ou’re Try just running findmnt, the type may be ntfs instead of ntfs-3g. The -t switch is saying to filter and only show ‘ntfs-3g’ mounts and it maybe called something else in your environment. The list will be a bit longer but you should be able to find the listing, it’ll be the directory where your drive is mounted.

I then unmounted Samsung and ran sudo mount -t ntfs-3g -o uid=1000,gid=1000,rw /dev/disk/by-label/Samsung /mnt/Samsung

Did Emby work after this? You said you had to remount, what did you use to do that?

If you can get Emby to work, you can then look at findmnt to see how the working directory is mounted (which options/etc) and then you can update your fstab to have those options so that it will mount on startup.

I’m guessing I could also open fstab and change it manually but I’m not quite there yet haha.

Now’s a good time!

Editing the fstab is the ‘correct’ Linux way of doing things, the GUIs can be convenient but they can also prevent you from learning how the system works (so when it later breaks, you have no idea where to start).

fstab is just a text file, there are 6 fields: <file system/device> <directory where the device will be mounted> <filesystem type> <options> <dump> <pass>. The way it tells the fields apart is through white space(a tab or space character).

/dev/sda1  (any amount of spaces or tabs)  /home/external-drive   options,seperated,by,comma    0(old way of marking a drive to be backed up)   2(fsck check order, 1 for your root directory, 2 for everything else, 0 if you don't want disk checking like for a swap device)

If you’re worried about breaking anything you can just make a backup:

sudo cp /etc/fstab /etc/fstab.bak

If you break something, just restore it:

sudo cp /etc/fstab.bak /etc/fstab

That fstab entry is correct (assuming the by-label path is valid), you could change auto to ntfs-3g but it is likely detecting the correct type so that isn’t needed.

If you can put it in the ‘not working’ and also in the ‘working’ state you can compare how it is mounted with

findmnt -t ntfs-3g

That’ll show the options that the mounts are currently using, a quick thing that you can try unmounting and remounting

sudo umount /mnt/Samsung
sudo mount /mnt/Samsung

That should mount it with the options in the fstab file, you can confirm with the findmnt command.

If not, that doesn’t work try umounting and then mounting with the options set explicitly:

sudo mount -t ntfs-3g -o uid=1000,gid=1000,rw /dev/disk/by-label/Samsung /mnt/Samsung

and see if they show in findmnt.

EDIT 2: Spoke too soon. After a reboot they were ro for Emby again. I added rw to the drive options and it worked again. But after another reboot, no go again. Turns out after each reboot unmounting then remounting works. Not sure if there’s a way to make this work. Maybe they mount too soon? I dunno, but for now it’s functional.

You shouldn’t need to remount if you added the options, they should be in the fstab file now so they’ll be applied at every reboot. If it isn’t doing that check the fstab file

cat /etc/fstab

To see what the relevant entry says. The options block is the block of text just before the two numbers at the end of the file.

However, I think I got it set. I added uid=1000,gid=1000 to the end of the options string and Emby now has control.

Yup, that’ll do it.

Now I just gotta look into why that worked haha.

Adding uid=1000,gid=1000 tells the NTFS mount driver “Present every file on this NTFS volume as if it is owned by UID 1000 and GID 1000.” The account running Emby is probably just your user account (guessing) so this did what you were trying to do with chown. Chown won’t work without the ‘permissions’ option and a usermapping file to translate linux IDs to the SID’s that NTFS uses, probably way more configuration than you need.

You should try to get off of NTFS ASAP. It’ll be fine for media storage but some things (pc games, esp) really hate running off of NTFS. I know how unwieldy it is to shift the data around, but it will save you a lot of headaches going forward.

Next, I have ntfs-3g installed and plan to run mount -t ntfs3 /dev/sdxY /mnt though I think I either need to create a drivename folder in /mnt or add mount -t ntfs3 /dev/sdxY /mnt/DRIVENAME the drive name to the end and it’ll create it maybe?

mount -m -t ntfs-3g /dev/sdx /mnt/mountname

Will create /mountname in /mnt and then mount the device on the created directory.

Instead of doing that you could just edit the fstab file and change the fileystem type to ntfs-3g(the package may be just ntfs on CachyOS so ymmv). Although, I suspect this is already done since you’re able to access and read the drive.

You may need to add -o permissions so that standard Linux permissions are applied (TFM: https://linux.die.net/man/8/ntfs-3g), alternatively you can add ‘permissions’ to your current fstab entries, then remount the drives.

ex:

/dev/sda1                             /mnt/windowsdrive           ntfs-3g            permissions,rw     0 0

I can now access my drives, so I began setting up my server. No problem this time! Emby sees the directories and scans them. Finally! However even though I can rw on the drives, Emby sees them as read only. Bummer.

