<?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 UNIX performance: RHEL5 server is slow in Operating System - Linux</title>
    <link>https://community.hpe.com/t5/operating-system-linux/unix-performance-rhel5-server-is-slow/m-p/4009589#M28494</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;The developers here are acusing that the RHEL5 servers are slow.  They run weblogic tests on these servers.  They have shown from their logs that weblogic server on RHEL5 takes a long time (3 minutes) to start in comparison to that of a SuSE 10 server (1 minute).&lt;BR /&gt;&lt;BR /&gt;This is not a problem related to only one RHEL5 server.  We have complains regarding all the RHEL5 servers here (both x86_64 and IA64).&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;From the output of top, I can see that a lot of processes are sleeping.  At any time, I would see only one or 2 processes in the run state.  Is this a normal thing even on a multiprocessor server?&lt;BR /&gt;&lt;BR /&gt;I went through the sar (system activity report) on one of the server during the start of weblogic.  Following are my observations:&lt;BR /&gt; a. The RAM is almost completely used; however there is no swap usage.&lt;BR /&gt; b. A lot of processes being created.&lt;BR /&gt; c. A lot of context switches happening.&lt;BR /&gt; d. Seems like the process that is running; is hopping from one processor to another, and I see about 15% of each processor's time is spent waiting on I/O during that time. &lt;BR /&gt; e. I also see some spike in NFS client activity during this time.&lt;BR /&gt;&lt;BR /&gt;I want to find out the root cause.  Please suggest.&lt;BR /&gt;&lt;BR /&gt;Thanks &amp;amp; Regards,</description>
    <pubDate>Wed, 30 May 2007 03:49:26 GMT</pubDate>
    <dc:creator>GnanaShekar</dc:creator>
    <dc:date>2007-05-30T03:49:26Z</dc:date>
    <item>
      <title>UNIX performance: RHEL5 server is slow</title>
      <link>https://community.hpe.com/t5/operating-system-linux/unix-performance-rhel5-server-is-slow/m-p/4009589#M28494</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;The developers here are acusing that the RHEL5 servers are slow.  They run weblogic tests on these servers.  They have shown from their logs that weblogic server on RHEL5 takes a long time (3 minutes) to start in comparison to that of a SuSE 10 server (1 minute).&lt;BR /&gt;&lt;BR /&gt;This is not a problem related to only one RHEL5 server.  We have complains regarding all the RHEL5 servers here (both x86_64 and IA64).&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;From the output of top, I can see that a lot of processes are sleeping.  At any time, I would see only one or 2 processes in the run state.  Is this a normal thing even on a multiprocessor server?&lt;BR /&gt;&lt;BR /&gt;I went through the sar (system activity report) on one of the server during the start of weblogic.  Following are my observations:&lt;BR /&gt; a. The RAM is almost completely used; however there is no swap usage.&lt;BR /&gt; b. A lot of processes being created.&lt;BR /&gt; c. A lot of context switches happening.&lt;BR /&gt; d. Seems like the process that is running; is hopping from one processor to another, and I see about 15% of each processor's time is spent waiting on I/O during that time. &lt;BR /&gt; e. I also see some spike in NFS client activity during this time.&lt;BR /&gt;&lt;BR /&gt;I want to find out the root cause.  Please suggest.&lt;BR /&gt;&lt;BR /&gt;Thanks &amp;amp; Regards,</description>
      <pubDate>Wed, 30 May 2007 03:49:26 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/unix-performance-rhel5-server-is-slow/m-p/4009589#M28494</guid>
      <dc:creator>GnanaShekar</dc:creator>
      <dc:date>2007-05-30T03:49:26Z</dc:date>
    </item>
    <item>
      <title>Re: UNIX performance: RHEL5 server is slow</title>
      <link>https://community.hpe.com/t5/operating-system-linux/unix-performance-rhel5-server-is-slow/m-p/4009590#M28495</link>
      <description>Waiting for I/O could indicate a problem with the disk or network subsystem.&lt;BR /&gt;&lt;BR /&gt;This could be a problem with the RAID configuration, problem with the driver/module or a problem with the network driver/configuration/tuning.&lt;BR /&gt;&lt;BR /&gt;You menthioned something about NFS client activity, how is the infraestructure?</description>
      <pubDate>Wed, 30 May 2007 08:09:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/unix-performance-rhel5-server-is-slow/m-p/4009590#M28495</guid>
      <dc:creator>Ivan Ferreira</dc:creator>
      <dc:date>2007-05-30T08:09:05Z</dc:date>
    </item>
  </channel>
</rss>

