<?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/"
	xmlns:georss="http://www.georss.org/georss" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#" xmlns:media="http://search.yahoo.com/mrss/"
		>
<channel>
	<title>Comments on: Upgrade Argh</title>
	<atom:link href="http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/feed/" rel="self" type="application/rss+xml" />
	<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/</link>
	<description>Just another Oracle weblog</description>
	<lastBuildDate>Sat, 18 May 2013 11:04:10 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.com/</generator>
	<item>
		<title>By: Jonathan Lewis</title>
		<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/#comment-41867</link>
		<dc:creator><![CDATA[Jonathan Lewis]]></dc:creator>
		<pubDate>Tue, 27 Sep 2011 22:53:38 +0000</pubDate>
		<guid isPermaLink="false">http://jonathanlewis.wordpress.com/?p=7274#comment-41867</guid>
		<description><![CDATA[Just landed and got to the hotel.
Strangely the problem doesn&#039;t apply to the recycle cache, or if you use the default cache with the nocache option.]]></description>
		<content:encoded><![CDATA[<p>Just landed and got to the hotel.<br />
Strangely the problem doesn&#8217;t apply to the recycle cache, or if you use the default cache with the nocache option.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jonathan Lewis</title>
		<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/#comment-41854</link>
		<dc:creator><![CDATA[Jonathan Lewis]]></dc:creator>
		<pubDate>Tue, 27 Sep 2011 08:21:10 +0000</pubDate>
		<guid isPermaLink="false">http://jonathanlewis.wordpress.com/?p=7274#comment-41854</guid>
		<description><![CDATA[We have to change parameters some time - otherwise we don&#039;t get all the &quot;new, improved&quot; bits.
The problems appear when your pre-launch testing doesn&#039;t happen to pick up the issue.
This particular extreme case was during a materialized view refresh - and it&#039;s easy for a job like that to not be run through a test suite (think of the number of times OTN has seen requests from people who discovered too late the side effects of the &#039;atomic&#039; refresh kicking in on a process that had worked perfectly for years).]]></description>
		<content:encoded><![CDATA[<p>We have to change parameters some time &#8211; otherwise we don&#8217;t get all the &#8220;new, improved&#8221; bits.<br />
The problems appear when your pre-launch testing doesn&#8217;t happen to pick up the issue.<br />
This particular extreme case was during a materialized view refresh &#8211; and it&#8217;s easy for a job like that to not be run through a test suite (think of the number of times OTN has seen requests from people who discovered too late the side effects of the &#8216;atomic&#8217; refresh kicking in on a process that had worked perfectly for years).</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Jonathan Lewis</title>
		<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/#comment-41853</link>
		<dc:creator><![CDATA[Jonathan Lewis]]></dc:creator>
		<pubDate>Tue, 27 Sep 2011 08:18:35 +0000</pubDate>
		<guid isPermaLink="false">http://jonathanlewis.wordpress.com/?p=7274#comment-41853</guid>
		<description><![CDATA[It seems as if Oracle used serial direct path read because the tables were assigned to a non-existent keep cache.
I didn&#039;t check to see what would happen if they were defined as NOCACHE (or if they were assigned to a non-existent recycle cache).
I&#039;ve got my laptop with me and a 13 hour journey ahead of me, so I might check that out - but I&#039;m sure there are plenty of readers who will beat me to it before I land.]]></description>
		<content:encoded><![CDATA[<p>It seems as if Oracle used serial direct path read because the tables were assigned to a non-existent keep cache.<br />
I didn&#8217;t check to see what would happen if they were defined as NOCACHE (or if they were assigned to a non-existent recycle cache).<br />
I&#8217;ve got my laptop with me and a 13 hour journey ahead of me, so I might check that out &#8211; but I&#8217;m sure there are plenty of readers who will beat me to it before I land.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Hemant K Chitale</title>
		<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/#comment-41851</link>
		<dc:creator><![CDATA[Hemant K Chitale]]></dc:creator>
		<pubDate>Tue, 27 Sep 2011 08:00:50 +0000</pubDate>
		<guid isPermaLink="false">http://jonathanlewis.wordpress.com/?p=7274#comment-41851</guid>
		<description><![CDATA[Excuse my obtuseness... Oracle did used Serial Direct Path Read for these tables not because they were &quot;sufficiently large&quot;  but because they were &quot;not to be cached&quot; ?  Would this behaviour be similar if we set &quot;NOCACHE&quot; for these tables ?]]></description>
		<content:encoded><![CDATA[<p>Excuse my obtuseness&#8230; Oracle did used Serial Direct Path Read for these tables not because they were &#8220;sufficiently large&#8221;  but because they were &#8220;not to be cached&#8221; ?  Would this behaviour be similar if we set &#8220;NOCACHE&#8221; for these tables ?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: laimis (lnd)</title>
		<link>http://jonathanlewis.wordpress.com/2011/09/26/upgrade-argh/#comment-41850</link>
		<dc:creator><![CDATA[laimis (lnd)]]></dc:creator>
		<pubDate>Tue, 27 Sep 2011 06:42:50 +0000</pubDate>
		<guid isPermaLink="false">http://jonathanlewis.wordpress.com/?p=7274#comment-41850</guid>
		<description><![CDATA[That´s a good one. The &quot;Top N&quot; thought is to exclaim why one trusts automated things in the first Oracle release ?
But the next thought is: why were parameters changed during upgrade in the first place ? Just to give a try ?
Discarding the established methods results in well hidden issues which require ingenious minds like that of Jonathan. Which is good for him and for other DBA&#039;s but really shatters confidence in Oracle...]]></description>
		<content:encoded><![CDATA[<p>That´s a good one. The &#8220;Top N&#8221; thought is to exclaim why one trusts automated things in the first Oracle release ?<br />
But the next thought is: why were parameters changed during upgrade in the first place ? Just to give a try ?<br />
Discarding the established methods results in well hidden issues which require ingenious minds like that of Jonathan. Which is good for him and for other DBA&#8217;s but really shatters confidence in Oracle&#8230;</p>
]]></content:encoded>
	</item>
</channel>
</rss>
