[OpenIndiana-discuss] iostat

[Jonathan Adams]

:( there really shouldn't be any problem with that system then ...

do you have the latest firmware installed on the controller/drives?

You'll need someone (other than me) with more dtrace experience to help any
further.


On 11 October 2013 19:43, jason matthews <jason at broken.net> wrote:

>
> the system config is as follows:
>
> 4x 4650L, 576GB of RAM, 8x 800GB DC S3700 load balanced over 2x LSI 9207-8i
>
> there is no swap. i am only running filebench on the pool so far.
>
>
> j.
>
> On Oct 11, 2013, at 9:28 AM, Jonathan Adams <t12nslookup at gmail.com> wrote:
>
> > Ouch! they are _some_ drives ...
> >
> > do you have "swap" running on them? how much ram do you have in that
> > machine?
> >
> >
> > On 11 October 2013 16:44, Jason Matthews <jason at broken.net> wrote:
> >
> >>
> >> i would say so. they are 800gb intel dc s3700s
> >>
> >> j.
> >>
> >> Sent from Jasons' hand held
> >>
> >> On Oct 11, 2013, at 1:08 AM, Jonathan Adams <t12nslookup at gmail.com>
> wrote:
> >>
> >>> do you have "Enterprise" grade disks in your machine?  We saw a lot of
> %b
> >>> and %w associated with "Consumer" Seagate disks ... not all the time,
> and
> >>> not for over a year in production, but once it started happening across
> >>> multiple servers we cottoned on pretty quickly.
> >>>
> >>>
> >>> On 10 October 2013 21:39, jason matthews <jason at broken.net> wrote:
> >>>
> >>>>
> >>>> I am not sure when reporting zpool stats was introduced into iostat
> but
> >> i
> >>>> love it. however, i am not sure what to make of it.
> >>>>
> >>>> For instance, what does a wsvc time of 20 seconds tell me? seems like
> my
> >>>> queue is too big but that is not reflected in individual device
> >> statistics.
> >>>> in the case of zpools, is this some sort of reflection on transaction
> >>>> groups?
> >>>>
> >>>> can anyone explain why i see such large time associated with zpool in
> >> the
> >>>> iostat output?
> >>>>
> >>>> thanks,
> >>>> j.
> >>>>
> >>>>                   extended device statistics
> >>>>   r/s    w/s        Mr/s   Mw/s    wait     actv  wsvc_t asvc_t  %w
>  %b
> >>>> device
> >>>>   0.0 3958.7    0.0  300.7       0.1     18.8   0.0       4.7        0
> >>>> 153   c3
> >>>>   0.0   11.8       0.0    0.1         0.0      0.0    0.0       0.6
> >>>>   0       0    c3t55CD2E400012A66Cd0
> >>>>   0.0    0.0        0.0    0.0         0.0      0.0    0.0       0.0
> >>>>   0       0    c3t55CD2E400012B753d0
> >>>>   0.0 2679.7    0.0  149.1       0.1      9.9    0.0       3.7
>  5
> >>>>   80    c3t50015178F36579E8d0
> >>>>   0.0 1267.2    0.0  151.5       0.0      8.8    0.0       7.0
>  3
> >>>>   73    c3t50015178F3657B74d0
> >>>>   0.0    0.0        0.0    0.0          0.0      0.0    0.0       0.0
> >>>>    0       0    c3t50015178F3657AECd0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c3t50015178F36579E1d0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c3t50015178F3657B85d0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c3t50015178F36579BAd0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c3t50015178F3657A05d0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c3t50015178F36579FEd0
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c5
> >>>>   0.0    0.0        0.0    0.0         0.0       0.0    0.0       0.0
> >>>>    0       0    c5t4d0
> >>>>   0.0 3946.1    0.0  300.6  80582.9 26.2 20421.0  6.6  88    89 data
> >>>>   0.0   11.6    0.0    0.1  0.0  0.0    1.1    0.8   0   0 rpool
> >>>>
> >>>>
> >>>> _______________________________________________
> >>>> OpenIndiana-discuss mailing list
> >>>> OpenIndiana-discuss at openindiana.org
> >>>> http://openindiana.org/mailman/listinfo/openindiana-discuss
> >>> _______________________________________________
> >>> OpenIndiana-discuss mailing list
> >>> OpenIndiana-discuss at openindiana.org
> >>> http://openindiana.org/mailman/listinfo/openindiana-discuss
> >>
> >> _______________________________________________
> >> OpenIndiana-discuss mailing list
> >> OpenIndiana-discuss at openindiana.org
> >> http://openindiana.org/mailman/listinfo/openindiana-discuss
> >>
> > _______________________________________________
> > OpenIndiana-discuss mailing list
> > OpenIndiana-discuss at openindiana.org
> > http://openindiana.org/mailman/listinfo/openindiana-discuss
>
>
> _______________________________________________
> OpenIndiana-discuss mailing list
> OpenIndiana-discuss at openindiana.org
> http://openindiana.org/mailman/listinfo/openindiana-discuss
>