I am expiriencing an issue where my entire cache pool is somehow running as read only, and via proxy, all VMs and dockers seem to be suffering too.

 

Can someone please help me troubleshoot?

 

 

EDIT: A reboot seems to have fixed it temporarily, but it is there anything in the logs indicating what caused this?

 

Thanks

 

Sent from my SM-N960W using Tapatalk

 

 

 

tower-diagnostics-20191130-1729.zip

Edited December 1, 20196 yr by bobo89
Added diagnostics

for some reason I can't download the zip to look.... is it an Intel SSD?  once Intels reach their max data written, they go read only.  Must replace it to fix if that's the case.

Your syslog shows;

Nov 25 17:30:22 Tower kernel: BTRFS warning (device loop2): block group 8083472384 has wrong amount of free space
Nov 25 17:30:22 Tower kernel: BTRFS warning (device loop2): failed to load free space cache for block group 8083472384, rebuilding it now

...

Nov 29 22:26:06 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 2696072 off 186834944 csum 0x80fe9b58 expected csum 0x2d132b24 mirror 1
Nov 29 22:26:06 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 2696072 off 186834944 csum 0x80fe9b58 expected csum 0x2d132b24 mirror 2

.. repeating

 

 

 

I recently had my cache go read-only which has 2x PCIe NVMe drives.

The cause was a stalled balance operation, not the drive itself.  After a reboot, the balance completed quickly.

 

BTRFS is buggy and is still getting updates, so i'm guessing this is a general problem when using PCIe devices.

 

Check the health of your nvme0n1 disk and think about doing cache backups.

 

I use the "CA Backup / Restore Appdata" to backup my cache only appdata, usb, and system directories to a parity protected share.

Next time please upload diags directly to the forum.

 

There are checksum errors with the cache pool:

 

Nov 26 02:44:13 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 1870791 off 45502005248 csum 0x4253eb6e expected csum 0xe5b40d03 mirror 1
Nov 26 02:44:13 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 1870791 off 45502005248 csum 0x4253eb6e expected csum 0xe5b40d03 mirror 2

 

These are usually the result of bad RAM, you should run memtest.

Next time please upload diags directly to the forum.
 
There are checksum errors with the cache pool:
 

Nov 26 02:44:13 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 1870791 off 45502005248 csum 0x4253eb6e expected csum 0xe5b40d03 mirror 1Nov 26 02:44:13 Tower kernel: BTRFS warning (device nvme0n1p1): csum failed root 5 ino 1870791 off 45502005248 csum 0x4253eb6e expected csum 0xe5b40d03 mirror 2

 
These are usually the result of bad RAM, you should run memtest.

Thanks this was the issue. Memory was overclocked too high,bringing it down to the xmp profile seemed to have worked! Thanks

Sent from my SM-N960W using Tapatalk