<?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: TCPIP socket error in Operating System - OpenVMS</title>
    <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635343#M50125</link>
    <description>Jack,&lt;BR /&gt;&lt;BR /&gt;a memory leak in DECC$RTL would probably be in user mode process memory and not affect the mbufs maintained by TCPIP in kernel mode.&lt;BR /&gt;&lt;BR /&gt;Which TCPIP V5.3 patch do you currently have installed ?&lt;BR /&gt;&lt;BR /&gt;Volker.</description>
    <pubDate>Sat, 01 Oct 2005 02:06:49 GMT</pubDate>
    <dc:creator>Volker Halle</dc:creator>
    <dc:date>2005-10-01T02:06:49Z</dc:date>
    <item>
      <title>TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635330#M50112</link>
      <description>After several days, telent, ftp and other tcpip related apps produce the following message on a 7.3-1 system:&lt;BR /&gt;&lt;BR /&gt;Internet interface error&lt;BR /&gt;socket ()&lt;BR /&gt;process quota exceeded&lt;BR /&gt;&lt;BR /&gt;I tried to increase the device sockets but this version of UCX (5.3) says it no longer supports it.  "ucx show comm" also no longer shows peak values of anything.  "ucx show comm/mem" shows how many mbufs are used, whatever they are.  I've increased the CHANNELCNT from 256 to 512 and RJOBLIM from 16 to 32, and autogened the system to see if it helps.&lt;BR /&gt;&lt;BR /&gt;Any other ideas?&lt;BR /&gt;</description>
      <pubDate>Tue, 27 Sep 2005 01:02:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635330#M50112</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-09-27T01:02:40Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635331#M50113</link>
      <description>Can you give more detail on the type of processes - interactive, batch? Have these been active all that time, or does it suddenly happen that all go down that way?&lt;BR /&gt;Anyway, check the process quota in UAF and SYSGEN: I think in particular BYTLM and TQELM &lt;BR /&gt;&lt;BR /&gt;Willem</description>
      <pubDate>Tue, 27 Sep 2005 01:23:35 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635331#M50113</guid>
      <dc:creator>Willem Grooters</dc:creator>
      <dc:date>2005-09-27T01:23:35Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635332#M50114</link>
      <description>HAve you looked at the errant processes with SDA (i.e. Anal/sys)? &lt;BR /&gt;&lt;BR /&gt;q&lt;BR /&gt;</description>
      <pubDate>Tue, 27 Sep 2005 01:45:07 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635332#M50114</guid>
      <dc:creator>Peter Quodling</dc:creator>
      <dc:date>2005-09-27T01:45:07Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635333#M50115</link>
      <description>The problem is first noticed when someone tries to telnet to the server from another system, typically a PC.  Then when they log on the server via DECnet, they try to telnet out of the server to another one and get the error message.  Same with ftp.&lt;BR /&gt;&lt;BR /&gt;No I haven't used SDA but I will now that you reminded me.  I've already tried ACC but not much I could find there.  Thanks.&lt;BR /&gt;</description>
      <pubDate>Tue, 27 Sep 2005 01:54:00 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635333#M50115</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-09-27T01:54:00Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635334#M50116</link>
      <description>If I get it right:&lt;BR /&gt;&lt;BR /&gt;PC --telnet --&amp;gt; VMS &lt;BR /&gt;PC --decnet --&amp;gt; VMS --telnet/ftp--&amp;gt; another server&lt;BR /&gt;&lt;BR /&gt;or:&lt;BR /&gt;&lt;BR /&gt;PC --telnet--&amp;gt; VMS --decnet--&amp;gt; VMS --telnet/ftp--&amp;gt; another server&lt;BR /&gt;&lt;BR /&gt;In both cases: what type of system is that? if VMS, check the user quota on that system.&lt;BR /&gt;&lt;BR /&gt;Check quota on the system reached by DECNet, as said: BYTLM, TQELM, but it might be you have to raise other ones as well. I could think of BIOLM and ASTLM as well. Take a look to the UAF settings of users TCPIP$FTP and TCPIP$TELNET</description>
