rbd: spell out image features unsupported by the kernel

[idryomov]

/sys/bus/rbd/supported_features is available on kernels starting with
4.11. Example output:

$ sudo rbd map x
rbd: sysfs write failed
RBD image feature set mismatch. This image cannot be mapped because the following immutable features are unsupported by the kernel: data-pool.
In some cases useful info is found in syslog - try "dmesg | tail" or so.
rbd: map failed: (6) No such device or address

$ sudo rbd map -p foo y
rbd: sysfs write failed
RBD image feature set mismatch. You can disable features unsupported by the kernel with "rbd feature disable foo/y object-map fast-diff deep-flatten".
In some cases useful info is found in syslog - try "dmesg | tail" or so.
rbd: map failed: (6) No such device or address

or, if /sys/bus/rbd/supported_features isn't there:

$ sudo rbd map z
rbd: sysfs write failed
RBD image feature set mismatch. Try disabling features unsupported by the kernel with "rbd feature disable".
In some cases useful info is found in syslog - try "dmesg | tail" or so.
rbd: map failed: (6) No such device or address

Fixes: http://tracker.ceph.com/issues/19095
Signed-off-by: Ilya Dryomov idryomov@gmail.com

[idryomov]

The kernel patch has to go in first.

[idryomov]

retest this please

[shinobu-x]

LGTM

[dillaman]

lgtm. While you are in here -- can you drop the " or so" portion of the 'try "dmesg | tail" or so' message?

[idryomov]

@dillaman I think that's a verbatim copy from mount(8), which means it's at least a dozen years old. Is there something wrong with it?

[dillaman]

@idryomov It just reads like a non-native English speaker wrote that clause. The best I can figure is that perhaps the goal was to say "or similar". It's just a trivial thing that bugs me every time I see a user paste that error message in a message or IRC chat. Not that big of a deal, just a suggestion to clean it up.

[idryomov]

Dropped " or so" and took off DNM.

[dillaman]

lgtm