[Orca-users] RE: emc dmx/clarion native fabric luns.

[leo.mccoy at kbm1.com]

Maybe this will help you.  I had a similar problem with my SAN disks.
 
My problem was in the orcallator.cfg file's plot parameters.  The original
data line was:
 
data
disk_runp_((?:c\d+t\d+d\d+)|(?:c\d+d\d+)|(?:[ms]d\d+)|(?:c\d+t([A-Z0-9]{32})
d\d+))

I changed it to:

data
disk_runp_((?:c\d+t\d+d\d+)|(?:c\d+d\d+)|(?:[ms]d\d+)|(?:c\d+t([A-Z0-9]{16,3
2})d\d+))

This better reflects the possible lengths of the SAN disk device names.

-----Original Message-----
From: orca-users-bounces+leo.mccoy=kbm1.com at orcaware.com
[mailto:orca-users-bounces+leo.mccoy=kbm1.com at orcaware.com]
Sent: Wednesday, April 06, 2005 11:20 AM
To: orca-users at orcaware.com
Subject: [Orca-users] RE: emc dmx/clarion native fabric luns. 



actually i realize this was an incomplete description of my issue - 

the orcallator daily file contains a column header of the title disk_runp_
c8t50060482CAFDE243d47 (for example) - 

irrespective of the values being 0 - these devices do  not appear on the
disk run percent graph. 


  _____  


From: Saladin, Miki 
Sent: Tuesday, April 05, 2005 5:06 PM
To: 'orca-users at orcaware.com'
Subject: emc dmx/clarion native fabric luns. 

 

setoolkit 3.3

orcallator.se  version  1.36

 

Implementation on EMC dmx/clarion with native fabric luns results in the
following /dev/disk indicators -  see the following example - 

 

ssd0     -> ssd0         -> c8t50060482CAFDE243d47       0

ssd0,c   -> ssd0,c       -> c8t50060482CAFDE243d47s2     0

ssd0,d   -> ssd0,d       -> c8t50060482CAFDE243d47s3     0

ssd0,e   -> ssd0,e       -> c8t50060482CAFDE243d47s4     0

ssd1     -> ssd1         -> c8t50060482CAFDE243d46       0

ssd1,c   -> ssd1,c       -> c8t50060482CAFDE243d46s2     0

ssd1,d   -> ssd1,d       -> c8t50060482CAFDE243d46s3     0

ssd1,e   -> ssd1,e       -> c8t50060482CAFDE243d46s4     0

ssd2     -> ssd2         -> c8t50060482CAFDE243d45       0

ssd2,a   -> ssd2,a       -> c8t50060482CAFDE243d45s0     0

ssd2,b   -> ssd2,b       -> c8t50060482CAFDE243d45s1     0

ssd2,c   -> ssd2,c       -> c8t50060482CAFDE243d45s2     0

ssd2,g   -> ssd2,g       -> c8t50060482CAFDE243d45s6     0

ssd3     -> ssd3         -> c8t50060482CAFDE243d44       0

ssd3,a   -> ssd3,a       -> c8t50060482CAFDE243d44s0     0

ssd3,b   -> ssd3,b       -> c8t50060482CAFDE243d44s1     0

ssd3,c   -> ssd3,c       -> c8t50060482CAFDE243d44s2     0

ssd3,g   -> ssd3,g       -> c8t50060482CAFDE243d44s6     0

 

disk_runp metrics are 0 - despite iostat -xnM returning non zero metrics. 

in the examples directory - dmx_disks.se, dmx_iost.se and dmx_siostat.se
work fine from the command line.  

 

my ? is how do i  now incorporate the use of these dmx_ ses into the
orcallator.se file ?  

 

the only reference to iostat in orcallator.se is 

#include <p_iostat_class.se>

 

there is no class file for dmx iostat. 

 

i'm sure i am not the only one who has done this. 

any hints would be appreciated! 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20050406/d1327966/attachment.html>