<?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: Linux kernel free mem usage clarification in Operating System - Linux</title>
    <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358268#M35233</link>
    <description>Shalom,&lt;BR /&gt;&lt;BR /&gt;nfslock service may not be running. That could cause this problem.&lt;BR /&gt;&lt;BR /&gt;You have sufficient memory.&lt;BR /&gt;&lt;BR /&gt;SEP</description>
    <pubDate>Fri, 13 Feb 2009 14:29:35 GMT</pubDate>
    <dc:creator>Steven E. Protter</dc:creator>
    <dc:date>2009-02-13T14:29:35Z</dc:date>
    <item>
      <title>Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358266#M35231</link>
      <description>&lt;BR /&gt;Red Hat Enterprise Linux AS release 3 (Taroon Update 9)&lt;BR /&gt;&lt;BR /&gt;Linux hostname 2.4.21-53.ELsmp #1 SMP Wed Nov 14 03:54:12 EST 2007 i686 i686 i386 GNU/Linux&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;# free -mt&lt;BR /&gt;             total       used       free     shared    buffers     cached&lt;BR /&gt;Mem:         10024      10004         19          0         61       8540&lt;BR /&gt;-/+ buffers/cache:       1401       8622&lt;BR /&gt;Swap:         4095        218       3877&lt;BR /&gt;Total:       14120      10223       3897&lt;BR /&gt;&lt;BR /&gt;getting following messages while doing "service nfd start". nfsd is failed to start.&lt;BR /&gt;&lt;BR /&gt;Feb 12 17:14:56 shdbl2p nfs: Starting NFS services:  succeeded&lt;BR /&gt;Feb 12 17:14:56 shdbl2p nfs: rpc.rquotad startup succeeded&lt;BR /&gt;Feb 12 17:14:56 shdbl2p nfsd[16431]: nfssvc: Cannot allocate memory&lt;BR /&gt;Feb 12 17:14:56 shdbl2p nfs: rpc.nfsd startup failed&lt;BR /&gt;Feb 12 17:14:56 shdbl2p nfs: rpc.mountd startup succeeded&lt;BR /&gt;&lt;BR /&gt;This memory error looks contradictory my understanding on the LINUX free memmory (MEM free=free+ cached)calculation. As of now i hav 8GB under cached.&lt;BR /&gt;&lt;BR /&gt;Could some one explain? OR is understaning wrong free mem?</description>
      <pubDate>Fri, 13 Feb 2009 13:57:38 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358266#M35231</guid>
      <dc:creator>skt_skt</dc:creator>
      <dc:date>2009-02-13T13:57:38Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358267#M35232</link>
      <description>I think that there should be another problem. I'm reading the free command as you are, but anyway, the system should report out of memory only when RAM + SWAP is consumed totally. This is not the case.&lt;BR /&gt;&lt;BR /&gt;Maybe, some kernel parameter or shell limit is in place.</description>
      <pubDate>Fri, 13 Feb 2009 14:19:23 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358267#M35232</guid>
      <dc:creator>Ivan Ferreira</dc:creator>
      <dc:date>2009-02-13T14:19:23Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358268#M35233</link>
      <description>Shalom,&lt;BR /&gt;&lt;BR /&gt;nfslock service may not be running. That could cause this problem.&lt;BR /&gt;&lt;BR /&gt;You have sufficient memory.&lt;BR /&gt;&lt;BR /&gt;SEP</description>
      <pubDate>Fri, 13 Feb 2009 14:29:35 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358268#M35233</guid>
      <dc:creator>Steven E. Protter</dc:creator>
      <dc:date>2009-02-13T14:29:35Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358269#M35234</link>
      <description>You're running a 32-bit kernel on a server that has multiple gigabytes of memory. There are two limits in that: &lt;BR /&gt;&lt;BR /&gt;- the design limit of 32-bit Linux memory management: the standard maximum userspace size without any highmem workarounds is "1 GB minus all memory-mapped hardware (like VideoRAM of the display adapter)". &lt;BR /&gt;&lt;BR /&gt;- a 32-bit processor simply cannot directly access more than 4 GB: it does not have enough bits on its address bus for that. A workaround is PAE, which is essentially a form of bank-switching. &lt;BR /&gt;&lt;BR /&gt;The side effect of the combination of these workarounds is that all memory is not equal. The first 1 GB or so of the memory is considered "low" memory. The rest is "high" memory. For some purposes (like I/O buffers), high memory is not usable: low memory must be used. &lt;BR /&gt;&lt;BR /&gt;(For example, to transfer data from one segment of high memory to another, the system may need to copy it from the source location to low memory, then switch high memory banks and copy the data from low memory to the target segment.)