolf 10TB on SATA port 1 yields exactly the same result - blank screen with a blinking cursor - after reboot and trying to boot from it.
Next try: simply try to install Jessie