
                   comp.os.os2.networking.tcp_ip    (Usenet)

                 Saturday, 28-Aug-1999 to Friday, 03-Sep-1999

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         27-Aug-99 21:52:27
  To: All                                               28-Aug-99 03:31:24
Subj: Re: multiple subnet support: Win9x vs OS/2

From: ad354@FreeNet.Carleton.CA (James Owens)

Henk kelder (nospam_hkelder@capgemini.nl) writes:
> If I remember correct you must besides adding the route entry defining
> yourself as the router to both networks, also let ROUTED run...

Still no go. . .

In all I use TCPCFG to set LAN0 to DHCP, turn IP forwarding on, and
run "routed" in the background, then add the following to SETUP.CMD (in this
order):

ifconfig lan0 n.n.99.2 netmask 255.255.255.0 alias
route add subnet n.n.99.0 n.n.99.2 0  
route add subnet n.n.60.0 n.n.99.2 0  

(where n.n is the network number)

After rebooting, I see from netstat -r that I have two routing paths:

n.n.60.0   n.n.60.66   255.255.255.0
n.n.99.0   n.n.99.2    255.255.255.0

When I use the route add command from a prompt to say, for example:

route add subnet n.n.60.0 n.n.99.2

or for that matter

route add subnet n.n.99.2 n.n.60.66

it says it cannot add the route because "entry exists."

I'm coming to the conclusion that OS/2 just can't do this for some reason.

I downloaded latest41.exe, but when I installed it (replaced some sys
files in MPTN/PROTOCOL) my system trapped at startup.




--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: abstewa@ibm.net                                   27-Aug-99 23:16:22
  To: All                                               28-Aug-99 03:31:25
Subj: Re: Where do you put a hosts or lmosts file

From: Alon Stewart <abstewa@ibm.net>

That is not what the resolv files are for.  They are for name resolution in
the
domain.  The hosts file is where you want to put the IP's of servers or
workstations you want to attach to.  The TCP/IP configuration is the best
place
to add IP's because you need to check the box telling it to look through the
hosts file before looking for a name server.  Lookups will be faster.

Kenny Trussell wrote:

> Thanks. I'll try that. Since I left the post, I read some of the other
> messages here and got the idea that the "resolv" and "resolv2" files were
> equivalent to a hosts file. I edited them and added my server name and IP,
> but it didn't fix the problem. I really appreciate your tip. I guess I just
> haven't seen the TCP/IP configuration program. I'll look for it tomorrow.
>
> James Knott <jknott@ibm.net> wrote in message
> news:e2st3odSR87O089yn@ibm.net...
> > In article <D6pt3.732$d5.65968@eagle.america.net>,
> > "Kenny Trussell" <kenny.trussell@thielekaolin.com> wrote:
> > >I have one OS/2 Warp 3 machine to get working on a TCP/IP network. I only
> > >need the machine to connect to one server. I would like to use a hosts
> file
> > >for the name to IP address resolution. I cannot find in any documentation
> > >where to put the file and whether it needs to be named "hosts" or
> "lmhosts".
> >
> > While you could create one manually, the best way is to use TCP/IP
> > Configuration.  Go to the 2nd page of the Hostnames tab.  You can
> > enter a list of hosts there and there is also an option to check that
> > list first.
> >
> >
> >  --
> > E-mail jknott@ca.ibm.com
> > _________________________________________________________________________
> > The above opinions are my own and not those of ISM Corp., a subsidiary of
> > IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: andre@grueneberg.de                               28-Aug-99 01:05:09
  To: All                                               28-Aug-99 10:43:13
Subj: Safe way to backlevel TCP/IP

From: andre@grueneberg.de (Andre Grueneberg)

  -=( Hi Kris )=-

Kris Kadela schrieb:

 > The only new thing still here is FP11. I am wondering if that is the
 > root of the problem as the traps still accour every 2-3 hours but
 > only during network activity.

Do you use ProcessCommander? I had some trouble after installing FP40 on Warp3 


that caused problems after some hours...after deinstalling PC everything 
worked okay again...

  -=( CU Andre )=-         E-Mail: andre@grueneberg.de

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Gateway of Andre Grueneberg (1:109/42)

+----------------------------------------------------------------------------+

From: nospam_hkelder@capgemini.nl                       28-Aug-99 10:19:28
  To: All                                               28-Aug-99 10:43:13
Subj: Re: multiple subnet support: Win9x vs OS/2

From: Henk kelder <nospam_hkelder@capgemini.nl>

James Owens wrote:
> 
> Henk kelder (nospam_hkelder@capgemini.nl) writes:
> > If I remember correct you must besides adding the route entry defining
> > yourself as the router to both networks, also let ROUTED run...
> 
> Still no go. . .
> 
> In all I use TCPCFG to set LAN0 to DHCP, turn IP forwarding on, and
> run "routed" in the background, then add the following to SETUP.CMD (in this
> order):
> 
> ifconfig lan0 n.n.99.2 netmask 255.255.255.0 alias
> route add subnet n.n.99.0 n.n.99.2 0
> route add subnet n.n.60.0 n.n.99.2 0
> 

I don't understand. First you state the you use DHCP for LAN0, then you
assign a specific number to LAN0.


> (where n.n is the network number)
> 
> After rebooting, I see from netstat -r that I have two routing paths:
> 
> n.n.60.0   n.n.60.66   255.255.255.0
> n.n.99.0   n.n.99.2    255.255.255.0
> 
> When I use the route add command from a prompt to say, for example:
> 
> route add subnet n.n.60.0 n.n.99.2
> 
> or for that matter
> 
> route add subnet n.n.99.2 n.n.60.66
> 
> it says it cannot add the route because "entry exists."
> 
> I'm coming to the conclusion that OS/2 just can't do this for some reason.
> 
> I downloaded latest41.exe, but when I installed it (replaced some sys
> files in MPTN/PROTOCOL) my system trapped at startup.
> 
> --
> 

I am not sure, but you could try to add a route like:
route add host n.n.60.66 n.n.99.2 netmask 255.255.255.0

I my office I am using a OS/2 warp 3 PC to route between our local lan
and a separate Tokenring subnet. The OS/2 machine has an ethernet and a
tokenring adapter installed.
It works there perfectly. I must however check how the box is
configured.

Please send me a mail to remember me to check the configuration when I
am in. (will be wednesday).

Henk

-- 
Remove nospam when replying..

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: capgemini.nl (1:109/42)

+----------------------------------------------------------------------------+

From: mail@ridax.se                                     28-Aug-99 08:37:12
  To: All                                               28-Aug-99 10:43:13
Subj: Re: Can I telnet into OS/2 and auto start a program?

From: mail@ridax.se (Mikael Wahlgren)

:>        It works. But, sadly, all the messages are shown in the window which
:>	telnet daeman runs. The remote window seems to be dead?!
:>
:>	How to correct such problem? Does the User ID, Group ID, or Shell
:>	setup by passwd.exe have any effects on it?

You can use OS2You as your telnet daemon instead...

Mikael Wahlgren - mail@ridax.se
Ridax programutveckling - PM2You/OS2You Remote Control for OS/2
FTP 194.52.57.138 - http://welcome.to/ridax
WIN2You - Remote Control for Windows 95/NT

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Ridax programutveckling (1:109/42)

+----------------------------------------------------------------------------+

From: kris@dgraph.com                                   28-Aug-99 03:40:06
  To: All                                               28-Aug-99 10:43:13
Subj: Re: Safe way to backlevel TCP/IP

From: Kris Kadela <kris@dgraph.com>


Andre Grueneberg wrote:
> 
>   -=( Hi Kris )=-
> 
> Kris Kadela schrieb:
> 
>  > The only new thing still here is FP11. I am wondering if that is the
>  > root of the problem as the traps still accour every 2-3 hours but
>  > only during network activity.
> 
> Do you use ProcessCommander? I had some trouble after installing FP40 on
Warp3
> 
> that caused problems after some hours...after deinstalling PC everything
> worked okay again...

Nope. never had the need for PC.
Anyway, I went back to the 16 bit stack and all is fine now.


> 
>   -=( CU Andre )=-         E-Mail: andre@grueneberg.de

-- 

**********************
DigiGraph Technical
http://www.dgraph.com
**********************

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: DigiGraph Technical (1:109/42)

+----------------------------------------------------------------------------+

From: tomlins@cam.org                                   28-Aug-99 10:52:20
  To: All                                               28-Aug-99 10:43:13
Subj: a Hung system with PEER and TCPBEUI

From: tomlins@cam.org

From: tomlins@cam.org
Newsgroups: comp.os.os2.networking.tcp-ip;,comp.os.os2.bugs
Subject: a hung system with PEER...
Date: Sat, 28 Aug 1999 03:34:44 GMT
Message-ID: <c1.2c.2RzS2s$09y@TOMLINS.HIP.CAM.ORG>
Reply-To: tomlins@cam.org

Hi,

I am using W4, FP11, mts8420, peer 8412.  When I try to configure peer using 
tcpbeui the system hangs, during the next boot, when setting the IP address.
This happens with both ipconfig and dhcpcd.  I have included protocol.ini,
ibmlan.ini and the section of the config.sys added by mpts and peerinst.  I
have
tried everything I can think of...

When the box hangs its solid, no numlock keyboard resp, no trap, no trace,
just
a boat anchor untill reset is pressed.  Works great if TCPBEUI is not in the
picture, NETBEUI alone is ok too.

