<?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: Performance Issues in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783969#M263672</link>
    <description>Thanks for ur reply.  &lt;BR /&gt;&lt;BR /&gt;Post _rate processis utilizing most of CPU power and that what is expected.  &lt;BR /&gt;&lt;BR /&gt;3-4 Days before we were running four strams of post_ rate and each was utilizing 80-90%   However now suddenly I dont know what has happened and it is utilizing only 40-50 %.&lt;BR /&gt;&lt;BR /&gt;I am sending you the output of the commands u have given. But still I am unable to get where my cpu power is going.  Kindly help me on this</description>
    <pubDate>Mon, 08 May 2006 01:34:29 GMT</pubDate>
    <dc:creator>xcvzxvzxdv</dc:creator>
    <dc:date>2006-05-08T01:34:29Z</dc:date>
    <item>
      <title>Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783966#M263669</link>
      <description>I need help on performance issues that are coming in my server.  When I issue top command the systems in 0.0 % idle.  Howver if I see at the processes the CPU consumption is less than what I is it showing in avg of TOP.  I am sending the output of TOP command.  &lt;BR /&gt;&lt;BR /&gt;I am not getting which processes are utilizing the CPU power and that are not getting displayed even with TOP command</description>
      <pubDate>Mon, 08 May 2006 00:58:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783966#M263669</guid>
      <dc:creator>xcvzxvzxdv</dc:creator>
      <dc:date>2006-05-08T00:58:38Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783967#M263670</link>
      <description>Hi, &lt;BR /&gt;&lt;BR /&gt;Sometimes, top doesnt help you. You can try the following, &lt;BR /&gt;&lt;BR /&gt;1) UNIX95= ps -ef -o cpu,pcpu,pid,comm&lt;BR /&gt;&lt;BR /&gt;Execute this and see what is happening in your system. &lt;BR /&gt;&lt;BR /&gt;2) &lt;A href="http://www.hpux.ws/system.perf.sh" target="_blank"&gt;http://www.hpux.ws/system.perf.sh&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;-Arun</description>
      <pubDate>Mon, 08 May 2006 01:10:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783967#M263670</guid>
      <dc:creator>Arunvijai_4</dc:creator>
      <dc:date>2006-05-08T01:10:37Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783968#M263671</link>
      <description>hi,&lt;BR /&gt;&lt;BR /&gt;from the top output, I think the "post_rate" used most of the CPU esource....&lt;BR /&gt;&lt;BR /&gt;u can also use ps to list the %CPU and sort it.&lt;BR /&gt;UNIX95= ps -ef -opcpu,pid,ppid,ruser,args|sort -k1&lt;BR /&gt;&lt;BR /&gt;GOOD LUCK!!</description>
      <pubDate>Mon, 08 May 2006 01:16:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783968#M263671</guid>
      <dc:creator>Warren_9</dc:creator>
      <dc:date>2006-05-08T01:16:05Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783969#M263672</link>
      <description>Thanks for ur reply.  &lt;BR /&gt;&lt;BR /&gt;Post _rate processis utilizing most of CPU power and that what is expected.  &lt;BR /&gt;&lt;BR /&gt;3-4 Days before we were running four strams of post_ rate and each was utilizing 80-90%   However now suddenly I dont know what has happened and it is utilizing only 40-50 %.&lt;BR /&gt;&lt;BR /&gt;I am sending you the output of the commands u have given. But still I am unable to get where my cpu power is going.  Kindly help me on this</description>
      <pubDate>Mon, 08 May 2006 01:34:29 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783969#M263672</guid>
      <dc:creator>xcvzxvzxdv</dc:creator>
      <dc:date>2006-05-08T01:34:29Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783970#M263673</link>
      <description>It seems that these process where cpu hungry,&lt;BR /&gt;I presume they run on Oracle 9i.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;19.89 14337 12105 bsnlprod post_rate SWITCH=MH PERSPECTIVE=INCOM&lt;BR /&gt;24.00  2529  1528 bsnlprod post_rate PERSPECTIVE=INCOM SWITCH=MPCHG&lt;BR /&gt;28.64 14108 14103 root     compress /user_sun/oracle9i/oradata/iobasprd/archive/archive_30&lt;BR /&gt;29.28 11243     1 oracle9i oracleiobasprd (LOCAL=NO)&lt;BR /&gt;42.90  2340  1528 bsnlprod post_rate PERSPECTIVE=INCOM SWITCH=GJ&lt;BR /&gt;54.02  1775  1528 bsnlprod post_rate PERSPECTIVE=INCOM SWITCH=WTR&lt;BR /&gt;&lt;BR /&gt;Try to find out if the are doing something on the OS or if they are doing something in &lt;BR /&gt;Oracle.&lt;BR /&gt;&lt;BR /&gt;Maybe you can check in the v$sqlarea&lt;BR /&gt;to find the number of blocks read/write &lt;BR /&gt;for these processses :&lt;BR /&gt;&lt;BR /&gt;select s.username, s.sid, s.machine, s.osuser, n.statistic#,&lt;BR /&gt;substr(s.program,1,25),&lt;BR /&gt;t.value "Aantal reads - writes"&lt;BR /&gt;from v$session s, v$sesstat t, v$statname n&lt;BR /&gt;where s.sid = &amp;amp;SID_FROM_PID&lt;BR /&gt;and s.sid = t.sid&lt;BR /&gt;and t.statistic# = n.statistic#&lt;BR /&gt;and n.name like '%physical%'&lt;BR /&gt;and not s.username is null&lt;BR /&gt;&lt;BR /&gt;You can find &amp;amp;SID_FROM_PID,&lt;BR /&gt;by querying v$process against v$session first.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 08 May 2006 01:52:11 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783970#M263673</guid>
      <dc:creator>Frank de Vries</dc:creator>
      <dc:date>2006-05-08T01:52:11Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783971#M263674</link>
      <description>yes we have oracle 9i installed , and post rate is the process which pusses the data in the Oracle 9i.</description>
      <pubDate>Mon, 08 May 2006 02:23:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783971#M263674</guid>
      <dc:creator>xcvzxvzxdv</dc:creator>
      <dc:date>2006-05-08T02:23:38Z</dc:date>
    </item>
    <item>
      <title>Re: Performance Issues</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783972#M263675</link>
      <description>hi,&lt;BR /&gt;&lt;BR /&gt;as you can see from the top, the nice value of the post_rate is 24, therefore those processes have lower priority than others.&lt;BR /&gt;&lt;BR /&gt;the %CPU drop may due to other processes which have higher priority... like the "oracleiobasprd"&lt;BR /&gt;&lt;BR /&gt;reset the nice value of the post_rate may help.&lt;BR /&gt;&lt;BR /&gt;GOOD LUCK!!</description>
      <pubDate>Mon, 08 May 2006 02:38:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/performance-issues/m-p/3783972#M263675</guid>
      <dc:creator>Warren_9</dc:creator>
      <dc:date>2006-05-08T02:38:44Z</dc:date>
    </item>
  </channel>
</rss>

