<?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: File table is full in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723868#M945528</link>
    <description>Hi:&lt;BR /&gt;&lt;BR /&gt;What's recommended is what works.  You can increase 'maxusers' or 'nproc' (which is based on 'maxusers' anyway), or you can simply substitute an appropriate value for 'nfile' that meets your needs.&lt;BR /&gt;&lt;BR /&gt;The formulae are generally designed to keep balance among kernel parameters and thereby optimize kernel structures.  They do not always work for all environments.&lt;BR /&gt;&lt;BR /&gt;Remember that you can track 'nfile' with 'sar -v'.  The current, high-water mark and number of overflows will be recorded:&lt;BR /&gt;&lt;BR /&gt;# sar -v 10 20 &lt;BR /&gt;&lt;BR /&gt;...would take 20-samples 10-seconds apart, for instance.&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt;&lt;BR /&gt;...JRF...</description>
    <pubDate>Tue, 14 May 2002 20:12:15 GMT</pubDate>
    <dc:creator>James R. Ferguson</dc:creator>
    <dc:date>2002-05-14T20:12:15Z</dc:date>
    <item>
      <title>File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723866#M945526</link>
      <description>I'm getting this error in my syslog file. &lt;BR /&gt;&lt;BR /&gt; Env : Oracle 8i. N-Classe HP_UX 11, 8G Phy Mem. &lt;BR /&gt;&lt;BR /&gt;MAXUSERS=600&lt;BR /&gt;nfile=20000 ( (16*(NPROC+16+MAXUSERS)/10+32+2*(NPTY+NSTRPTY+NSTRTEL))&lt;BR /&gt;&lt;BR /&gt;What is recomended number here to solve this problem&lt;BR /&gt;</description>
      <pubDate>Tue, 14 May 2002 20:05:32 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723866#M945526</guid>
      <dc:creator>Rushank</dc:creator>
      <dc:date>2002-05-14T20:05:32Z</dc:date>
    </item>
    <item>
      <title>Re: File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723867#M945527</link>
      <description>Rushank&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Change it to atleast 50000 and monitor using sar -v 2 10 , the same was the issue wiht one of our production boxes and i have seen even 50000 getting full actually is is realtive to the application being used , but I recommned chaging to 50000 and then monitoring it.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Manoj Srivastava</description>
      <pubDate>Tue, 14 May 2002 20:08:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723867#M945527</guid>
      <dc:creator>MANOJ SRIVASTAVA</dc:creator>
      <dc:date>2002-05-14T20:08:40Z</dc:date>
    </item>
    <item>
      <title>Re: File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723868#M945528</link>
      <description>Hi:&lt;BR /&gt;&lt;BR /&gt;What's recommended is what works.  You can increase 'maxusers' or 'nproc' (which is based on 'maxusers' anyway), or you can simply substitute an appropriate value for 'nfile' that meets your needs.&lt;BR /&gt;&lt;BR /&gt;The formulae are generally designed to keep balance among kernel parameters and thereby optimize kernel structures.  They do not always work for all environments.&lt;BR /&gt;&lt;BR /&gt;Remember that you can track 'nfile' with 'sar -v'.  The current, high-water mark and number of overflows will be recorded:&lt;BR /&gt;&lt;BR /&gt;# sar -v 10 20 &lt;BR /&gt;&lt;BR /&gt;...would take 20-samples 10-seconds apart, for instance.&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt;&lt;BR /&gt;...JRF...</description>
      <pubDate>Tue, 14 May 2002 20:12:15 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723868#M945528</guid>
