<?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: Oracle trace in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880002#M774052</link>
    <description>I did lsof to find the remote pid. It is a java process on the app server.</description>
    <pubDate>Fri, 13 Oct 2006 14:51:38 GMT</pubDate>
    <dc:creator>KapilRaj</dc:creator>
    <dc:date>2006-10-13T14:51:38Z</dc:date>
    <item>
      <title>Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879997#M774047</link>
      <description>Guys,&lt;BR /&gt;&lt;BR /&gt;I have two database processes  [ LOCAL=NO ] on the production database server. lsof shows  they are from the apps server. But the apps does not know what that is :( same reply from the DBAs. These processes are consuming 20 % of CPU each. They are running for last 3 days !!. &lt;BR /&gt;There is also a file in "bdump" directory called , ${ORACLE_SID}_ora_&lt;PID&gt;.trc Does it mean something ? It looks like a coredump file though the information inside is plain text !. &lt;BR /&gt;&lt;BR /&gt;Can I kill this processes ?.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;Kaps&lt;/PID&gt;</description>
      <pubDate>Fri, 13 Oct 2006 13:12:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879997#M774047</guid>
      <dc:creator>KapilRaj</dc:creator>
      <dc:date>2006-10-13T13:12:07Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879998#M774048</link>
      <description>Hi Kaps,&lt;BR /&gt;&lt;BR /&gt;To see the underlying SQL statements for these processes, substitute your PIDs in the following statement:&lt;BR /&gt;&lt;BR /&gt;SELECT sql_text&lt;BR /&gt;FROM v$sqlarea sa, v$session s, v$process p&lt;BR /&gt;WHERE sa.address = s.sql_address&lt;BR /&gt;AND s.paddr = p.addr&lt;BR /&gt;AND p.spid = '&lt;OS_PID&gt;';&lt;BR /&gt;&lt;BR /&gt;"[LOCAL=NO]" means a process is not one of Oracle's background processes.  Therefore, it should be safe to kill them.  Just realize that this may cause unexpected results on your application server.&lt;BR /&gt;&lt;BR /&gt;The preferred order for killing an Oracle process:&lt;BR /&gt;&lt;BR /&gt;1) Close the session from the application server.&lt;BR /&gt;&lt;BR /&gt;2) SQL&amp;gt; alter system kill session 'sid,serial#';&lt;BR /&gt;&lt;BR /&gt;3) SQL&amp;gt; alter system kill session 'sid,serial#' immediate;&lt;BR /&gt;&lt;BR /&gt;4) # kill &lt;PID&gt;&lt;BR /&gt;&lt;BR /&gt;5) Usual kill switch progression (-1, -2, -3, -11, -9).&lt;BR /&gt;&lt;BR /&gt;Trace files in bdump are commonplace, and may not be related to the issue at hand.  These files are ASCII data.&lt;BR /&gt;&lt;BR /&gt;PCS&lt;/PID&gt;&lt;/OS_PID&gt;</description>
      <pubDate>Fri, 13 Oct 2006 14:10:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879998#M774048</guid>
      <dc:creator>spex</dc:creator>
      <dc:date>2006-10-13T14:10:15Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879999#M774049</link>
      <description>Having a .trc file in the bdump directory does not mean an *error ? Is that what you mean ?&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;Kaps</description>
      <pubDate>Fri, 13 Oct 2006 14:29:24 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3879999#M774049</guid>
      <dc:creator>KapilRaj</dc:creator>
      <dc:date>2006-10-13T14:29:24Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880000#M774050</link>
      <description>If you run:&lt;BR /&gt;&lt;BR /&gt;select s.process from v$session s, v$process p&lt;BR /&gt;where s.paddr = p.addr and p.spid = &lt;PID of="" local="NO" process=""&gt;&lt;BR /&gt;&lt;BR /&gt;you should get something like 1180:1023.&lt;BR /&gt;&lt;BR /&gt;The first number in that pair should be the process ID on the app server. Go to that server and see what that process is. If it is a windows box use taskman or the following command:&lt;BR /&gt;&lt;BR /&gt;tasklist -FI "PID eq 1180" (or whatever your pid is)&lt;BR /&gt;&lt;BR /&gt;HTH&lt;BR /&gt;&lt;/PID&gt;</description>
      <pubDate>Fri, 13 Oct 2006 14:35:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880000#M774050</guid>
      <dc:creator>Jonathan Fife</dc:creator>
      <dc:date>2006-10-13T14:35:07Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880001#M774051</link>
      <description>Of course a trace file means there was an error.  However, the file you see may represent a different error--unrelated to the errant application server processes.  Look at the timestamp of the file to determine when it was created.  Better yet, look at the contents.&lt;BR /&gt;&lt;BR /&gt;In general, trace files just happen.  As long as their number is small and you review them regularly, you should be fine.</description>
      <pubDate>Fri, 13 Oct 2006 14:48:52 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880001#M774051</guid>
      <dc:creator>spex</dc:creator>
      <dc:date>2006-10-13T14:48:52Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880002#M774052</link>
      <description>I did lsof to find the remote pid. It is a java process on the app server.</description>
      <pubDate>Fri, 13 Oct 2006 14:51:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880002#M774052</guid>
      <dc:creator>KapilRaj</dc:creator>
      <dc:date>2006-10-13T14:51:38Z</dc:date>
    </item>
    <item>
      <title>Re: Oracle trace</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880003#M774053</link>
      <description>hi Kaps,&lt;BR /&gt;&lt;BR /&gt;Allow me to add that in case, this happens too often you may wish to log a Service Request with Oracle.&lt;BR /&gt;&lt;BR /&gt;The trace file will also allow you to identify the culprit user. If you query your alert.log, you can also get the exact Oralce error message.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;hope this helps too!&lt;BR /&gt;&lt;BR /&gt;kind regards&lt;BR /&gt;yogeeraj</description>
      <pubDate>Sat, 14 Oct 2006 04:08:46 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oracle-trace/m-p/3880003#M774053</guid>
      <dc:creator>Yogeeraj_1</dc:creator>
      <dc:date>2006-10-14T04:08:46Z</dc:date>
    </item>
  </channel>
</rss>

