btrfs: Reset `mount -o rw,degraded` count With btrfs, it seems that there is a single opportunity to use: -o rw,degraded (eg, if there is a missing disk in a RAID1 configuration) * How can one force the degraded mount to proceed `rw` a second time? * How can one edit the filesystem device so as to pretend that it has not yet been mounted `rw,degraded`? I understand there are kernel patches which remove various mount checks, however I want to patch the filesystem instead, and have all the checks still in place.

Not really an answer, but perhaps some leads...

There is a known gotcha in the wiki which affects btrfs RAID1:

> When a raid1 volume is stuck as read-only [...] it can only be recovered by dumping its contents, recreating it and restoring the contents

But that's just the offical line - there is a simple kernel patch which effectively comments out the line:


goto fail_sysfs;


Which will allow multiple `rw,degraded` mounts.

Please shout if you find a filesystem patch!