<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: Defunct processes using all the CPU in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949219#M931595</link>
    <description>Hi Mott,&lt;BR /&gt;&lt;BR /&gt;I've seen similar situations where defunct processes were consuming massive amounts of cpu time.  At this point, you need to get rid of them.  If they have a parent PID of 1, the only way to clear them out is to reboot.  If the PPID is other than 1, you can try killing the parent.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Pete</description>
    <pubDate>Fri, 11 Apr 2003 12:12:06 GMT</pubDate>
    <dc:creator>Pete Randall</dc:creator>
    <dc:date>2003-04-11T12:12:06Z</dc:date>
    <item>
      <title>Defunct processes using all the CPU</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949218#M931594</link>
      <description>A lot of defunct processes (shown below) associated with an Oracle database application were using all the CPU on a T-600 with 8 CPU's.  I thought a defunct process was a zombie process that used no CPU time.  The software that I am using, BMC's Patrol for UNIX Perform Predict was reporting that the defunct processes were using all the CPU time - is this a measurement error from the Patrol data?  Also why would so many processes crash like this at the same time?&lt;BR /&gt;&lt;DEFUNCT&gt; "/usr/bin/sh \./process_award_idx\.sh"&lt;BR /&gt;&lt;DEFUNCT&gt; "cat testdata\.lst"&lt;BR /&gt;&lt;DEFUNCT&gt; "cut -f6 -d\*"&lt;BR /&gt;&lt;DEFUNCT&gt; "cut -f8 -d|"&lt;BR /&gt;&lt;DEFUNCT&gt; "cut -f9 -d|"&lt;BR /&gt;&lt;DEFUNCT&gt; "diff /dla_pr/rep_ecf3/020705/abstract/Abstract_d_YPE02158000"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep -n \^N1\\\* 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep -v FREIGHT"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \\\*BD\\\*"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \\\*ST\\\*"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \\\*XE\\\*"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \^N1\\\* 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \^NTE\\\* 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \^PER\\\* 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "grep \^REF\\\* 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "process_sol_seed process_sol_seed"&lt;BR /&gt;&lt;DEFUNCT&gt; "pwd"&lt;BR /&gt;&lt;DEFUNCT&gt; "sed -n 12p 850stsegs0001"&lt;BR /&gt;&lt;DEFUNCT&gt; "sed -n 16p 850stsegs0004"&lt;BR /&gt;&lt;DEFUNCT&gt; "sed -n 18p 850stsegs0001"&lt;BR /&gt;&lt;BR /&gt;Mott Given&lt;BR /&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;</description>
      <pubDate>Fri, 11 Apr 2003 12:09:39 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949218#M931594</guid>
      <dc:creator>Mott Given</dc:creator>
      <dc:date>2003-04-11T12:09:39Z</dc:date>
    </item>
    <item>
      <title>Re: Defunct processes using all the CPU</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949219#M931595</link>
      <description>Hi Mott,&lt;BR /&gt;&lt;BR /&gt;I've seen similar situations where defunct processes were consuming massive amounts of cpu time.  At this point, you need to get rid of them.  If they have a parent PID of 1, the only way to clear them out is to reboot.  If the PPID is other than 1, you can try killing the parent.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Pete</description>
      <pubDate>Fri, 11 Apr 2003 12:12:06 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949219#M931595</guid>
      <dc:creator>Pete Randall</dc:creator>
      <dc:date>2003-04-11T12:12:06Z</dc:date>
    </item>
    <item>
      <title>Re: Defunct processes using all the CPU</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949220#M931596</link>
      <description>Can you kill those process??&lt;BR /&gt;&lt;BR /&gt; ps -ef | grep defunct  would give you list of defunct process and most of them should be having a ? and 1 under ppid showing as a runaway process. &lt;BR /&gt;&lt;BR /&gt;Kill the process and see it makes any difference.&lt;BR /&gt;The following command would show you top 10 cpu intensive process&lt;BR /&gt;UNIX95= ps -e -o "vsz pcpu ruser pid stime time state args" | sort -rn |head -10&lt;BR /&gt;&lt;BR /&gt;Goodluck,&lt;BR /&gt;-USA..&lt;BR /&gt;</description>
      <pubDate>Fri, 11 Apr 2003 12:23:25 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/defunct-processes-using-all-the-cpu/m-p/2949220#M931596</guid>
      <dc:creator>Uday_S_Ankolekar</dc:creator>
      <dc:date>2003-04-11T12:23:25Z</dc:date>
    </item>
  </channel>
</rss>

