<?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: FailOver in MCSG in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545964#M700764</link>
    <description>Forgot...&lt;BR /&gt;It is very easy if you use this software, you can do it in GUI:&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://software.hp.com/portal/swdepot/displayProductInfo.do?productNumber=B8325BA" target="_blank"&gt;http://software.hp.com/portal/swdepot/displayProductInfo.do?productNumber=B8325BA&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Regard,&lt;BR /&gt;Hoang Chi Cong&lt;BR /&gt;</description>
    <pubDate>Tue, 17 May 2005 20:01:03 GMT</pubDate>
    <dc:creator>Hoang Chi Cong_1</dc:creator>
    <dc:date>2005-05-17T20:01:03Z</dc:date>
    <item>
      <title>FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545958#M700758</link>
      <description>Hi,&lt;BR /&gt;   I have setup a MCSG cluster in HPUX operating system and configured a package. But i am not sure about the failover. Suppose if my machine(on which package is primary) gets rebooted, then will the secondary node take over automatically and start working as  primary? or do i need to start that manually. What are the settings i am suppose to do, for automatic failover. Also what happens if i helt the package on primary node? will the secondary node take over the control with any manual intervention. Please clarify this to me.</description>
      <pubDate>Tue, 17 May 2005 05:56:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545958#M700758</guid>
      <dc:creator>Siva Prasad</dc:creator>
      <dc:date>2005-05-17T05:56:55Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545959#M700759</link>
      <description>HI Siva,&lt;BR /&gt;It totally depends on the POLICY you have configured in cluster configuration ascii file. If you have set the Cluster as failover then if your sys1 faults or reboots then all the packages running on sys1 will get transferred on sys2 and start running on sys2.&lt;BR /&gt;If you halt the package manually then it won't automatically start on sys2. You need to manually start them on sys2.&lt;BR /&gt;&lt;BR /&gt;If you give some configuration input from /etc/cmcluster/cmclconf.ascii then picture would be more clearer.&lt;BR /&gt;&lt;BR /&gt;Hope that helps.&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 17 May 2005 06:02:29 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545959#M700759</guid>
      <dc:creator>Bharat Katkar</dc:creator>
      <dc:date>2005-05-17T06:02:29Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545960#M700760</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;This depends on your failover policy and failback policy. If your failback policy is Automatic then when failed node is restored your package will be shifted to that node again. This should not be done unless specific needs because this will give two switchovers to your package which can be avaided by setting it manual and reverting package to corresponding node manually at a proper time when load/users are less which are going to be affected by revert back.&lt;BR /&gt;&lt;BR /&gt;If you halt it manually it will not start on other node bacause system understands that you are doing it intentionally for some maintenance puposes etc.&lt;BR /&gt;&lt;BR /&gt;You can test switch over by terminating the services associated with package.&lt;BR /&gt;&lt;BR /&gt;HTH,&lt;BR /&gt;Devender</description>
      <pubDate>Tue, 17 May 2005 06:08:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545960#M700760</guid>
      <dc:creator>Devender Khatana</dc:creator>
      <dc:date>2005-05-17T06:08:38Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545961#M700761</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;There are several factors that will inturn decide the package automatic failover.&lt;BR /&gt;&lt;BR /&gt;1. run 'cmviewcl -v' if the package shows enabled on both nodes whenever your primary is rebooted or taken out of network, package will switch over to next node.&lt;BR /&gt;&lt;BR /&gt;If you see a package is not enabled to run on a particular node, use cmmodpkg.&lt;BR /&gt;cmmodpkg -e -n &lt;NODENAME&gt; &lt;PACKAGE_NAME&gt; &lt;BR /&gt;&lt;BR /&gt;2. If you halt the package using cmhaltpkg it will not fail over. But you can manually start the package on second node.&lt;BR /&gt;&lt;BR /&gt;Hope this helps you&lt;BR /&gt;&lt;BR /&gt;Sudeesh&lt;BR /&gt;&lt;/PACKAGE_NAME&gt;&lt;/NODENAME&gt;</description>
