<?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: dmesg in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435235#M4688</link>
    <description>Hi, (new in this forum)&lt;BR /&gt;Perhaps it could be the number of inodes, which is too high ??&lt;BR /&gt;see it with bdf -i . Often occours if you have a lot of small files to access.&lt;BR /&gt;I am not 100% sure if this will cause an output which can be viewed by dmesg.&lt;BR /&gt;Anyway an idea to take a look at.</description>
    <pubDate>Tue, 08 Aug 2000 08:14:43 GMT</pubDate>
    <dc:creator>Lars Mousten</dc:creator>
    <dc:date>2000-08-08T08:14:43Z</dc:date>
    <item>
      <title>dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435232#M4685</link>
      <description>/var and /home file systems are 46% and 30% used respectively. There is nothing in syslog, but dmesg buffer is being filled up with messages: &lt;BR /&gt;/home: file system full&lt;BR /&gt;/var: file system full&lt;BR /&gt;What is the reason? How can I stop this?&lt;BR /&gt;&lt;BR /&gt;Regards, &lt;BR /&gt;Rumen&lt;BR /&gt;</description>
      <pubDate>Tue, 08 Aug 2000 05:48:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435232#M4685</guid>
      <dc:creator>Rumen Ginev</dc:creator>
      <dc:date>2000-08-08T05:48:47Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435233#M4686</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;have a look at /var/adm/syslog/syslog.log&lt;BR /&gt;There you can see when the mesages appeared.&lt;BR /&gt;I think your full filesys messages are in the past. When /var goes full this could happen, when someone does a vi with a big file. When opening the file vi stores temporary on /var. If the user leaves then vi the /var files are removed.&lt;BR /&gt;Within /home it could be that a user has dumped a core file an possibly you have a cron job that removes periodically core files so that the filesystem ist freed.&lt;BR /&gt;&lt;BR /&gt;Regards&lt;BR /&gt;&lt;BR /&gt;Andrew</description>
      <pubDate>Tue, 08 Aug 2000 06:01:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435233#M4686</guid>
      <dc:creator>Andreas Voss</dc:creator>
      <dc:date>2000-08-08T06:01:40Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435234#M4687</link>
      <description>sounds like they are old messages.  Is there a time stamp on the dmesg entries?  And unless you've restarted syslogd since /var filled up, syslogd is no longer logging error messages.  Look at the time stamp of the last message in /var/adm/syslog/syslog.log.  &lt;BR /&gt;stop and start syslogd:&lt;BR /&gt;/sbin/init.d/syslogd stop and&lt;BR /&gt;/sbin/init.d/syslogd start&lt;BR /&gt;&lt;BR /&gt;you should get a read of the syslog.conf in the syslog.log file.</description>
      <pubDate>Tue, 08 Aug 2000 06:05:24 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435234#M4687</guid>
      <dc:creator>curt larson</dc:creator>
      <dc:date>2000-08-08T06:05:24Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435235#M4688</link>
      <description>Hi, (new in this forum)&lt;BR /&gt;Perhaps it could be the number of inodes, which is too high ??&lt;BR /&gt;see it with bdf -i . Often occours if you have a lot of small files to access.&lt;BR /&gt;I am not 100% sure if this will cause an output which can be viewed by dmesg.&lt;BR /&gt;Anyway an idea to take a look at.</description>
      <pubDate>Tue, 08 Aug 2000 08:14:43 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435235#M4688</guid>
      <dc:creator>Lars Mousten</dc:creator>
      <dc:date>2000-08-08T08:14:43Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435236#M4689</link>
