fabioz23 / maczfs

Automatically exported from code.google.com/p/maczfs
Other
0 stars 0 forks source link

Pools disappear after 10.9.2 update #134

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago
What steps will reproduce the problem?
1.  Update to 10.9.2
2.  Reboot
3.  Pool doesn't mount

What is the expected output? 

A:  To see my pool mounted

What do you see instead?
A:  It doesn't mount

What version of the product are you using? On what operating system?

OS 10.9.2
ZFS 74.3.3

Please provide any additional information below.
Zpool command works and is recognized however when I go to import my pool it 
doesn't recognize any pools as being available.

Original issue reported on code.google.com by buse...@gmail.com on 1 Mar 2014 at 3:00

GoogleCodeExporter commented 9 years ago
Summary:

I see the same problem. However, in my case, this is just after installing 
74.3.3. I was previously using 74.1.0 with no problem on 10.9.2.
A forced import "zpool import -f {ID|NAME}" will mount the volume.
Every following reboot (even if you use "zpool export") I still have to do an 
explicit "zpool import -f" to mount the volume.

Details:

$ zpool status
You must be root in order to load the ZFS kext
internal error: failed to initialize ZFS library
$ sudo -s
Password:
root# zpool status
no pools available
    ====== PROMPTED that the kext is NOT signed.
root# zpool import
  pool: splish
    id: 4450072898083926272
 state: ONLINE
action: The pool can be imported using its name or numeric identifier.
config:

    splish       ONLINE
      raidz2     ONLINE
        disk4s2  ONLINE
        disk5s2  ONLINE
        disk3s2  ONLINE
        disk6s2  ONLINE
root# zpool status
no pools available
root# zpool import 4450072898083926272
cannot import 'splish': pool may be in use from other system
use '-f' to import anyway
root# zpool import -f 4450072898083926272
root# zpool status
  pool: splish
 state: ONLINE
 scrub: none requested
config:

    NAME         STATE     READ WRITE CKSUM
    splish       ONLINE       0     0     0
      raidz2     ONLINE       0     0     0
        disk4s2  ONLINE       0     0     0
        disk5s2  ONLINE       0     0     0
        disk3s2  ONLINE       0     0     0
        disk6s2  ONLINE       0     0     0

errors: No known data errors

================================
The file system is now mounted.

Reboot.

After a reboot, the file system does not automatically mount.
A "zpool import -f splish" is required again.

root# zpool status
no pools available
root# zpool import
  pool: splish
    id: 4450072898083926272
 state: ONLINE
action: The pool can be imported using its name or numeric identifier.
config:

    splish       ONLINE
      raidz2     ONLINE
        disk4s2  ONLINE
        disk3s2  ONLINE
        disk5s2  ONLINE
        disk6s2  ONLINE
root# zpool status
no pools available
root# zpool import splish
cannot import 'splish': pool may be in use from other system
use '-f' to import anyway
root# zpool import -f splash

======================
Now, the volume is accessible again.

To see if I could workaround the problem, I thought that maybe "zpool export" 
would fix things.
Nope, "zpool import" still runs into the same problem.
"zpool import -f splish" is required to mount the volume.

Original comment by sbyt...@gmail.com on 1 Mar 2014 at 3:03

GoogleCodeExporter commented 9 years ago
Thanks for the report.
The code handling mount/umount and import/export is unchanged between 74.1.0 
(reported as working by @sbytnar in comment #1) and 74.3.3, so the change is 
probably somewhere else.
@sbytnar : Can you run our data collection script collect-maczfs-state.sh 
(http://downloads.maczfs.org/detail=21) and attach its output here and also run 
"zdb -vv -l /dev/diskXsY" (replace "diskXsY" with a device in your pool) once 
after a failed import and once after a (forced) successful import and post the 
result here?  The zdb utility should be at /usr/sbin/zdb

Original comment by googlelogin@bjoern-kahl.de on 2 Mar 2014 at 12:52