<?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 query in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674592#M712450</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;What do u mean by halting the primary node without halting package or cluster ?&lt;BR /&gt;&lt;BR /&gt;If your clustering is configured properly, the unavailability of primary node, will cause a package shift to the secondary.&lt;BR /&gt;&lt;BR /&gt;HTH,&lt;BR /&gt;Shiju</description>
    <pubDate>Fri, 01 Mar 2002 15:50:51 GMT</pubDate>
    <dc:creator>Helen French</dc:creator>
    <dc:date>2002-03-01T15:50:51Z</dc:date>
    <item>
      <title>cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674591#M712449</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I have a 2 node MC/SG cluster.  If I halt the primary node, without halting package or the cluster, will the package fail over to the secondary node ?&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;&lt;BR /&gt;Hilary</description>
      <pubDate>Fri, 01 Mar 2002 15:47:29 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674591#M712449</guid>
      <dc:creator>BFA6</dc:creator>
      <dc:date>2002-03-01T15:47:29Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674592#M712450</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;What do u mean by halting the primary node without halting package or cluster ?&lt;BR /&gt;&lt;BR /&gt;If your clustering is configured properly, the unavailability of primary node, will cause a package shift to the secondary.&lt;BR /&gt;&lt;BR /&gt;HTH,&lt;BR /&gt;Shiju</description>
      <pubDate>Fri, 01 Mar 2002 15:50:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674592#M712450</guid>
      <dc:creator>Helen French</dc:creator>
      <dc:date>2002-03-01T15:50:51Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674593#M712451</link>
      <description>Hi Hilary,&lt;BR /&gt;&lt;BR /&gt;If you have two node cluster and if there is a package running on node1. It will failover to the other cluster node node2 if node2 is configured as the alternate node for the package and if the node1 is shutdown or taken out of the cluster. However if package switching is disbaled for the package, the package will not failover to the other node node2. To check if the package switching is enabled on not do a cmviewcl,&lt;BR /&gt;&lt;BR /&gt;cmviewcl -v -p package_name&lt;BR /&gt;&lt;BR /&gt;Hope this helps.&lt;BR /&gt;&lt;BR /&gt;Regds&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Mar 2002 15:51:14 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674593#M712451</guid>
      <dc:creator>Sanjay_6</dc:creator>
      <dc:date>2002-03-01T15:51:14Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674594#M712452</link>
      <description>Certainly should IF pkg switching is enabled.&lt;BR /&gt;&lt;BR /&gt;Jeff</description>
      <pubDate>Fri, 01 Mar 2002 15:52:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674594#M712452</guid>
      <dc:creator>Jeff Schussele</dc:creator>
      <dc:date>2002-03-01T15:52:51Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674595#M712453</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;Package switching is enabled.&lt;BR /&gt;&lt;BR /&gt;If I do a shutdown -h 0 on the primary node, without doing anything else, will the package switch, or will the secondary node form a reform the cluster, with just itself as a member ?&lt;BR /&gt;&lt;BR /&gt;Hilary&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Mar 2002 15:58:36 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674595#M712453</guid>
      <dc:creator>BFA6</dc:creator>
      <dc:date>2002-03-01T15:58:36Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674596#M712454</link>
      <description>Hilary,&lt;BR /&gt;&lt;BR /&gt; Should do both - reform the cluster as a single node AND start the pkg on it.&lt;BR /&gt;&lt;BR /&gt;Jeff</description>
      <pubDate>Fri, 01 Mar 2002 16:01:21 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674596#M712454</guid>
      <dc:creator>Jeff Schussele</dc:creator>
      <dc:date>2002-03-01T16:01:21Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674597#M712455</link>
