Internet Standards/CD1.mdf/winsock/hacker/93-12/000001_ICH211@ZAM001.Z…KFA-JUELICH.DE_Thu Dec 9 13:01:45 1993.msg

Internet Message Format | 1993-12-30 | 6KB

Received: from zam001.zam.kfa-juelich.de by SunSITE.Unc.EDU (5.65c+IDA/FvK-1.06) with SMTP
          id AA11249; Thu, 9 Dec 1993 06:06:01 -0500
Message-Id: <[e-mail]>
Received: from DJUKFA11 by ZAM001.ZAM.KFA-JUELICH.DE (IBM VM SMTP V2R2)
   with BSMTP id 2720; Thu, 09 Dec 93 12:04:37 +0100
Date:         Thu, 09 Dec 93 12:01:45 +0100
From: Thomas Heil <[e-mail]>
Organization: Forschungszentrum Juelich GmbH - Institut fuer Chemie 2
Subject:      How to reset conn. with LWP winsock (SO_LINGER unchangeable)
To: [e-mail]

Hello!

Perhaps someone can enlighten me:
Novell's LanWorkPlace WINSOCK.DLL does not allow the changing of the
SO_LINGER socket option (it's defaulted to SO_DONTLINGER, and that's the
way it stays). So ... how can I reset a connection when using this
implementation (which is normally done by setting SO_LINGER with a
zero timeout)?

/Thomas

---------------------------------------+--------------------------------------
Mail: Thomas Heil                      | EMail: BITNET: [e-mail]
      Forschungszentrum Juelich - ICG2 |      Internet: [e-mail]
      52425 Juelich                    | Phone: [phone]
      Germany                          | Fax:   [phone]
From [e-mail] Mon Dec 13 04:21:10 1993
Received: from usage.csd.unsw.OZ.AU by SunSITE.Unc.EDU (5.65c+IDA/FvK-1.06) with SMTP
          id AA08119; Sun, 12 Dec 1993 17:10:47 -0500
Received: by usage.csd.unsw.OZ.AU id AA04744
  (5.65c/IDA-1.4.4 for winsock-hackers%sunsite.unc.edu); Mon, 13 Dec 1993 09:10:45 +1100
Received: by atlas (4.1/1.35)
    id AA18486; Mon, 13 Dec 93 09:21:10 EST
Date: Mon, 13 Dec 93 09:21:10 EST
From: [e-mail] (Paul Brooks)
Message-Id: <9312122221.AA18486@atlas>
To: [e-mail]
Subject: Re: Valid socket
Newsgroups: alt.winsock
In-Reply-To: <[e-mail]>
Organization: TurboSoft Pty Ltd, Sydney, Australia
Cc: [e-mail]

In article <[e-mail]> you write:
|Hi:
|
|INVALID_SOCKET is defined as (SOCKET)(~0). Is the value 0 a valid socket? I
|have checked some winsock implementions. They all seemed to start from 1 to
|whatever.

>From my reading of the spec, yes, 0 is a valid number for a socket. The ONLY
value not allowed for a socket is ~0, which is defined that way because
a SOCKET is unsigned, so '-1' is not a valid value. In binary, -1 and ~0
both contain all '1' bits.

    Note that the value of sockets need not be integers starting from '1'.
It is perfectly possible that some implementations will use other types of
identifiers for sockets, such as near pointers to a descripter block, or
even far pointers in 32-bit versions. You cannot rely on
SOCKET values being simple positive integers, and testing the return result
for being ' <0 ' like much ported Unix code does will bite you badly.

--
Paul Brooks              |[e-mail]       |Emerging Standard:
TurboSoft Pty Ltd        |[e-mail]|  one that has not yet
579 Harris St., Ultimo   |                         |  been superseded.
Sydney Australia 2007    |ph: [phone]       |
From [e-mail] Fri Dec 13 00:55:46 1993
Received: from farallon.com (farallon.farallon.com) by SunSITE.Unc.EDU (5.65c+IDA/FvK-1.06) with SMTP
          id AA17677; Mon, 13 Dec 1993 09:53:28 -0500
Received: from waygate.farallon.com ([163.176.8.12]) by farallon.com (4.1/SMI-4.1)
    id AA11379; Mon, 13 Dec 93 06:53:06 PST
Message-Id: <[e-mail]>
Date: 13 Dec 1993 08:55:46 -0800
From: "Ping He" <[e-mail]>
Subject: RE: Valid socket
Return-Receipt-To: "Ping He" <[e-mail]>
To: "Paul Brooks" <[e-mail]>
Cc: [e-mail]

Thank you for your info.

Ping
_______________________________________________________________________________
To: Ping He
Cc: [e-mail]
From: Paul Brooks on Sun, Dec 12, 1993 2:10 PM
Subject: Re: Valid socket
RFC Header:Received: by waygate.farallon.com with SMTP;12 Dec 1993 14:10:29
-0800
Received: from usage.csd.unsw.OZ.AU by farallon.com (4.1/SMI-4.1)
    id AA08359; Sun, 12 Dec 93 14:10:41 PST
Received: by usage.csd.unsw.OZ.AU id AA04744
  (5.65c/IDA-1.4.4 for phe%farallon.com); Mon, 13 Dec 1993 09:10:45 +1100
Received: by atlas (4.1/1.35)
    id AA18486; Mon, 13 Dec 93 09:21:10 EST
Date: Mon, 13 Dec 93 09:21:10 EST
From: [e-mail] (Paul Brooks)
Message-Id: <9312122221.AA18486@atlas>
To: ping_he.kansas_mail@waygate
Subject: Re: Valid socket
Newsgroups: alt.winsock
In-Reply-To: <[e-mail]>
Organization: TurboSoft Pty Ltd, Sydney, Australia
Cc: [e-mail]

In article <[e-mail]> you write:
|Hi:
|
|INVALID_SOCKET is defined as (SOCKET)(~0). Is the value 0 a valid socket? I
|have checked some winsock implementions. They all seemed to start from 1 to
|whatever.

>From my reading of the spec, yes, 0 is a valid number for a socket. The ONLY
value not allowed for a socket is ~0, which is defined that way because
a SOCKET is unsigned, so '-1' is not a valid value. In binary, -1 and ~0
both contain all '1' bits.

    Note that the value of sockets need not be integers starting from '1'.
It is perfectly possible that some implementations will use other types of
identifiers for sockets, such as near pointers to a descripter block, or
even far pointers in 32-bit versions. You cannot rely on
SOCKET values being simple positive integers, and testing the return result
for being ' <0 ' like much ported Unix code does will bite you badly.

--
Paul Brooks              |[e-mail]       |Emerging Standard:
TurboSoft Pty Ltd        |[e-mail]|  one that has not yet
579 Harris St., Ultimo   |                         |  been superseded.
Sydney Australia 2007    |ph: [phone]       |