      <pubDate>Tue, 27 Sep 2005 02:13:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635334#M50116</guid>
      <dc:creator>Willem Grooters</dc:creator>
      <dc:date>2005-09-27T02:13:41Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635335#M50117</link>
      <description>Anywhere --telnet --&amp;gt; VMS_A --&amp;gt; telnet anywhere works OK for a few days then one day the first telnet from anywhere won't connect.  Then:&lt;BR /&gt;&lt;BR /&gt;VMS_B --SET HOST --&amp;gt; VMS_A --telnet --&amp;gt; anywhere gives the error message&lt;BR /&gt;&lt;BR /&gt;At other times telnet from VMS_A is OK.&lt;BR /&gt;&lt;BR /&gt;I do recall one day I could telnet from VMS_A but not to it!</description>
      <pubDate>Tue, 27 Sep 2005 02:33:40 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635335#M50117</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-09-27T02:33:40Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635336#M50118</link>
      <description>could you be reaching the limit on the telnet service - see&lt;BR /&gt;$ TCPIP SHOW SERVICE TELNET/FULL&lt;BR /&gt;&lt;BR /&gt;with previous sessions not ended properly.</description>
      <pubDate>Tue, 27 Sep 2005 03:51:44 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635336#M50118</guid>
      <dc:creator>Ian Miller.</dc:creator>
      <dc:date>2005-09-27T03:51:44Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635337#M50119</link>
      <description>Nothing obvious yet.  Iâ  ll probably have to wait for the problem to happen again then see the evidence.  What I have found is when I do a tcpip show comm/mem every so often, the following keep increasing.  I wonder if this is a sign of a memory leak.&lt;BR /&gt;&lt;BR /&gt;9568 mbufs in use:&lt;BR /&gt;        3148 mbufs allocated to socket structures&lt;BR /&gt;        6290 mbufs allocated to protocol control blocks&lt;BR /&gt;</description>
      <pubDate>Thu, 29 Sep 2005 08:04:52 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635337#M50119</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-09-29T08:04:52Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635338#M50120</link>
      <description>Jack,&lt;BR /&gt;&lt;BR /&gt;how did you 'get rid of the problem' in the past ? Reboot ? If so, it could indeed be a memory leak. Does nonpaged pool usage increase as well (SHOW MEM/POOL/FULL) ?&lt;BR /&gt;&lt;BR /&gt;Volker.</description>
      <pubDate>Thu, 29 Sep 2005 08:23:17 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635338#M50120</guid>
      <dc:creator>Volker Halle</dc:creator>
      <dc:date>2005-09-29T08:23:17Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635339#M50121</link>
      <description>could you post &lt;BR /&gt;$ ana/sys&lt;BR /&gt;tcpip sh inetcb/stat&lt;BR /&gt;&lt;BR /&gt;it may show some stuff</description>
      <pubDate>Thu, 29 Sep 2005 08:27:56 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635339#M50121</guid>
      <dc:creator>labadie_1</dc:creator>
      <dc:date>2005-09-29T08:27:56Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635340#M50122</link>
      <description>Jack,&lt;BR /&gt;&lt;BR /&gt;which version of TCPIP are you running ($ TCPIP SHOW VERSION) ?&lt;BR /&gt;&lt;BR /&gt;There is a solution description in TCPIP V5.3 ECO 4 to a problem, which sounds similar to yours:&lt;BR /&gt;&lt;BR /&gt;After upgrade to V5.3, system runs for a while then see %SYSTEM-F-EXQUOTA process quota exceeded when trying to use a socket application.  $ TCPIP SHOW COMM/MEM shows close to 10,000 socket structures allocated.&lt;BR /&gt;&lt;BR /&gt;Volker.</description>
