Posts by HiassofT

NFS over UDP is disabled in LE (like in most other distributions as well), so just use TCP instead.

See the kernel config for background info why UDP isn't supported anymore:

linux/Kconfig at master · torvalds/linux

Linux kernel source tree. Contribute to torvalds/linux development by creating an account on GitHub.

github.com

   default y
   help
     Choose Y here to disable the use of NFS over UDP. NFS over UDP
     on modern networks (1Gb+) can lead to data corruption caused by
     fragmentation during high loads.

so long,

Hias

Quote from blueribb

Can I go back to 11 if I need to?

Only with a clean install and restoring the backup.

Best to test it by booting from a USB pendrive in live or run mode.

so long,

Hias

I've added the fix to LE11 and LE12, it'll be in the next nightly builds.

so long,

Hias

Make sure you safely remove the drive in LE / Kodi befoce disconnecting it - just like you would/should do in Windows or any other OS

so long,

Hias

On LE11 and before use "vcdbg log msg" - vclog is only available in LE12.

so long,

Hias

I'm no expert in kodi caching details but quite likely it had cached more audio data than video so audio could continue for a while.

As for the USB drive: might be worth to just ditch the case, there are tons of reports of cheap USB-SATA/NVMe adapters/cases being crap.

so long,

Hias

The log shows that your USB (SSD?) drive frequently stops responding and the kernel needs to reset it (after the 30 second timeout elapsed) so no wonders you have issues playing media from it.

May 07 07:49:49.601266 RPi4 kernel: usb 2-2: reset SuperSpeed USB device number 2 using xhci_hcd
May 07 07:49:49.617935 RPi4 kernel: sd 0:0:0:0: [sda] tag#0 UNKNOWN(0x2003) Result: hostbyte=0x03 driverbyte=DRIVER_OK cmd_age=30s
May 07 07:49:49.618535 RPi4 kernel: sd 0:0:0:0: [sda] tag#0 CDB: opcode=0x28 28 00 03 3b 5d e2 00 01 00 00
May 07 07:49:49.618913 RPi4 kernel: I/O error, dev sda, sector 54222306 op 0x0:(READ) flags 0x80700 phys_seg 6 prio class 2

Insufficient power and broken cables/connectors/... are very often causing such errors, so first try to look into that and connect via a powered USB3 hub.

Lots of USB devices, especially the cheap no-name ones, have broken USB implementations and sometimes it's possible to work around those with usb-storage.quirks - in your case it's not the rather frequent UAS issue though, it's already using the usb-storage driver.

Quick search for the USB IDs (idVendor=13fd, idProduct=3456) didn't show up anything useful, searching for "sage 3639S" did neither so not sure what device you actually have.

so long,

Hias

Using a btrfs filesystem as /storage is not supported, the btrfs kernel driver is built as a module (so not available during initramfs stage) and also the initramfs doesn't contain fsck.btrfs - this is only available via the btrfs-progs addon (so in case anything goes wrong you won't be able to fix the issue on LE).

Stick to ext4 for /storage - you can use btrfs on additional drives though.

so long,

Hias

You need to add "dtoverlay=cma,cma-256" to the end of config.txt, otherwise LE won't boot on devices with 512MB RAM.

so long,

Hias

Do you have a space in your password? It looks like LE will trip over that (and maybe other special characters, too) - so avoid them.

so long,

Hias

Settings->System->Display->Video Calibration.

And, no, 640x480 is not correct - 720x480 is the correct resolution for NTSC. This is the same as on DVDs and includes overscan (like on DVDs).

so long,

Hias

Mario77 most RPi2s are 32bit SOCs (Cortex A7) - only the very latest RPi2 production runs used the same (64bit capable, Cortex-A53) SOC as RPi3.

so long,

Hias

Thanks for checking, you have plenty of space free so we can tick that idea off the list.

so long,

Hias

Is your storage partition full? That would explain why copying the database doesn't work.

ssh in, run "df -h" and post the output

so long,

Hias

This is already supported and if you do a clean installation of LE11 config.txt will contain the instructions how to enable it

###############################################################################
# Use distroconfig-composite.txt instead of distroconfig.txt to enable
# composite video output.
# The composite video mode needs to be configured in cmdline.txt:
# For PAL add: video=Composite-1:720x576@50ie
# For NTSC add: video=Composite-1:720x480@60ie
################################################################################
include distroconfig.txt
#include distroconfig-composite.txt

so long,

Hias

One interesting point in the ntfs-3g docs though is that it mounted the filesystem as read-only when it was dirty or had not too severe errors whereas the ntfs3 kernel driver completely refuses to mount the filesystem in that case.

This could mean users with problems (or problematic behavior / setups) would not have noticed that they had issues in the past whereas they notice the issues now.

With LE10 and before the issues would only show up when trying to write to the drive with LE (eg via SMB), but as lots of those users seem to prefer to write media from their Windows boxes, and use LE only to read the media files, that rarely happened.

so long,

Hias

If the drives worked before, on LE10, and failed at the first attempt with LE11 then one possibility is that the filesystems were already unclean but the old driver didn't care about that - whilst the new driver is more cautious (just a guess though).

Run the filesystem checks/repairs on Windows and keep an eye on it. If it happens again we would need to know what triggered that.

so long,

Hias

Quote from Mario77

Dont take me wrong but as posts about ntfs formated HDD's not mounting are popping up allover the forum..does it really looks like we have to go to the ext4 way or is this some issue that might get fixed one day?

NTFS has always been troublesome in Linux and never was a good choice. It's best to use ext4 and copy stuff via SMB or, if you really really need to connect the drive to other non-linux devices use EXFAT.

No idea if/when anything gets fixed, and so far we don't even know what the users with problem reports did (or what happened on their devices) that caused the NTFS partitions to become dirty so we are completely in the dark here (and my guess is some percentage of the reports are caused by user errors like unplugging the disk or hard pulling the plug while the drive was mounted - in that case any filesystem will get marked dirty).

so long,

Hias