On 10/19/08 03:49, Audun Frøysaa wrote:
> Ok, maybe i should write some more information :-).
> 
> I am wondering if it could be a problem with the 2 external USB drives, but 
> when i had LVM setup i did not have this problem..
> 
> My system setup is like this:
> 4 Sata drives: 1 TB from Samsung and 3 500Gb from Samsung
> 2 External USB drives from WD ( 500gb elements ).
> 
> -bash-3.2# zpool list 
> NAME    SIZE   USED  AVAIL    CAP  HEALTH  ALTROOT
> Media  1.91T   128G  1.78T     6%  ONLINE  -
> 
> -bash-3.2# zpool status Media
>   pool: Media
>  state: ONLINE
>  scrub: resilver completed after 0h0m with 0 errors on Sat Oct 18 10:52:36 
> 2008
> config:
> 
>         NAME        STATE     READ WRITE CKSUM
>         Media       ONLINE       0     0     0
>           raidz2    ONLINE       0     0     0
>             c2t0d0  ONLINE       0     0     0
>             c3t0d0  ONLINE       0     0     0
>             c0d1    ONLINE       0     0     0
>             c4t0d0  ONLINE       0     0     0
>             c4t1d0  ONLINE       0     0     0
>             c5t0d0  ONLINE       0     0     0
>             c5t1d0  ONLINE       0     0     0
> 
> errors: No known data errors
> 
> -bash-3.2# zpool iostat  
>                capacity     operations    bandwidth
> pool         used  avail   read  write   read  write
> ----------  -----  -----  -----  -----  -----  -----
> Media        128G  1.78T      3      3   395K   375K
> 
> Is there some way to find the error. 
> 
> Thanks
> --
> This message posted from opensolaris.org
> _______________________________________________
> networking-discuss mailing list
> [email protected]

Hi Audun,

There is no indication here of what you are seeing. It is not clear to 
me whether your are seeing these 'gaps' on the wire (thus networking) or 
on the system with the ZFS pool (thus ZFS). Keep in mind that ZFS will 
write to disk every five (5) seconds by default, unless there is a lot 
of data to commit or some other operation requires immediate commit to 
disk (an NFS write, for example--don't know whether the SMB protocol has 
similar synchronous write requirement (NFS write does it for data 
integrity in case the server crashed)).

Also, I can't tell whether 120mb is Megabit (Mb) since you are 
mentioning networking, or MegaByte (MB) since you are mentioning disk 
(ZFS). Since that is almost an order order of magnitude difference, it 
may be important, as the question is regarding performance.

Steffen

(reply-to is set to alias, no need to reply-all)
_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to