From: "Roderick Klein" Received: from mxout2.mailhop.org ([63.208.196.166] verified) by 2rosenthals.com (CommuniGate Pro SMTP 5.0.9) with ESMTP id 269983 for os2-wireless_users@2rosenthals.com; Fri, 04 Aug 2006 11:57:53 -0400 Received: from mxin1.mailhop.org ([63.208.196.175]) by mxout2.mailhop.org with esmtp (Exim 4.51) id 1G923c-000DRb-Cy for os2-wireless_users@2rosenthals.com; Fri, 04 Aug 2006 11:57:52 -0400 Received: from mail.mensys.nl ([194.109.204.136] helo=mensys.nl) by mxin1.mailhop.org with esmtp (Exim 4.51) id 1G923b-000FRL-Sn for os2-wireless_users@2rosenthals.com; Fri, 04 Aug 2006 11:57:51 -0400 Received: from [213.84.84.66] (HELO THINKPADRW) by mensys.nl (CommuniGate Pro SMTP 5.0.1) with SMTP id 1646572 for os2-wireless_users@2rosenthals.com; Fri, 04 Aug 2006 17:57:45 +0100 To: "OS/2 Wireless Users Mailing List" Date: Fri, 04 Aug 2006 18:01:20 +0100 Reply-To: "Roderick Klein" Priority: Normal X-Mailer: PMMail 1.96a For OS/2 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Subject: Re: [OS2Wireless]Re: SOCKETS.SYS Error Message-ID: X-Mail-Handler: MailHop by DynDNS X-Spam-Score: -2.5 (--) On Fri, 04 Aug 2006 16:50:35 +0100 (BST), Dave Saville wrote: >On Fri, 04 Aug 2006 11:15:16 -0400, Lewis G Rosenthal wrote: > >>On 08/04/06 06:41 am, Roderick Klein thus wrote : >>> I have not realy followed this thread but the stack of OS/2 is not that bad. >>> You get traps in sockets.sys ? >>> >>> Have you the the file \mptn\etc\inetcfg.ini ? >>> At last line does it have the line >>> SACK 1 0 0 1 >>> Change the first 0 into a one. >>> >>> >>Good point, Roderick. I forgot all about that tip. > >Which means and does what? I hate it when the fix is "change x to y" with no >explanation. I mean change the first 1 into a zero. This file was distributed on eCS 1.1 CD #2 "TCP/IP tuner. SACK stands for Selective acknowlegdement. It should make TCP/IP communication more efficient. However duo to a bug in the OS/2 TCP/IP stack this option results in the IP stack eating up more and more resources until the machine has no more kernel resources. Roderick Klein