      <pubDate>Thu, 29 Sep 2005 08:29:22 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635340#M50122</guid>
      <dc:creator>Volker Halle</dc:creator>
      <dc:date>2005-09-29T08:29:22Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635341#M50123</link>
      <description>Volker&lt;BR /&gt;&lt;BR /&gt;We have the same idea in mind I guess :-)</description>
      <pubDate>Thu, 29 Sep 2005 08:33:26 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635341#M50123</guid>
      <dc:creator>labadie_1</dc:creator>
      <dc:date>2005-09-29T08:33:26Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635342#M50124</link>
      <description>Discovered that patch kit VMS731_ACRTL-V0400 has the following fix, which may fix the problem.  Before I apply this patch Iâ  ll first apply the latest ECO for TCPIP services to see if it has any impact the accumulating mbufs.&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;6.2.13  socket_fd() memory leak&lt;BR /&gt;&lt;BR /&gt;6.2.13.1  Problem Description:&lt;BR /&gt;&lt;BR /&gt;A memory leak can occur in programs using socket_fd.&lt;BR /&gt;&lt;BR /&gt;Images Affected:&lt;BR /&gt;&lt;BR /&gt;-  [SYSLIB]DECC$SHR.EXE&lt;BR /&gt;&lt;BR /&gt;-  [SYSLIB]DECC$SHR_EV56.EXE&lt;BR /&gt;&lt;BR /&gt;</description>
      <pubDate>Fri, 30 Sep 2005 23:26:34 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635342#M50124</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-09-30T23:26:34Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635343#M50125</link>
      <description>Jack,&lt;BR /&gt;&lt;BR /&gt;a memory leak in DECC$RTL would probably be in user mode process memory and not affect the mbufs maintained by TCPIP in kernel mode.&lt;BR /&gt;&lt;BR /&gt;Which TCPIP V5.3 patch do you currently have installed ?&lt;BR /&gt;&lt;BR /&gt;Volker.</description>
      <pubDate>Sat, 01 Oct 2005 02:06:49 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635343#M50125</guid>
      <dc:creator>Volker Halle</dc:creator>
      <dc:date>2005-10-01T02:06:49Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635344#M50126</link>
      <description>No patch/ECO for TCPIP is installed at the moment.&lt;BR /&gt;</description>
      <pubDate>Sat, 01 Oct 2005 03:31:33 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635344#M50126</guid>
      <dc:creator>Jack Churchill</dc:creator>
      <dc:date>2005-10-01T03:31:33Z</dc:date>
    </item>
    <item>
      <title>Re: TCPIP socket error</title>
      <link>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635345#M50127</link>
      <description>install the ECO kit for V5.3&lt;BR /&gt;&lt;A href="ftp://ftp.itrc.hp.com/openvms_patches/layered_products/alpha/DEC-AXPVMS-TCPIP_ECO-V0503-184-4.txt" target="_blank"&gt;ftp://ftp.itrc.hp.com/openvms_patches/layered_products/alpha/DEC-AXPVMS-TCPIP_ECO-V0503-184-4.txt&lt;/A&gt;&lt;BR /&gt;&lt;A href="ftp://ftp.itrc.hp.com/openvms_patches/layered_products/alpha/DEC-AXPVMS-TCPIP_ECO-V0503-184-4.PCSI-DCX_AXPEXE" target="_blank"&gt;ftp://ftp.itrc.hp.com/openvms_patches/layered_products/alpha/DEC-AXPVMS-TCPIP_ECO-V0503-184-4.PCSI-DCX_AXPEXE&lt;/A&gt;&lt;BR /&gt;&lt;BR /&gt;it does look like a memory leak. The ECO kit release notes do not mention all the fixes contained.</description>
      <pubDate>Sat, 01 Oct 2005 09:06:41 GMT</pubDate>
      <guid>https://community.hpe.com/t5/operating-system-openvms/tcpip-socket-error/m-p/3635345#M50127</guid>
      <dc:creator>Ian Miller.</dc:creator>
      <dc:date>2005-10-01T09:06:41Z</dc:date>
    </item>
  </channel>
</rss>