I really *would* like to be able to share files with the other boxes here.

protocol.ini:

[PROT_MAN]

   DRIVERNAME = PROTMAN$

[IBMLXCFG]

   tcpbeui_nif = tcpbeui.nif
   tcpip_nif = tcpip.nif
   KTC100_nif = KTC100.nif
   DLKFETI_nif = DLKFETI.nif

[NETBIOS]

   DriverName = netbios$
   ADAPTER0 = tcpbeui$,0

[tcpbeui_nif]

   DriverName = tcpbeui$
   Bindings = KTC100_nif
   NODETYPE = "B-Node"
   OS2TRACEMASK = 0x0
   SESSIONS = 130
   NCBS = 225
   NAMES = 21
   SELECTORS = 15
   USEMAXDATAGRAM = "NO"
   NETBIOSTIMEOUT = 500
   NETBIOSRETRIES = 2
   NAMECACHE = 1000
   PRELOADCACHE = "NO"
   NAMESFILE = 0
   DATAGRAMPACKETS = 20
   PACKETS = 50
   INTERFACERATE = 300

[tcpip_nif]

   DriverName = TCPIP$
   Bindings = KTC100_nif,DLKFETI_nif
   IfType = 6,6

[KTC100_nif]

   DriverName = KTC100$
   SIA_Mode = "AUTOSENSE"

[DLKFETI_nif]

   DriverName = DLKFET$


ibmlan.ini: (some comments removed)

; OS/2 Peer initialization file

[networks]

  net1 = tcpbeui$,0,LM10,34,70,14
; This information is read by the redirector at device initialization time.

[requester]

  COMPUTERNAME = OSCAR
  DOMAIN = FAMILY
; The following parameters generally do not need to be
; changed by the user.
  charcount = 16
  chartime = 250
  charwait = 3600
  keepconn = 600
  keepsearch = 600
  maxcmds = 16
  maxerrorlog = 100
  maxthreads = 10
  maxwrkcache = 64
  numalerts = 12
  numcharbuf = 10
  numservices = 7
  numworkbuf = 15
  numdgrambuf = 14
  othdomains = 
  printbuftime = 90
  sesstimeout = 45
  sizcharbuf = 512
  sizerror = 1024
  sizworkbuf = 4096
  useallmem = no
; The wrkheuristics parameter sets a variety of requester fine-tuning
; options.  Brief descriptions are provided here for each digit.  More
; complete descriptions, along with explanations of relationships between
; digits can be found in the OS/2 Peer Technical Reference.
;
; The next lines help you to locate bits in the wrkheuristics entry.
;                           1         2         3         4
;                 012345678901234567890123456789012345678901
  wrkheuristics = 111111112131111111000101112011122100121110
  wrkservices = MESSENGER,PEER
  wrknets = NET1

[messenger]

  logfile = messages.log
  sizmessbuf = 4096

[peer]

  auditing = resource
; The following parameters generally do not need to be
; changed by the user. NOTE : srvnets= is represented in
; the server info struct as a 16-bit lan mask. Srvnet names
; are converted to indexes within [networks] for the named nets.
  guestacct = guest
  autodisconnect = -1
  forwardauth = no
  maxauditlog = 100
  maxchdevjob = 6
  maxchdevq = 2
  maxchdevs = 2
  maxconnections = 128
  maxlocks = 64
  maxopens = 160
  maxsearches = 150
  maxsessopens = 120
  maxsessreqs = 25
  maxsessvcs = 1
  maxshares = 64
  maxusers = 32
  numbigbuf = 6
  numfiletasks = 1
  numreqbuf = 48
  sizreqbuf = 4096
  srvanndelta = 3000
  srvannounce = 180
  srvcomment = Ed
  srvhidden = no
; The srvheuristics parameter sets a variety of server fine-tuning
; options.  Brief descriptions are provided here for each digit.  More
; complete descriptions, along with explanations of relationships between
; digits can be found in the OS/2 Peer Technical Reference.
;
; The next lines help you to locate bits in the srvheuristics entry.
;                           1         2
;                 012345678901234567890
  srvheuristics = 111101411113110013311
  SRVSERVICES = 
  srvnets = NET1

[replicator]

  replicate = IMPORT
  importpath = D:\ibmlan\repl\import
  tryuser = yes
  password = 

[services]

; Correlates name of service to pathname of service program.
; The pathname must be either
;       1) an absolute path (including the drive specification)
;                       OR
;       2) a path relative to the IBMLAN root
  messenger = services\msrvinit.exe
  peer = services\peerinit.exe
  replicator = services\replicat.exe
  requester = services\wksta.exe

config.sys:

LIBPATH=D:\MUGLIB\DLL;E:\apps\comm\PROGRAM;.;D:\IBMLAN\NETLIB;D:\OS2\DLL;D:\MPT
N\DLL;D:\ibmcom\DLL;D:\IBMI18N\DLL;D:\OS2\MDOS;D:\;D:\OS2\APPS\DLL;D:\JAVA11\DL
L;D:\MMOS2\DLL;D:\IBMINST;D:\NSC\DLL;D:\TCPIP\DLL;D:\TCPIP\PCOMOS2;D:\TCPIP\UMA
IL;E:\BONUSPAK\IBMWORKS;E:\EMX\DLL;E:\XFREE86\LIB;D:\OS2\FS\EXT2;E:\DIALS;
SET
PATH=D:\MPTN\BIN;D:\ibmcom;E:\apps\comm\PROGRAM;D:\IBMLAN\NETPROG;D:\MUGLIB;D:\
OS2;D:\OS2\SYSTEM;D:\OS2\MDOS\WINOS2;D:\OS2\INSTALL;D:\;D:\OS2\MDOS;D:\OS2\APPS
;D:\MMOS2;.\.;E:\TINYED;E:\UTILS\FM2;E:\UTILS\ZIP;D:\NSC;D:\TCPIP\BIN;D:\TCPIP\
PCOMOS2;D:\TCPIP\UMAIL;E:\BONUSPAK\IBMWORKS;D:\SIO;E:\EMX\BIN;E:\EMX\LESS;E:\XF
REE86\BIN;D:\OS2\FS\EXT2;E:\UTILS\GO;E:\DIALS;D:\Java11\rmi-iiop\bin;D:\JAVA11\
BIN;
..
CALL=D:\ibmcom\PROTOCOL\NETBIND.EXE
RUN=D:\ibmcom\LANMSGEX.EXE
SET NLSPATH=D:\MPTN\MSG\NLS\%N;
SET ETC=D:\MPTN\ETC
DEVICE=D:\MPTN\PROTOCOL\SOCKETS.SYS
DEVICE=D:\MPTN\PROTOCOL\AFOS2.SYS
DEVICE=D:\MPTN\PROTOCOL\AFINET.SYS
DEVICE=D:\MPTN\BIN\VDOSTCP.VDD
RUN=D:\MPTN\BIN\CNTRL.EXE
CALL=D:\OS2\CMD.EXE /Q /C D:\MPTN\BIN\MPTSTART.CMD >NUL
RUN=D:\ibmcom\PROTOCOL\NBTCP.EXE
RUN=D:\MPTN\BIN\VDOSCTL.EXE
DEVICE=D:\IBMCOM\PROTOCOL\TCPBEUI.OS2
DEVICE=D:\IBMLAN\NETPROG\RDRHELP.200
IFS=D:\IBMLAN\NETPROG\NETWKSTA.200 /I:D:\IBMLAN /N
DEVICE=D:\IBMCOM\PROTOCOL\NETBIOS.OS2
DEVICE=D:\IBMCOM\MACS\KTC100.OS2
DEVICE=D:\IBMCOM\MACS\DLKFET.OS2
RUN=D:\IBMLAN\NETPROG\LSDAEMON.EXE
SET DLSINI=D:\IBMLAN\NETPROG\NETGUI.INI
SET INIT_FILE_NAMES=netgui
SET INIT_FILE_RANGES=200
SET WPS_COMMUNICATION=YES
SET LANINSTEP=



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: me (1:109/42)

+----------------------------------------------------------------------------+

From: jknott@ibm.net                                    28-Aug-99 07:34:11
  To: All                                               28-Aug-99 10:43:14
Subj: Re: How do I Change IP Address info

From: jknott@ibm.net (James Knott)

In article <HGTw3.547$H.32280@news.uswest.net>,
"Justin Greywolf" <diego@cryogen.com> wrote:
>
>James Knott wrote in message <9vmw3odSRYfO089yn@ibm.net>...
>
>>>It could not hurt to also change the IP statement in the MPTCONFG.ini,
>>>located in the same directory.
>>
>>Why would you want to do that?  Changing setup.cmd, either manually or
>>via tcpcfg does what's required.
>
>
>Not if you are changing the IP.  You need to change it in both files or it
>will lead to problems.

What problems.  According to the "Network Adapters and Protocol 
Services Guide" (quoted below), if TCP/IP is installed, most of that 
file, including the part containing the IP address is ignored.



" The Sockets MPTS section contains the TCP/IP configuration
parameters.  When used, this section generates an  MPTCONFIG.INI file
with the values specified.  During MPTS installation/configuration (if
there is no IBM TCP/IP  product on the workstation), the values
specified in the Sockets MPTS section are used to generate the
SETUP.CMD file. 

The MPTS section contains several segments that are [CONTROL],
[IFCONFIG], [ROUTE], [DHCP], and [NETBIOS].  If  the workstation
currently has an IBM TCP/IP Base Kit product on it, only the [CONTROL]
section needs to be  specified (if required) to configure Local IPC
sockets, NetBIOS Sockets, or TCP/IP Sockets access.  If the 
Workstation does not have the IBM TCP/IP product on the system, a
complete Sockets MPTS section (with all  segments properly completed)
must be specified to enable TCP/IP sockets access. 

You can specify a valid MPTCONFG.INI file within the Sockets MPTS
section.  The information specified in this  section overwrites an
existing MPTCONFG.INI file.  To update a currently configured
MPTCONFG.INI file, use the  MPTS Update Section.  The Sockets MPTS
section should be placed after any NetBIOS sections."



As shown above, only the control part of the file is used when TCP/IP 
is installed.  The IP address is in the ipconfig section, and 
therefore ignored.


