03-11-2010
Extremely high kernel CPU Usage (Solaris 10 SPARC)
I've got a domain running on a few boards of a 25k. I'm seeing very high kernel cpu usage in top and cant' quite explain it. System runs a large number of smallish Oracle 10g2 databases (30), used mainly for development.
load average: 36.63, 36.68, 37.42
2489 processes: 2452 sleeping, 21 running, 2 zombie, 1 stopped, 13 on cpu
CPU states: 15.8% idle, 21.1% user, 63.1% kernel, 0.0% iowait, 0.0% swap
Memory: 192G real, 85G free, 93G swap in use, 67G swap free
I have to transcribe the figures by hand, but all 15 processes listed with top are oracle, running anywhere from 2% cpu to .5% cpu each (should be making up the userland process usage anyway).
What's the best way to troubleshoot unknown kernel usage issues?
Thanks much,
utopiajoe
10 More Discussions You Might Find Interesting
1. UNIX for Advanced & Expert Users
Hi
I have a lot of scripts running on a Sun Solaris server, which are constantly running in a loop looking for work to do. When they have no work they sleep for a certain amount of time (60secs normally). I have 13 of these scripts running the number of sleep command issued can be in the order... (4 Replies)
Discussion started by: handak9
4 Replies
2. UNIX for Advanced & Expert Users
Hi,
I am seeing very high kernel usage and very high load averages on my system (Although we are not loading much data to our database). Here is the output of top...does anyone know what i should be looking at?
Thanks,
Lorraine
last pid: 13144; load averages: 22.32, 19.81, 16.78 ... (4 Replies)
Discussion started by: lorrainenineill
4 Replies
3. Ubuntu
I am running a Dell PE R815 with 4 x AMD 12 core CPUs with 128GB of RAM and a RAID 5 array of 6 SAS disks. This is an HPC application and is definitely CPU bound, however once I run 16 of these processes (thus pinning 16 cores) the work performed slows down dramatically, to maybe 5 or 10% of what... (2 Replies)
Discussion started by: mowmentous
2 Replies
4. UNIX for Advanced & Expert Users
Hi all,
I have a solaris box, and I would like to know if anyone has commands to check kernel usage's.
Scenario:
solaris box is having cpu 100 % issue. .
I have used sar -u 10 5 it shows 35 % kernel usage.
If you can guide with some docs that would be good..
waiting for... (1 Reply)
Discussion started by: SunSolars_admin
1 Replies
5. AIX
ibm tivoli monitor's kuxagent is taking up a lot of cpu usage.
anybody has any idea? i attatched a screenshot. (3 Replies)
Discussion started by: curtis911
3 Replies
6. Red Hat
Hi team
I have three physical servers running on Red Hat Enterprise Linux Server release 6.2 with the following memory conditions:
# cat /proc/meminfo | grep -i mem
MemTotal: 8062888 kB
MemFree: 184540 kB
Shmem: 516 kB
and the following swap conditions:
... (6 Replies)
Discussion started by: hedkandi
6 Replies
7. Linux
Hi All,
Yesterday my Linux server went panic and even a small command took a lot of time to run.
When i monitored pl find the below output
Cpu(s): 0.1%us, 98.4%sy, 0.0%ni, 1.5%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
the time spent on kernel mode is 98 % and also idle time is around 1.5 %...... (4 Replies)
Discussion started by: jegaraman
4 Replies
8. AIX
Dear All,
One production Server is affected by high usage.
Application is slow now. Please guide me how to solve it?
NMON Report shows full cpu usage.
Here im posting some server details.
bash-3.2# lparstat -i
Node Name : *********
Partition Name ... (6 Replies)
Discussion started by: Thala
6 Replies
9. Shell Programming and Scripting
I have created one script and i have added it into cron to run after 10 mins. However I have noted that whenever that script runs, It causes CPU utilization of server to increase about 10-20 %
I have rechecked script and there is no way i can make changes, Script contain only 2-3 commands. So... (4 Replies)
Discussion started by: Nakul_sh
4 Replies
10. Solaris
HI
In my M5000 , one of domain is having SAp installed. from today onwards we are facing some stange issue.
when we start SAP application, that particular user is taking 95 % of system CP and renaming 5 % is taken by system . because of this reason application is slow.
i have 4 CPU(32... (4 Replies)
Discussion started by: bentech4u
4 Replies
LEARN ABOUT DEBIAN
sg_get_cpu_stats_diff
sg_get_cpu_stats(3) Library Functions Manual sg_get_cpu_stats(3)
NAME
sg_get_cpu_stats, sg_get_cpu_stats_diff, sg_get_cpu_percents - get cpu usage
SYNOPSIS
#include <statgrab.h>
sg_cpu_percents *sg_get_cpu_percents(void);
sg_cpu_stats *sg_get_cpu_stats(void);
sg_cpu_stats *sg_get_cpu_stats_diff(void);
DESCRIPTION
sg_get_cpu_stats() and sg_get_cpu_stats_diff() both return static pointers of type sg_cpu_stats. sg_get_cpu_stats() returns the total
amount of "ticks" the operating system has spent in each of the different states. sg_get_cpu_stats_diff() returns the difference in "ticks"
for each of the states since last time sg_get_cpu_stats_diff() or sg_get_cpu_stats() was called. If it has never been called, it will
return the result of sg_get_cpu_stats().
The value stored (the "ticks") will vary between operating systems. For example Solaris has a total of 100 per second, while Linux has sub-
stantially more. Also, different operating systems store different information - you won't find nice cpu on Solaris for example.
Because of this, you will ideally always want to work on a scale against the total, or in percentages.
sg_get_cpu_percents() returns a pointer to a static sg_cpu_percents. The function calls sg_get_cpu_stats_diff() and changes the values into
percentages. If it has never been called before (and nor has sg_get_cpu_stats() or sg_get_cpu_stats_diff()), the returned percentages will
be the systems total ever since its uptime. (Unless the counters have cycled)
RETURN VALUES
There are two structures returned by the CPU statistics functions.
typedef struct{
long long user;
long long kernel;
long long idle;
long long iowait;
long long swap;
long long nice;
long long total;
time_t systime;
}sg_cpu_stats;
typedef struct{
float user;
float kernel;
float idle;
float iowait;
float swap;
float nice;
time_t time_taken;
}sg_cpu_percents;
user kernel idle iowait swap nice
The different CPU states.
systime time_taken
The time taken in seconds since the last call of the function, or the system time.
SEE ALSO
statgrab(3)
WEBSITE
http://www.i-scream.org/libstatgrab/
i-scream $Date: 2005/04/25 11:25:45 $ sg_get_cpu_stats(3)