<?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: Cluster setup in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341871#M686286</link>
    <description>For 3 nodes with to packages i prefer using asl failover policy min_package_node:&lt;BR /&gt;&lt;BR /&gt;# "FAILOVER_POLICY" is the policy to be applied when package fails.&lt;BR /&gt;# The alternative policy is "MIN_PACKAGE_NODE". This policy means&lt;BR /&gt;# Serviceguard will select from the list of "NODE_NAME" entries the&lt;BR /&gt;# node, which is running fewest packages when this package needs to&lt;BR /&gt;# start.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
    <pubDate>Wed, 21 Jan 2009 13:22:00 GMT</pubDate>
    <dc:creator>likid0</dc:creator>
    <dc:date>2009-01-21T13:22:00Z</dc:date>
    <item>
      <title>Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341867#M686282</link>
      <description>We have 3 servers node1, node2 and node3.&lt;BR /&gt;I want to create a scenario like:&lt;BR /&gt;If node2 fails, package app1 should failover to node3 and then package app2 running on node3 should move to failover server node1.&lt;BR /&gt;What are the things I need to consider for this setup?&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;Srini</description>
      <pubDate>Wed, 21 Jan 2009 12:14:29 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341867#M686282</guid>
      <dc:creator>Srinikalyan</dc:creator>
      <dc:date>2009-01-21T12:14:29Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341868#M686283</link>
      <description>Config package app1 to use nodes 2 and 3. After failing on node 2 - it will automatically switch to 3. &lt;BR /&gt;For package app2 use the following - in the control script of the package app1 you can add some checks - is package app2 is tunning on the same node, move this package to another (1 for example).&lt;BR /&gt;&lt;BR /&gt;See package failover configuration - &lt;A href="http://docs.hp.com/en/B3936-90105/ch03s03.html" target="_blank"&gt;http://docs.hp.com/en/B3936-90105/ch03s03.html&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;regards,&lt;BR /&gt;ivan</description>
      <pubDate>Wed, 21 Jan 2009 12:39:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341868#M686283</guid>
      <dc:creator>Ivan Krastev</dc:creator>
      <dc:date>2009-01-21T12:39:38Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341869#M686284</link>
      <description>yes u can do this .. &lt;BR /&gt;&lt;BR /&gt;the basic idea is to check if the package app1 has failed over from node2 that is not running at node 2, if so then enable pkg app1 to run on node3 and if the package app1 is running on node3 then halt the package app2 if running on node3 , enable that to start on node1 and start on node1.&lt;BR /&gt;&lt;BR /&gt;this intelligence can be set into the package app2, yes , u can see the following link for this..&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1301744" target="_blank"&gt;http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1301744&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;see what Rita Workman and Stephen Doud have to say...&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;what packages is the node1 running?&lt;BR /&gt;&lt;BR /&gt;what do u wish in the event of node 1 failure ?&lt;BR /&gt;&lt;BR /&gt;anf failures egarding if the package aap1 fails when running on node3 that is if node3 fails ?&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;regards&lt;BR /&gt;Sujit</description>
      <pubDate>Wed, 21 Jan 2009 12:41:16 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341869#M686284</guid>
      <dc:creator>sujit kumar singh</dc:creator>
      <dc:date>2009-01-21T12:41:16Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341870#M686285</link>
      <description>Ivan and Sujit,&lt;BR /&gt;Thanks for the suggestion.&lt;BR /&gt;Nodel is merely a failover node. This node is used only in case of any failure of the primary nodes.&lt;BR /&gt;</description>
      <pubDate>Wed, 21 Jan 2009 13:12:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341870#M686285</guid>
      <dc:creator>Srinikalyan</dc:creator>
      <dc:date>2009-01-21T13:12:38Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341871#M686286</link>
      <description>For 3 nodes with to packages i prefer using asl failover policy min_package_node:&lt;BR /&gt;&lt;BR /&gt;# "FAILOVER_POLICY" is the policy to be applied when package fails.&lt;BR /&gt;# The alternative policy is "MIN_PACKAGE_NODE". This policy means&lt;BR /&gt;# Serviceguard will select from the list of "NODE_NAME" entries the&lt;BR /&gt;# node, which is running fewest packages when this package needs to&lt;BR /&gt;# start.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Wed, 21 Jan 2009 13:22:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341871#M686286</guid>
      <dc:creator>likid0</dc:creator>
      <dc:date>2009-01-21T13:22:00Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341872#M686287</link>
      <description>Hi,&lt;BR /&gt;In my requirement, do we need hearbeat connections for the three nodes ?&lt;BR /&gt;Thanks,&lt;BR /&gt;Srini</description>
      <pubDate>Wed, 21 Jan 2009 15:05:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341872#M686287</guid>
      <dc:creator>Srinikalyan</dc:creator>
      <dc:date>2009-01-21T15:05:37Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341873#M686288</link>
      <description>Do I need the heartbeat connections between the three nodes for the above mentioned scenario?&lt;BR /&gt;Thanks,&lt;BR /&gt;Srini</description>
      <pubDate>Thu, 22 Jan 2009 11:54:12 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341873#M686288</guid>
      <dc:creator>Srinikalyan</dc:creator>
      <dc:date>2009-01-22T11:54:12Z</dc:date>
    </item>
    <item>
      <title>Re: Cluster setup</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341874#M686289</link>
      <description>hi Srini&lt;BR /&gt;&lt;BR /&gt;u must have the heartbeat to all the nodes in the cluster as that is how the cluster manager node understands that the node is alive when it receives the heart beat signals from the nodes.&lt;BR /&gt;&lt;BR /&gt;if u configure a node or try to config a node for which u have not heartbeat defined, cmcheckconf and cmapplyconf shall simply throw u warnings and errors regarding the configuration.&lt;BR /&gt;&lt;BR /&gt;so definitely u should have a bridged heartbeat network onfigured and defined for all the three nodes and also that with the heartbeat failovers.&lt;BR /&gt;&lt;BR /&gt;regards&lt;BR /&gt;sujit</description>
      <pubDate>Thu, 22 Jan 2009 12:29:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-setup/m-p/4341874#M686289</guid>
      <dc:creator>sujit kumar singh</dc:creator>
      <dc:date>2009-01-22T12:29:51Z</dc:date>
    </item>
  </channel>
</rss>

