<?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: OAS and kernel tuning in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843307#M936631</link>
    <description>Hello,&lt;BR /&gt;&lt;BR /&gt;By OAS, do you mean Oracle 9iAS?  If so, I have it running on a couple of my K-460 servers, one with 1GB and another with 3.75GB RAM.  This is what we did kernelwise and it has worked fairly well:&lt;BR /&gt;&lt;BR /&gt;Went into sam and applied the General OLTP/Database Monolithic System tuned parameter set.  Then we verified/changed the following parameters:&lt;BR /&gt;&lt;BR /&gt;MAXUSERS=400 minimum&lt;BR /&gt;SHMMAX=0x00000004&lt;BR /&gt;MAXDSIZ=0x00000004&lt;BR /&gt;SEMAEM=16384&lt;BR /&gt;SEMMNS=2048&lt;BR /&gt;SEMMNI=1400&lt;BR /&gt;SEMMAP=1402 &lt;BR /&gt;SHMSEG=200&lt;BR /&gt;SHMMNI=1024&lt;BR /&gt;SEMMNU=700&lt;BR /&gt;SEMUME=300&lt;BR /&gt;SEMVMX=32767&lt;BR /&gt;&lt;BR /&gt;Make NBUF and BUFPAGES=0&lt;BR /&gt;&lt;BR /&gt;Hope this helps&lt;BR /&gt;&lt;BR /&gt;Chris&lt;BR /&gt;</description>
    <pubDate>Tue, 12 Nov 2002 17:53:43 GMT</pubDate>
    <dc:creator>Christopher McCray_1</dc:creator>
    <dc:date>2002-11-12T17:53:43Z</dc:date>
    <item>
      <title>OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843305#M936629</link>
      <description>I don't know if this the exact place for this question, but let's go anyway...&lt;BR /&gt;&lt;BR /&gt;We have here two servers running OAS Version: RM 4.0.8.2 HP-UX B.11.00 U (as stated by OAS daemons) and suddenly, one of them has started to hang up. No kernel modifications have been made, and our customer has reinstalled the OAS twice since these hangups have been detected.&lt;BR /&gt;&lt;BR /&gt;Today, I checked kernel parameters and found this:&lt;BR /&gt;&lt;BR /&gt;dbc_min_pct = 2&lt;BR /&gt;dbc_max_pct = 2&lt;BR /&gt;nbuf = 0&lt;BR /&gt;bufpages = 1992&lt;BR /&gt;&lt;BR /&gt;As the server in question is an A500 with 1GB RAM with 2GB of swap space, I thought that the values of dbc_min_pct and dbc_max_pct were dangerously low, as even the Configurable Kernel Parameters Help advises against too low values to dbc_min_pct.&lt;BR /&gt;&lt;BR /&gt;So, I configured dbc_min_pct as 10 and dbc_max_pct as 25, but I hadn't noticed that bufpages &amp;lt;&amp;gt; 0, and thus the machine panic'ed when booting the new kernel. Old kernel boots the server normally (though I don't dare to start the OAS daemons).&lt;BR /&gt;&lt;BR /&gt;Is this a correct approach for this problem?&lt;BR /&gt;&lt;BR /&gt;TIA,&lt;BR /&gt;Paulo Fessel</description>
      <pubDate>Tue, 12 Nov 2002 17:22:59 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843305#M936629</guid>
      <dc:creator>Paulo A G Fessel</dc:creator>
      <dc:date>2002-11-12T17:22:59Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843306#M936630</link>
      <description>What was nbuf set to when it was panicking?&lt;BR /&gt;&lt;BR /&gt;nbuf is the parm that controls if a static number if buff pages is used or dynamic buffer space.  if nbuff is set to 0, then it is dynamic and the kernel uses dbc_min_pct and dbc_max_pct to set the size of the buffer area. so if tha min_pct is set to 10 then at the minimum your buffer area will be set to 10% of the amount of memory you have, and it can grow all the way up to your max_pct.&lt;BR /&gt;&lt;BR /&gt;if nbuf = 1, then you are using the same anount of buffer pages all the time which is defined by bufpages.&lt;BR /&gt;&lt;BR /&gt;HTH&lt;BR /&gt;&lt;BR /&gt;-Rusty</description>
      <pubDate>Tue, 12 Nov 2002 17:53:26 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843306#M936630</guid>