How is Emby installed? In a container or are you running it on bare metal?

When you say it sees them as read only, what indicates this? A log? A GUI element?

So you read a bunch of articles detailing exactly how they plan on using AI and they have specifically said that they’re not using it for vibe coding.

And your conclusion after reading all of this is that they’re going to be using it for vibe coding.

It seems like you’re doing some vibe reading

Rsyslog comes standard on most linux distros, yes. It didn’t get that way because the development team are idiots who don’t know what they’re doing.

So, on one hand we have the developers of a piece of software used by a majority of Linux users saying that, after 2 years of evaluation and testing, that these tools are useful and effective and on the other hand we have random social media people who’s only knowledge on the topic is that they know AI is bad and therefore this cannot be anything but bad.

I, like the other commenters here, am not privy to the 2 years of evaluation and experimentation.

However, I won’t let that stop me from condemning this in no uncertain terms because AI bad and that’s more correct than any amount of “observational data” or “experimental evidence”.

We may not be users of the software, contributors to the project or were even aware of what rsyslog is prior to seeing this announcement but as people who are chronically outraged we have valid concerns about this development which will ruin the software that we only recently discovered existed.

Amcrest cameras have a RTSP stream open by default. If you know the camera’s IP address you can just open the stream in a media player, like VLC, and set it to save the stream to disk.

Also, you could probably save the stream without viewing using ffmpeg.

Or Frigate

Run pw-top while you’re having the issue and post the output.

It’s probably a bad default setting that’s setting the buffer/quantum incorrectly.

I just prepend everything in the home directory with a dot every 6 months or so, no problems so far

Disabling Secure Boot fixed it? That is very interesting.

Glad you got it solved! :)

it seems to happen primarily when I try to move the mouse regardless of the situation (if I’m in dialogue or exploring, but it’s the worst when fighting), and during cutscenes where they always play slowly and voices go out of sync

There was an issue with the Steam overlay that happened when they added the screen recording feature. After about 25 minutes any input will cause a delay in rendering the current frame.

You can disable the overlay or add LD_PRELOAD=“” to your launch options, if this is the problem

Spectacle

Flameshot is pretty good too

About [Secure Boot], here’s the related settings I found in the BIOS settings, but I forgot one more option and it’s like a boot list with two entries about openSUSE, one has “secureboot” on the name and the other doesn’t, if it helps I’ll add a picture of it to that album.

Try booting with the non-secure boot option then

cat /sys/firmware/ipl/secure

1 means secure boot is enabled, 0 means disabled.

If it’s 0 then you could use any distro, regardless of secure boot support.

There’s a config file somewhere to make it permanent, if you can’t find it just let me know and I’ll dig through my notes (if you find out, post it here for future googlers)

It’s annoying, but it isn’t feeding my data into the AIs.

default.clock.quantum

That probably didn’t do anything. I think it sets the quantum if it isn’t otherwise set.

Check pw-top while you’re playing from a source where you experience crackling. You can see the quantum value, if it’s low (usually 1) then that source is using the minimum quantum.

You can change it, temporarily (until reboot/pipewire service restart) with

pw-metadata -n settings 0 clock.min-quantum <value>

Try 256 to start with increase if you still get crackling.

Here’s the documentation on pipewire buffering: https://gitlab.freedesktop.org/pipewire/pipewire/-/wikis/FAQ#pipewire-buffering-explained

I’ll leave finding the config file to make this permanent as an exercise for the reader.

None of the problems that they’re describing are caused by using an Nvidia card.

The default proton is bad, using GE-Proton will often improve things.

CS2 getting 40fps and no sound. A quick trip over to https://www.protondb.com/app/730 will show that this is a problem with CS2 as it happens on every graphics card. Setting SDL_AUDIO_DRIVER=pipewire resolves the issue for a lot of people.

Helldivers 2 low FPS: Protondb has people with both graphics cards reporting the low fps issues. Setting a framerate cap will reduce FPS variance, using GE-Proton10 fixes HDR issues. In my experience disabling vsync improved performance significantly, this is true of a lot of games if you’re using adaptive sync.

I see this all the time in the Linux communities. If someone is having an issue with a game and they use an Nvidia card then community responses are often to do the blame Nvidia meme rather than troubleshoot the issue.

If it’s the same bug as what I’m seeing, I just wrote a script that changes the display settings and sets them back then bound it to a hotkey.

Any monitor that stays blank after waking up is fixed if you adjust the display settings (tricky to do with blank monitors, hence the script).

So fake.

That version of Luxo lamp wasn’t invented until 1937

smh my head