<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[system.cpu.usage parameter of HPUX subagent problem]]></title><description><![CDATA[<p dir="auto">Hi:<br />
    I  found the data collected by system.cpu.usage parameter of HPUX subagents  inconsistent with the other performance tools like glance,top,vmstat and so on.<br />
    I tried to trace the CpuUsageCollector function in src/agent/subagent/HPUX/system.cpp and found the reason:<br />
    The HPUX subagent counts the cpu wait  statment as cpu usage:<br />
    m_cpuUsage[m_currentSlot++] = 100 - ((float)((psd.psd_cpu_time[CP_IDLE] - m_idle) * 100) / delta);<br />
    but other performace tools counts the cpu wait statment as cpu idle.<br />
    I modified the souce code like this:<br />
    m_cpuUsage[m_currentSlot++] =  ((float)(((psd.psd_cpu_time[CP_USER] - m_user) + (psd.psd_cpu_time[CP_SYS] - m_sys) ) * 100) / delta)   (mybe adds the CPU nice statment is better);<br />
    As a result this metric  was consistent  with other performance tools. <br />
    I wonder which way calculates the CPU usage is more helpful.</p>
]]></description><link>https://forum.netxms.org/topic/2980/system.cpu.usage-parameter-of-hpux-subagent-problem</link><generator>RSS for Node</generator><lastBuildDate>Fri, 13 Nov 2026 15:51:23 GMT</lastBuildDate><atom:link href="https://forum.netxms.org/topic/2980.rss" rel="self" type="application/rss+xml"/><pubDate>Wed, 05 Mar 2014 02:29:52 GMT</pubDate><ttl>60</ttl></channel></rss>