Critical Warning: Volume is ¿healthy?

[RoboKaren]

This is strange. I just received a critical warning that one of my volumes is healthy. I'd understand a warning that it was degraded, or that it was a 9.2.x that needing upgrading, but that it is healthy?

The GUI shows a checksum error with the volume scrub but doesn't give a warning.

It's only when I drop down to the command line do I find out that the zpool command is much more concerned with the checksum error than the GUI is:

Code:

[nakamura@bethel] ~% zpool status
  pool: betheltime
state: ONLINE
status: One or more devices has experienced an unrecoverable error.  An
    attempt was made to correct the error.  Applications are unaffected.
action: Determine if the device needs to be replaced, and clear the errors
    using 'zpool clear' or replace the device with 'zpool replace'.
   see: http://illumos.org/msg/ZFS-8000-9P
  scan: scrub repaired 0 in 31h30m with 0 errors on Thu Nov 27 15:30:08 2014
config:

    NAME                                            STATE     READ WRITE CKSUM
    betheltime                                      ONLINE       0     0     0
      mirror-0                                      ONLINE       0     0     0
        gptid/6bf9ffa3-dbcb-11e3-b567-9cb65407d831  ONLINE       0     0     0
        gptid/6c724f8a-dbcb-11e3-b567-9cb65407d831  ONLINE       0     0     2

TL;DR: It seems to me that the GUI could do a better job of warning the user when the volume is 'healthy' but actually has a disk that may be on the verge of failing.

 

[Ericloewe]

I'd put in a ticket for a more informative error message.

I have the feeling it might be the kind of thing that'll have to wait for 10.1, but might as well try.

 

[BigDave]

This sounds similar to this earlier post
https://forums.freenas.org/index.php?threads/new-warning.24339/

 

[Ericloewe]

This sounds similar to this earlier post
https://forums.freenas.org/index.php?threads/new-warning.24339/

It is similar, but that one seems to be a real bug, whereas this one is probably a consequence of all the layers that go between ZFS's internals and the GUI. That's precisely one of the areas in which 10.1 is supposed to receive work.

 

[RoboKaren]

@eric - should I write a ticket or can you?

Happy to comply either way.

 

[Ericloewe]

@eric - should I write a ticket or can you?

Happy to comply either way.

It's best if you do, since you know the situation and any steps involved in reaching this point.

 

[RoboKaren]

Done. Posted as bug #6875: https://bugs.freenas.org/issues/6875