<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: Identifying shared memory segment users using lsof</title>
	<atom:link href="http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/feed/" rel="self" type="application/rss+xml" />
	<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/?utm_source=rss&#038;utm_medium=rss&#038;utm_campaign=identifying-shared-memory-segment-users-using-lsof</link>
	<description>Oracle, Exadata, Performance, Troubleshooting - Mobile Life and Productivity.</description>
	<lastBuildDate>Wed, 08 Feb 2012 08:03:50 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3.1</generator>
	<item>
		<title>By: jc nars</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-9988</link>
		<dc:creator>jc nars</dc:creator>
		<pubDate>Fri, 16 Dec 2011 14:39:26 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-9988</guid>
		<description>Hi Tanel,
Thanks much. Following the note &quot;Retrieve deleted files on Unix / Linux using File Descriptors [ID 444749.1]&quot; we did a:
head -100 /proc/16927/fd/44 &gt; /tmp/file1

Basically we had wanted to upload to Support the first few lines of the huge file...but  after a few mins, the space was back in the filesystem !

I thank you again for taking time to respond to the blog comment.</description>
		<content:encoded><![CDATA[<p>Hi Tanel,<br />
Thanks much. Following the note &#8220;Retrieve deleted files on Unix / Linux using File Descriptors [ID 444749.1]&#8221; we did a:<br />
head -100 /proc/16927/fd/44 &gt; /tmp/file1</p>
<p>Basically we had wanted to upload to Support the first few lines of the huge file&#8230;but  after a few mins, the space was back in the filesystem !</p>
<p>I thank you again for taking time to respond to the blog comment.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tanel Poder</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-9968</link>
		<dc:creator>Tanel Poder</dc:creator>
		<pubDate>Fri, 16 Dec 2011 08:20:48 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-9968</guid>
		<description>&lt;a href=&quot;#comment-9958&quot; rel=&quot;nofollow&quot;&gt;@jc nars&lt;/a&gt; 

But yes, the next time I&#039;d just truncate the file with &quot;&gt; filename.trc&quot; ... or make sure such traces aren&#039;t dumped at all...
</description>
		<content:encoded><![CDATA[<p><a href="#comment-9958" rel="nofollow">@jc nars</a> </p>
<p>But yes, the next time I&#8217;d just truncate the file with &#8220;> filename.trc&#8221; &#8230; or make sure such traces aren&#8217;t dumped at all&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tanel Poder</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-9967</link>
		<dc:creator>Tanel Poder</dc:creator>
		<pubDate>Fri, 16 Dec 2011 08:19:18 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-9967</guid>
		<description>&lt;a href=&quot;#comment-9958&quot; rel=&quot;nofollow&quot;&gt;@jc nars&lt;/a&gt; 

What you can also do is to identify SMON&#039;s spid (16927) and then:

1) ORADEBUG SETOSPID 16927
2) ORADEBUG CLOSE_TRACE

As this is SMON, I&#039;m not too comfortable sending oradebug commands to it (wouldn&#039;t want to crash it by any chance! :) but you can use this as your last option instead of restarting the instance (But do this at your own risk! :)</description>
		<content:encoded><![CDATA[<p><a href="#comment-9958" rel="nofollow">@jc nars</a> </p>
<p>What you can also do is to identify SMON&#8217;s spid (16927) and then:</p>
<p>1) ORADEBUG SETOSPID 16927<br />
2) ORADEBUG CLOSE_TRACE</p>
<p>As this is SMON, I&#8217;m not too comfortable sending oradebug commands to it (wouldn&#8217;t want to crash it by any chance! :) but you can use this as your last option instead of restarting the instance (But do this at your own risk! :)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: jc nars</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-9958</link>
		<dc:creator>jc nars</dc:creator>
		<pubDate>Fri, 16 Dec 2011 05:11:26 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-9958</guid>
		<description>Hi Tanel,
How do u do...one of the background processes SMON ran up a HUUUUUUUUUUGE trace file that just left 5% free space. And, the rookie dba rm-ed the file...instead of doing something like echo &#039;&#039;&gt; bigtracefile.

