• 0 Posts
  • 161 Comments
Joined 3 years ago
cake
Cake day: June 13th, 2023

help-circle

  • You already tested with a standard Windows 10/11 install ISO? Put that on a bootable USB along with your exe but instead of installing you go into the recovery options and should see a way to get to the cmd prompt where you can test run that .exe. It might have the same results as Windows PE but it’s worth a try and downloading the Windows ISO is free anyway.

    Worst case if you have a spare HDD/SSD you can put that into your system, temporarily install Windows 10/11 onto it (I don’t think you even need to worry about activation), run your .exe, then shutdown and swap your drives back to your normal setup and be done with it.

    But yeah I get what you’re saying, ideally there’s a better way but I’m not too sure what else to suggest within Linux itself.


  • A little research got me to a “systemrescue” iso and that one worked fine. The live environment fired up and I was able to save all my data by mounting the partition via terminal into /mnt/mountfolder/.

    Nice. I always keep an ISO of systemrescue on a bootable USB for these occasions, it’s gotten me out of jams in both Windows and Linux situations.

    Not sure what to make of your issue with Ubuntu stopping from working, including the live boot, only for it to work again for you in the end. My hunch is wonky hardware but can’t really say.



  • Will be curious what solution you come up with. I tried to do this with my current Debian installation but never quite got it to boot off the RAID-1 array. In the end just went without RAID on the drive with Debian installed, maybe will re-attempt next time I do an OS install. I do have mdadm RAID-1 working normally for my data drives, just not the boot drive… you technically could just do that if you want to have your RAID-1 data drives separate from your OS boot drive.

    Can’t comment on the Calameres installer but the regular Debian installer does detect RAID configurations from other mdadm setups. So you could either create your RAID-1 configuration in the shell during the Debian install or even create it in another Linux boot shell before jumping into the Debian installer. e.g. booting any live Linux with mdadm in it, configure the RAID-1 there, then boot into the Debian installer - the Debian installer will know there’s a prior RAID-1 on those disks and allow you to proceed with installing on the array if you wanted.

    What tripped me up afterwards was trying to get it to boot off that RAID-1 afterwards, that part is not so straightforward. The link in the other comment does go into that so maybe it’ll helpful.



  • Some ideas for the future

    Xrdp fail… plain and simple…

    Xrdp usually works fine, you should try to find any specific error messages or logs. Xrdp also runs a service so you could also see if the service itself is running or what it’s status is (systemctl status xrdp).

    For me Xrdp did fail when I initially tried to run it. I don’t remember the exact error being produced but there was something wrong with the port number xrdp wanted to use… in the end I had to stop the service, edit /etc/xrdp/xrdp.ini and set the port to a specific port number without using vsock. xrdp by default was set to use vsock ports which wasn’t working for for me.


  • Set up SSH on it and make sure you’re able to connect into it while it works normally, that way when the issue occurs you can do a quick test to see if the system itself is still up and running.

    I’m not on Linux Mint / Cinnamon but I’ve occasionally seen GNOME sort of hang/freeze so the screen display becomes non-responsive. After a couple of times of that happening I ended up setting up SSH on the system and configured a SSH client on my phone so I can do a SSH connection into the desktop and force-logout my user (which apparently fixes the issue and brings my main desktop back to a normal login screen). I haven’t quite figured out if it’s Gnome issues or something to do with my Nvidia GPU… though with Linux if Nvidia is involved then it’s usually Nvidia, ugh.

    Also if you’re physically at the computer when it happens try unplugging/replugging in the monitor cable, maybe there’s something wonky going on there or with the display connection.

    Just some ideas to help you along :)





  • I’ve never needed to do this but have you looked into creating a Reverse SSH Tunnel? Maybe that can accomplish what you want https://www.howtogeek.com/428413/what-is-reverse-ssh-tunneling-and-how-to-use-it/

    Similar to what you are trying to do - A while back when I needed to remote connect to a firewalled Windows computer I set up a reverse VNC connection on that Windows computer that would get initiated whenever I sync’d a file over to the Windows system via Dropbox/Syncthing (those work without port forwarding). Reverse VNC, like Reverse SSH or other connections basically try to initiate the connection out of the firewalled system, it’s an interesting workaround when you have no incoming port forwards. Not sure if I’d recommend that type of set up but it is more secure than sending emails so there’s that.

    If you’re able to set up a mesh VPN that might work better but you do have other options if you need them.


  • Feels like this should have already been resolved with standard Linux permissions, any user on your system should be able to read/traverse that folder and files with the r-x permission on world.