-- 
E-mail jknott@ca.ibm.com
_________________________________________________________________________
The above opinions are my own and not those of ISM Corp., a subsidiary of
IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: jknott@ibm.net                                    28-Aug-99 07:48:12
  To: All                                               28-Aug-99 10:43:14
Subj: Re: multiple subnet support: Win9x vs OS/2

From: jknott@ibm.net (James Knott)

In article <7q5slt$np7@freenet-news.carleton.ca>,
ad354@FreeNet.Carleton.CA (James Owens) wrote:
>
>James Knott (jknott@ibm.net) writes:
>> In article <7q4046$12u@freenet-news.carleton.ca>,
>> ad354@FreeNet.Carleton.CA (James Owens) wrote:
>>>
>>>Using an ip dump, I have found that a Windows 9x machine on two
>>>subnets over the same physical network will provide the
>>>correct source address for the subnet (for example, if pinging n.n.99.2
>>>it will provide a source address of, say,n.n.99.5, but if pinging
>>>n.n.60.66 it will provide source address n.n.60.59).
>>>
>>>My OS/2 Warp 4 machine won't. It always uses the first address assigned
>>>(in this case n.n.60.66) as the source address. This means that messages
>>>sent to the other subnet contain the wrong source address and
>>>communication fails (unless the receiving machine happens to be on
>>>both subnets also.)
>>>
>>>Is there a fix?
>> 
>> Is the routing configured correctly for both subnets?
>
>You'll have to forgive my ignorance -- as I say, I can't get the admin to
>do this as he's given up on OS/2. Are you saying I need one or more route
>entries on my OS/2 machine?
>
>If so, what would they look like?  I believe I tried adding a line to
>route destination n.n.99.1 through router n.n.99.2 (me) with netmask
>255.255.255.0, with ipgate on, but this didn't work. 
>
>(I say I "believe" I tried it because I tried about a zillion things. . .)

Since you have two subnets, I assume you're using an alias on one NIC 
or two NICS.  Either way, OS/2 sees them as two separate networks. 
You use the TCP/IP configuration, routing page to set up routing for 
each network.  You should have a DEFAULT route where you want off 
network traffic to go, and a NET route for each network (or in your 
case subnet).  If you don't have a second NET route, you can create 
one on that Routing page, by clicking on Add, and filling in the 
details.


-- 
E-mail jknott@ca.ibm.com
_________________________________________________________________________
The above opinions are my own and not those of ISM Corp., a subsidiary of
IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         28-Aug-99 14:15:02
  To: All                                               28-Aug-99 14:21:24
Subj: Re: multiple subnet support: Win9x vs OS/2

From: ad354@FreeNet.Carleton.CA (James Owens)


Henk kelder (nospam_hkelder@capgemini.nl) writes:

> James Owens wrote:

>> In all I use TCPCFG to set LAN0 to DHCP, turn IP forwarding on, and
>> run "routed" in the background, then add the following to SETUP.CMD (in
this
>> order):
>> 
>> ifconfig lan0 n.n.99.2 netmask 255.255.255.0 alias
>> route add subnet n.n.99.0 n.n.99.2 0
>> route add subnet n.n.60.0 n.n.99.2 0
>> 
> 
> I don't understand. First you state the you use DHCP for LAN0, then you
> assign a specific number to LAN0.

Unfortunately one subnet uses dynamic addressing and one uses static
addressing. Both are on the same physical network.

>> After rebooting, I see from netstat -r that I have two routing paths:
>> 
>> n.n.60.0   n.n.60.66   255.255.255.0
>> n.n.99.0   n.n.99.2    255.255.255.0
>> 
>> When I use the route add command from a prompt to say, for example:
>> 
>> route add subnet n.n.60.0 n.n.99.2

>> it says it cannot add the route because "entry exists."

> I am not sure, but you could try to add a route like:
> route add host n.n.60.66 n.n.99.2 netmask 255.255.255.0

I have tried this (with two lines, routing both ways). I believe the
routing entries are added successfully; I will check next time I am at
work (possibly before Monday.) But I still couldn't ping n.n.99.1 -- the
packets go out with source address n.n.60.66. 

> I my office I am using a OS/2 warp 3 PC to route between our local lan
> and a separate Tokenring subnet. The OS/2 machine has an ethernet and a
> tokenring adapter installed.
> It works there perfectly. I must however check how the box is
> configured.
> 
> Please send me a mail to remember me to check the configuration when I
> am in. (will be wednesday).

Thanks, but my problem is that I am trying to use two subnets on a single
interface.  As I say, Windows can do it, which is very annoying. . . 

--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         28-Aug-99 14:25:25
  To: All                                               28-Aug-99 16:46:25
Subj: Re: multiple subnet support: Win9x vs OS/2

From: ad354@FreeNet.Carleton.CA (James Owens)

James Knott (jknott@ibm.net) writes:

> ad354@FreeNet.Carleton.CA (James Owens) wrote:

>>. . . Are you saying I need one or more route
>>entries on my OS/2 machine?

>>If so, what would they look like? . . .
> 
> Since you have two subnets, I assume you're using an alias on one NIC 
> or two NICS.  Either way, OS/2 sees them as two separate networks. 
> You use the TCP/IP configuration, routing page to set up routing for 
> each network.  You should have a DEFAULT route where you want off 
> network traffic to go, and a NET route for each network (or in your 
> case subnet).  If you don't have a second NET route, you can create 
> one on that Routing page, by clicking on Add, and filling in the 
> details.

Yes, I am using an alias on a single NIC to take a static address on a
second subnet. The same NIC has already been provided with a dynamic
address on the first network. (Note that if I connect to the second
network alone and boot, it works fine.)

If I just add the alias after the dhcpstrt command, and do nothing else
(ie no routing under TCPCFG or in SETUP.CMD) then immediately after
boot, netstat -r shows two routes:

n.n.60.0  n.n.60.66  255.255.255.0 (for my dynamic address)
n.n.99.0  n..n.99.2  255.255.255.0  (for my static address)

As I understand it, that should mean if I ping n.n.99.1, the OS should
send the packet on the network n.n.99.0, presumably through the
appropriate adapter (in this case the appropriate alias on a single
adapter.) 

When I check the packet (through iptrace and ipformat) it gives
a destination n.n.99.1, but the source address is n.n.60.66. 

Am I missing something, or is this a limitation of OS/2?

--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: 72764.2166@csi.com                                28-Aug-99 10:45:23
  To: All                                               28-Aug-99 16:46:25
Subj: resolv2 vs hosts

From: "Barry" <72764.2166@csi.com>

Can someone explain the difference between the resolv2 file and the hosts
configuration tab in TCPIP?  Is resov2 for the local network and hosts for
remote networks?  Resolv2 has input for Nameserver addresses, does hosts
have/need Nameserver addresses?

Barry 


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MindSpring Enterprises (1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     28-Aug-99 15:31:10
  To: All                                               28-Aug-99 16:46:26
Subj: Re: resolv2 vs hosts

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Sat, 28 Aug 1999 10:45:46 -0400 (EDT), Barry wrote:

>Can someone explain the difference between the resolv2 file and the hosts
>configuration tab in TCPIP?  Is resov2 for the local network and hosts for
>remote networks?  Resolv2 has input for Nameserver addresses, does hosts
>have/need Nameserver addresses?

resolv2:

Holds your domain and your nameservers.

hosts:

Holds IP addresses for certain hosts which you don't want to, or can't, go to
the nameservers.  For example, if a certain machine isn't IN the nameserver,
you can give it a nickname here.

In my home, I have a LAN but no nameserver.  So I use my ISP's nameserver for
anything outside of my LAN, and just put all my machines in one another's
hosts file.  Eventually I'll put a nameserver internally so I don't need to
do that anymore.
---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: drowelf@nospamvnet.net                            28-Aug-99 10:42:03
  To: All                                               28-Aug-99 16:46:26
Subj: DHCP Problem

From: "Eric A. Erickson" <drowelf@nospamvnet.net>

My new notebook can not seem to receive the DHCP Offer from my Warp Server 
Advanced (TcpIp 4.1) Server. I can see the DHCP Server offering the address to
the client, but it never seems to accept it. I am running Warp 4 FP11 and
TcpIp V4.0 
with MPTS IP 8424 stack.  I know that DHCP is working properly since I can get 

a DHCP Served address when I'm at a client site. 

So how can I debug this problem?




Elvish Software Foundry, Inc.    		- Internet:  drowelf@vnet.net
IBM Certified OS/2 Warp Engineer 	- IBMLink:   HONE81(ESFISA1)
IBM Certified OS2/ Warp Developer & Associate Visual Age C++ Developer
'Already where I want to be Today 	- Voice/Fax: (281)-398-2625 <-Newe'


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Elvish Software Foundry, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: IHateSPAM@montypython.org                         28-Aug-99 11:23:14
  To: All                                               28-Aug-99 16:46:26
Subj: What does INETWAIT.EXE do?

From: "Larry Belan" <IHateSPAM@montypython.org>

Can anyone point me to the answer of what INETWAIT does?  It's called in the
\mptn\bin\mptstart.cmd file, and if there's an errorcode, it passes up the
setup.cmd which init's the lan interfaces.

I ask, since I'm trying to use my laptop with two networks (Ether pcmcia and
ISA  TR in a dock) and if I boot without the dock, the ethenet card never
starts.

The ethernet is lan0 and the TR is lan1.  I never use BOTH at the same time.

Thanks for any pointers!
--
Larry Belan
PennState - Fayette
www.fe.psu.edu


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: PennState - Fayette Computer Center (1:109/42)

+----------------------------------------------------------------------------+

From: fm24@rummelplatz.uni-mannheim.de                  28-Aug-99 19:19:16
  To: All                                               28-Aug-99 21:13:09
Subj: Nslookup works but Netscape can not find hosts.

From: fm24@rummelplatz.uni-mannheim.de (Oliver Klimek)

I just installed Warp 4 and internet access via ISDN using isdnpm. When
connected to the net, Netscape (or any other internet program) is unable
to resolve hostnames, but when I start nslookup, all hostnames can be 
resolved. I have both resolv and resolv2 with correct configurations. What
is the reason for this problem ?

Oliver

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Rechenzentrum Uni Mannheim (1:109/42)

+----------------------------------------------------------------------------+

From: fBeythien@gmx.de                                  28-Aug-99 19:52:17
  To: All                                               28-Aug-99 21:13:09
Subj: Re: DHCP Problem

From: fBeythien@gmx.de (Frank Beythien)

On Sat, 28 Aug 1999 15:42:06, "Eric A. Erickson" 
<drowelf@nospamvnet.net> wrote:

> My new notebook can not seem to receive the DHCP Offer from my Warp Server 
> Advanced (TcpIp 4.1) Server. I can see the DHCP Server offering the address
to
> the client, but it never seems to accept it. I am running Warp 4 FP11 and
TcpIp V4.0 
> with MPTS IP 8424 stack.  I know that DHCP is working properly since I can
get 
> a DHCP Served address when I'm at a client site. 
> 
> So how can I debug this problem?

run DHCPMON and activate logging in  DHCPcd.cfg.

