<?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: 4 node Cluster without cluster lock in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4833967#M657814</link>
    <description>&lt;P&gt;the attached should help answer your questions&lt;/P&gt;</description>
    <pubDate>Wed, 20 Jul 2011 14:31:05 GMT</pubDate>
    <dc:creator>donna hofmeister</dc:creator>
    <dc:date>2011-07-20T14:31:05Z</dc:date>
    <item>
      <title>4 node Cluster without cluster lock</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4833281#M657813</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;I have found several articles about this configuration,4 cluster node without cluster lock with redudant heartbeat connections between them.&lt;/P&gt;&lt;P&gt;My questions are:&lt;/P&gt;&lt;P&gt;If 2 nodes goes down , toc shutdown or init 0, the other 2 node (50%)&amp;nbsp; remain up or goes in toc all cluster ?&lt;/P&gt;&lt;P&gt;During a complete startup at leat two nodes have to do up and running to start the cluster ?&lt;/P&gt;&lt;P&gt;Thanks&lt;/P&gt;</description>
      <pubDate>Wed, 20 Jul 2011 08:10:57 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4833281#M657813</guid>
      <dc:creator>CRbosupportohp</dc:creator>
      <dc:date>2011-07-20T08:10:57Z</dc:date>
    </item>
    <item>
      <title>Re: 4 node Cluster without cluster lock</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4833967#M657814</link>
      <description>&lt;P&gt;the attached should help answer your questions&lt;/P&gt;</description>
      <pubDate>Wed, 20 Jul 2011 14:31:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4833967#M657814</guid>
      <dc:creator>donna hofmeister</dc:creator>
      <dc:date>2011-07-20T14:31:05Z</dc:date>
    </item>
    <item>
      <title>Re: 4 node Cluster without cluster lock</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4834383#M657815</link>
      <description>&lt;P&gt;Cluster arbitration (cluster lock, lock LUN or Quorum server) is REQUIRED for a split-brain situation.&amp;nbsp; If an arbitration device is not available, the remaining functioning nodes will TOC (memory dump and reboot).&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;&lt;STRONG&gt;&lt;U&gt;Quorum rules and cluster membership&lt;/U&gt;&lt;/STRONG&gt;&lt;BR /&gt;Serviceguard response to an abrupt departure of a node is determined by&lt;BR /&gt;quorum rules.&amp;nbsp; An abrupt departure can be the result of HB network failure,&lt;BR /&gt;or reboot/panic/power-failure.&lt;BR /&gt;&lt;BR /&gt;&lt;U&gt;Greater-than-50% of quorum (&amp;gt;50%)&lt;/U&gt;&lt;BR /&gt;If the number of nodes remaining in HB contact with one another comprise&lt;BR /&gt;more than 50% of the original membership, these nodes automatically reform&lt;BR /&gt;a cluster.&amp;nbsp; Packages that were orphaned as a result of the reformation are&lt;BR /&gt;adopted by the next-in-line active node if package "global" (AUTO_RUN) and&lt;BR /&gt;"node switching" are enabled for that package and node.&lt;BR /&gt;&lt;BR /&gt;&lt;U&gt;Equal-to-50% of quorum (=50%)&lt;/U&gt;&lt;BR /&gt;If a HB protocol failure occurs between half of the cluster membership and&lt;BR /&gt;the other half, cluster arbitration(6) is required to prevent split-brain&lt;BR /&gt;clusters.&amp;nbsp; The nodes seek out the arbitration device to gain authorization to&lt;BR /&gt;form a new cluster.&amp;nbsp; The first half to contact the arbitration device and&lt;BR /&gt;receive authorization reforms a reduced cluster.&lt;/P&gt;&lt;P&gt;If the other half of the cluster membership is still active, but late in&lt;BR /&gt;addressing the arbitration device, it is forced to TOC (reboot) to preserve&lt;BR /&gt;data integrity (it's packages could possibly be adopted by the reforming&lt;BR /&gt;sub-cluster).&lt;BR /&gt;&lt;BR /&gt;&lt;U&gt;Less-than-50% of quorum (&amp;lt;50%)&lt;/U&gt;&lt;BR /&gt;A node or nodes that find themselves in a minority of the original cluster&lt;BR /&gt;membership are forced to TOC to preserve data integrity (see previous&lt;BR /&gt;paragraph).&amp;nbsp;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;</description>
      <pubDate>Wed, 20 Jul 2011 18:42:30 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4834383#M657815</guid>
      <dc:creator>Stephen Doud</dc:creator>
      <dc:date>2011-07-20T18:42:30Z</dc:date>
    </item>
    <item>
      <title>Re: 4 node Cluster without cluster lock</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4834437#M657816</link>
      <description>&lt;P&gt;Shalom,&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;With a properly set up quorum server, at least one node should not TOC.&amp;nbsp; In the event of a total loss of heartbeat, the quorum server decides which nodes TOC and which nodes stay up.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;It is certainly possible for unforseen events to cause all nodes to boot, but unlikely.&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;SEP&lt;/P&gt;</description>
      <pubDate>Wed, 20 Jul 2011 19:25:11 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/4-node-cluster-without-cluster-lock/m-p/4834437#M657816</guid>
      <dc:creator>Steven E. Protter</dc:creator>
      <dc:date>2011-07-20T19:25:11Z</dc:date>
    </item>
  </channel>
</rss>