      <description>&lt;BR /&gt;The errors in dmesg mean that since the server was last rebooted you have had those filesystems full, at some point (maybe a long time ago). The only ways to find out when they filled up is to check /var/adm/syslog/syslog.log for any full messages (grep -i full /var/adm/syslog/syslog.log). If you cant see any messages there then another possibility is if you are running measureware/perfview then fire up perfview (pv) and display a graph and change a metric to space util for the lvol which has filled up then scroll back in time on the graph to see when it was full. &lt;BR /&gt; &lt;BR /&gt;A more common way to capture dmesg output is a cron job running say every 5/10 minutes which does a dmesg - &amp;gt;&amp;gt;/var/adm/messages. This way a timestamp is put into the messages file when an error is logged in dmesg, so by simply looking at /var/adm/messages for the full errors you will see a date/timestamp when they ocurred.&lt;BR /&gt;&lt;BR /&gt;The only way to clear these errors out when doing a dmesg command is to reboot your server.&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 08 Aug 2000 08:26:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435236#M4689</guid>
      <dc:creator>Stefan Farrelly</dc:creator>
      <dc:date>2000-08-08T08:26:33Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435237#M4690</link>
      <description>dmesg reads a small buffer which is wtitten cyclically by the kernel.&lt;BR /&gt;&lt;BR /&gt;Are you just running dmesg?  If so, it will simply print the contents of the buffer every time.&lt;BR /&gt;&lt;BR /&gt;dmesg -  should be used to display any new messages since the last time it was run see man dmesg.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;&lt;BR /&gt;John</description>
      <pubDate>Tue, 08 Aug 2000 08:33:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435237#M4690</guid>
      <dc:creator>John Palmer</dc:creator>
      <dc:date>2000-08-08T08:33:00Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435238#M4691</link>
      <description>dmesg is a kernel buffer that keeps only the last couple of pages of messages but with no time stamp. To log these messages, use a cron job such as:&lt;BR /&gt;&lt;BR /&gt;1,6,16,26,36,46,56 * * * * /usr/sbin/dmesg - &amp;gt;&amp;gt; /var/adm/dmesg.log&lt;BR /&gt;&lt;BR /&gt;Once this is in root's crontab, all dmesg entries will be logged and time stamped every 5 minutes.  dmesg - will produce no output (and no additional log entries) if there are no new messages.</description>
      <pubDate>Tue, 08 Aug 2000 14:00:48 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435238#M4691</guid>
      <dc:creator>Bill Hassell</dc:creator>
      <dc:date>2000-08-08T14:00:48Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435239#M4692</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;the dmesg buffer is often missleading. There is no time stamp with the error&lt;BR /&gt;messages. So the error might have happend a day before. dmesg is in memory&lt;BR /&gt;only and the low level driver (starting with the boot sequenz) are writing error&lt;BR /&gt;messages there.&lt;BR /&gt;&lt;BR /&gt;Regards Martin Schark&lt;BR /&gt;mailto:mschark@gmx.net</description>
      <pubDate>Wed, 09 Aug 2000 20:29:20 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435239#M4692</guid>
      <dc:creator>Martin Schark</dc:creator>
      <dc:date>2000-08-09T20:29:20Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435240#M4693</link>
      <description>/home can get filled up by the "core files" caused by the application failiure or a compilation error.&lt;BR /&gt;/var can get filled up my the root's mail box file name /var/mail/root or the /var/adm/sa/sa*** files may occupy space. &lt;BR /&gt;or /var/adm/crash has a core directory.&lt;BR /&gt;The above mentioned files are the root causes for the file systems to get filled up&lt;BR /&gt;clear them accordingly</description>
      <pubDate>Thu, 10 Aug 2000 04:26:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435240#M4693</guid>
      <dc:creator>R D SUNDAR RAJAN</dc:creator>
      <dc:date>2000-08-10T04:26:33Z</dc:date>
    </item>
    <item>
      <title>Re: dmesg</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435241#M4694</link>
      <description>Hi,&lt;BR /&gt;dmesg doesn't show any time stamps, So I advice you to look at /var/adm/syslog/syslog.log to keep track of any errors.&lt;BR /&gt;If ur filesystems are not shown full in bdf output, then obviously it is a old message.</description>
      <pubDate>Tue, 15 Aug 2000 06:42:30 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/dmesg/m-p/2435241#M4694</guid>
      <dc:creator>Ramesh Donti</dc:creator>
      <dc:date>2000-08-15T06:42:30Z</dc:date>
    </item>
  </channel>
</rss>

