<?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: kill ghost processes in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481540#M213440</link>
    <description>If these are zombies &lt;DEFUNCT&gt; processes then nothing other than a reboot will remove them but again, they do no harm. You might try a shutdown abort on your database but that is about as drastic as a kill -9. Kill -9 is one of those things that should be used as a weapon of last resort because it is the very thing that can cause the kinds of problems that you are seeing. When a process receives a SIGKILL (kill -9) absolutely no cleanup (detaching shared memory, removing temporary files, etc.) can be done.&lt;BR /&gt;&lt;/DEFUNCT&gt;</description>
    <pubDate>Tue, 08 Feb 2005 18:04:41 GMT</pubDate>
    <dc:creator>A. Clay Stephenson</dc:creator>
    <dc:date>2005-02-08T18:04:41Z</dc:date>
    <item>
      <title>kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481534#M213434</link>
      <description>Hi all,&lt;BR /&gt;how I can kill ghost processes?I know kill -9 but it fail...Is there a specific command for kill this processes?? I have trouble with oracle process&lt;BR /&gt;&lt;BR /&gt;Thanks!!&lt;BR /&gt;Ciao</description>
      <pubDate>Tue, 08 Feb 2005 17:30:09 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481534#M213434</guid>
      <dc:creator>DIEGO_40</dc:creator>
      <dc:date>2005-02-08T17:30:09Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481535#M213435</link>
      <description>I assume that by "ghost" processes you actually mean "zombie" processes. You can't kill them (because they are already dead and can certainly never respond to a signal which is all a kill is). Zombie's consume no resources (other than a process table slot) and do no real harm. You may be talking about a process that is waiting on an io event; these too cannot respond to a signal because they are waiting on a higher priority event --- like an io request to complete.</description>
      <pubDate>Tue, 08 Feb 2005 17:36:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481535#M213435</guid>
      <dc:creator>A. Clay Stephenson</dc:creator>
      <dc:date>2005-02-08T17:36:33Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481536#M213436</link>
      <description>Diego,&lt;BR /&gt;&lt;BR /&gt;If you  are refering to &lt;DEFUNCT&gt; processes  then you may  be  out  of luck.  To  clean these  up  you  would  need  to kill the parent proceses  so  if the parent process ID (PPID)  is  1 a  reboot  is your  only  recourse.  &lt;BR /&gt;&lt;BR /&gt;Perhaps  you  could provide  an an output of 'ps -ef'.&lt;BR /&gt;&lt;BR /&gt;Also,  'kill  -9'  should  be  discouraged.  Using it  is  often what causes  these &lt;DEFUNCT&gt; processes since it  doesn't  clean  up.&lt;BR /&gt;&lt;BR /&gt;David&lt;/DEFUNCT&gt;&lt;/DEFUNCT&gt;</description>
      <pubDate>Tue, 08 Feb 2005 17:40:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481536#M213436</guid>
      <dc:creator>David Child_1</dc:creator>
      <dc:date>2005-02-08T17:40:00Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481537#M213437</link>
      <description>Ghost processes, do you mean the ones that have a parent of 1 or defuncts. Either way, if kill -9 doesn't do it a reboot will and sometimes that's the only way.&lt;BR /&gt;&lt;BR /&gt;The Great Almighty Wizard of 1z, does a lot of cool stuff, then goes burp and becomes a nudnick. Nothing to do at that point but put 'em out to pasture. &lt;BR /&gt;&lt;BR /&gt;If the ghost processes aren't taking up any resources, let them hang out until you can schedule a reboot.&lt;BR /&gt;&lt;BR /&gt;L8tr</description>
      <pubDate>Tue, 08 Feb 2005 17:41:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481537#M213437</guid>
      <dc:creator>Robert Salter</dc:creator>
      <dc:date>2005-02-08T17:41:51Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481538#M213438</link>
      <description>When I kill them with the option -9 they still running (ps -ef|grep &lt;PID_NAME&gt;),only the day after I don't have more the problem, but in meantime I can't shutdown the oracle istance...I don't want reboot the machine (this is the last chance) because a lot of people is connected to this machine(partition of superdome). My friend who worked with me knew a method for kill this processes, is possible??&lt;BR /&gt;&lt;BR /&gt;Thanks a lot!&lt;BR /&gt;Ciao&lt;/PID_NAME&gt;</description>
      <pubDate>Tue, 08 Feb 2005 17:53:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481538#M213438</guid>
      <dc:creator>DIEGO_40</dc:creator>
      <dc:date>2005-02-08T17:53:07Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481539#M213439</link>
      <description>It sounds like  they are waiting on I/O and therefore don't  die  right  away.  Unfortunately   there is  no  way  to  kill  a process  waiting  on I/O. From what you said  it appears  the  processes  finally die so I would recommend against  using  the '-9'. Just kill it and wait for I/O to  finish.&lt;BR /&gt;&lt;BR /&gt;You  may want to  take a look at your I/O  subsystem to   see how its running.  Does  'sar 5  5'   show  high   WIO%? &lt;BR /&gt;&lt;BR /&gt;David</description>
      <pubDate>Tue, 08 Feb 2005 18:01:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481539#M213439</guid>
      <dc:creator>David Child_1</dc:creator>
      <dc:date>2005-02-08T18:01:47Z</dc:date>
    </item>
    <item>
      <title>Re: kill ghost processes</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481540#M213440</link>
      <description>If these are zombies &lt;DEFUNCT&gt; processes then nothing other than a reboot will remove them but again, they do no harm. You might try a shutdown abort on your database but that is about as drastic as a kill -9. Kill -9 is one of those things that should be used as a weapon of last resort because it is the very thing that can cause the kinds of problems that you are seeing. When a process receives a SIGKILL (kill -9) absolutely no cleanup (detaching shared memory, removing temporary files, etc.) can be done.&lt;BR /&gt;&lt;/DEFUNCT&gt;</description>
      <pubDate>Tue, 08 Feb 2005 18:04:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/kill-ghost-processes/m-p/3481540#M213440</guid>
      <dc:creator>A. Clay Stephenson</dc:creator>
      <dc:date>2005-02-08T18:04:41Z</dc:date>
    </item>
  </channel>
</rss>

