[Orca-users] inflated disk stats (not ignoring disksuite metadevices?)

Jeff Wasilko Jeff.Wasilko at Tufts.EDU
Mon May 21 10:20:55 PDT 2007


We're running a snapshot version from 2005 (revision 490), and I've  
noticed that a couple of the disk stats seem to be massively inflated:

Disk System Wide Reads/Writes Per Second
Disk System Wide Transfer Rate

Comparing the orcallator file to iostat for the same time (midnight):

Orcallator:

disk_rd/s	disk_wr/s	disk_rK/s	disk_wK/s
0.1		8534.2		0.6		69475

iostat:

Sun May 20 12:00:04 2007
      cpu
us sy wt id
17 73  1  9
                     extended device statistics
     r/s    w/s   Mr/s   Mw/s wait actv wsvc_t asvc_t  %w  %b device
     0.0    1.8    0.0    0.0  0.0  0.0    3.3    8.3   1   2 d40
     0.0    1.8    0.0    0.0  0.0  0.0    0.0    8.2   0   1 d41
     0.0    1.8    0.0    0.0  0.0  0.0    0.0    7.1   0   1 d42
2640.6    0.0   20.7    0.0  0.0 59.2    0.0   22.4   1  99 d100
2640.6    0.0   20.7    0.0  0.0 59.1    0.0   22.4   1  99 d101
     0.0    2.4    0.0    0.0  0.0  0.0    0.0    8.5   0   2 c1t0d0
     0.0    2.4    0.0    0.0  0.0  0.0    0.0    7.6   0   2 c1t1d0
1315.2    0.0   10.3    0.0  0.0 29.8    0.0   22.7   1  99  
c5t600C0FF0000000000067B72FC4055300d0
1326.9    0.0   10.4    0.0  0.0 29.3    0.0   22.1   1  98  
c5t600C0FF0000000000067B74C71D24700d0
     0.0  335.5    0.0   21.0  0.0  0.3    0.0    0.8   1  27 rmt/1
     0.0  459.1    0.0   28.7  0.0  0.4    0.0    0.8   1  38 rmt/2

In this case, it looks like it's not ignoring the disksuite  
metadevices. Here's what the metadevice looks like:

jeffw at alexandrite:~# metastat -p d100
d100 -m d101 1
d101 1 2 c5t600C0FF0000000000067B72FC4055300d0s0  
c5t600C0FF0000000000067B74C71D24700d0s0 -i 128b

Is this something that is fixed in later releases?

Thanks!

-j



More information about the Orca-users mailing list