<?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: Basic Cluster Question in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864613#M696758</link>
    <description>A service is a trigger for Serviceguard.&lt;BR /&gt;If the service has failed (the monitor script exitted because it did not detect a critical process), it will cause (after the sole restart allowed) Serviceguard to move the package to the other server (conditional on AUTO_RUN and Node Switching parameter settings at the time).&lt;BR /&gt;&lt;BR /&gt;Whether you configure multiple process names in a single Service, or 3 Services each monitoring it's own process makes no difference - if any of the critical processes ceases to exist, SG will react the same way and fail the package over.  &lt;BR /&gt;Therefore, for ease of managment - let one monitor watch all 3 processes.</description>
    <pubDate>Mon, 18 Sep 2006 11:14:50 GMT</pubDate>
    <dc:creator>Stephen Doud</dc:creator>
    <dc:date>2006-09-18T11:14:50Z</dc:date>
    <item>
      <title>Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864612#M696757</link>
      <description>Running HP Openview OVO 8.x on MCSG 11.16&lt;BR /&gt;The Cluster control files were created by the Openview Installation Software. Everything is working fine. But the cntl file was created with just one service&lt;BR /&gt;&lt;BR /&gt;SERVICE_NAME[0]="test_service"&lt;BR /&gt;SERVICE_CMD[0]="/etc/cmcluster/test/test.mon"&lt;BR /&gt;SERVICE_RESTART[0]="-r 1"&lt;BR /&gt;&lt;BR /&gt;The test.mon points to three diff monitor files one for oracle and two are openview relates with lists of processes.&lt;BR /&gt;&lt;BR /&gt;In this scenario, is it a good practice to have multiple services instead of just one? What is the advantage / disadvantage of having multiple services Vs Single Service?&lt;BR /&gt;&lt;BR /&gt;I know very little about clustering, Just looking for some best practice and whether I should continue with one service or go for multiple services...&lt;BR /&gt;&lt;BR /&gt;-Kathir&lt;BR /&gt;</description>
      <pubDate>Mon, 18 Sep 2006 10:03:37 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864612#M696757</guid>
      <dc:creator>Kathir Lakshmanan</dc:creator>
      <dc:date>2006-09-18T10:03:37Z</dc:date>
    </item>
    <item>
      <title>Re: Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864613#M696758</link>
      <description>A service is a trigger for Serviceguard.&lt;BR /&gt;If the service has failed (the monitor script exitted because it did not detect a critical process), it will cause (after the sole restart allowed) Serviceguard to move the package to the other server (conditional on AUTO_RUN and Node Switching parameter settings at the time).&lt;BR /&gt;&lt;BR /&gt;Whether you configure multiple process names in a single Service, or 3 Services each monitoring it's own process makes no difference - if any of the critical processes ceases to exist, SG will react the same way and fail the package over.  &lt;BR /&gt;Therefore, for ease of managment - let one monitor watch all 3 processes.</description>
      <pubDate>Mon, 18 Sep 2006 11:14:50 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864613#M696758</guid>
      <dc:creator>Stephen Doud</dc:creator>
      <dc:date>2006-09-18T11:14:50Z</dc:date>
    </item>
    <item>
      <title>Re: Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864614#M696759</link>
      <description>Shalom,&lt;BR /&gt;&lt;BR /&gt;My answer is that it depends.&lt;BR /&gt;&lt;BR /&gt;If you have three processes that need the be monitored and all three are required then all three should be monitored by one monitor that tries to correct any problems it finds.&lt;BR /&gt;&lt;BR /&gt;Your run requirements determine how to set up the monitor.&lt;BR /&gt;&lt;BR /&gt;SEP</description>
      <pubDate>Mon, 18 Sep 2006 11:34:24 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864614#M696759</guid>
      <dc:creator>Steven E. Protter</dc:creator>
      <dc:date>2006-09-18T11:34:24Z</dc:date>
    </item>
    <item>
      <title>Re: Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864615#M696760</link>
      <description>Thanks, As I understood, all the mon service  require is a return code of 0. If the return code is non zero then it tries to restart and if the restart attempt is failed then it fails over to the other node.&lt;BR /&gt;&lt;BR /&gt;Two questions, &lt;BR /&gt;&lt;BR /&gt;When I specify '-r 1' for restart where it does it look for the restart script? &lt;BR /&gt;&lt;BR /&gt;If the one service I defined calls 3 monitor scripts and those three scripts are monitoring roughly 6 Oracle Processes and 20 Openview Processes and 1 OV communication process. &lt;BR /&gt;&lt;BR /&gt;Instead of this one service, If I group the processes and create multiple services, is there any advantage like, the individual services can be restarted so that the whole application doesnt have to be restarted?&lt;BR /&gt;&lt;BR /&gt;-Kathir&lt;BR /&gt;</description>
      <pubDate>Mon, 18 Sep 2006 12:11:59 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864615#M696760</guid>
      <dc:creator>Kathir Lakshmanan</dc:creator>
      <dc:date>2006-09-18T12:11:59Z</dc:date>
    </item>
    <item>
      <title>Re: Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864616#M696761</link>
      <description>The service cmd is the script that restarts.. when processes are not found to be running, the service cmd exits, and then restarts per the parameter until it is 0, then fails and halts the package.</description>
      <pubDate>Mon, 18 Sep 2006 14:18:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864616#M696761</guid>
      <dc:creator>Kevin Wright</dc:creator>
      <dc:date>2006-09-18T14:18:44Z</dc:date>
    </item>
    <item>
      <title>Re: Basic Cluster Question</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864617#M696762</link>
      <description>Thanks Kevin..</description>
      <pubDate>Mon, 18 Sep 2006 14:28:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/basic-cluster-question/m-p/3864617#M696762</guid>
      <dc:creator>Kathir Lakshmanan</dc:creator>
      <dc:date>2006-09-18T14:28:41Z</dc:date>
    </item>
  </channel>
</rss>

