After updating from 7.2.1 to 7.3.0 I get this message concerning my cache pool:

Id Path           Write errors Read errors Flush errors Corruption errors Generation errors
-- -------------- ------------ ----------- ------------ ----------------- -----------------
 3 /dev/nvme0n1p1            0           0            0               255                 0
 4 /dev/nvme1n1p1            0           0            0               255                 0

I have been trying to find out what the specific errors are, but I can't find them anywhere. I would like to know which specific files or data is affected.

Both devices successfully complete an extended SMART self test. But when I do a scrub, it aborts with these errors:

May 18 18:01:52 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 18 18:01:52 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 18 18:01:52 BigNAS kernel: BTRFS info (device nvme0n1p1): scrub: not finished on devid 4 with status: -5
May 18 18:02:03 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 18 18:02:03 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 18 18:02:03 BigNAS kernel: BTRFS info (device nvme0n1p1): scrub: not finished on devid 3 with status: -5

What can I do about this? And how do I get more detailed information?

Edit: I've added the diagnostics.

NAS-diagnostics-20260521-0909.zip

Edited May 21May 21 by BreakfastPurrito
Added diagnostics

7.3 has a new pool monitoring feature, and btrfs is detecting data corruption on both devices is most often bad RAM. Suggest as a first step to run memtest.

I'm seeing similar on my cache pool devices. After upgrading there were thousands of corruption errors.

I have two pools, one for apps cache and one for cache

I reset both counters, and the apps cache stopped increasing, however the cache continued to increase slowly.

I turned off spindown on both drives and it stopped increasing for a while, however started to slowly climb again over the following 24h.

Memtest passes without error.

The drive passes short and extended testing without error.

Identification: Samsung_SSD_990_PRO_2TB_S7DNNU0Y714679D (nvme1n1)

pool device stats:

Id Path           Write errors Read errors Flush errors Corruption errors Generation errors
-- -------------- ------------ ----------- ------------ ----------------- -----------------
 1 /dev/nvme1n1p1            0           0            0                16                 0

Have not tried a scrub or format yet

Any advice would be greatly appreciated!

Cheers

4 minutes ago, SirBitey said:

seeing similar

Start your own thread with your Diagnostics.

On 5/19/2026 at 6:22 AM, JorgeB said:

7.3 has a new pool monitoring feature, and btrfs is detecting data corruption on both devices is most often bad RAM. Suggest as a first step to run memtest.

I ran Memtest86 and it passed with no errors.

Scrub the pool and post the output from the GUI

On 5/21/2026 at 10:11 AM, JorgeB said:

Scrub the pool and post the output from the GUI

This is the output from the GUI:

Scrub started:    Sat May 23 22:14:24 2026
Status:           aborted
Duration:         0:06:14
Total to scrub:   2.71TiB
Rate:             4.08GiB/s
Error summary:    no errors found

Followed in the logs by:

May 23 22:20:32 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 23 22:20:32 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 23 22:20:32 BigNAS kernel: BTRFS info (device nvme0n1p1): scrub: not finished on devid 4 with status: -5
May 23 22:20:38 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 23 22:20:38 BigNAS kernel: BTRFS critical (device nvme0n1p1): unable to find chunk map for logical 6180170305658470400 length 16384
May 23 22:20:38 BigNAS kernel: BTRFS info (device nvme0n1p1): scrub: not finished on devid 3 with status: -5

Since the scrub is aborting, the filesystem has some issues. I recommend backing it up, reformatting, restoring the data, then monitoring the pool for any new errors.

What filesystem should I be reformatting? Just the cache or the array as well?

Just that pool