      <dc:creator>Rusty Sapper</dc:creator>
      <dc:date>2002-11-12T17:53:26Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843307#M936631</link>
      <description>Hello,&lt;BR /&gt;&lt;BR /&gt;By OAS, do you mean Oracle 9iAS?  If so, I have it running on a couple of my K-460 servers, one with 1GB and another with 3.75GB RAM.  This is what we did kernelwise and it has worked fairly well:&lt;BR /&gt;&lt;BR /&gt;Went into sam and applied the General OLTP/Database Monolithic System tuned parameter set.  Then we verified/changed the following parameters:&lt;BR /&gt;&lt;BR /&gt;MAXUSERS=400 minimum&lt;BR /&gt;SHMMAX=0x00000004&lt;BR /&gt;MAXDSIZ=0x00000004&lt;BR /&gt;SEMAEM=16384&lt;BR /&gt;SEMMNS=2048&lt;BR /&gt;SEMMNI=1400&lt;BR /&gt;SEMMAP=1402 &lt;BR /&gt;SHMSEG=200&lt;BR /&gt;SHMMNI=1024&lt;BR /&gt;SEMMNU=700&lt;BR /&gt;SEMUME=300&lt;BR /&gt;SEMVMX=32767&lt;BR /&gt;&lt;BR /&gt;Make NBUF and BUFPAGES=0&lt;BR /&gt;&lt;BR /&gt;Hope this helps&lt;BR /&gt;&lt;BR /&gt;Chris&lt;BR /&gt;</description>
      <pubDate>Tue, 12 Nov 2002 17:53:43 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843307#M936631</guid>
      <dc:creator>Christopher McCray_1</dc:creator>
      <dc:date>2002-11-12T17:53:43Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843308#M936632</link>
      <description>Sorry,&lt;BR /&gt;&lt;BR /&gt;dbc_min_pct = 5&lt;BR /&gt;dbc_max_pct = 20&lt;BR /&gt;&lt;BR /&gt;Chris&lt;BR /&gt;</description>
      <pubDate>Tue, 12 Nov 2002 17:54:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843308#M936632</guid>
      <dc:creator>Christopher McCray_1</dc:creator>
      <dc:date>2002-11-12T17:54:44Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843309#M936633</link>
      <description>If you do mean Oracle Application Server, then we had no end of trouble with it. I believe that Oracle don't support it any longer. &lt;BR /&gt;&lt;BR /&gt;We've replaced it with IAS (based on Apache). This is a lot more reliable.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;John</description>
      <pubDate>Tue, 12 Nov 2002 18:03:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843309#M936633</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2002-11-12T18:03:47Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843310#M936634</link>
      <description>I'd be very surprised if the size of your buffer cache caused daemons to hang.&lt;BR /&gt;&lt;BR /&gt;Admittedly, 8Mb does sound rather small and as I don't know what OAS is, I can't really comment on whether it's big enough.&lt;BR /&gt;&lt;BR /&gt;Does a lot of disk I/O happen? If so then increasing the cache may have benefit but it may be offset by paging if your memory is already fully subscribed.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;John</description>
      <pubDate>Tue, 12 Nov 2002 18:06:14 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843310#M936634</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2002-11-12T18:06:14Z</dc:date>
    </item>
    <item>
      <title>Re: OAS and kernel tuning</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843311#M936635</link>
      <description>To anybody following this thread, my two previous posts have been transposed due to the time discrepancy between the ITRC servers. It makes more sense to read them in reverse order.</description>
      <pubDate>Tue, 12 Nov 2002 18:26:56 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/oas-and-kernel-tuning/m-p/2843311#M936635</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2002-11-12T18:26:56Z</dc:date>
    </item>
  </channel>
</rss>

