<?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: Monitoring System Calls................... in Operating System - HP-UX</title>
    <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598503#M33778</link>
    <description>Pandian,&lt;BR /&gt;&lt;BR /&gt;I am afraid you can capture that information through Measureware. You can find all the metrics that can be captured in /var/opt/perf/reptall. Capturing all the systems called related to a particular process would be too much to collect. Just try capturing them using tusc or trace for five mintutes, you will see the size of the log file. If you really need to troubleshoot, you need to use either tusc or trace depending on the OS version.&lt;BR /&gt;&lt;BR /&gt;-Sri</description>
    <pubDate>Mon, 22 Oct 2001 03:14:45 GMT</pubDate>
    <dc:creator>Sridhar Bhaskarla</dc:creator>
    <dc:date>2001-10-22T03:14:45Z</dc:date>
    <item>
      <title>Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598500#M33775</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;Using Measureware, is there a way to capture the name the sytem calls and their rates?. I could not find a metric for that. &lt;BR /&gt;&lt;BR /&gt;I can monitor the number of system calls in global metrics. &lt;BR /&gt;&lt;BR /&gt;I know Glance has the capability to monitor the name and the rate of system calls. &lt;BR /&gt;&lt;BR /&gt;Can measureware capture it??</description>
      <pubDate>Mon, 22 Oct 2001 02:41:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598500#M33775</guid>
      <dc:creator>R.M.PANDIAN</dc:creator>
      <dc:date>2001-10-22T02:41:51Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598501#M33776</link>
      <description>Hello,&lt;BR /&gt;I went through the measureware metric files.&lt;BR /&gt;I captured these two metrics&lt;BR /&gt;&lt;BR /&gt;PROC_CPU_SYSCALL_UTIL&lt;BR /&gt;PROC_CPU_SYSCALL_TIME    &lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 22 Oct 2001 02:54:25 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598501#M33776</guid>
      <dc:creator>T G Manikandan</dc:creator>
      <dc:date>2001-10-22T02:54:25Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598502#M33777</link>
      <description>Hello,&lt;BR /&gt;i also found the below metrics&lt;BR /&gt;GBL_CPU_SYSCALL_UTIL&lt;BR /&gt;  GBL_CPU_NORMAL_UTIL   &lt;BR /&gt;GBL_SYSCALL_RATE</description>
      <pubDate>Mon, 22 Oct 2001 02:57:24 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598502#M33777</guid>
      <dc:creator>T G Manikandan</dc:creator>
      <dc:date>2001-10-22T02:57:24Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598503#M33778</link>
      <description>Pandian,&lt;BR /&gt;&lt;BR /&gt;I am afraid you can capture that information through Measureware. You can find all the metrics that can be captured in /var/opt/perf/reptall. Capturing all the systems called related to a particular process would be too much to collect. Just try capturing them using tusc or trace for five mintutes, you will see the size of the log file. If you really need to troubleshoot, you need to use either tusc or trace depending on the OS version.&lt;BR /&gt;&lt;BR /&gt;-Sri</description>
      <pubDate>Mon, 22 Oct 2001 03:14:45 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598503#M33778</guid>
      <dc:creator>Sridhar Bhaskarla</dc:creator>
      <dc:date>2001-10-22T03:14:45Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598504#M33779</link>
      <description>TGM,&lt;BR /&gt;&lt;BR /&gt;None of the metrics that u suggested records the name of the system calls.&lt;BR /&gt;&lt;BR /&gt;Sri,&lt;BR /&gt;&lt;BR /&gt;I went through all metrics of Measureware, none of them refers to capturing the names of the system calls used on a system wide basis.&lt;BR /&gt;&lt;BR /&gt;Glance Plus is able to report the name of the systems calls executed on a online basis.&lt;BR /&gt;&lt;BR /&gt;trace or like utilities reports the stats for a specific process. I need the system wide stats.&lt;BR /&gt;&lt;BR /&gt;My requirement is that I need to be able to record the Name of the system calls and the rate at which they are being used by the processes on a system wide basis (Global).&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Any suggestions?&lt;BR /&gt;&lt;BR /&gt;- Pandian&lt;BR /&gt;</description>
