<?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: Automatically killing looping _orasrv process in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129060#M674622</link>
    <description>You could also renice the process so it still runs but others can get the CPU: renice -n 20 PID</description>
    <pubDate>Thu, 11 Sep 2008 14:16:28 GMT</pubDate>
    <dc:creator>Dennis Handly</dc:creator>
    <dc:date>2008-09-11T14:16:28Z</dc:date>
    <item>
      <title>Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129054#M674616</link>
      <description>Greetings all,&lt;BR /&gt;&lt;BR /&gt;Have a PHB who wants me to write a script to automatically kill off a "_orasrv" process which has utilised more than [X]% CPU, as its "obviously looping" at that point.&lt;BR /&gt;&lt;BR /&gt;Any idea how I can convince him otherwise without the use of firearms? Has anyone ever effectively monitored oracle server process for loop conditions?&lt;BR /&gt;&lt;BR /&gt;Share and Enjoy! Ian</description>
      <pubDate>Wed, 10 Sep 2008 08:57:52 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129054#M674616</guid>
      <dc:creator>Ian Dennison_1</dc:creator>
      <dc:date>2008-09-10T08:57:52Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129055#M674617</link>
      <description>It's probably hopeless, but you might try&lt;BR /&gt;asking how to tell the difference between a&lt;BR /&gt;process using X% of the CPU productively,&lt;BR /&gt;and a process wasting X% of the CPU.  It's&lt;BR /&gt;hard to automate a procedure which can't be&lt;BR /&gt;defined.&lt;BR /&gt;&lt;BR /&gt;A debugger might let you trace the program&lt;BR /&gt;execution, but how to distinguish a useful&lt;BR /&gt;loop from a useless loop might still be&lt;BR /&gt;difficult.&lt;BR /&gt;&lt;BR /&gt;Or, you could just write a procedure to look&lt;BR /&gt;at, say, "top" output, and whack the thing&lt;BR /&gt;when it gets high.  Then wait for the&lt;BR /&gt;complaints to roll in.</description>
      <pubDate>Wed, 10 Sep 2008 14:14:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129055#M674617</guid>
      <dc:creator>Steven Schweda</dc:creator>
      <dc:date>2008-09-10T14:14:38Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129056#M674618</link>
      <description>If Oracle is looping then a very bad thing to do is to kill it -- you will likely corrupt the database. It is trivial to create very bad SQL code that will cause Oracle to consume CPU cycles. Is it in a meaningless loop? No. It is doing what it was told to do. If the data is worthless, then by all means, kill off processes that take too much time. But how do you define "too much time"? Then plan on how to restore the corrupted database files.&lt;BR /&gt; &lt;BR /&gt;It is impossible to determine that Oracle is in an endless loop without Oracle debug tools.</description>
      <pubDate>Wed, 10 Sep 2008 15:20:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129056#M674618</guid>
      <dc:creator>Bill Hassell</dc:creator>
      <dc:date>2008-09-10T15:20:15Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129057#M674619</link>
      <description>usual suggestion for such situations is&lt;BR /&gt;&lt;BR /&gt;1) get it in writing&lt;BR /&gt;2) respond that this is a bad idea, for all the reasons noted.  copy dba (if other than PHB)&lt;BR /&gt;&lt;BR /&gt;if forced:&lt;BR /&gt;3) implement, preferably kill -9&lt;BR /&gt;4) arrange to go on vacation...&lt;BR /&gt;&lt;BR /&gt;watch the fun insue.&lt;BR /&gt;&lt;BR /&gt;BTW: I really don't recommend the above, but....</description>
      <pubDate>Wed, 10 Sep 2008 17:01:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129057#M674619</guid>
      <dc:creator>OldSchool</dc:creator>
      <dc:date>2008-09-10T17:01:55Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129058#M674620</link>
      <description>Hi Ian:&lt;BR /&gt;&lt;BR /&gt;For your sake, if you are forced into a 'kill', do so with increasing prejudice.  Begin with a 'kill -hup'; escalate to a 'kill -term' [the default]; and only as a *last* restort, use 'kill -9'.  You certainly don't want orphaned shared memory segments lingering around because you didn't give your process a chance to cleanup.&lt;BR /&gt;&lt;BR /&gt;This all aside, Bill's comments are very germane.&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt;&lt;BR /&gt;...JRF...</description>
      <pubDate>Wed, 10 Sep 2008 17:09:16 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129058#M674620</guid>
      <dc:creator>James R. Ferguson</dc:creator>
      <dc:date>2008-09-10T17:09:16Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129059#M674621</link>
      <description>Thanks everyone for their feedback. I will definitely get it in writing and include the escalating "kill" values. "Holiday" - whats that?&lt;BR /&gt;&lt;BR /&gt;Share and enjoy! Ian</description>
      <pubDate>Thu, 11 Sep 2008 06:41:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129059#M674621</guid>
      <dc:creator>Ian Dennison_1</dc:creator>
      <dc:date>2008-09-11T06:41:40Z</dc:date>
    </item>
    <item>
      <title>Re: Automatically killing looping _orasrv process</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129060#M674622</link>
      <description>You could also renice the process so it still runs but others can get the CPU: renice -n 20 PID</description>
      <pubDate>Thu, 11 Sep 2008 14:16:28 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/automatically-killing-looping-orasrv-process/m-p/5129060#M674622</guid>
      <dc:creator>Dennis Handly</dc:creator>
      <dc:date>2008-09-11T14:16:28Z</dc:date>
    </item>
  </channel>
</rss>

