Xymon Mailing List Archive search

netapp.pl & cpu-test

2 messages in this thread

list Martin Flemming · Fri, 23 Nov 2007 14:19:57 +0100 (CET) ·
Hi !

I've got a problem with or don't understand  the cpu-test ...

We've delete many data on the netapps,
and if we do an syststat -m local on the machnies
we could see 

 ANY  AVG  CPU0 CPU1 CPU2 CPU3
 93%  27%    2%  16%   0%  91%
 97%  31%    2%  31%   0%  92%
 89%  28%    2%  23%   0%  86%
 61%  20%    3%  24%   0%  54%
 53%  17%    2%  20%   0%  45%
 93%  28%    2%  16%   0%  92%


but with the cpu-test we can see 

Fri Nov 23 13:48:07 2007 up 134 days, 23:56 CPU Usage= 15%

System clock is 0 seconds off

Detailed CPU Summary
Total	= 46%
Avg.	= 15%
CPU0	= 2%
CPU1	= 17%
CPU2	= 0%
CPU3	= 39%

Ok, of course the time-interval is problem,
the default local on the machine is 15 sec and for netapp.pl 5 minutes ...

So, my Question 

is  there a possibilty to make a graph for each of the cpu's ?

and which is best/optimize frequency to detect this cpu-usage ?

or can we have a graph for Average and Actual (maybe ANY ?) CPU-Usage for 
all together and each cpu for his own ? 

cheers,

       Martin 


Martin Flemming
DESY / IT          office : Building 2b / 008a
Notkestr. 85       phone  : XXX - XXXX - XXXX
22603 Hamburg      mail   : user-f286aaa49a76@xymon.invalid
list Martin Flemming · Wed, 12 Dec 2007 11:50:09 +0100 (CET) ·
Hi, once again ...

.. is it possible to make a mulit-graph for the cpu-test ...

for each cpu in one graph .. maybe also inclusive "Avg"-graph ...
System clock is 0 seconds off

Detailed CPU Summary
Total   = 79%
Avg.    = 30%
CPU0    = 4%
CPU1    = 50%
CPU2    = 0%
CPU3    = 68%
NetApp Release 7.2.1.1: Tue Jan 23 01:01:31 PST 2007
<!--bb-xsnmp.pl-->

netapp.pl version 1.10 - column cpu lifetime 60, tested in ~ 00:00:02 (max 00:02:00)

Thanks & cheers

 	Martin
quoted from Martin Flemming


On Fri, 23 Nov 2007, user-f286aaa49a76@xymon.invalid wrote:
Hi !

I've got a problem with or don't understand  the cpu-test ...

We've delete many data on the netapps,
and if we do an syststat -m local on the machnies
we could see

ANY  AVG  CPU0 CPU1 CPU2 CPU3
93%  27%    2%  16%   0%  91%
97%  31%    2%  31%   0%  92%
89%  28%    2%  23%   0%  86%
61%  20%    3%  24%   0%  54%
53%  17%    2%  20%   0%  45%
93%  28%    2%  16%   0%  92%


but with the cpu-test we can see

Fri Nov 23 13:48:07 2007 up 134 days, 23:56 CPU Usage= 15%

System clock is 0 seconds off

Detailed CPU Summary
Total	= 46%
Avg.	= 15%
CPU0	= 2%
CPU1	= 17%
CPU2	= 0%
CPU3	= 39%

Ok, of course the time-interval is problem,
the default local on the machine is 15 sec and for netapp.pl 5 minutes ...

So, my Question

is  there a possibilty to make a graph for each of the cpu's ?

and which is best/optimize frequency to detect this cpu-usage ?

or can we have a graph for Average and Actual (maybe ANY ?) CPU-Usage for
all together and each cpu for his own ?

cheers,

      Martin


Martin Flemming
DESY / IT          office : Building 2b / 008a
Notkestr. 85       phone  : XXX - XXXX - XXXX
22603 Hamburg      mail   : user-f286aaa49a76@xymon.invalid