ZFS-8000-8A: assistance needed

[Gary Palmer]

On Thu, Sep 08, 2016 at 03:35:50PM +0300, Ruslan Makhmatkhanov wrote:
> Gary Palmer wrote on 09/08/2016 15:16:
> > On Thu, Sep 08, 2016 at 02:46:04PM +0300, Ruslan Makhmatkhanov wrote:
> >> Maurizio Vairani wrote on 09/08/2016 10:39:
> >>> Hi Ruslan,
> >>>
> >>> Il 06/09/2016 22:00, Ruslan Makhmatkhanov ha scritto:
> >>>>     NAME              STATE     READ WRITE CKSUM
> >>>>     storage_ssd       ONLINE       0     0 59.3K
> >>>>       mirror-0        ONLINE       0     0     0
> >>>>         gpt/drive-06  ONLINE       0     0     0
> >>>>         gpt/drive-07  ONLINE       0     0     9
> >>>>       mirror-1        ONLINE       0     0  119K
> >>>>         gpt/drive-08  ONLINE       0     0  119K
> >>>>         gpt/drive-09  ONLINE       0     0  119K
> >>>>     cache
> >>>>       mfid5           ONLINE       0     0     0
> >>>>       mfid6           ONLINE       0     0     0
> >>>>
> >
> >> So, I'm interested how you detect that particular drives (with gpt
> >> labels drive-07, drive-08 and drive-09) are failing?
> >
> > The CKSUM errors from the zpool status output.  They are
> > indicative of underlying problems.  gpt/drive-07 has a low count
> > so it's possibly not a major problem, but gpt/drive-08 and
> > gpt/drive-09 both have a significant number.  I find it a bit
> > curious that both drives in the mirror have similar number of
> > errors.  Possibly indicating data pathing errors.
> 
> Ah, ok, thanks. I wasn't able to find meaning of CKSUM field description 
> in zpool manpage, so didn't got it at the time. Not sure about data 
> pathing errors though.

I mean something in the data path for both drives that is common
to those two drives and not the others.  If they're all attached to
the same controller, maybe they're on a different SAS port or cable
or something to the other drives.

Regards,

Gary

> >
> > I presume you are using ECC memory?
> 
> Yes, it's ECC.
> 
> 
> -- 
> Regards,
> Ruslan
> 
> T.O.S. Of Reality
>