      <description>Hi Hilary,&lt;BR /&gt;&lt;BR /&gt;You can test the package switching by just taking the node1 out of the cluster.&lt;BR /&gt;&lt;BR /&gt;Run this on node1,&lt;BR /&gt;&lt;BR /&gt;cmhaltnode&lt;BR /&gt;&lt;BR /&gt;The node2 should form a cluster with itself as the only node in the cluster and the package that was running on node1 should start on node2. You can also try this by simply doing a shutdown on node1,&lt;BR /&gt;&lt;BR /&gt;shutdown -h now &lt;BR /&gt;or shutdown -h 0&lt;BR /&gt;&lt;BR /&gt;This will do a shutdown on node1 and then reform the cluster with only node2 and move the package running on node1 to node2.&lt;BR /&gt;&lt;BR /&gt;Hope this helps.&lt;BR /&gt;&lt;BR /&gt;Regds&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Mar 2002 16:02:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674597#M712455</guid>
      <dc:creator>Sanjay_6</dc:creator>
      <dc:date>2002-03-01T16:02:51Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674598#M712456</link>
      <description>Actually both should occur.&lt;BR /&gt;As part of the shutdowmn of the primary node, the secondary node will be informed the primary is leaving the clsuter, and a clsuter reform will take place that results in a single node cluster.&lt;BR /&gt;During the shutdown, the primary node will halt the package, and the secondary will then start the package up, PROVIDED that the package variables are enabled for both LOCAL and GLOBAL switching. Check cmviewcl to confirm&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Mar 2002 16:03:59 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674598#M712456</guid>
      <dc:creator>melvyn burnard</dc:creator>
      <dc:date>2002-03-01T16:03:59Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674599#M712457</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;Yes, it will reform the cluster and the package will shift to the secondary node (provided the cluster configuration is perfect).&lt;BR /&gt;&lt;BR /&gt;Btw, when you do the shutdown -h 0, the rc scripts will shutdown the cluster services in the primary node.&lt;BR /&gt;&lt;BR /&gt;HTH,&lt;BR /&gt;Shiju</description>
      <pubDate>Fri, 01 Mar 2002 16:04:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674599#M712457</guid>
      <dc:creator>Helen French</dc:creator>
      <dc:date>2002-03-01T16:04:33Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674600#M712458</link>
      <description>Thanks for all the replies.&lt;BR /&gt;&lt;BR /&gt;The reason for the query is that we are having a powerdown and I have issued instructions for the package &amp;amp; cluster to be halted, then both machines halted.  A colleague has suggested that this is not necessary &amp;amp; the boxes can just be halted.&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;&lt;BR /&gt;Hilary&lt;BR /&gt;</description>
      <pubDate>Fri, 01 Mar 2002 16:11:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674600#M712458</guid>
      <dc:creator>BFA6</dc:creator>
      <dc:date>2002-03-01T16:11:03Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674601#M712459</link>
      <description>well the answer to that is that you could indeed just shutdown both nodes.&lt;BR /&gt;The issue to keep in mind is that if you shutdown the primary node first, the package will start to switch, or even complete switching, and then you want to shut the second server down.&lt;BR /&gt;If you are looking at powering down BOTH nodes at the same time, it might be better to do cmhaltcl -f first to ensure the application shuts down and stays down, then halt the nodes.&lt;BR /&gt;Otherwise, issue the shutdown on the node that is NOT running the package, and after a short period then do the same on the other node.&lt;BR /&gt;This just saves the time for an unnecessary packag eswitch.&lt;BR /&gt;&lt;BR /&gt;HTH</description>
      <pubDate>Fri, 01 Mar 2002 19:00:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674601#M712459</guid>
      <dc:creator>melvyn burnard</dc:creator>
      <dc:date>2002-03-01T19:00:05Z</dc:date>
    </item>
    <item>
      <title>Re: cluster query</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674602#M712460</link>
      <description>Hi Hilary:&lt;BR /&gt;&lt;BR /&gt;Since you indicate you will be shutting down both nodes, remember that you may find on bootup that a cluster does not automatically form.  This will occur is both nodes are not booted within a 10-minute (AUTO_START_TIMEOUT) window.  If this is the case, allow the TIMEOUT to expire and manually issue a 'cmruncl'.  'cmrunnode' will not start a one-node cluster.&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt;&lt;BR /&gt;...JRF...</description>
      <pubDate>Fri, 01 Mar 2002 19:41:28 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/cluster-query/m-p/2674602#M712460</guid>
      <dc:creator>James R. Ferguson</dc:creator>
      <dc:date>2002-03-01T19:41:28Z</dc:date>
    </item>
  </channel>
</rss>

