<?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: / filesystem in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568912#M30271</link>
    <description>One thing you may wish to do is to check for large files.  The following command will accomplish this&lt;BR /&gt;find / -xdev -size +1000000c -exec ll {} \;&lt;BR /&gt;&lt;BR /&gt;There could also be processes that are still writing to files even though space has been made free.  To check:&lt;BR /&gt;Run lsof and redirect the output &amp;gt; /tmp/lsof.out &lt;BR /&gt;Look for entries that have /dev/dsk/vg## in the last column these are processes that can still be writing even though space has been cleared up. In my case it was a glance process. An example of the lsof is &lt;BR /&gt;emsagent 1502 root 0w VREG 64,0x2 47387 1884 / (/dev/vg00/l &lt;BR /&gt;vol3) &lt;BR /&gt;where=emsagent is the process name &lt;BR /&gt;1502= is the process number &lt;BR /&gt;/ = the filesystem that is being written to &lt;BR /&gt;/dev/vg00/lvol3=gives an inidication that the problem could be there. &lt;BR /&gt;&lt;BR /&gt;By killing the process giving me problems I went from 97% full to 65%. &lt;BR /&gt;&lt;BR /&gt;Hope this helps. &lt;BR /&gt;nancy</description>
    <pubDate>Thu, 23 Aug 2001 10:00:34 GMT</pubDate>
    <dc:creator>nancy rippey</dc:creator>
    <dc:date>2001-08-23T10:00:34Z</dc:date>
    <item>
      <title>/ filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568909#M30268</link>
      <description>Guys please help me out here. I am administerin high availability systems. Somehow my / filesystem keeps filling up. I have a hard time tracing the problem. Where do u think i could look and possibly re-claim my space?</description>
      <pubDate>Thu, 23 Aug 2001 05:03:42 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568909#M30268</guid>
      <dc:creator>Olebile Molefi</dc:creator>
      <dc:date>2001-08-23T05:03:42Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568910#M30269</link>
      <description>Hi&lt;BR /&gt;   &lt;BR /&gt;1) Do not have too many directories  on the /  filesystem.&lt;BR /&gt;2) Check if any process is generating core dumps, these can fill up the space.&lt;BR /&gt;3) Check if any processes are opening files and writing into them on the root filesystem. You can use glance to monitor the open files.&lt;BR /&gt;&lt;BR /&gt;Hope this helps.&lt;BR /&gt;&lt;BR /&gt;...BPK...</description>
      <pubDate>Thu, 23 Aug 2001 05:21:46 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568910#M30269</guid>
      <dc:creator>Praveen Bezawada</dc:creator>
      <dc:date>2001-08-23T05:21:46Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568911#M30270</link>
      <description>I would suggest to run a cronjob every 5 mins to run du -x / . After sometime, compare the outputs and observe which directory is growing. Also, it would be better if you can bring down the server in singler user mode with only / and /stand mounted and check if any of the mounted directories are already having some data. If so, they will be hidden once the file systems are mounted during later run levels and be included in the used space.&lt;BR /&gt;&lt;BR /&gt;Also search for core files&lt;BR /&gt;&lt;BR /&gt;find / -xdev -name core -print &lt;BR /&gt;&lt;BR /&gt;Find files of more than 5MB size&lt;BR /&gt;&lt;BR /&gt;find / -xdev -size 10000 -print &lt;BR /&gt;&lt;BR /&gt;-Sri&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Thu, 23 Aug 2001 05:38:54 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568911#M30270</guid>
      <dc:creator>Sridhar Bhaskarla</dc:creator>
      <dc:date>2001-08-23T05:38:54Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568912#M30271</link>
