So i did the update.
and now i get this upon ssh login
“[ INFO ] DietPi-Login | DietPi-Software is already running on another terminal/screen. Please complete or exit, before continuing.”
That keeps spamming until i Ctrl+C m then im on root again . any ideas on how to fix this ? its really annoying.
Okey , i just got f**ked by the update.
gave me some stuff about not being a complete install , and reverted to a nice clean installation!
Everything on it is gone! woop dee doo , thxs alot .
Yeah similar problem, tried the update twice.
First time I got this after the reboot…
/DietPi/dietpi/login: line 160: ((: > 0 : syntax error: operand expected (error token is "> 0 ")
The version number remained the same… the update did not work.
Tried to update again and after reboot got
[FAILED] DietPi-Software | Root privileges required. Please run the command with "G_SUDO", or, "sudo -i".
mv: failed to access '/var/tmp/dietpi/logs/dietpi-firstrun-setup.log': Permission denied
[FAILED] DietPi-Software | Root privileges required. Please run the command with "G_SUDO", or, "sudo -i".
mv: failed to access '/var/tmp/dietpi/logs/dietpi-firstrun-setup.log': Permission denied
[FAILED] DietPi-Software | Root privileges required. Please run the command with "G_SUDO", or, "sudo -i".
mv: failed to access '/var/tmp/dietpi/logs/dietpi-firstrun-setup.log': Permission denied
[FAILED] DietPi-Software | Root privileges required. Please run the command with "G_SUDO", or, "sudo -i".
mv: failed to access '/var/tmp/dietpi/logs/dietpi-firstrun-setup.log': Permission denied
Hi, I had the same problem. DietPi wants to finish the update, but you probably did not ssh as root.
Try running
sudo /DietPi/dietpi/login
It then finishes the update with the required Root priviliges.
So if I reboot and login in using root user it appears to do some more install then takes me into dietpi-software, asking if I want to change the default passwords for root and dietpi and will not let me exit the software install page saying I have to select some software before I can exit. I chose a new software package to install, that completed. Now it seems to be back to normal.
Jep sorry guys, there was an issue were in some cases (and we are still not why why/how this can happen), DietPi-RAMdisk was not syncing the new code to disk on shutdown, since it was running after the /boot mount was already unmounted.
I pushed a fix with v6.20.6 to implement some failsafe steps and revert some change to DietPi-RAMdisk that was responsible for this.
This caused several issues in combination with a install stage change, making the system believe it would have not yet done first run setup and such.
Where ever you hang, please cancel/interrupt any update or dietpi-software execution.
Also assure that there is non still active on a different terminal.
Check htop for active dietpi-* processes and stop them by navigating there with up-down keys and press F9 and return.
Then do the following to update to v6.20.6 and skip any possible unwanted first run update after reboot:
echo 1 > /DietPi/dietpi/.install_stage
echo 1 > /boot/dietpi/.install_stage
dietpi-update 1
DebianPlebian
Do not worry about your installed software titles, data and such. Due to the above issue the system after reboot believes it is a fresh image (first run setup), but it’s not. If you follow the above to apply v6.20.6, everything should be setup as before, no data or installed software lost.
This did not work for me, I get these two lines;
/DietPi/dietpi/login: line 41: /DietPi/dietpi/.dietpi-autostart_index: No such file or directory
/DietPi/dietpi/login: line 160: ((: > 0 : syntax error: operand expected (error token is "> 0 ")
Does not update and when I run it again I get this error;
chmod: cannot access ‘/etc/profile.d/dietpi-*.sh’: No such file or directory
Ran touch /etc/profile.d/dietpi-test.sh and it updated, but then I am back to this after a restart;
[FAILED] DietPi-Software | Root privileges required. Please run the command with “G_SUDO”, or, “sudo -i”.
mv: failed to access ‘/var/tmp/dietpi/logs/dietpi-firstrun-setup.log’: Permission denied
Ran sudo -i, but then NTPD runs forever;
[ INFO ] DietPi-Run_ntpd | NTPD: Waiting for completion of systemd-timesyncd (11/60)
NTP is set to pool.ntp.org and works fine on 6.17
trembler
Did this happen just now, so when attempting to update to v6.20.6?
First of all, login via root account. Ignore possible error messages.
Then run the following:
echo 1 > /DietPi/dietpi/.install_stage
echo 1 > /DietPi/dietpi/.update_stage
> /etc/profile.d/dietpi-unsupported_terminal.sh
dietpi-update 1
If it goes through, everything should be fine, as well after reboot.
If the chmod: cannot access ‘/etc/profile.d/dietpi-*.sh’: No such file or directory prompts. Cancel the update and run:
sync
sleep 5
dietpi-update 1
This assures the new update code has really been synced to disk and reruns the new update code which should not throw the error again.
It seems to have sorted itself out, I just did an override on the NTP, opened a root shell and it went through as a fresh install, rebooted, but everything was there and NTP was fine.
I probably brought some of this on myself for not using the root account.
Will this be sorted in a new release?
I have my second pihole on a raspberry pi, so I dont have snapshots to rely on, so I will hold off on that one.
Thanks for swift reply!
trembler
Okay, this is fine then.
The first run setup, as if it was a fresh image, usually should not occur without root account, since on first boot no individual user exist 
.
It was indeed a nasty issue with v6.20.5 that resulted in wrong applied system state, so the system (more precisely, the DietPi login script) believed it is first run, even it’s not. And then of course, if you login with our non-root account, the root permissions are missing.
But yeah running through the first run setup again results without any harm, system otherwise will be as before.
However this should not occur anymore when updating from v6.19.7 directly to v6.20.6 which is now offered. So it this DID happen for you when attempting to update e.g. today to v6.20.6 directly, this means we didn’t sort it fully out.
Sorry for the late reply, this was on 6.20.6
Thanks, this fixed the same issue I had when upgrading to v6.20.6.