load is not the same as cpu usage, load is the number of queued processes in a period. maybe there are processes waiting for some events, such as I/O.
2011/7/19 Javier González <[email protected]>: > Hello, > I have seen a lot peaks on the CPU LOAD of my nagios machine (See pnp4nagios > graphs on pdf document attached). > I don´t know which process/processes are the ones that use too much CPU on > that periods because the PC is only used for Nagios (And plugins used with > it) and all the plugins defined have a check_period of 1 minute and there is > a lot of periods where the load is near to 0 (so what happens every 2 hours > more or less that use so much CPU???) > I have tried to find the process that causes this behavior with the "top" > command but %Cpu idle is always (also on the peaks of the CPU LOAD) more han > 70% and no process uses more than 4% of CPU (See three examples of TOP > command where %CPU idle is always high but Load vary a lot on each case) > ****************************************************************** > top - 11:52:27 up 9:51, 1 user, load average: 0.09, 0.56, 0.62 > Tasks: 102 total, 1 running, 101 sleeping, 0 stopped, 0 zombie > Cpu(s): 7.9%us, 1.5%sy, 0.0%ni, 88.4%id, 1.0%wa, 0.3%hi, 0.8%si, > 0.0%st > Mem: 2050176k total, 675704k used, 1374472k free, 132628k buffers > Swap: 2096472k total, 0k used, 2096472k free, 333384k cached > top - 13:30:05 up 11:29, 2 users, load average: 1.53, 1.27, 0.83 > Tasks: 110 total, 3 running, 107 sleeping, 0 stopped, 0 zombie > Cpu(s): 9.2%us, 1.5%sy, 0.0%ni, 88.6%id, 0.0%wa, 0.0%hi, 0.7%si, > 0.0%st > Mem: 2050176k total, 708540k used, 1341636k free, 139020k buffers > Swap: 2096472k total, 0k used, 2096472k free, 334668k cached > top - 13:16:39 up 1 day, 11:15, 2 users, load average: 2.45, 2.03, 1.96 > Tasks: 106 total, 1 running, 105 sleeping, 0 stopped, 0 zombie > Cpu(s): 7.8%us, 1.4%sy, 0.0%ni, 89.1%id, 1.0%wa, 0.2%hi, 0.5%si, > 0.0%st > Mem: 2050176k total, 845668k used, 1204508k free, 167244k buffers > Swap: 2096472k total, 0k used, 2096472k free, 351224k cached > > > > ******************************************************************** > > Is there some way to find wich process cause this high CPU LOAD ? (Which > proccess use sometimes a lot of the CPU time so queuing other processes). > Why sometimes the CPU Load is near to 0 and other times Load_1_min is near > to 4 (On the example 2,45 but there are higher values) if the processes are > always the same ??? > Thank you very much > ***************************************************************************** > Environment Description. > Intel Core 2 DUO E7500 / 2,93GHz > DDR2 2GB / 667MHz > Centos 5.5 > Nagios 3.2.3 > nagios-plugins-1.4.11 > net-snmp-5.6 > snmptt-1.3 > mklivestatus-1.1.8 > nagvis-1.5.6 > pnp4nagios-0.6.10 > ------------------------------------------------------------------------------ > Magic Quadrant for Content-Aware Data Loss Prevention > Research study explores the data loss prevention market. Includes in-depth > analysis on the changes within the DLP market, and the criteria used to > evaluate the strengths and weaknesses of these DLP solutions. > http://www.accelacomm.com/jaw/sfnl/114/51385063/ > _______________________________________________ > Nagios-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/nagios-users > ::: Please include Nagios version, plugin version (-v) and OS when reporting > any issue. > ::: Messages without supporting info will risk being sent to /dev/null > ------------------------------------------------------------------------------ Magic Quadrant for Content-Aware Data Loss Prevention Research study explores the data loss prevention market. Includes in-depth analysis on the changes within the DLP market, and the criteria used to evaluate the strengths and weaknesses of these DLP solutions. http://www.accelacomm.com/jaw/sfnl/114/51385063/ _______________________________________________ Nagios-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/nagios-users ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. ::: Messages without supporting info will risk being sent to /dev/null
