readonly fs bug

❯ sudo journalctl -xb | tail -1000 | ai are there any issues that would explain why i get the 'read only filesystem' everythere now Error: error returned from database: (code: 8) attempt to write a readonly database

Caused by: (code: 8) attempt to write a readonly database

Location: crates/atuin/src/command/client/history.rs:321:9

tl;dr: Your root filesystem has been remounted read-only due to a disk (BTRFS) error.

Let's break down your logs and why everything is read-only:

1. Symptoms

You have dozens of errors like EROFS: read-only file system, open '/home/jumski/.config/discord/logs/renderer_js.log'
Every desktop app (Firefox, Slack, Discord, Wireplumber, Klipper and more) is unable to write their normal files in your home directory.
You tried sudo mount -o remount,rw / and it made no difference.

The first mention of real trouble is around:

cze 11 16:43:03 pc kernel: page: refcount:4 mapcount:0 mapping:000000001216a813 ...
...
cze 11 16:43:06 pc discord[7883]: ... EROFS: read-only file system ...

Everything goes read-only after that.

2. The Root Cause: Filesystem/Hardware Error

When Linux (actually, the kernel) detects a serious disk or filesystem error, especially on btrfs, ext4, or similar filesystems, it will remount the affected partition as read-only to protect data/stop further corruption.

The tell-tale clues

The log line at 16:43:03 is from the kernel and is not a user application. (The kernel itself is reporting a bad memory/disk page.)
All write failures start right after this event.
EROFS ("Error Read-Only File System") is being returned by your kernel to all applications trying to write.
Attempts to remount as rw without fixing the underlying problem won't work (mount -o remount,rw / after kernel error does nothing until reboot + fsck).

Filesystem in your case

The log snippet shows aops:btree_aops [btrfs] ino:1 -- so your root is running on BTRFS.
Btrfs is more likely to remount root as read-only on corruption than, say, ext4.

3. How this happens

Bad sectors, power loss, dying SSD/HDD, or BTRFS bugs can cause this.
When the kernel can't successfully write/read parts of its filesystems, or encounters a critical error, it protects itself by remounting all or part of the FS as ro.
Especially common with BTRFS on older kernels or on physically failing disks.

4. What should you do?

a) Check Disk Health

First, check if your disk is failing!

For SSD/HDD (substitute sda for your actual disk):

sudo smartctl -a /dev/sda

Look for Reallocated_Sector_Ct, Current_Pending_Sector, Uncorrectable_Error_Cnt, etc.

b) Look for more kernel errors

Check for further BTRFS or I/O (Input/Output) errors:

sudo dmesg | egrep -i 'error|fail|btrfs|io'

c) Reboot and Run Filesystem Check

For BTRFS, you need to check the filesystem from a rescue system, or at least a root shell after reboot.

VERY IMPORTANT: Do NOT run btrfs check --repair on a mounted read-write (or even read-only) root. Best is to boot from a live USB system, then:

sudo btrfs check /dev/sdaX
# If it returns errors, then only if you understand the risks:
# sudo btrfs check --repair /dev/sdaX

Be sure /dev/sdaX is the partition your root BTRFS is on.

d) Backup anything important, immediately!

If the disk is failing, copying files (even in read-only mode!) is your best chance at saving irreplaceable data.

5. Summary / Why This Happens

You encountered a disk or filesystem error.
The kernel, to prevent data loss or further damage, remounted your root (btrfs) filesystem as read-only.
This instantly breaks anything that needs to write to disk, including application logs, browser caches, settings, etc.
Situation will only fix after you (1) fix the underlying issue (disk/FS) and (2) reboot the system.

What next? (step-by-step)

Immediately back up anything valuable while you still can!
Check your disk health (smartctl etc.)
Review dmesg/journalctl for hardware/BTRFS errors.
Reboot into a recovery/live USB (if needed) and run btrfs check.
Replace the disk if hardware is failing. (Often the root cause.)
Restore from backups if necessary.
Update your kernel/BTRFS tools. Some BTRFS issues are fixed in recent updates.

References

If you want specific help:

Run:

sudo smartctl -a /dev/sda
sudo dmesg | tail -100 | egrep -i 'btrfs|error|fail'
sudo btrfs check /dev/sdaX

And paste the relevant output (replace device names with your actual disk/partition).

In summary: Your root filesystem (BTRFS on Manjaro) encountered a serious error/hardware failure and was automatically remounted read-only for your safety. Investigate and fix the hardware/filesystem; only then will normal operations resume. Back up your stuff right now!

jumski/readonly-fs.md