On 12 August 2011 18:51, subrata majumder <[email protected]> wrote:
> I am using Net SNMP 5.4.2.1.
> snmpget returns 0 for UCD-SNMP-MIB::ssCpuRawKernel.0 . However, I am
> getting correct value for other CPU data.
>
> Does anybody know any issue related to UCD-SNMP-MIB::ssCpuRawKernel in
> 5.4.2.1? I try to compare this behavior with SNMP(5.3.3) on the same
> machine. I get correct value from snmpd(5.3.3).
There was a significant re-working of the code used to calculate the
various CPU statistics between the 5.3.x and 5.4 releases. As part
of this, we took the opportunity to rationalise exactly what these
various objects monitored.
If you look at the code used to calculate these values in 5.3.x
relevant fragment is as follows:
v5.3.x:
[ucd-snmp/vmstat.c]
case CPURAWSYSTEM:
long_ret = vmstat(cpurawsystem)+vmstat(cpurawinter)+vmstat(cpurawsoft);
return ((u_char *) (&long_ret));
case CPURAWKERNEL:
long_ret = vmstat(cpurawsystem);
return ((u_char *) (&long_ret));
So the value returned as cpuRawKernel monitors the single 'system' value,
while cpuRawSystem monitors a combination of three separate metrics.
Starting with the 5.4.x line, these individual values are separated out.
So cpuRawSystem reports the same value that was previously used as
cpuRawKernel, while the other two elements are available as
cpuRaw{Intr,Softirq} respectively.
So although the cpuRawKernel object is now 0 (on Linux systems at least),
you should find that cpuRawSystem provides exactly the same data.
Dave
------------------------------------------------------------------------------
EMC VNX: the world's simplest storage, starting under $10K
The only unified storage solution that offers unified management
Up to 160% more powerful than alternatives and 25% more efficient.
Guaranteed. http://p.sf.net/sfu/emc-vnx-dev2dev
_______________________________________________
Net-snmp-users mailing list
[email protected]
Please see the following page to unsubscribe or change other options:
https://lists.sourceforge.net/lists/listinfo/net-snmp-users