CU/2
-- 
Frank Beythien   fBeythien@gmx.de

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Prima e.V. - Dortmund - Germany (1:109/42)

+----------------------------------------------------------------------------+

From: bhk@dsl.co.uk                                     28-Aug-99 21:10:22
  To: All                                               29-Aug-99 05:35:21
Subj: Re: Problems with using NT resources from OS/2

From: bhk@dsl.co.uk (Brian {Hamilton Kelly})

On Thursday, in article
     <1rnx3.2243$x5.161120@eagle.america.net>
     kenny.trussell@thielekaolin.com "Kenny Trussell" wrote:

> Thanks every one for the help. I got it working. I needed to do several
> things that different ones of you suggested. The final thing that got it
> working was editing the RFCHOSTS and RFCBCST files. Thanks!

Glad to have been of help!

Seriously, *I* had never heard[1] of that information on Frank's (?was it
he?) site before, and it got *me* going at work after one whole building
was switched to a different subnet: so I was pleased to pass on the info.

[1] Not strictly true: I'd read that "saga" of MIT about two years ago,
but didn't need those files with our network architecture at that time; I
was very pleased when someone else (article now expired here, so I cannot
credit him) posted the pointer to those pages, and I then went and reread
them.

-- 
Brian {Hamilton Kelly}                                         bhk@dsl.co.uk
    "But we're a university.  We /have/ to have a library!..."said Ridcully,
         "What sort of people would we be if we didn't go into the library?"
    "Students", said the Senior Wrangler, morosely. [TP: The Last Continent]

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Dragonhill Systems Ltd (1:109/42)

+----------------------------------------------------------------------------+

From: evsi@naverex.kiev.ua                              29-Aug-99 00:16:20
  To: All                                               29-Aug-99 10:42:25
Subj: Re: TCP/IP V4.1 SDK?

From: evsi@naverex.kiev.ua (Sergey I. Yevtushenko)

On Wed, 25 Aug 1999 08:46:58, "Joerg Huber" <hubi@theoffice.net> 
wrote:

Hi,
> 
> I have the TCP/IP V4.1 OS/2 runtime DLL's (from the Developer Connection
> CD's) and need the SDK (Header + Libs) for development. Where can I get/buy
> the SDK?

Have you ever tried to look at IBM's developer connection web site?
You may find there an update for TOOLKIT 4.0 which conatins info you 
are looking for.
Another source - Aurora's TOOLKIT, but it requires higher level of 
subscription,
AFAIK. 
Anyway, if you meet any problems with getting required files, drop me 
a note.

Regards,
	Sergey.

*--------------------------------------
ES@Home

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: ES@Nome (1:109/42)

+----------------------------------------------------------------------------+

From: Jeff.davis.no@spam.ibm.net                        28-Aug-99 23:40:13
  To: All                                               29-Aug-99 10:42:26
Subj: Problem with loopback and hostname

From: "Jeff Davis" <Jeff.davis.no@spam.ibm.net>

I have TCP 4.1 on my Warp 4 PC.  I am not LAN connected.  I installed DB2 and
it needs loopback for one of its services.  I configured this in the notebook
and the ping localhost works fine.  Unfortunately the HOSTNAME command
returns nothing whereas it needs to return the host name.  The set command is
there.  When I connect to the Internet the HOSTNAME command returns a value
slip129-37-170-156.pq.ca.ibm.net so it seems as if something is not exactly
kosher with my loopback setup.  Any suggestion will be appreciated.

Jeff

// ------------------------------------
// Jeff Davis
// Prince Edward Island, Canada



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     29-Aug-99 03:06:06
  To: All                                               29-Aug-99 15:49:07
Subj: Re: Problem with loopback and hostname

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Sat, 28 Aug 1999 23:40:27 -0300 (ADT), Jeff Davis wrote:

>I have TCP 4.1 on my Warp 4 PC.  I am not LAN connected.  I installed DB2 and
>it needs loopback for one of its services.  I configured this in the notebook
>and the ping localhost works fine.  Unfortunately the HOSTNAME command
>returns nothing whereas it needs to return the host name.  The set command is
>there.  When I connect to the Internet the HOSTNAME command returns a value
>slip129-37-170-156.pq.ca.ibm.net so it seems as if something is not exactly
>kosher with my loopback setup.  Any suggestion will be appreciated.

Under "Host Names", do you have "This Computer's Host Name" set to something?
 Let's say it's "foo".  At the bottom, you'll see a "Hosts" tab.  Select it. 
Add 127.0.0.1 as localhost with an alias of "foo".

I think that'll do it, but I don't have a disconnected machine to try.

---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: chadpaul@netexpress.net                           29-Aug-99 03:47:00
  To: All                                               29-Aug-99 15:49:07
Subj: warp 4 and two net cards 

From: chadpaul@netexpress.net

Hello,

I am trying to install and configure two netcards.
I am using OS/2 v.4.

I have tried pci with isa, two pci and two isa. 
The last attempt I put two 3com 3c905 cards in.
one at irq 10 and 300h,  the other is at irq 12 and 380h.
WHile installing, the system only saw one card in the installation.
when it rebooted, it seemed to detect more than one card.

Is there a config file to edit and say what card is using what address and
irq?

IS anyone using two cards in a os/2 machine.

Has anyone ever tried using the Intel server dual port ethernet card or any of 
the adaptec dual or quad port ethernet cards in a os/2 4.0 machine?

Any info is appreciated.

chadpaul@netexpress.net

Chad Pauli

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network (1:109/42)

+----------------------------------------------------------------------------+

From: Jeff.davis.no@spam.ibm.net                        29-Aug-99 00:44:26
  To: All                                               29-Aug-99 15:49:07
Subj: Re: Problem with loopback and hostname

From: "Jeff Davis" <Jeff.davis.no@spam.ibm.net>

On Sun, 29 Aug 1999 03:06:12 GMT, Darin McBride wrote:

>On Sat, 28 Aug 1999 23:40:27 -0300 (ADT), Jeff Davis wrote:
>
>>I have TCP 4.1 on my Warp 4 PC.  I am not LAN connected.  I installed DB2
and
>>it needs loopback for one of its services.  I configured this in the
notebook
>>and the ping localhost works fine.  Unfortunately the HOSTNAME command
>>returns nothing whereas it needs to return the host name.  The set command
is
>>there.  When I connect to the Internet the HOSTNAME command returns a value
>>slip129-37-170-156.pq.ca.ibm.net so it seems as if something is not exactly
>>kosher with my loopback setup.  Any suggestion will be appreciated.
>
>Under "Host Names", do you have "This Computer's Host Name" set to something?
> Let's say it's "foo".  At the bottom, you'll see a "Hosts" tab.  Select it. 
>Add 127.0.0.1 as localhost with an alias of "foo".
>
>I think that'll do it, but I don't have a disconnected machine to try.
>
>---
>Disclaimer: unless explicitly mentioned otherwise, I do not speak,
>nor have I ever spoken, for the company I work for.
>
Unfortunately, been there and done that.  My local name (not being too
original) is jeff (lower case).  The DB2 set up instructions are quite clear
for OS/2 and told me to add the alias if I chose to give a machine name.  I
can ping either localhost or jeff so I know the alias is working.  Thanks
though.  Initially I had not read down to the alias part in the instruction
and didn't have it in.  When I saw this I thought eureka but was sadly
mistaken.  Any other thoughts.

Jeff

// ------------------------------------
// Jeff Davis
// Prince Edward Island, Canada



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                29-Aug-99 04:37:09
  To: All                                               29-Aug-99 15:49:07
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37C64ECA.1755AFFC@businessnet.net>, Sacha Prins <sprins@businessnet.net>
writes:
>Baden Kudrenecky wrote:
>
>> In <37C599EB.87A3C267@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
>>
>>    AFAIK, the 86xx MPTS is really buggy, get an 8423 or 8424.
>
>How does 842x stack up against 86xx in performance? I'm afraid I need the 32
bit
>performance for the stuff I'm doing.
>
>Sacha

   AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
one computer with 8424, and it is definitely a 32 bit stack, and
the syslevel states that the previous version was a 8421, which
I believe also was a 32 bit stack, as I went 32 bit in spring
1998, with the previous January MPTS (8421?), and it
significantly improved the TCP/IP performance.

baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: kris@dgraph.com                                   28-Aug-99 22:47:20
  To: All                                               29-Aug-99 15:49:07
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: Kris Kadela <kris@dgraph.com>


Baden Kudrenecky wrote:
> 
> In <37C64ECA.1755AFFC@businessnet.net>, Sacha Prins <sprins@businessnet.net> 
writes:
> >Baden Kudrenecky wrote:
> >
> >> In <37C599EB.87A3C267@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
> >>
> >>    AFAIK, the 86xx MPTS is really buggy, get an 8423 or 8424.
> >
> >How does 842x stack up against 86xx in performance? I'm afraid I need the
32 bit
> >performance for the stuff I'm doing.
> >
> >Sacha
> 
>    AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
> one computer with 8424, and it is definitely a 32 bit stack, and
> the syslevel states that the previous version was a 8421, which
> I believe also was a 32 bit stack, as I went 32 bit in spring
> 1998, with the previous January MPTS (8421?), and it
> significantly improved the TCP/IP performance.
> 
> baden
> 
> baden@unixg.ubc.ca
> http://baden.nu/
> OS/2, Solaris & Linux

I did not know that 8421+ were 32 bit. Are you sure?
Is this upgrade hell or what? I don't even know what level of MPTS I am
at now.
Does anyone have a stable MPTS 86xx install?? How?

