So, I have this:
├─nvme0n1p4 259:4 0 30G part bcache 8fd98dbd-4e87-4faa-8cec-d5f1e3910daa ae581b32-8c62-41a9-bfcf-ca8cec66e46a
│ └─bcache0 252:0 0 1.1T disk crypto_LUKS d6c98ca4-23b4-4aa3-819b-b384d4ac1b6a
│ └─pool2 251:3 0 1.1T crypt btrfs /mnt/btrfs_p 5ed9d54d-2f12-4735-85c2-e0445a316088
├─sda7 8:7 0 1.1T part bcache 6d88cea7-6cfc-4d30-8cf5-8c8b216e019d 753dca8b-07
│ └─bcache0 252:0 0 1.1T disk crypto_LUKS d6c98ca4-23b4-4aa3-819b-b384d4ac1b6a
│ └─pool2 251:3 0 1.1T crypt btrfs /mnt/btrfs_p 5ed9d54d-2f12-4735-85c2-e0445a316088
├─sda6 8:6 0 65G part bcache 21c07a2c-6ed2-42c2-aafb-8e1bbfa471a5 753dca8b-06
└─sdb1 8:17 0 1.8T part bcache b324cfc4-a5ae-4888-ad6c-a5ba1002fdce 89ba6da8-01
└─bcache1 252:1 0 1.8T disk crypto_LUKS b519d6ce-291a-4500-937e-a5957e6ece6b
└─pool3 251:4 0 1.8T crypt btrfs /mnt/btrfs_p 5874cad5-d6dd-4b3d-a455-4e0ae2611f09
bcache0 is fine
bcache1 ended up in a bad state, I'm not sure how.
No biggie, right, just delete/recreate/re-register sda6, except how?
https://www.kernel.org/doc/Documentation/bcache.txt
SYSFS - CACHE DEVICE:
says nothing about how I can free/release sda6
saruman:~# echo /dev/sda6 > /sys/fs/bcache/unregister
-su: /sys/fs/bcache/unregister: Permission denied
is not helpful, nothing in kern.log
saruman:~# wipefs -a /dev/sda6
wipefs: error: /dev/sda6: probing initialization failed: Device or resource busy
saruman:~# l /dev/bcache/by-uuid/
total 0
drwxr-xr-x 2 root root 80 Feb 17 20:20 ./
drwxr-xr-x 3 root root 60 Feb 17 20:20 ../
lrwxrwxrwx 1 root root 13 Feb 17 20:20 6d88cea7-6cfc-4d30-8cf5-8c8b216e019d -> ../../bcache0
lrwxrwxrwx 1 root root 13 Feb 17 20:20 b324cfc4-a5ae-4888-ad6c-a5ba1002fdce -> ../../bcache1
Those UUIDs don't show up in lsblk
Oh my, what is this?
saruman:~# l /sys/block/sda/sda6/bcache/set
lrwxrwxrwx 1 root root 0 Feb 18 22:28 /sys/block/sda/sda6/bcache/set -> ../../../../../../../../../../../fs/bcache/f419195d-184e-44b7-9f88-a834a5458f8c/
This other UUID still does not show up in lsblk, by now I've lost count
of why I have 3 different kinds of UUIDs instead of 2.
Mmmh, now I'm starting to get it, I have to look at caching devices by UUID, find the right one, and stop it:
saruman:/sys/fs/bcache# l
total 0
drwxr-xr-x 5 root root 0 Feb 18 22:32 ./
drwxr-xr-x 10 root root 0 Feb 17 20:20 ../
drwxr-xr-x 7 root root 0 Feb 18 22:07 6698753e-78d0-4f6a-a8cb-fbb6a20ddae8/
drwxr-xr-x 7 root root 0 Feb 18 22:07 b7ba27a1-2398-4649-8ae3-0959f57ba128/
drwxr-xr-x 7 root root 0 Feb 18 22:32 f419195d-184e-44b7-9f88-a834a5458f8c/
Ok, let's try
saruman:/sys/fs/bcache/f419195d-184e-44b7-9f88-a834a5458f8c# echo 1 > stop
Ah, success:
saruman:~# wipefs -a /dev/sda6
/dev/sda6: 16 bytes were erased at offset 0x00001018 (bcache): c6 85 73 f6 4e 1a 45 ca 82 65 f5 7f 48 ba 6d 81
This isn't super obvious :)
Is there a howto document where those steps could be documented in a howto fashion?
(removing a caching device and replacing it with another one, maybe you got a bigger/faster SSD)
In https://www.kernel.org/doc/Documentation/bcache.txt or otherwise?
Thanks,
Marc
--
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
.... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/ | PGP 1024R/763BE901
--
To unsubscribe from this list: send the line "unsubscribe linux-bcache" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