      <dc:creator>James R. Ferguson</dc:creator>
      <dc:date>2002-05-14T20:12:15Z</dc:date>
    </item>
    <item>
      <title>Re: File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723869#M945529</link>
      <description>We got an L-class (4GB RAM) running the same thing with MAXUSERS=1000.&lt;BR /&gt;There is not definite answer to your question. It depends on how much the system table space is being used at the time it hit the limit. If I were you, I would double maxusers for a start.</description>
      <pubDate>Tue, 14 May 2002 20:14:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723869#M945529</guid>
      <dc:creator>S.K. Chan</dc:creator>
      <dc:date>2002-05-14T20:14:05Z</dc:date>
    </item>
    <item>
      <title>Re: File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723870#M945530</link>
      <description>Out put of sar -v 10 20 &lt;BR /&gt;&lt;BR /&gt;16:23:48 text-sz  ov  proc-sz  ov  inod-sz  ov  file-sz  ov &lt;BR /&gt;16:23:58   N/A   N/A 526/4000  0  4457/5672  0  15925/20010 0&lt;BR /&gt;16:24:08   N/A   N/A 527/4000  0  4128/5672  0  15894/20010 0&lt;BR /&gt;16:24:18   N/A   N/A 528/4000  0  3913/5672  0  15900/20010 0&lt;BR /&gt;16:24:28   N/A   N/A 527/4000  0  3673/5672  0  15911/20010 0&lt;BR /&gt;16:24:38   N/A   N/A 529/4000  0  3462/5672  0  15922/20010 0&lt;BR /&gt;16:24:48   N/A   N/A 529/4000  0  3242/5672  0  15926/20010 0&lt;BR /&gt;16:24:58   N/A   N/A 528/4000  0  3006/5672  0  15900/20010 0&lt;BR /&gt;16:25:08   N/A   N/A 533/4000  0  3011/5672  0  15896/20010 0&lt;BR /&gt;16:25:18   N/A   N/A 532/4000  0  3002/5672  0  15854/20010 0&lt;BR /&gt;16:25:28   N/A   N/A 527/4000  0  3010/5672  0  15846/20010 0&lt;BR /&gt;16:25:38   N/A   N/A 527/4000  0  3007/5672  0  15815/20010 0&lt;BR /&gt;16:25:48   N/A   N/A 525/4000  0  2986/5672  0  15812/20010 0&lt;BR /&gt;16:25:58   N/A   N/A 527/4000  0  2974/5672  0  15809/20010 0&lt;BR /&gt;16:26:08   N/A   N/A 527/4000  0  2980/5672  0  15820/20010 0&lt;BR /&gt;16:26:18   N/A   N/A 527/4000  0  2975/5672  0  15835/20010 0&lt;BR /&gt;16:26:28   N/A   N/A 528/4000  0  2959/5672  0  15842/20010 0&lt;BR /&gt;16:26:38   N/A   N/A 528/4000  0  3006/5672  0  15812/20010 0&lt;BR /&gt;16:26:48   N/A   N/A 531/4000  0  3008/5672  0  15856/20010 0&lt;BR /&gt;16:26:58   N/A   N/A 534/4000  0  2986/5672  0  15862/20010 0&lt;BR /&gt;16:27:08   N/A   N/A 527/4000  0  2982/5672  0  15852/20010 0&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 14 May 2002 20:25:56 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723870#M945530</guid>
      <dc:creator>Rushank</dc:creator>
      <dc:date>2002-05-14T20:25:56Z</dc:date>
    </item>
    <item>
      <title>Re: File table is full</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723871#M945531</link>
      <description>First line from your data ..&lt;BR /&gt;proc-sz 526/4000 gives 13% utilization of nproc&lt;BR /&gt;inod-sz 4457/5672 gives 78% utilization of ninode&lt;BR /&gt;file-sz 15925/20010 gives 79% utlization of nfile.&lt;BR /&gt;&lt;BR /&gt;So no changes is needed for nproc. You might want to bump up ninode and nfile so that it works out to about 50% utilization. To get 50%..&lt;BR /&gt;ninode will be about 9000&lt;BR /&gt;nfile will be about 32000</description>
      <pubDate>Tue, 14 May 2002 20:48:18 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/file-table-is-full/m-p/2723871#M945531</guid>
      <dc:creator>S.K. Chan</dc:creator>
      <dc:date>2002-05-14T20:48:18Z</dc:date>
    </item>
  </channel>
</rss>