      <description>One thing you may wish to do is to check for large files.  The following command will accomplish this&lt;BR /&gt;find / -xdev -size +1000000c -exec ll {} \;&lt;BR /&gt;&lt;BR /&gt;There could also be processes that are still writing to files even though space has been made free.  To check:&lt;BR /&gt;Run lsof and redirect the output &amp;gt; /tmp/lsof.out &lt;BR /&gt;Look for entries that have /dev/dsk/vg## in the last column these are processes that can still be writing even though space has been cleared up. In my case it was a glance process. An example of the lsof is &lt;BR /&gt;emsagent 1502 root 0w VREG 64,0x2 47387 1884 / (/dev/vg00/l &lt;BR /&gt;vol3) &lt;BR /&gt;where=emsagent is the process name &lt;BR /&gt;1502= is the process number &lt;BR /&gt;/ = the filesystem that is being written to &lt;BR /&gt;/dev/vg00/lvol3=gives an inidication that the problem could be there. &lt;BR /&gt;&lt;BR /&gt;By killing the process giving me problems I went from 97% full to 65%. &lt;BR /&gt;&lt;BR /&gt;Hope this helps. &lt;BR /&gt;nancy</description>
      <pubDate>Thu, 23 Aug 2001 10:00:34 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568912#M30271</guid>
      <dc:creator>nancy rippey</dc:creator>
      <dc:date>2001-08-23T10:00:34Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568913#M30272</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;You could do &lt;BR /&gt;du -kx / |sort -n &lt;BR /&gt;(this would list the disk usage in KBytes )&lt;BR /&gt;&lt;BR /&gt;Look for core files&lt;BR /&gt;&lt;BR /&gt;Look in /dev/ and /etc/cmcluster directories.&lt;BR /&gt;The package log files in a cluster might get really huge.&lt;BR /&gt;du -ks /etc/cmcluster/* |sort -n&lt;BR /&gt;&lt;BR /&gt;-HTH&lt;BR /&gt;Ramesh</description>
      <pubDate>Thu, 23 Aug 2001 11:29:45 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568913#M30272</guid>
      <dc:creator>linuxfan</dc:creator>
      <dc:date>2001-08-23T11:29:45Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568914#M30273</link>
      <description>Hi Olebile,&lt;BR /&gt;You must having somewhere core file.&lt;BR /&gt;As other says run &lt;BR /&gt;du -sk /* &amp;gt; file1&lt;BR /&gt;then when you have system fill up run again&lt;BR /&gt;du -sk /* &amp;gt; file2 compare two file and find out why and where that directory is having large file.&lt;BR /&gt;&lt;BR /&gt;Sachin</description>
      <pubDate>Thu, 23 Aug 2001 13:02:25 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568914#M30273</guid>
      <dc:creator>Sachin Patel</dc:creator>
      <dc:date>2001-08-23T13:02:25Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568915#M30274</link>
      <description>A lot depends on what filesystems you have on the system.  For example, if you did not create a /var filesystem, you could be filling up there.  Of course I would do as above first and check for core file.  Then I would do something like this:&lt;BR /&gt;&lt;BR /&gt;cd /&lt;BR /&gt;touch /tmp/flist&lt;BR /&gt;cat /etc/fstab | cut -f2 -d" " | grep "/" &amp;gt; /tmp/flist&lt;BR /&gt;ls | while read line&lt;BR /&gt;do&lt;BR /&gt;if grep $line /tmp/flist &amp;gt; /dev/null 2&amp;gt; /dev/null&lt;BR /&gt;then&lt;BR /&gt;   #do nothing&lt;BR /&gt;else&lt;BR /&gt;   du -sk $line&lt;BR /&gt;fi&lt;BR /&gt;done&lt;BR /&gt;&lt;BR /&gt;Hope this helps&lt;BR /&gt;&lt;BR /&gt;...jcd...</description>
      <pubDate>Thu, 23 Aug 2001 13:29:13 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568915#M30274</guid>
      <dc:creator>Joseph C. Denman</dc:creator>
      <dc:date>2001-08-23T13:29:13Z</dc:date>
    </item>
    <item>
      <title>Re: / filesystem</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568916#M30275</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;You obviously have a problem; the / filesystem should not be growing at all! It's very common for someone to accidently recreate device files&lt;BR /&gt;like /dev/null or /dev/rmt/1m or to tar to /dev/rmt/Om (the letter) when they meant 0m (the number). That's the first thing to look for. You could be creating core files for processes running in the / filesystem. Use the find command to look for large files or files that have been modified in the last day. Only a few files in the / filesystem should actually be modified in a properly configured system.</description>
      <pubDate>Thu, 23 Aug 2001 15:39:47 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/filesystem/m-p/2568916#M30275</guid>
      <dc:creator>A. Clay Stephenson</dc:creator>
      <dc:date>2001-08-23T15:39:47Z</dc:date>
    </item>
  </channel>
</rss>

