[zfs-discuss] Re: Ubuntu 12.04 zfs hang

gerrit.vanklaveren at gmail.com gerrit.vanklaveren at gmail.com
Sun May 5 17:00:59 EDT 2013


I uninstalled ubuntu-zfs then tried 'zpool import -d /dev/disk/by-id 
volume1' and it mounted. I then rebooted and i couldn't get it to mount. I 
am lost. Any other suggestions. I really don't want to take my data off of 
the zpool again. I am a little confused on if I took all 5 disks (3X1TB and 
2X3TB) Raidz them together if auto expand will allow full use of the 3TB 
drives or are they going to at as 1TB. I need 1 drive fault tolerance and 
as much space out of these drives as possible.

On Sunday, May 5, 2013 12:47:06 AM UTC-7, gerrit.va... at gmail.com wrote:
>
> https://gist.github.com/anonymous/5520048
>
> I did have it mounted at one point in Ubuntu so I know my setup 
> is capable of doing it, I have a non stranded zpool setup 2X3TB and 3X1TB 
> they are partitioned in 1TBs then are raidz together across the 2X3TB and 
> then one whole 1TB drive. Giving me the max possible storage. I do have 
> data i need on this pool. I would have to use more drives to create a pool 
> out of. 
>
> On Saturday, May 4, 2013 8:00:28 PM UTC-7, Darik Horn wrote:
>>
>> On Sat, May 4, 2013 at 8:30 PM,  <gerrit.va... at gmail.com> wrote: 
>> > 1. Hung 
>> > 2. How do i unset the Share properties i booted back into freenas and 
>> run 
>> > "#zpool get all volume1" and i don't see any share properties 
>> > 3.  https://gist.github.com/anonymous/5519354 
>>
>> Please post the output of `parted --list` too.  If possible, then 
>> create and import a new pool on this computer to isolate for something 
>> particular to FreeNAS. 
>>
>> TLDR:  This looks like a non-trivial or compound bug.  You probably 
>> won't get a fix until Brian looks at it. 
>>
>> ... 
>>
>> The first thing that jumps out at me are these lines after a reboot: 
>>
>> [    3.326080] kernel tried to execute NX-protected page - exploit 
>> attempt? (uid: 0) 
>> [    3.326085] BUG: unable to handle kernel paging request at 
>> ffff880424fe8d18 
>>
>> That should not happen on a healthy system. Run a memtest diagnostic 
>> on this computer to check for bad hardware. 
>>
>> The second thing that I notice is a cranky sleep event followed by a 
>> ZFS glitch near these lines: 
>>
>> [  820.546724] PM: Syncing filesystems ... done. 
>> [  820.547821] PM: Preparing system for mem sleep 
>>
>> Power management could be a secondary problem, but it is difficult to 
>> tell because the logs are in one big paste.  Be sure to do a hard 
>> reset before the import.  Try installing one of the Linux 3.8 backport 
>> packages, which could have better AMD A4-5300 support. 
>>
>> And this is interesting because I haven't seen a problem near the 
>> elevator routines in a while: 
>>
>> [ 2274.805939] Call Trace: 
>> [ 2274.805942]  [<ffffffff8169d8b9>] schedule+0x29/0x70 
>> [ 2274.805946]  [<ffffffff8169db7e>] schedule_preempt_disabled+0xe/0x10 
>> [ 2274.805951]  [<ffffffff8169c797>] __mutex_lock_slowpath+0xd7/0x150 
>> [ 2274.805956]  [<ffffffff81326c22>] ? noop_init_queue+0x32/0x60 
>> [ 2274.805963]  [<ffffffff8169c3aa>] mutex_lock+0x2a/0x50 
>> [ 2274.805967]  [<ffffffff81307d54>] elevator_exit+0x24/0x60 
>> [ 2274.805972]  [<ffffffff813084b5>] elevator_switch+0xa5/0x160 
>> [ 2274.805978]  [<ffffffff813085e6>] elevator_change+0x76/0xd0 
>> [ 2274.805982]  [<ffffffffa0759f8a>] vdev_disk_open+0x17a/0x3c0 [zfs] 
>> [ 2274.806039]  [<ffffffff81085f3a>] ? finish_task_switch+0x4a/0xf0 
>> [ 2274.806045]  [<ffffffffa0757b45>] vdev_open+0xf5/0x490 [zfs] 
>> [ 2274.806102]  [<ffffffffa07588c6>] vdev_open_child+0x26/0x40 [zfs] 
>> [ 2274.806173]  [<ffffffffa06476c6>] taskq_thread+0x236/0x4b0 [spl] 
>> [ 2274.806183]  [<ffffffff8108a360>] ? try_to_wake_up+0x200/0x200 
>> [ 2274.806198]  [<ffffffffa0647490>] ? task_done+0x160/0x160 [spl] 
>> [ 2274.806203]  [<ffffffff81077b73>] kthread+0x93/0xa0 
>> [ 2274.806210]  [<ffffffff816a8324>] kernel_thread_helper+0x4/0x10 
>> [ 2274.806216]  [<ffffffff81077ae0>] ? flush_kthread_worker+0xb0/0xb0 
>> [ 2274.806221]  [<ffffffff816a8320>] ? gs_change+0x13/0x13 
>>
>> -- 
>> Darik Horn <daj... at vanadac.com> 
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.zfsonlinux.org/pipermail/zfs-discuss/attachments/20130505/85132fb6/attachment.html>


More information about the zfs-discuss mailing list