Sophie

Sophie

distrib > Fedora > 13 > i386 > by-pkgid > fd1445e4ec50ea0254985a386bb83a07 > files > 263

pacemaker-doc-1.1.1-1.fc13.i686.rpm

<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /><title>9.3.2. Moving Resources Due to Failure</title><link rel="stylesheet" href="./Common_Content/css/default.css" type="text/css" /><meta name="generator" content="publican 1.6" /><meta name="package" content="Pacemaker-Pacemaker_Explained-1.1-en-US-1-0" /><link rel="home" href="index.html" title="Configuration Explained" /><link rel="up" href="s-moving-resource.html" title="9.3. Moving Resources" /><link rel="prev" href="s-moving-resource.html" title="9.3. Moving Resources" /><link rel="next" href="ch09s03s03.html" title="9.3.3. Moving Resources Due to Connectivity Changes" /></head><body class=""><p id="title"><a class="left" href="https://fedorahosted.org/publican"><img src="Common_Content/images/image_left.png" alt="Product Site" /></a><a class="right" href="https://fedorahosted.org/publican"><img src="Common_Content/images/image_right.png" alt="Documentation Site" /></a></p><ul class="docnav"><li class="previous"><a accesskey="p" href="s-moving-resource.html"><strong>Prev</strong></a></li><li class="next"><a accesskey="n" href="ch09s03s03.html"><strong>Next</strong></a></li></ul><div class="section" title="9.3.2. Moving Resources Due to Failure"><div class="titlepage"><div><div><h3 class="title" id="s-failure-migration">9.3.2. Moving Resources Due to Failure</h3></div></div></div><div class="para">
				New in 1.0 is the concept of a migration threshold <sup>[<a id="id2724469" href="#ftn.id2724469" class="footnote">9</a>]</sup>. Simply define <code class="literal">migration-threshold=N</code> for a resource and it will migrate to a new node after N failures. There is no threshold defined by default. To determine the resource's current failure status and limits, use <code class="command">crm_mon --failcounts</code>
			</div><div class="para">
				By default, once the threshold has been reached, node will no longer be allowed to run the failed resource until the administrator manually resets the resource's failcount using <code class="command">crm_failcount</code> (after hopefully first fixing the failure's cause). However it is possible to expire them by setting the resource's <code class="literal">failure-timeout</code> option.
			</div><div class="para">
				So a setting of <code class="literal">migration-threshold=2</code> and <code class="literal">failure-timeout=60s</code> would cause the resource to move to a new node after 2 failures and potentially allow it to move back (depending on the stickiness and constraint scores) after one minute.
			</div><div class="para">
				There are two exceptions to the migration threshold concept and occur when a resource either fails to start or fails to stop. Start failures cause the failcount to be set to <code class="literal">INFINITY</code> and thus always cause the resource to move immediately.
			</div><div class="para">
				Stop failures are slightly different and crucial. If a resource fails to stop and STONITH is enabled, then the cluster will fence the node in order to be able to start the resource elsewhere. If STONITH is not enabled, then the cluster has no way to continue and will not try to start the resource elsewhere, but will try to stop it again after the failure timeout.
			</div><div class="important"><h2>Important</h2><div class="para">
					Please read <a class="xref" href="s-rules-recheck.html" title="8.6. Ensuring Time Based Rules Take Effect">Section 8.6, “Ensuring Time Based Rules Take Effect”</a> before enabling this option.
				</div></div><div class="footnotes"><br /><hr /><div class="footnote"><p><sup>[<a id="ftn.id2724469" href="#id2724469" class="para">9</a>] </sup>
					The naming of this option was unfortunate as it is easily confused with true migration, the process of moving a resource from one node to another without stopping it. Xen virtual guests are the most common example of resources that can be migrated in this manner.
				</p></div></div></div><ul class="docnav"><li class="previous"><a accesskey="p" href="s-moving-resource.html"><strong>Prev</strong>9.3. Moving Resources</a></li><li class="up"><a accesskey="u" href="#"><strong>Up</strong></a></li><li class="home"><a accesskey="h" href="index.html"><strong>Home</strong></a></li><li class="next"><a accesskey="n" href="ch09s03s03.html"><strong>Next</strong>9.3.3. Moving Resources Due to Connectivity Chang...</a></li></ul></body></html>