      <pubDate>Mon, 22 Oct 2001 04:14:05 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598504#M33779</guid>
      <dc:creator>R.M.PANDIAN</dc:creator>
      <dc:date>2001-10-22T04:14:05Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598505#M33780</link>
      <description>HP has a new tool (not sure if its supported or not) called Prospect which does exactly what you're looking to do.  It only works on 11.x though.  Although I have to warn you, it will generate lots of data, especially if you're looking to monitor system calls globally.&lt;BR /&gt;&lt;BR /&gt;&lt;A href="ftp://ftp.cup.hp.com/dist/networking/tools/prospect/" target="_blank"&gt;ftp://ftp.cup.hp.com/dist/networking/tools/prospect/&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;Hope this helps.&lt;BR /&gt;&lt;BR /&gt;-Santosh</description>
      <pubDate>Mon, 22 Oct 2001 07:56:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598505#M33780</guid>
      <dc:creator>Santosh Nair_1</dc:creator>
      <dc:date>2001-10-22T07:56:00Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598506#M33781</link>
      <description>Hi Pandian,&lt;BR /&gt;&lt;BR /&gt;You can use Glance in "adviser only" mode to capture data in a log file.  You just need to create "adviser syntax" containing the metrics that you want to capture.  For example, save this to a file, called syscalls.syntax in your current directory:&lt;BR /&gt;&lt;BR /&gt;/Begin/&lt;BR /&gt;&lt;BR /&gt;PRINT "=============================="&lt;BR /&gt;PRINT "           ", gbl_stattime, "           "&lt;BR /&gt;PRINT "=============================="&lt;BR /&gt;SYSTEMCALL LOOP&lt;BR /&gt;        PRINT syscall_call_name|12, " ", syscall_call_rate&lt;BR /&gt;&lt;BR /&gt;/End/&lt;BR /&gt;&lt;BR /&gt;Then run:&lt;BR /&gt;&lt;BR /&gt;glance -adviser_only -syntax syscalls.syntax -j 60&lt;BR /&gt;&lt;BR /&gt;You can check glance man pages for more command line options.  Also, the document /opt/perf/paperdocs/gp/C/adviser.pdf contains a more interesting example that uses systemcall loop.  More adviser syntax examples are available under a subdirectory of /opt/perf/examples.&lt;BR /&gt;&lt;BR /&gt;Can Measureware capture this?&lt;BR /&gt;&lt;BR /&gt;Well, not directly, but if you use Measureware DSI logs, you can collect any Glance metrics to a particular type of a log file which can then be analyzed by Measureware tools, PerfView, etc.  For more information, see the document /opt/perf/paperdocs/mwa/C/mwadsi.pdf .&lt;BR /&gt;Before reading any of these documents in depth, I would highly recommend the following:&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://devresource.hp.com/devresource/Docs/TechPapers/PakPerform.pdf" target="_blank"&gt;http://devresource.hp.com/devresource/Docs/TechPapers/PakPerform.pdf&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;I hope this is helpful.&lt;BR /&gt;&lt;BR /&gt;Regards ... Mladen</description>
      <pubDate>Tue, 23 Oct 2001 03:51:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598506#M33781</guid>
      <dc:creator>Mladen Despic</dc:creator>
      <dc:date>2001-10-23T03:51:33Z</dc:date>
    </item>
    <item>
      <title>Re: Monitoring System Calls...................</title>
      <link>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598507#M33782</link>
      <description>It looks like the above link is outdated.  Here is the current one:&lt;BR /&gt;&lt;BR /&gt;&lt;A href="http://h21007.www2.hp.com/dspp/files/unprotected/PakPerform.pdf" target="_blank"&gt;http://h21007.www2.hp.com/dspp/files/unprotected/PakPerform.pdf&lt;/A&gt;</description>
      <pubDate>Tue, 23 Oct 2001 04:00:51 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-hp-ux/monitoring-system-calls/m-p/2598507#M33782</guid>
      <dc:creator>Mladen Despic</dc:creator>
      <dc:date>2001-10-23T04:00:51Z</dc:date>
    </item>
  </channel>
</rss>