-- 

**********************
DigiGraph Technical
http://www.dgraph.com
**********************

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: DigiGraph Technical (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                29-Aug-99 05:16:28
  To: All                                               29-Aug-99 15:49:08
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37C8BB6C.3CFC2787@dgraph.com>, Kris Kadela <kris@dgraph.com> writes:
>
>
>Baden Kudrenecky wrote:
>> 
>> In <37C64ECA.1755AFFC@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
>> >Baden Kudrenecky wrote:
>> >
>> >> In <37C599EB.87A3C267@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
>> >>
>> >>    AFAIK, the 86xx MPTS is really buggy, get an 8423 or 8424.
>> >
>> >How does 842x stack up against 86xx in performance? I'm afraid I need the
32 bit
>> >performance for the stuff I'm doing.
>> >
>> >Sacha
>> 
>>    AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
>> one computer with 8424, and it is definitely a 32 bit stack, and
>> the syslevel states that the previous version was a 8421, which
>> I believe also was a 32 bit stack, as I went 32 bit in spring
>> 1998, with the previous January MPTS (8421?), and it
>> significantly improved the TCP/IP performance.
>> 
>> baden
>> 
>> baden@unixg.ubc.ca
>> http://baden.nu/
>> OS/2, Solaris & Linux
>
>I did not know that 8421+ were 32 bit. Are you sure?

   Definitely!  My IPSpeed has two versions, 16 bit and 32 bit,
so I can test it with that.

>Is this upgrade hell or what? I don't even know what level of MPTS I am
>at now.

   No, I  downloaded the images from:
ftp.software.ibm.com/ps/products/mpts/fixes/english-us

   I have included the two readme's below.  I used SVD to put
all the images in one directory, and I installed MPTS from that.
It looks like the 8424 package uses the Corrective Service Tool.

To find out your MPTS version, type "syslevel" at a command 
prompt.

>Does anyone have a stable MPTS 86xx install?? How?

   My friend who supports large OS/2 LANs and WANs, told me to
stay away from the 86xx stacks as they are way too buggy.

baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux


IBM MPTS LAN Adapter and Protocol Support FixPak WR08424 : English - US


  *******************************************************************
  *                                                                 *
  * This package is not a cumulative FixPak or a product refresh.   *
  *                                                                 *
  * It requires that you have either WR08421 or WR08423 installed.  *
  *                                                                 *
  * You must use SERVICE or FSERVICE, not MPTS.EXE, to install this *
  * package.                                                        *
  *******************************************************************

WR08424.INF, located on the first FixPak diskette, contains
installation, and other useful information about this FixPak.
Please review before installation of service.


WARNING : This FixPak is installed using Corrective Service
          Facility (CSF) 2-B Version 1.39. Once you use this
          version of the FixTool you will not be able to use
          previous versions of the FixTool.


*******************************************************
 Diskettes can be created using the LOADDSKF utility.
 There are 3 files/images included in this package.
*******************************************************
     FIXT139  DSK   FixPak Tool (CSF) V 1.39
     W08424B1 DSK   MPTS DISKETTE #1
     W08824B2 DSK   MPTS DISKETTE #2



 IBM MPTS LAN Adapter and Protocol Support FixPak WR08421 : U.S.-English


 This FixPak is a refresh of the MPTS LAN Adapter and Protocol Support
 program and is intended to replace (upgrade) only LICENSED instances
 of MPTS.


 This FixPak supersedes WR08419, and adds the following APAR fix :


 IC17544 : USING TCPBEUI, CONNECTION BEING DROPPED DURING FILE TRANSFER
 (NET  OP MOVE FAILS)


 This version of MPTS is installed using the MPTS Program Product
 Installation Aid, MPTS.EXE. There is NO selective install of MPTS
 sub-components. All components of MPTS are installed, including all
 of the MAC drivers. Please refer to the MPTS Configuration Guide on
 Diskette 6 for complete Installation and Configuration details.


************************************************************************
 Diskettes can be created using the LOADDSKF utility.
 There are 6 files/images included in this package.
************************************************************************
     W08421B1 DSK MPTS DISKETTE #1  3.50 1.44MB
     W08421B2 DSK MPTS DISKETTE #2  3.50 1.44MB
     W08421B3 DSK MPTS DISKETTE #3  3.50 1.44MB
     W08421B4 DSK MPTS DISKETTE #4  3.50 1.44MB
     W08421B5 DSK MPTS DISKETTE #5  3.50 1.44MB
     W08421B6 DSK MPTS DISKETTE #6  3.50 1.44MB


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: bklein@earthlink.net                              29-Aug-99 05:56:10
  To: All                                               29-Aug-99 15:49:08
Subj: PPP Over Ethernet

From: bklein@earthlink.net

I just got my DSL line installed via Earthlink and Pac-Bell. Earthlink's
connection requires WinPoET (PPP over ethernet) so at the moment I have to
use Win 9x. Does anyone know of a PPPOE driver for OS/2? TIA

  Brian

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: EarthLink Network, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: popp.bbs@bbs.lit.edu.tw                           29-Aug-99 16:02:25
  To: All                                               29-Aug-99 15:49:08
Subj: Re: Wissen pur

From: "ui֨k" <popp.bbs@bbs.lit.edu.tw>

25656532323215848
Boris Lipinski <BorisLipinski@datacom-gmbh.de> wrote in message
news:7q3i8n$lpn$2@news.netcologne.de...
> An alle Suchenden,
>
> unter www.datacom-bookstore.de findet man das beste Lexikon zur Daten- und
> Telekommunikation.
> Noch nicht mal im englischsprachigen findet man ein solches Lexikon.
>
> mfg Boris
>
>
>
>
>


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Biglobe Taiwan NEWS (1:109/42)

+----------------------------------------------------------------------------+

From: cannon@sonic.net                                  29-Aug-99 00:20:25
  To: All                                               29-Aug-99 15:49:08
Subj: setting up connectivity between NT and OS/2

From: Andrew Cannon <cannon@sonic.net>

My employer has installed an ISDN line to my home for remote access to
work.  This is the first time I have attempted to setup a network on
OS/2(always had used a modem to ISP).

There doesn't seem to be any instructions on setting this up to an NT
network, and I have struggled with this for quite some time, so I
would appreciate any advice.  Here is what I have done so far:

1)installed the NIC and ISDN hardware
2) TCP/IP Configuration
    network - enabled lan0 manually (IP,subnet mask)
    routing - pc to isdn (treated as a gateway)
        (note: "create default ... a given host" button seems to
        be disabled all the time, so I imagine there will be a conflict
        when connecting to my ISP)
    hostnames - added
        my host name
        local domain
        several nameservers (all NT)
        added several important hosts (NT, UNIX machines at work)
    autostart- routed
    everything else is the default
3)MPTN -
    LAPS - removed No Network Adapter configuration
        added my LAN adapter
        added protocols
        0 - ... NETBIOS over TCP/IP
        0 TCP/IP

during bootup
ifconfig reports that the interface cannot be found (I believe
lan0 should be the default interface)

TME 10 NetFinity Network Interface reports
   Network Interface 'NETBIOS' atartup failed

netstat -n reports no interfaces

Seems like there is is fundamental problem here.  Any help or
suggestions would be great, Thanks

--
Andy Cannon   http://www.sonic.net/~cannon



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Sonic,Santa Rosa CA,http://www.sonic.net (1:109/42)

+----------------------------------------------------------------------------+

From: halazry@omantel.net.om                            29-Aug-99 08:55:05
  To: All                                               29-Aug-99 15:49:09
Subj: LPD Ports

From: "Hamed" <halazry@omantel.net.om>

I am using OS/2 Warp Server 4.0.  I am facing a problem configuring LPD
ports.

I have enable 8 LPD ports in TCP/IP Printing Configuration and selected
Autostart LPDPORTD.

During printer configuration (Printer template), when I view the printer
ports I cannot see any LPD ports; all  other local ports (LPTx, COMx) are
shown.

Any ideas?

Hamed

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Oman Telecommunications Company (1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     29-Aug-99 14:14:02
  To: All                                               29-Aug-99 15:49:09
Subj: Re: Problem with loopback and hostname

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Sun, 29 Aug 1999 00:44:53 -0300 (ADT), Jeff Davis wrote:

>Unfortunately, been there and done that.  My local name (not being too
>original) is jeff (lower case).  The DB2 set up instructions are quite clear
>for OS/2 and told me to add the alias if I chose to give a machine name.  I
>can ping either localhost or jeff so I know the alias is working.  Thanks
>though.  Initially I had not read down to the alias part in the instruction

[I wrote most of the instructions... :->]

>and didn't have it in.  When I saw this I thought eureka but was sadly
>mistaken.  Any other thoughts.

If you type "nslookup jeff", does it return your address (while not
connected)?  You may not have the "Use hosts first" box checked.  That's
critical, too.

I assume what you're trying to do is get the DB2 UDB v5.x HTML search server
(Net.Question) working.  DB2 itself should work without all this extra stuff.
 The version of Net.Question in DB2 UDB v6.1 doesn't need quite so much of
this setup, I think.  Mind you, by the time I started playing with it, I
already had everything set up from v5.

---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: stevens@                                          29-Aug-99 10:21:20
  To: All                                               29-Aug-99 16:52:17
Subj: Sendmail problems

From: stevens@

Since installing Warp on a new machine, I can't get sendmail to process
messages,
in or out.  Other parts of the TCP/IP stack seem to work fine: inet/ftp/www,
etc.