&lt;BR /&gt;&lt;BR /&gt;It is possible to run out of low memory while still having plenty of high memory. I have a suspicion this is happening in your case.&lt;BR /&gt;&lt;BR /&gt;Please run:&lt;BR /&gt;egrep 'High|Low' /proc/meminfo&lt;BR /&gt;&lt;BR /&gt;to see your high/low memory situation.&lt;BR /&gt;&lt;BR /&gt;If your workload demands more than 1 GB of low memory, I'm afraid the only solution is to upgrade to a 64-bit version of the OS.&lt;BR /&gt;&lt;BR /&gt;For quite a while, I've been following this policy: if a new server is installed with more than 2 GB of memory or it seems likely that its memory will be expanded beyond that limit in the foreseeable future, I'd strongly recommend installing a 64-bit version of the OS from the beginning.&lt;BR /&gt;&lt;BR /&gt;MK</description>
      <pubDate>Fri, 13 Feb 2009 14:59:58 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358269#M35234</guid>
      <dc:creator>Matti_Kurkela</dc:creator>
      <dc:date>2009-02-13T14:59:58Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358270#M35235</link>
      <description>starting nfslock did not make any diff.&lt;BR /&gt;&lt;BR /&gt;# egrep 'High|Low' /proc/meminfo&lt;BR /&gt;HighTotal:     9567024 kB&lt;BR /&gt;HighFree:        18976 kB&lt;BR /&gt;LowTotal:       697836 kB&lt;BR /&gt;LowFree:         22064 kB&lt;BR /&gt;&lt;BR /&gt;How can i check the cpu is 32bit?Just curious. I know my linux kernel is 32bit&lt;BR /&gt;&lt;BR /&gt;Does nfsd uses high or low memory? In my case free is low on both cases.</description>
      <pubDate>Fri, 13 Feb 2009 15:53:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358270#M35235</guid>
      <dc:creator>skt_skt</dc:creator>
      <dc:date>2009-02-13T15:53:33Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358271#M35236</link>
      <description>I'm not sure of the 1 GB limitation on nowdays kernels. It's suppossed that a HIGHMEM enabled kernel can use all the system memory.&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://kerneltrap.org/node/2450" target="_blank"&gt;http://kerneltrap.org/node/2450&lt;/A&gt;</description>
      <pubDate>Fri, 13 Feb 2009 18:12:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358271#M35236</guid>
      <dc:creator>Ivan Ferreira</dc:creator>
      <dc:date>2009-02-13T18:12:44Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358272#M35237</link>
      <description>when run on a server, there is hardly any excuse for running the 32bit version on 64bit hardware.</description>
      <pubDate>Mon, 16 Feb 2009 09:27:30 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358272#M35237</guid>
      <dc:creator>dirk dierickx</dc:creator>
      <dc:date>2009-02-16T09:27:30Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358273#M35238</link>
      <description>&lt;!--!*#--&gt;Why not install and run the hugemem kernel?  Doesn't that allow using the first 4GB as lowmem and the rest as highmem or is my understanding of that kernel off?</description>
      <pubDate>Wed, 18 Feb 2009 07:16:21 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358273#M35238</guid>
      <dc:creator>JHKCMO</dc:creator>
      <dc:date>2009-02-18T07:16:21Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358274#M35239</link>
      <description>he must already be running hugemem otherwise he wouldn't be able to recognize all his ram.</description>
      <pubDate>Wed, 18 Feb 2009 09:56:32 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358274#M35239</guid>
      <dc:creator>dirk dierickx</dc:creator>
      <dc:date>2009-02-18T09:56:32Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358275#M35240</link>
      <description>&lt;!--!*#--&gt;Looks like uname is saying he's running on SMP:&lt;BR /&gt;&lt;BR /&gt;Linux hostname 2.4.21-53.ELsmp #1 SMP &lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Wed, 18 Feb 2009 15:00:03 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358275#M35240</guid>
      <dc:creator>JHKCMO</dc:creator>
      <dc:date>2009-02-18T15:00:03Z</dc:date>
    </item>
    <item>
      <title>Re: Linux kernel free mem usage clarification</title>
      <link>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358276#M35241</link>
      <description>i dont see any issue in recongizing the high/low memory on the system. see the post above</description>
      <pubDate>Wed, 18 Feb 2009 19:26:57 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-linux/linux-kernel-free-mem-usage-clarification/m-p/4358276#M35241</guid>
      <dc:creator>skt_skt</dc:creator>
      <dc:date>2009-02-18T19:26:57Z</dc:date>
    </item>
  </channel>
</rss>

