So I woke up early this morning, had to go pee, checked my phone and UnRAID had notified me of corruption on my cache pool again. 

It's been a long while since I last had an issue with BTFRS - the proposed problem back then was my memory but I was never able to find any faults with it. Of course it has to happen the one time I'm away on vacation.. but that's just a law of nature at this point

I built an entirely new server and everything seemed to be going OK, until today. 

I set up the userscript to monitor as recommended by JorgeB..

Output:

root@FortyTwo:~# btrfs dev stats /mnt/cache
[/dev/sdb1].write_io_errs    0
[/dev/sdb1].read_io_errs     0
[/dev/sdb1].flush_io_errs    0
[/dev/sdb1].corruption_errs  6
[/dev/sdb1].generation_errs  0
[/dev/sdc1].write_io_errs    0
[/dev/sdc1].read_io_errs     0
[/dev/sdc1].flush_io_errs    0
[/dev/sdc1].corruption_errs  8
[/dev/sdc1].generation_errs  0

 

Scrub didn't see anything:

 

Which just seems.. odd. 

So as per Jorge's recommendations I used btrfs dev stats -z /mnt/cache and am now rebooting with my fingers crossed and my cheeks clenched. 

 

Please advise!

 

 

fortytwo-diagnostics-20230301-0534-anon.zip

There was some crashing going on before btrfs detected the issues, e.g.:

 

Feb 28 08:19:14 FortyTwo kernel: Cannot map memory with base addr 0x14c878600000 and size of 0x200 pages
Feb 28 08:28:39 FortyTwo kernel: BUG: Bad page map in process Plex Media Scan  pte:ffff8881e1deb1e0 pmd:48ffcc067
Feb 28 08:28:39 FortyTwo kernel: addr:000015204cc93000 vm_flags:00100073 anon_vma:ffff8881b4476680 mapping:0000000000000000 index:15204cc93
Feb 28 08:28:39 FortyTwo kernel: file:(null) fault:0x0 mmap:0x0 read_folio:0x0
Feb 28 08:28:39 FortyTwo kernel: CPU: 0 PID: 10693 Comm: Plex Media Scan Tainted: P        W  O      5.19.17-Unraid #2
Feb 28 08:28:39 FortyTwo kernel: Hardware name: Supermicro Super Server/X11SSL-F, BIOS 2.6 06/12/2021
Feb 28 08:28:39 FortyTwo kernel: Call Trace:
Feb 28 08:28:39 FortyTwo kernel: <TASK>
Feb 28 08:28:39 FortyTwo kernel: dump_stack_lvl+0x44/0x5c
Feb 28 08:28:39 FortyTwo kernel: print_bad_pte+0x1cb/0x1e5
Feb 28 08:28:39 FortyTwo kernel: vm_normal_page+0x84/0x9e
Feb 28 08:28:39 FortyTwo kernel: unmap_page_range+0x376/0x66e
Feb 28 08:28:39 FortyTwo kernel: ? __down_write_common+0x31/0x4e9
Feb 28 08:28:39 FortyTwo kernel: unmap_vmas+0x87/0xbb
Feb 28 08:28:39 FortyTwo kernel: unmap_region+0xab/0x101
Feb 28 08:28:39 FortyTwo kernel: __do_munmap+0x275/0x2e2
Feb 28 08:28:39 FortyTwo kernel: __vm_munmap+0x69/0xb7
Feb 28 08:28:39 FortyTwo kernel: __x64_sys_munmap+0x17/0x1e
Feb 28 08:28:39 FortyTwo kernel: do_syscall_64+0x68/0x81
Feb 28 08:28:39 FortyTwo kernel: entry_SYSCALL_64_after_hwframe+0x63/0xcd
Feb 28 08:28:39 FortyTwo kernel: RIP: 0033:0x15205a192895
Feb 28 08:28:39 FortyTwo kernel: Code: 00 00 00 0f 05 9b 48 89 c7 e9 27 55 fe ff 41 56 53 50 49 89 f6 48 89 fb e8 88 3d 02 00 b8 0b 00 00 00 48 89 df 4c 89 f6 0f 05 <9b> 48 89 c7 48 83 c4 08 5b 41 5e e9 fb 54 fe ff 31 c0 83 fa 04 74
Feb 28 08:28:39 FortyTwo kernel: RSP: 002b:00007ffe7a5e0af0 EFLAGS: 00000206 ORIG_RAX: 000000000000000b
Feb 28 08:28:39 FortyTwo kernel: RAX: ffffffffffffffda RBX: 000015204c614000 RCX: 000015205a192895
Feb 28 08:28:39 FortyTwo kernel: RDX: 0000000000000000 RSI: 0000000000901000 RDI: 000015204c614000
Feb 28 08:28:39 FortyTwo kernel: RBP: 0000000000000000 R08: 000000000000003f R09: 000000000000001d
Feb 28 08:28:39 FortyTwo kernel: R10: 0000000000000001 R11: 0000000000000206 R12: 000015204cf14ffc
Feb 28 08:28:39 FortyTwo kernel: R13: 0000000000000001 R14: 0000000000901000 R15: 0000000000901000
Feb 28 08:28:39 FortyTwo kernel: </TASK>
Feb 28 08:28:39 FortyTwo kernel: BUG: Bad rss-counter state mm:0000000024d10fb9 type:MM_ANONPAGES val:1

 

But not sure what's causing them, see if new errors are found soon.