Sendmail.err gives me
79 99/08/29 10:16:33 554 buildaddr: no net

while the sendmail debugging output yields

dropenvelope fe7b440: id=<null>, flags=0x0

EXE Version IBM OS/2 SENDMAIL VERSION 2.02
DLL Version IBM OS/2 SENDMAIL VERSION 2.02
DLL Pathname E:\TCPIP\DLL\SENDMAIL.DLL

getauthinfo: quito@localhost
setoption Q=e:\mptn\etc\mqueue
SYSTEM IDENTITY (after readcf):
            (short domain name) $w = QUITO.cee.usu.edu
        (canonical domain name) $j = QUITO.cee.usu.edu
               (subdomain name) $m = cee.usu.edu
                    (node name) $k = quito.cee.usu.edu
dropenvelope fe7b440: id=<null>, flags=0x0
    79  Child Process Starting
queuename: assigned id KAA000.79, env=fe7b440
setsender(<stevens@quito.cee.usu.edu>)

--parseaddr(<stevens@quito.cee.usu.edu>)
syserr: ExitStat = 70

and sendmail.log yields

79 S> 220-QUITO.cee.usu.edu Sendmail IBM OS/2 SENDMAIL VERSION 2.02 ready at
Sun
, 29 Aug 1999 10:16:33 -0600
220 ESMTP spoken here
79 S< HELO quito.cee.usu.edu
79 S> 250 QUITO.cee.usu.edu Hello quito.cee.usu.edu, pleased to meet you
79 S< MAIL FROM: <stevens@quito.cee.usu.edu>
79 S> 554 buildaddr: no net
79 S< RCPT TO: <stevens@quito.cee.usu.edu>

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Utah State University (1:109/42)

+----------------------------------------------------------------------------+

From: duanec@eyebm.net                                  29-Aug-99 14:14:26
  To: All                                               29-Aug-99 16:52:17
Subj: Re: multiple subnet support: Win9x vs OS/2

From: "Duane A. Chamblee" <duanec@eyebm.net>

On 27 Aug 1999 11:26:53 GMT, James Owens wrote:

>>>My OS/2 Warp 4 machine won't. It always uses the first address assigned
>>>(in this case n.n.60.66) as the source address. This means that messages
>>>sent to the other subnet contain the wrong source address and
>>>communication fails (unless the receiving machine happens to be on
>>>both subnets also.)
>>>
>>>Is there a fix?

this may be address in the MPTS updates and fixpaks. Are you running the
basic installation from the Warp4 CD?

If so, You may want to update your stack to the latest 16-bit (MPTS 5.1x)
  WR08423(install) + WR08424(fix)

or, install the 32 bit stack (not really recommended for a client machine)
  WR08610 + WR08620

The are here:
ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us/
--------------------------------------
Change the eye to an i when you reply!


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: lampsstgt@aol.com                                 29-Aug-99 18:51:10
  To: All                                               29-Aug-99 16:52:17
Subj: Netscape API ?

From: lampsstgt@aol.com (Lamps stgt)

Is there an C/C++ API for the Netscape Communicator? I want to send a
html-page
to the Netscape Communicator from a C-application.
Thomas

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: AOL Bertelsmann Online GmbH & Co. KG http://www.g
(1:109/42)

+----------------------------------------------------------------------------+

From: sprins@businessnet.net                            29-Aug-99 23:16:06
  To: All                                               29-Aug-99 19:53:19
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: Sacha Prins <sprins@businessnet.net>

Baden Kudrenecky wrote:

> >>    AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
> >> one computer with 8424, and it is definitely a 32 bit stack, and
> >> the syslevel states that the previous version was a 8421, which
> >> I believe also was a 32 bit stack, as I went 32 bit in spring
> >> 1998, with the previous January MPTS (8421?), and it
> >> significantly improved the TCP/IP performance.
> >>
> >> baden
> >>
> >> baden@unixg.ubc.ca
> >> http://baden.nu/
> >> OS/2, Solaris & Linux
> >
> >I did not know that 8421+ were 32 bit. Are you sure?
>
>    Definitely!  My IPSpeed has two versions, 16 bit and 32 bit,
> so I can test it with that.

Well, it could also be that 8421+ have a TCPIP32.dll which thunks to TCPIP.dll 
thich could be
16 bit. However, I'm less and less impressed by performance and more and more
by stability
each day. I might try to downlevel if IBM doesn't come up with any solution
soon. Are 8421+
for SMP machines? Or do I need a different version for that? I'll look around

> >Does anyone have a stable MPTS 86xx install?? How?

Peeps that do some webbrowsing only using the 86xx may do that without too
many problems...

>    My friend who supports large OS/2 LANs and WANs, told me to
> stay away from the 86xx stacks as they are way too buggy.

It also depends on the applications. I'm splitting up services over different
servers now and
that way I get to choose the stablest stack for each server. My mail server
(InetMail) e.g.
benefits a lot from 8610. So does Lotus Go (which is a terrible buggy program
by the way).
Apache, Xitami, other EMX apps and the JVM apparently don't.

What I don't understand is that the MPTS team doesn't make fixing this a
priority (or maybe
they do, and it only seems that they don't). From my understanding WSeB with
the 8620 stack
also has all the problems we are having and then some. So much for (b)leeding
edge e-business
tools...

Sacha


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: A2000 Kabeltelevisie en Telecommunicatie (1:109/42)

+----------------------------------------------------------------------------+

From: "operagost"@e-mail.com (remove t...               29-Aug-99 22:34:16
  To: All                                               30-Aug-99 03:42:11
Subj: Re: warp 4 and two net cards

Message sender: "operagost"@e-mail.com (remove the - )

From: Stephen Eickhoff <"operagost"@e-mail.com (remove the - )>

Maybe I should put this in a FAQ somewhere!

I have two 3c900 in my web server. I had to go into the settings in MPTS and
specify which PCI slot # it was or else both drivers initialized the same
card. You may have the same problem. Note which PCI slot the driver reports
for the card. Assuming you have both cards in adjacent slots, the next
higher number should be the correct slot for card number 2.

chadpaul@netexpress.net wrote:

> Hello,
>
> I am trying to install and configure two netcards.
> I am using OS/2 v.4.
>
> I have tried pci with isa, two pci and two isa.
> The last attempt I put two 3com 3c905 cards in.
> one at irq 10 and 300h,  the other is at irq 12 and 380h.
> WHile installing, the system only saw one card in the installation.
> when it rebooted, it seemed to detect more than one card.
>
> Is there a config file to edit and say what card is using what address and
irq?
>
> IS anyone using two cards in a os/2 machine.
>
> Has anyone ever tried using the Intel server dual port ethernet card or any
of the adaptec dual or quad port ethernet cards in a os/2 4.0 machine?
>
> Any info is appreciated.
>
> chadpaul@netexpress.net
>
> Chad Pauli

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: 72764.2166@csi.com                                29-Aug-99 19:49:01
  To: All                                               30-Aug-99 03:42:11
Subj: Re: resolv2 vs hosts

From: "Barry" <72764.2166@csi.com>

What exactly is a Nameserver?  Is it a server with certain software that is
able to resolve ip addresses?

Barry


Sat, 28 Aug 1999 15:31:21 GMT, Darin McBride wrote:

>On Sat, 28 Aug 1999 10:45:46 -0400 (EDT), Barry wrote:
>
>>Can someone explain the difference between the resolv2 file and the hosts
>>configuration tab in TCPIP?  Is resov2 for the local network and hosts for
>>remote networks?  Resolv2 has input for Nameserver addresses, does hosts
>>have/need Nameserver addresses?
>
>resolv2:
>
>Holds your domain and your nameservers.
>
>hosts:
>
>Holds IP addresses for certain hosts which you don't want to, or can't, go to
>the nameservers.  For example, if a certain machine isn't IN the nameserver,
>you can give it a nickname here.
>
>In my home, I have a LAN but no nameserver.  So I use my ISP's nameserver for
>anything outside of my LAN, and just put all my machines in one another's
>hosts file.  Eventually I'll put a nameserver internally so I don't need to
>do that anymore.
>---
>Disclaimer: unless explicitly mentioned otherwise, I do not speak,
>nor have I ever spoken, for the company I work for.
>
>
>



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MindSpring Enterprises (1:109/42)

+----------------------------------------------------------------------------+

From: 72764.2166@csi.com                                29-Aug-99 19:57:21
  To: All                                               30-Aug-99 03:42:11
Subj: Peer network vs server

From: "Barry" <72764.2166@csi.com>

Is there any difference in setup of a peer network vs a network with a server
that all workstations connect to?  Right now I have an OS/2 computer
connected peer to peer to two WIN98 computers.  What would be the setup
difference If I would make one of the WIN98 computers an NT server, that the
other 2 workstations connect to?

Barry


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MindSpring Enterprises (1:109/42)

+----------------------------------------------------------------------------+

From: Jeff.davis.no@spam.ibm.net                        29-Aug-99 21:06:19
  To: All                                               30-Aug-99 03:42:11
Subj: Re: Problem with loopback and hostname

From: "Jeff Davis" <Jeff.davis.no@spam.ibm.net>

On Sun, 29 Aug 1999 14:14:04 GMT, Darin McBride wrote:

>On Sun, 29 Aug 1999 00:44:53 -0300 (ADT), Jeff Davis wrote:
>
>[I wrote most of the instructions... :->]
>
And a good job you did too.
>
>If you type "nslookup jeff", does it return your address (while not
>connected)?  You may not have the "Use hosts first" box checked.  That's
>critical, too.
>
Use hosts first is checked but nslookup returns nothing (gives an error about
not fonding a server actually)
Ping works whether online or not...  I looked in my CONFIG.SYS and found
SET USE_HOSTS_FIRST=1
Which I think means use hosts first in IBMese 
>I assume what you're trying to do is get the DB2 UDB v5.x HTML search server
>(Net.Question) working.  DB2 itself should work without all this extra stuff.
> The version of Net.Question in DB2 UDB v6.1 doesn't need quite so much of
>this setup, I think.  Mind you, by the time I started playing with it, I
>already had everything set up from v5.
>
Actually it's the DB2 UDB v6.1 HTML search server (Net.Question)..  I
downloaded the whole thing at work (personal edition) and have JAVA 1.1.8 as
well..  My goal is to create a small application in JAVA using DB2 as the
data store.  It's sort of a proof of concept.  In fact I have a funny feeling
that JAVA uses the same search engine.  I suspect that it is a TCP issue.  I
went cruising looking for updates and there is a fix set for TCP 4.1.  In the
readme there is a mention of localhost updates.  I will putter around this
week and will apply the fixes next weekend after backing everything up.