      <pubDate>Tue, 17 May 2005 06:14:11 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545961#M700761</guid>
      <dc:creator>Sudeesh</dc:creator>
      <dc:date>2005-05-17T06:14:11Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545962#M700762</link>
      <description>Serviceguard is an automation package which manages virtualization of applications.  For a good explanation of the concepts of Serviceguard - have a look at the "Managing Serviceguard" manual:&lt;BR /&gt;&lt;A href="http://docs.hp.com/en/ha.html#Serviceguard" target="_blank"&gt;http://docs.hp.com/en/ha.html#Serviceguard&lt;/A&gt;&lt;BR /&gt;SG operates in an active/active environment where each server is permitted to operate critical applications as well as standby for other servers operating critical applications.&lt;BR /&gt;In the event of a disaster on a server, SG is designed to cause the remaining servers to adopt packages which were running on a the server that crashed unexpectedly.  &lt;BR /&gt;&lt;BR /&gt;When the administrator manually halts a package (cmhaltpkg), SG considers this an Administrative event - and does NOT cause the package to fail over to another server.&lt;BR /&gt;However when the admin forces a node out of the cluster (cmhaltnode -f) even though the node is still running packages, SG causes the packages to failover to their adoptive node(s).&lt;BR /&gt;To understand this is more detail I recommend the Managing Serviceguard manual -to get a broader understanding of how Serviceguard works under various conditions.&lt;BR /&gt;Particularly this section:&lt;BR /&gt;"Responses to Failures" --&amp;gt;&amp;gt; &lt;A href="http://docs.hp.com/en/B3936-90079/ch03s07.html" target="_blank"&gt;http://docs.hp.com/en/B3936-90079/ch03s07.html&lt;/A&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 17 May 2005 08:06:52 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545962#M700762</guid>
      <dc:creator>Stephen Doud</dc:creator>
      <dc:date>2005-05-17T08:06:52Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545963#M700763</link>
      <description>Hi&lt;BR /&gt;&lt;BR /&gt;First, in the package configuration file, you need to set "PACKAGE_TYPE" parameter with "FAILOVER".&lt;BR /&gt;&lt;BR /&gt;Then use cmmodpkg command to enable switching attributes for a high availability package:&lt;BR /&gt;#cmmodpkg -e -n &lt;NODE_NAMES&gt; &lt;PACKAGE_NAME&gt; &lt;BR /&gt;&lt;BR /&gt;See on the cmmodpkg manpage for more detail!&lt;BR /&gt;&lt;BR /&gt;Regard,&lt;BR /&gt;Hoang Chi Cong&lt;BR /&gt;&lt;/PACKAGE_NAME&gt;&lt;/NODE_NAMES&gt;</description>
      <pubDate>Tue, 17 May 2005 19:59:39 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545963#M700763</guid>
      <dc:creator>Hoang Chi Cong_1</dc:creator>
      <dc:date>2005-05-17T19:59:39Z</dc:date>
    </item>
    <item>
      <title>Re: FailOver in MCSG</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545964#M700764</link>
      <description>Forgot...&lt;BR /&gt;It is very easy if you use this software, you can do it in GUI:&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://software.hp.com/portal/swdepot/displayProductInfo.do?productNumber=B8325BA" target="_blank"&gt;http://software.hp.com/portal/swdepot/displayProductInfo.do?productNumber=B8325BA&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Regard,&lt;BR /&gt;Hoang Chi Cong&lt;BR /&gt;</description>
      <pubDate>Tue, 17 May 2005 20:01:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/failover-in-mcsg/m-p/3545964#M700764</guid>
      <dc:creator>Hoang Chi Cong_1</dc:creator>
      <dc:date>2005-05-17T20:01:03Z</dc:date>
    </item>
  </channel>
</rss>

