Re: Tried to replace a drive in a raid 1 and all hell broke loose

[efkf <efkf@firemail.cc>]

[-- Attachment #1: Type: text/plain, Size: 1195 bytes --]

On 5/27/22 16:25, Forza wrote
> On 2022-05-27 17:13, efkf@firemail.cc wrote:
>> I had mounted the fs with -o degraded and one drive a couple of times 
>> just as a sanity check to make sure the data really is in both drives, 
>> i assume this would mount it rw and fall into the category you described. 
> 
> With Btrfs, this is not good practice. If you mount RAID mirrors 
> independently as degraded, the metadata could be updated differently and 
> when you combine the drives again, the data would not be the same on 
> each device - which would lead to corruption. This is true even for ro 
> mounts. ro is a Linux VFS thing, Btrfs will still write to the disk for 
> its internal things.

I don't know much about advanced filesystems but I can imagine scenarious where it would be beneficial to reintroduce an old drive into an array that maybe even has changed without whiping it, maybe it's the only one storing the intact copy of an old file.
Anyway, is there a way to check the data is really redundant without trusting the filesystem telling me it's so?

> Good to hear that you got most of your data back. :)
:)

I tried another client now, hopefully the attachment has gone through

[-- Attachment #2: transfer_dmesg_full.xz --]
[-- Type: application/x-xz, Size: 44660 bytes --]