gstat [was: raid or not raid]

[Ian Smith]

Hey Beto,

On Fri, 25 May 2007 19:55:47 +1000 Norberto Meijome wrote:

 > replying to your email down the thread...but using this content...

hijacking the thread almost entirely off to the side ..

 > >     0      6      6    627    8.5      0      0    0.0    5.0| ad4s1g
 > 
 > the g partition in slice 1 of disk ad4.
 > 
 > They all get used (from the GEOM POV) when , for example, userland
 > access the fs located in ad4s1g . 
 >
 > clear as mud?  ;)

Not having played with RAID here, gstat was useful news, ta, especially
as vmstat & iostat don't show acd0 anymore .. however I'm a bit puzzled
by the below; nothing but 'find / -name xdfsfdxfx' running, so why does
it show so much write activity?  This is consistent (with find anyway).

dT: 10.005  flag_I 10000000us  sizeof 240  i -1
 L(q)  ops/s    r/s   kBps   ms/r    w/s   kBps   ms/w   %busy Name
    0      0      0      0    0.0      0      0    0.0    0.0| fd0
    0      0      0      0    0.0      0      0    0.0    0.0| cd0
    1    156    134    880    6.7     22    352   23.0   89.0| ad0
    0      0      0      0    0.0      0      0    0.0    0.0| acd0
    0      0      0      0    0.0      0      0    0.0    0.0| ad0s1
    1    156    134    880    6.7     22    352   23.5   89.3| ad0s2
[..]
    0      0      0      4   23.4      0      0    0.0    0.7| ad0s2a
    0      0      0      0    0.0      0      0    0.0    0.0| ad0s2b
    0      0      0      0    0.0      0      0    0.0    0.0| ad0s2c
    0      0      0      0    0.0      0      0    0.0    0.0| ad0s2d
    1    155    133    876    6.8     22    352   24.0   90.6| ad0s2e

I thought maybe updating file access times? but would that be so much of
a penalty, even on this 300MHz laptop with a UDMA33 5400rpm drive ..

BTW, this is on 5.5-STABLE, seeing you weren't sure about it on 5.x

 > Your reasoning is excellent -- it's only your basic assumptions that
 > are wrong. 

Probably very apt :)

Thanks again, Ian