If you have any more suggestions feed them to me.  I will try most anything
if it seems to make sense
>---
>Disclaimer: unless explicitly mentioned otherwise, I do not speak,
>nor have I ever spoken, for the company I work for.
>
Jeff

// ------------------------------------
// Jeff Davis
// Prince Edward Island, Canada



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     30-Aug-99 01:39:06
  To: All                                               30-Aug-99 05:29:12
Subj: Re: resolv2 vs hosts

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Sun, 29 Aug 1999 19:49:02 -0400 (EDT), Barry wrote:

>What exactly is a Nameserver?  Is it a server with certain software that is
>able to resolve ip addresses?

A nameserver is basically a database server that keeps track of which names
refer to which IP addresses, and can "speak" on a certain port (53) in a
certain way (as defined by certain internet standards) as to relay the
information in the database to any user asking for it.  It's more complicated
than this, but you have to realise that a certain amount of the internet's
names <==> IP addresses mapping is stored on the nameserver's system.  Most
especially, its own domain.

---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     30-Aug-99 01:40:26
  To: All                                               30-Aug-99 05:29:12
Subj: Re: Problem with loopback and hostname

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Sun, 29 Aug 1999 21:06:39 -0300 (ADT), Jeff Davis wrote:

>On Sun, 29 Aug 1999 14:14:04 GMT, Darin McBride wrote:
>
>>On Sun, 29 Aug 1999 00:44:53 -0300 (ADT), Jeff Davis wrote:
>>
>>[I wrote most of the instructions... :->]
>>
>And a good job you did too.

<blush>

>>If you type "nslookup jeff", does it return your address (while not
>>connected)?  You may not have the "Use hosts first" box checked.  That's
>>critical, too.
>>
>Use hosts first is checked but nslookup returns nothing (gives an error about
>not fonding a server actually)

<slaps forehead> Duh.  Of course.  Try "host jeff".  Also, if you would,
what's the output of running "sniffle /P" (the case is important on the
parameter)?

>Ping works whether online or not...  I looked in my CONFIG.SYS and found
>SET USE_HOSTS_FIRST=1
>Which I think means use hosts first in IBMese 

heheh.  Yeah, it's one of those rare cases where a programmer uses a variable
that actually means what it says.  :-)

>>I assume what you're trying to do is get the DB2 UDB v5.x HTML search server
>>(Net.Question) working.  DB2 itself should work without all this extra
stuff.
>> The version of Net.Question in DB2 UDB v6.1 doesn't need quite so much of
>>this setup, I think.  Mind you, by the time I started playing with it, I
>>already had everything set up from v5.
>>
>Actually it's the DB2 UDB v6.1 HTML search server (Net.Question)..  I
>downloaded the whole thing at work (personal edition) and have JAVA 1.1.8 as

Work?  Where?  Not too many people have v6.1 yet, I would have imagined.  We
only GA'd a bit ago...  If you're internal, reply via Notes.

>well..  My goal is to create a small application in JAVA using DB2 as the
>data store.  It's sort of a proof of concept.  In fact I have a funny feeling
>that JAVA uses the same search engine.  I suspect that it is a TCP issue.  I

Java doesn't use any search engine.  Visual Age for Java, however, does. 
Important question: which did you install first - VA Java, or DB2 PE?  Also,
which version of VA Java?

>went cruising looking for updates and there is a fix set for TCP 4.1.  In the
>readme there is a mention of localhost updates.  I will putter around this
>week and will apply the fixes next weekend after backing everything up.
>
>If you have any more suggestions feed them to me.  I will try most anything
>if it seems to make sense

What I'd really like to know is *what* exactly isn't working?  I assume both
installed fine, but just the search in DB2 doesn't work, and the help (the
ENTIRE help) in VA Java doesn't work.  I assume everything else works fine? 
Of course, if you're developing, the help is somewhat important.  :-)

---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: heloman@my-deja.com                               30-Aug-99 02:22:15
  To: All                                               30-Aug-99 05:29:13
Subj: "RoadRunner" cable service

From: heloman@my-deja.com

I am currently using a modem and due to lousy phone
service and lines cannot get better than 2300bps.
This makes almost everything today really sloooowwww.
At a computer store her in Tampa, Time Warner was
demonstrating their RoadRunner cable service. As
usual they provide, "NO support for OS/2!". While I
believe I can provide a lan type card, what really
has me bugged is they have a WINDOWS only login/on
program. According to the rep this program must be
utilized every time one logs on to the system. Is
anyone using RoadRunner and if so HOW did you get
around this lousy way of logging on? What lan card
would you recommend and where are the driver(s)
located? I look forward to your recommendations.....


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Deja.com - Share what you know. Learn what you do
(1:109/42)

+----------------------------------------------------------------------------+

From: stevens@quito.cee.usu.edu                         29-Aug-99 08:43:00
  To: All                                               30-Aug-99 05:29:13
Subj: Sendmail problems

From: "David K. Stevens" <stevens@quito.cee.usu.edu>

After installing Warp/fp11/mpts8424 on a new box, I can't seem to get
sendmail working -

ExitStat = 70

in sendmail debug window

291 S> 220-QUITO.cee.usu.edu Sendmail IBM OS/2 SENDMAIL VERSION 2.02
ready at Sun, 29 Aug     1999 08:15:51 -0600
220 ESMTP spoken here
291 S< HELO quito.cee.usu.edu
291 S> 250 QUITO.cee.usu.edu Hello quito.cee.usu.edu, pleased to meet
you
291 S< MAIL FROM: <stevens@quito.cee.usu.edu>
291 S> 554 buildaddr: no net
291 S< RCPT TO: <stevens@quito.cee.usu.edu>

in sendmail.log

and online info is no help.  inet/ftp/www etc. seem to work ok. Where do
I look for answers?

David K. Stevens

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                30-Aug-99 03:02:26
  To: All                                               30-Aug-99 05:29:13
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37C9B12D.49F371A6@businessnet.net>, Sacha Prins <sprins@businessnet.net>
writes:
>Baden Kudrenecky wrote:
>
>> >>    AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
>> >> one computer with 8424, and it is definitely a 32 bit stack, and

   It's actually 8423.

>> >> the syslevel states that the previous version was a 8421, which
>> >> I believe also was a 32 bit stack, as I went 32 bit in spring
>> >> 1998, with the previous January MPTS (8421?), and it
>> >> significantly improved the TCP/IP performance.
>> >>
>> >> baden
>> >>
>> >> baden@unixg.ubc.ca
>> >> http://baden.nu/
>> >> OS/2, Solaris & Linux
>> >
>> >I did not know that 8421+ were 32 bit. Are you sure?
>>
>>    Definitely!  My IPSpeed has two versions, 16 bit and 32 bit,
>> so I can test it with that.
>
>Well, it could also be that 8421+ have a TCPIP32.dll which thunks to
TCPIP.dll thich could be
>16 bit. However, I'm less and less impressed by performance and more and more 
by stability

   I just checked now on two machines:

   One has OS/2 Connect and FP_40 with syslevel.mpt 8000, and
the other has OS/2 3.0 Red and syslevel.mpt 8423 over 8421.

   Using Watchcat, the TCPIP.DLL for the 8000 machine is a 16
bit DLL, and for the 8423 machine, the TCPIP.DLL is 32 bit.
There are still 16 bit EXEs and DLLs on both machines, notably:

LANMSGEX.EXE
    type:        Windowable
    mod DLL(32)         DOSCALL1.DLL
    mod DLL(16)         LANMSGDL.DLL
    mod DLL(32)         MSG.DLL
    mod DLL(32)         VIOCALLS.DLL
AFNBINI.EXE
    type:        Windowable
CNTRL.EXE
    type:        Windowable

>each day. I might try to downlevel if IBM doesn't come up with any solution
soon. Are 8421+
>for SMP machines? Or do I need a different version for that? I'll look around

   No, if you go to
ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us,
you will see that there are separate packages for SMP:
16-May-97 00:00      - wr08502_smp     16-May-97 00:00      - 
26-Aug-97 00:00      - wr08503_smp     26-Aug-97 00:00      - 
03-Dec-97 00:00      - wr08504_smp     03-Dec-97 00:00      - 
15-May-98 00:00      - wr08506_smp     15-May-98 00:00      - 

>> >Does anyone have a stable MPTS 86xx install?? How?
>
>Peeps that do some webbrowsing only using the 86xx may do that without too
many problems...
>
>>    My friend who supports large OS/2 LANs and WANs, told me to
>> stay away from the 86xx stacks as they are way too buggy.
>
>It also depends on the applications. I'm splitting up services over different 
servers now and
>that way I get to choose the stablest stack for each server. My mail server
(InetMail) e.g.
>benefits a lot from 8610. So does Lotus Go (which is a terrible buggy program 
by the way).
>Apache, Xitami, other EMX apps and the JVM apparently don't.

   The only reason that went for the new MPTS 8423 (on my
laptop), was for the enhanced DHCP support, and the TCPIP speed
increase was an unexpected bonus.

>What I don't understand is that the MPTS team doesn't make fixing this a
priority (or maybe
>they do, and it only seems that they don't). From my understanding WSeB with
the 8620 stack
>also has all the problems we are having and then some. So much for (b)leeding 
edge e-business
>tools...

   That's why I stay away from the edge, never lone the bleeding
edge.  {:-) It was just a few weeks ago that I upgraded to OS/2
3.0 Connect from Red, and to Warp Server 4.0 from Lan Server 4.0
over 3.0 Red, and then applied my first OS/2 fix packs.


baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                30-Aug-99 03:38:04
  To: All                                               30-Aug-99 05:29:13
Subj: Re: setting up connectivity between NT and OS/2

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37C8DF53.1DACC59C@sonic.net>, Andrew Cannon <cannon@sonic.net> writes:
>My employer has installed an ISDN line to my home for remote access to
>work.  This is the first time I have attempted to setup a network on
>OS/2(always had used a modem to ISP).
>
>There doesn't seem to be any instructions on setting this up to an NT
>network, and I have struggled with this for quite some time, so I
>would appreciate any advice.  Here is what I have done so far:
>
>1)installed the NIC and ISDN hardware
>2) TCP/IP Configuration
>    network - enabled lan0 manually (IP,subnet mask)
>    routing - pc to isdn (treated as a gateway)
>        (note: "create default ... a given host" button seems to
>        be disabled all the time, so I imagine there will be a conflict
>        when connecting to my ISP)
>    hostnames - added
>        my host name
>        local domain
>        several nameservers (all NT)
>        added several important hosts (NT, UNIX machines at work)
>    autostart- routed
>    everything else is the default
>3)MPTN -
>    LAPS - removed No Network Adapter configuration
>        added my LAN adapter
>        added protocols
>        0 - ... NETBIOS over TCP/IP
>        0 TCP/IP

   I have not done this, however, it is my understanding that
you need to have both NETBIOS and TCP/IP on LAN0, and that
NETBIOS over TCP/IP has to be on a separate interface, i.e.
LAN1.