    From your earlier comment one of the parent directories is listing with “+” permission so it sounds like you are using extra permissions outside of the standard Linux, probably ACL permissions. I haven’t worked with ACL permissions on Linux enough to offer much advice but what you may want to do is run something like getfacl d64ea996-98aa-4eef-8a93-0a2ba2246f0f as well as run it on one of your media files, that might give you some clues on what the ACL permissions look like.

    Also - I’m assuming you already did this - but if you didn’t already make sure your jellyfin service is actually launching with the expected user and user group. Based on your setup you may want to be sure the “jellyfin” Linux user is added to the “video” user group, and then make sure that’s what is being referenced in jellyfin.service.conf (by default is usually in /etc/systemd/system/jellyfin.service.d). If you edit that file be sure to stop the jellyfin service first and then restart it after editing e.g. “systemctl stop jellyfin” and “systemctl start jellyfin”.

    Only other thing I can think of is that maybe there’s something wrong with the way your system is mounting the drive and folder. Since you seem to be using the autogenerated UUID for the mounted folder name (d64ea996-98aa-4eef-8a93-0a2ba2246f0f) maybe it’s better to edit fstab yourself and set up your mount as something more permanent with a normal name, and also double-check the permissions everything is mounting as.



  • Debian by default uses the Nouveau open source driver for Nvidia GPUs and that driver does support Pascal. Debian installations will continue to work just fine even without Nvidia’s development support.

    https://en.wikipedia.org/wiki/Nouveau_(software)

    I don’t know if that’s something that can be done on Arch but in theory you can test the fallback Intel driver vs Nouveau and see which fallback you prefer.

    Nouveau works well for day-to-day use and works with Wayland. I’m not a hardcore gamer but have played low-mid range Steam games without issue. I suspect it may not do well playing high end AAA games but then again if you’re rocking a Pascal era GPU it’s unlikely you’ve been playing those type of games anyway.

    EDIT: Just to add, pretty sure the built in Intel iGPU on your laptop is more power efficient vs the Nvidia GPU so it may be worthwhile to disable the Nvidia GPU entirely rather than worrying about software drivers.


  • Are the files within the folders the media folder also with permissions 775 (rwxrwxr-x)? The actual .mkv .mp4 or whatever files themselves should be the same permissions as the parent folder(s). Could be worth doing a chmod -R 0775 d64ea996-98aa-4eef-8a93-0a2ba2246f0f if you’re not sure.

    You probably already tried this but just in case: In the Jellyfin admin pages, in Libraries, try removing that media library, then re-add the media library and when adding Folders paste in your entire media folder path/name: /media/velummortis/d64ea996-98aa-4eef-8a93-0a2ba2246f0f

    The reason I mention that is that I did a quick test here with a bad permissioned folder, added the media folder, Jellyfin couldn’t see anything in it obviously. But even after I fixed the folder permissions Jellyfin still wasn’t seeing the media - I had to remove the library entirely and then re-add it to force Jellyfin to attempt to scan the folder with new 775 world permissions.


  • Also using Debian + Jellyfin normal install with basic permissions (in the future maybe I’ll try this with ACL permissions).

    Starting out I tinkered with creating a specific user group for jellyfin but it never quite worked right… in the end I reset the jellyfin user/service back to defaults and set permissions on my media folders to 0775 so that the jellyfin service can read/execute (no write) as the world permission bit. Also made sure to set permissions recursively (chmod -R). For me it’s fine though keep in mind setting 5 for world does mean all users with access to your system can read and traverse through those media folders.


  • I’ve been using Debian with the default GNOME on an old laptop and main desktop and have been very happy with it. Coming from Windows I love that it’s way simpler and I don’t need to set a million options.

    But remember the thing with Linux is you’re not locked into anything - So try GNOME or XFCE for a few weeks, then if you still want something else install and switch over to another desktop environment. You could even install all these desktop environments during the Debian install itself and just keep switching every time you log in.



  • Just to be sure, did you already test that the port is actually open and forwarded? e.g. with your torrent client running browse to a port test website like https://canyouseeme.org/ , https://www.yougetsignal.com/tools/open-ports/ , etc. put in your torrent client’s incoming port and check if the website can “see” your open port at your torrent client.

    And the ISP (or router) itself isn’t doing anything weird to block torrents, right? In your torrent client if you click any working public torrent, click on the Trackers tab, you should see DHT as working along with whatever open trackers are on the public torrent. In other words you won’t see anything like “waiting” something (I forget the exact message you’ll see when DHT is being blocked but it’ll definitely not be working).

    EDIT: Also if it’s a new ISP with new router it might have firewall rules set up that are slowing things down, something to check.