Forum Loading...

there used to be a way to bypass the release file requirement. You could drop into a shell and edit the /etc/apt/sources.list and point the installer to a debian mirror instead of the usb or cdrom. That might still work for you but you're network/ethernet has to be online.

Hello, Code: sudo service nut-server statusCode: Select all× nut-server.service - Network UPS Tools - power devices information server Loaded: loaded (/usr/lib/systemd/system/nut-server.service; enabled; preset: enabled) Active: failed (Result: exit-code) since Thu 2026-10-01 08:12:01 UTC; 11min ago Duration: 6ms Invocation: fd841a0a19674524b103df6816ad2658 Process: 1669 ExecStartPre=/usr/bin/systemd-tmpfiles --create /usr/lib/tmpfiles.d/nut-common-tmpfiles.conf (code=exited, status=0/SUCCESS) Process: 1672 ExecStart=/usr/sbin/upsd -F (code=exited, status=1/FAILURE) Main PID: 1672 (code=exited, status=1/FAILURE)Oct 01 08:12:01 [hostname] systemd[1]: nut-server.service: Scheduled restart job, restart counter is at 5.Oct 01 08:12:01 [hostname] systemd[1]: nut-server.service: Start request repeated too quickly.Oct 01 08:12:01 [hostname] systemd[1]: nut-server.service: Failed with result 'exit-code'.Oct 01 08:12:01 [hostname] systemd[1]: Failed to start nut-server.service - Network UPS Tools - power devices information server. You should verify why the /usr/sbin/upsd -F command fails in nut-server.service. You could temporary disable the service and start it manually in foreground (with -F option, see user manual ) to read error messages, if any. You can also increase the verbosity with the -D option.

Hello, I just upgraded a server running the Proxmox hypervision version 8 (PVE 8.4.19, Debian 12.14, kernel 6.8.12-30-pve) to version 9 (PVE 9.2.21, Debian 13.7, kernel 7.0.14-20-pve) and the NUT UPS server/monitor stopped working. It doesn't connect to the UPS anymore and doesn't even detect it with nut-scanner. Everything was working fine just before the upgrade. There were no physical changes or any change on the UPS, just the OS upgrade/reboot. I've checked the nut group assignments and udev rules and everything seems to be right. The UPS is connected through a Serial-USB converter, which is shown on dmesg and lsusb. I've seen some related threads about this same issue, but nothing worked for me so far. * https://forum.proxmox.com/threads/nut-m ... e9.173375/ * https://forum.proxmox.com/threads/nut-on-9-1.180749/ * Does anyone else have a clue on how to fix this? PS: I've already tried rebooting, physically disconnecting/reconnecting the USB adapter and serial cable, completely powering off/on the server, reinstalling the nut packages, reconfiguring using the default configuration, restoring the old configuration files... I've also followed the Debian Wiki ( https://wiki.debian.org/nut ) trying to reinstall and configure it from scratch but the problem persists. Here is my configuration (nothing has changed, I kept the old configuration files while upgrading): /etc/nut/nut.conf Code: MODE=standalone /etc/nut/ups.conf Code: maxretry = 3[UPS-NAME] driver = nutdrv_qx port =

Yes where are the 256/512 checksums? I don't want pgp or whatever.

These are screenshots from a dot-run installer: you persistently keep doing exactly what I and the wiki told you, is known NOT to work. I honestly can't possibly help you given this.

Hello @sunrat , I know it's possible to use AI models for this kind of thing but I don't know of any freely available software. Maybe you can scratch something up from this - https://github.com/pyannote/pyannote-audio . It uses PyTorch which is the basis for a stem splitting software Thank you so much for your advice. I will give it a try if it is compatible with my hardware resources. I use regularly, Ultimate Vocal Remover 5, which took some head scratching to set up on Linux but it works very well. Only uses 14GB of my drive too. PyTorch is huge. Even if it does not apply to my use case, this is an interesting piece of software to explore. Thanks. Aki.

What you are suggesting is commenting out all devices in the file. When updating a device, then identify what device you have. Uncomment that device. Do the updating. After updating then comment out the deice which got updated. Correct? Yes, I think so. Interesting project.

btw do i need install kernel modules and which one? Driver installation ask for them I now recall that dot-run installers ask which flavor of modules should be installed: why would you use a dot-run installer?? The wiki clearly states that Debian-13 is unsupported by these drivers... If that's the case, then the only thing left is to reinstall the system: dot-run installers are known to create random mess that is impossible to clean up. https://www.pasteboard.co/0oYV9zow0dsR.jpg https://www.pasteboard.co/vJ-oAhEYQpw1.jpg https://www.pasteboard.co/fDJE0b8UqxmO.jpg

Have you tried disabling any power saving features? Are you using mac address randomisation? if you are try disabling that?

Which nvidia driver should i use btw thanks

btw do i need install kernel modules and which one? Driver installation ask for them I now recall that dot-run installers ask which flavor of modules should be installed: why would you use a dot-run installer?? The wiki clearly states that Debian-13 is unsupported by these drivers... If that's the case, then the only thing left is to reinstall the system: dot-run installers are known to create random mess that is impossible to clean up.

comment out those lines when done. What you are suggesting is commenting out all devices in the file. When updating a device, then identify what device you have. Uncomment that device. Do the updating. After updating then comment out the deice which got updated. Correct? If a file is fully commented out it cannot present any security implications?