<?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 Safe usage of 'max_proc_per_user' in Operating System - Tru64 Unix</title>
    <link>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180726#M8520</link>
    <description>I have a system that is being pounded with oracle queries and at times has a load at or near 12.  I'm getting 'max_proc_per_user' exceeded messages, but I'm concerned that allowing more processes will continue to overload the 2 1gig CPUs I have in my ES-45 box with 4GB RAM.  &lt;BR /&gt;&lt;BR /&gt;Any thoughts?</description>
    <pubDate>Mon, 02 Feb 2004 18:04:24 GMT</pubDate>
    <dc:creator>Russ Park</dc:creator>
    <dc:date>2004-02-02T18:04:24Z</dc:date>
    <item>
      <title>Safe usage of 'max_proc_per_user'</title>
      <link>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180726#M8520</link>
      <description>I have a system that is being pounded with oracle queries and at times has a load at or near 12.  I'm getting 'max_proc_per_user' exceeded messages, but I'm concerned that allowing more processes will continue to overload the 2 1gig CPUs I have in my ES-45 box with 4GB RAM.  &lt;BR /&gt;&lt;BR /&gt;Any thoughts?</description>
      <pubDate>Mon, 02 Feb 2004 18:04:24 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180726#M8520</guid>
      <dc:creator>Russ Park</dc:creator>
      <dc:date>2004-02-02T18:04:24Z</dc:date>
    </item>
    <item>
      <title>Re: Safe usage of 'max_proc_per_user'</title>
      <link>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180727#M8521</link>
      <description>maybe there is a tuning mismatch on your system. be sure you have configured the database and the system as described in the oracle tuning guide.&lt;BR /&gt;&lt;BR /&gt;There are dependencies within the max_proc_per_user parameters. But maybe there is another problem leading to the increased processes. Please verify if tcp timeout is an issue here leading to processes in WAIT state.&lt;BR /&gt;</description>
      <pubDate>Tue, 03 Feb 2004 02:37:58 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180727#M8521</guid>
      <dc:creator>Ralf Puchner</dc:creator>
      <dc:date>2004-02-03T02:37:58Z</dc:date>
    </item>
    <item>
      <title>Re: Safe usage of 'max_proc_per_user'</title>
      <link>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180728#M8522</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I know, it makes you feel uneasy to see the machine at it's limit, however 'max_proc_per_user' exceeded messages are worse bacause you never know, when they hit. I would first of all check if oracle is running with good performance. We have a 4cpu/4gb 4100 and only after I tuned indexes and oracle configuration, the machine is now running smoothly.&lt;BR /&gt;&lt;BR /&gt;Michael&lt;BR /&gt;</description>
      <pubDate>Tue, 03 Feb 2004 02:52:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180728#M8522</guid>
      <dc:creator>Michael Schulte zur Sur</dc:creator>
      <dc:date>2004-02-03T02:52:40Z</dc:date>
    </item>
    <item>
      <title>Re: Safe usage of 'max_proc_per_user'</title>
      <link>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180729#M8523</link>
      <description>&lt;BR /&gt;The limit that is hit here is a per-user limit NOT a system limit.&lt;BR /&gt;So what user is this logged against ? Do all your clients connect with the same username ?&lt;BR /&gt;Can't you give different usergroups another collective username against which they could authenticate ?&lt;BR /&gt;&lt;BR /&gt;If you increase max_proc_per_user, you should also check max_threads_per user. By default it's 4 times max_proc_per_user and it should at least be larger than max_proc_per_user. (See man sys_attrs_proc)&lt;BR /&gt;&lt;BR /&gt;Probably the dependencies Ralf was thinking about are those that come with maxusers, which in turn influences taskmax, threadmax and min_free_vnodes, but nowadays maxusers is selfconfiguring at boot-time based on the available physical memory.&lt;BR /&gt;Rgds,&lt;BR /&gt;   Johan.</description>
      <pubDate>Tue, 03 Feb 2004 12:38:55 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-tru64-unix/safe-usage-of-max-proc-per-user/m-p/3180729#M8523</guid>
      <dc:creator>Johan Brusche</dc:creator>
      <dc:date>2004-02-03T12:38:55Z</dc:date>
    </item>
  </channel>
</rss>

