White Blank Screen After Pi-hole Freeze and Reboot Loop

[evelyn-freya]

Hi guys,

Blank "White screen" background used to confirm that my monitor was not the issue after my Pi-hole froze. The screen displayed perfectly, so I’m certain the problem lies with the system or disk.

I tried using the sysreq sequence to safely shut down my Pi-hole but could not connect via PiVPN (also installed). When I connected to a monitor, the screen was frozen.

White screen background helped me pinpoint the issue further. Upon restarting, Raspbian boots to a certain point, performs a disk check, and then reboots. This cycle keeps repeating (even after leaving it overnight).

Is there a way to skip the disk check or fix this within Raspbian? Could I disconnect the SSD and validate it on my Debian PC to resolve this? Or is a reinstall my only option?

Thanks for your time and help!

[deHakkelaar]

When your Pi freezes for no apparent reason:

500 Internal Server Error

Does below one show any Under-voltage messages? sudo journalctl --full --no-hostname --pager-end --grep voltage If so, have a look below: If you see warnings, switch to a higher quality power supply and cable. Low quality power supplies can corrupt storage or cause unpredictable behaviour within the Raspberry Pi. https://www.raspberrypi.com/documentation/computers/raspberry-pi.html#power-supply-warnings

@evelyn-freya

Is there a way to skip the disk check or fix this within Raspbian?

If you connect a screen, you should see the disk check being performed on screen during boot.
Or at least a mention of it being performed.
If you dont see any boot output on screen at all, the boot sequence didnt get that far yet to do the disk check.

@evelyn-freya

Could I disconnect the SSD and validate it on my Debian PC to resolve this?

Yes you can.
You can list detected block devices (disks or partitions) plus their mountpoints with below:

$ lsblk -p
NAME             MAJ:MIN RM  SIZE RO TYPE MOUNTPOINT
/dev/mmcblk0     179:0    0 14.8G  0 disk
├─/dev/mmcblk0p1 179:1    0  256M  0 part /boot
└─/dev/mmcblk0p2 179:2    0 14.6G  0 part /

Search on the net how to use the fsck command:

man fsck

EDIT: Dont forget to unmount first (umount) if it shows mounted with the lsblk command!

[deHakkelaar]

When the partitions are still mounted on the Debian PC, you can even inspect the systemd journal on the SSD eg:

sudo journalctl --full --no-hostname --pager-end --directory /<ROOT_MOUNTPOINT_FOLDER>/var/log/journal/

$ man journalctl
[..]
       -D DIR, --directory=DIR
           Takes a directory path as argument. If specified,
           journalctl will operate on the specified journal
           directory DIR instead of the default runtime and system
           journal paths.

But this is a bit out of scope here