This is in the first node of Exadata X2.
Oracle Support says no other option other than to kill SMON to free up the filesystem !
I will be hanged if I request a bounce...as we just moved from V1 to X2 2 months back.

Any &#039;unsupported&#039; tricks you guys at Enkitec know of?!!
Thanks

P.S.:
lsof -p 16927 &#124; egrep &#039;^COMMAND&#124;trc&#124;trm&#039;
COMMAND   PID   USER   FD   TYPE             DEVICE        SIZE       NODE NAME
oracle  16927 oracle   44w   REG              253,2 48599035904   12880302 /u01/app/oracle/diag/rdbms/edaprd/edaprd1/trace/edaprd1_smon_16927.trc (deleted)
oracle  16927 oracle   45w   REG              253,2  3661829175   12880303 /u01/app/oracle/diag/rdbms/edaprd/edaprd1/trace/edaprd1_smon_16927.trm (deleted)</description>
		<content:encoded><![CDATA[<p>Hi Tanel,<br />
How do u do&#8230;one of the background processes SMON ran up a HUUUUUUUUUUGE trace file that just left 5% free space. And, the rookie dba rm-ed the file&#8230;instead of doing something like echo &#8221;&gt; bigtracefile.</p>
<p>This is in the first node of Exadata X2.<br />
Oracle Support says no other option other than to kill SMON to free up the filesystem !<br />
I will be hanged if I request a bounce&#8230;as we just moved from V1 to X2 2 months back.</p>
<p>Any &#8216;unsupported&#8217; tricks you guys at Enkitec know of?!!<br />
Thanks</p>
<p>P.S.:<br />
lsof -p 16927 | egrep &#8216;^COMMAND|trc|trm&#8217;<br />
COMMAND   PID   USER   FD   TYPE             DEVICE        SIZE       NODE NAME<br />
oracle  16927 oracle   44w   REG              253,2 48599035904   12880302 /u01/app/oracle/diag/rdbms/edaprd/edaprd1/trace/edaprd1_smon_16927.trc (deleted)<br />
oracle  16927 oracle   45w   REG              253,2  3661829175   12880303 /u01/app/oracle/diag/rdbms/edaprd/edaprd1/trace/edaprd1_smon_16927.trm (deleted)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Nayyar Ahmad</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-886</link>
		<dc:creator>Nayyar Ahmad</dc:creator>
		<pubDate>Fri, 30 Jan 2009 09:45:39 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-886</guid>
		<description>Hi,

i am using Solaris 9, when i ran ipcs -ma command there were several output line, though Solaris does&#039;t mention STATUS column, how can i investigate in this situation ?</description>
		<content:encoded><![CDATA[<p>Hi,</p>
<p>i am using Solaris 9, when i ran ipcs -ma command there were several output line, though Solaris does&#8217;t mention STATUS column, how can i investigate in this situation ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Tanel Poder</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-851</link>
		<dc:creator>Tanel Poder</dc:creator>
		<pubDate>Fri, 23 Jan 2009 11:50:19 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-851</guid>
		<description>Thanks Chen :)

Python was reading some interesting stuff out of there ;)

But I&#039;ve had cases where an Oracle server process fails to die during shutdown and keeps being attached to SGA...</description>
		<content:encoded><![CDATA[<p>Thanks Chen :)</p>
<p>Python was reading some interesting stuff out of there ;)</p>
<p>But I&#8217;ve had cases where an Oracle server process fails to die during shutdown and keeps being attached to SGA&#8230;</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Chen Shapira</title>
		<link>http://blog.tanelpoder.com/2009/01/22/identifying-shared-memory-segment-users-using-lsof/comment-page-1/#comment-849</link>
		<dc:creator>Chen Shapira</dc:creator>
		<pubDate>Fri, 23 Jan 2009 02:11:21 +0000</pubDate>
		<guid isPermaLink="false">http://blog.tanelpoder.com/?p=192#comment-849</guid>
		<description>Very cool demo. What was python doing in your SGA?</description>
		<content:encoded><![CDATA[<p>Very cool demo. What was python doing in your SGA?</p>
]]></content:encoded>
	</item>
</channel>
</rss>