>during bootup
>ifconfig reports that the interface cannot be found (I believe
>lan0 should be the default interface)
>
>TME 10 NetFinity Network Interface reports
>   Network Interface 'NETBIOS' atartup failed
>
>netstat -n reports no interfaces
>
>Seems like there is is fundamental problem here.  Any help or
>suggestions would be great, Thanks
>
>--
>Andy Cannon   http://www.sonic.net/~cannon
>
>
>


baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: stefand@lcam.u-psud.fr                            30-Aug-99 08:45:17
  To: All                                               30-Aug-99 12:22:13
Subj: Re: resolv2 vs hosts

From: stefand@lcam.u-psud.fr (Stefan A. Deutscher)

On Sat, 28 Aug 1999 15:31:21 GMT, Darin McBride
<dmcbride@no.tower.spam.to.org> wrote:
>On Sat, 28 Aug 1999 10:45:46 -0400 (EDT), Barry wrote:
>
>>Can someone explain the difference between the resolv2 file and the
>>hosts configuration tab in TCPIP?  Is resov2 for the local network and
>>hosts for remote networks?  Resolv2 has input for Nameserver
>>addresses, does hosts have/need Nameserver addresses?
>
>resolv2:
>
>Holds your domain and your nameservers.
>
>hosts:
>
>Holds IP addresses for certain hosts which you don't want to, or can't,
>go to the nameservers.  For example, if a certain machine isn't IN the
>nameserver, you can give it a nickname here.


On top of that, you can put in nicknames for machines which are known to
your name servers for added ease of use.

A line like

my.ftp.host.uuu.edu  128.128.128.128  uuu # primary ftp host at U. of Uu

would allow you to type

 ftp uuu      or       telnet uuu          or       ping uuu [etc.]

instead of

 ftp my.ftp.host.uuuuu.edu

etc. Very nice when you deal with lots of machines, and also it
increases response time since the name server lookup over the net is not
needed. Only possible drawback: When the sysadmins of U. of Uu change
the IP numbers of their machines and the stuff in your hosts file points
to rubbish afterwards. This can be somewhat configured.

NB: Dunno off the top of my head whether it is 

  name ip nickname # comment

or

  ip name nickname # comment

but it's in the docs.

 Cheers,
               Stefan

-- 
=========================================================================
Stefan A. Deutscher                       | (+33-(0)1)   voice      fax
Laboratoire des Collisions Atomiques et   | LCAM :  6915-7699  6915-7671
Mol\'{e}culaires (LCAM), B\^{a}timent 351 | home :  5624-0992  call first
Universit\'{e} de Paris-Sud               | email:  sad@utk.edu 
91405 Orsay Cedex, France (Europe)        |         (forwarded to France)
=========================================================================
 Do you know what they call a quarter-pounder with cheese in Paris?

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Universite Paris-Sud, France. (1:109/42)

+----------------------------------------------------------------------------+

From: morgannalefey@my-deja.com                         30-Aug-99 12:36:00
  To: All                                               30-Aug-99 14:26:20
Subj: Re: warp 4 and two net cards

From: morgannalefey@my-deja.com

In article <Zziy3.507$g57.81129@typhoon2.gnilink.net>,
  Stephen Eickhoff <"operagost"@e-mail.com (remove the - )> wrote:
> Maybe I should put this in a FAQ somewhere!

Heh, would only work if people read the FAQ. ;D

> I have two 3c900 in my web server. I had to go into the settings in
MPTS and specify which PCI slot # it was or else both drivers
initialized the same
> card. You may have the same problem. Note which PCI slot the driver
reports for the card. Assuming you have both cards in adjacent slots,
the next
> higher number should be the correct slot for card number 2.

This is the solution that worked for me last week, Chad.  Figure out
what slot OS/2 is looking for and go to MPTS, select the first card
driver, then Edit (or is it options?) and there's a place to specify
the slot and the card ID.  I recommend entering both. :)

> chadpaul@netexpress.net wrote:
>
> > Hello,
> >
> > I am trying to install and configure two netcards.
> > I am using OS/2 v.4.
> >
> > I have tried pci with isa, two pci and two isa.
> > The last attempt I put two 3com 3c905 cards in.
> > one at irq 10 and 300h,  the other is at irq 12 and 380h.
> > WHile installing, the system only saw one card in the installation.
> > when it rebooted, it seemed to detect more than one card.
> >
> > Is there a config file to edit and say what card is using what
address and irq?
> >
> > IS anyone using two cards in a os/2 machine.
> >
> > Has anyone ever tried using the Intel server dual port ethernet
card or any of the adaptec dual or quad port ethernet cards in a os/2
4.0 machine?
> >
> > Any info is appreciated.
> >
> > chadpaul@netexpress.net
> >
> > Chad Pauli
>
>

--
Siobhan Perricone
PC Technician
Alltel Information Services
(I only speak for myself, not for Alltel)


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Deja.com - Share what you know. Learn what you do
(1:109/42)

+----------------------------------------------------------------------------+

From: diego@cryogen.com                                 30-Aug-99 08:01:16
  To: All                                               30-Aug-99 16:56:27
Subj: Re: How do I Change IP Address info

From: "Justin Greywolf" <diego@cryogen.com>

That may be what it says there...I am only going from experience with all of
the machines that we use here.  If we do not change the IP in MPTCONFG.INI,
the workstation will be confused.


James Knott wrote in message <+k8x3odSRE4J089yn@ibm.net>...
>In article <HGTw3.547$H.32280@news.uswest.net>,
>"Justin Greywolf" <diego@cryogen.com> wrote:
>>
>>James Knott wrote in message <9vmw3odSRYfO089yn@ibm.net>...
>>
>>>>It could not hurt to also change the IP statement in the MPTCONFG.ini,
>>>>located in the same directory.
>>>
>>>Why would you want to do that?  Changing setup.cmd, either manually or
>>>via tcpcfg does what's required.
>>
>>
>>Not if you are changing the IP.  You need to change it in both files or it
>>will lead to problems.
>
>What problems.  According to the "Network Adapters and Protocol
>Services Guide" (quoted below), if TCP/IP is installed, most of that
>file, including the part containing the IP address is ignored.
>
>
>
>" The Sockets MPTS section contains the TCP/IP configuration
>parameters.  When used, this section generates an  MPTCONFIG.INI file
>with the values specified.  During MPTS installation/configuration (if
>there is no IBM TCP/IP  product on the workstation), the values
>specified in the Sockets MPTS section are used to generate the
>SETUP.CMD file.
>
>The MPTS section contains several segments that are [CONTROL],
>[IFCONFIG], [ROUTE], [DHCP], and [NETBIOS].  If  the workstation
>currently has an IBM TCP/IP Base Kit product on it, only the [CONTROL]
>section needs to be  specified (if required) to configure Local IPC
>sockets, NetBIOS Sockets, or TCP/IP Sockets access.  If the
>Workstation does not have the IBM TCP/IP product on the system, a
>complete Sockets MPTS section (with all  segments properly completed)
>must be specified to enable TCP/IP sockets access.
>
>You can specify a valid MPTCONFG.INI file within the Sockets MPTS
>section.  The information specified in this  section overwrites an
>existing MPTCONFG.INI file.  To update a currently configured
>MPTCONFG.INI file, use the  MPTS Update Section.  The Sockets MPTS
>section should be placed after any NetBIOS sections."
>
>
>
>As shown above, only the control part of the file is used when TCP/IP
>is installed.  The IP address is in the ipconfig section, and
>therefore ignored.
>
>
>--
>E-mail jknott@ca.ibm.com
>_________________________________________________________________________
>The above opinions are my own and not those of ISM Corp., a subsidiary of
>IBM Canada Ltd.


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: U S WEST Interprise (1:109/42)

+----------------------------------------------------------------------------+

From: bgsmith/r6pnw_deschutes@fs.fed.xxus               30-Aug-99 08:09:29
  To: All                                               30-Aug-99 16:56:27
Subj: Re: "RoadRunner" cable service

From: "Bradley G. Smith" <bgsmith/r6pnw_deschutes@fs.fed.xxus>

On Mon, 30 Aug 1999 02:22:31 GMT, heloman@my-deja.com wrote:

>Is
>anyone using RoadRunner and if so HOW did you get
>around this lousy way of logging on? What lan card
>would you recommend and where are the driver(s)
>located? I look forward to your recommendations.....

Not all roadrunner systems use the log-on process. Mine does not for example,
and OS/2 works just fine. There is an OS/2 version of the log-on process,
checkout: http://Larkin.NuclearWinter.com/rros2/

Brad



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: US Forest Service (1:109/42)

+----------------------------------------------------------------------------+

From: sprins@businessnet.net                            30-Aug-99 18:32:04
  To: All                                               30-Aug-99 16:56:27
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: Sacha Prins <sprins@businessnet.net>


Baden Kudrenecky wrote:

> >each day. I might try to downlevel if IBM doesn't come up with any solution 
soon. Are 8421+
> >for SMP machines? Or do I need a different version for that? I'll look
around
>
>    No, if you go to
> ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us,
> you will see that there are separate packages for SMP:

What does 'converged' mean? Stacks wr08423_conv and wr08424_convare that. I
thought that that
means it can be installed on all OS/2 versions (including SMP). But I'm not
sure.

> >So much for (b)leeding edge e-business
> >tools...
>
>    That's why I stay away from the edge, never lone the bleeding
> edge.  {:-)

A wise decision.

Sacha


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: A2000 Kabeltelevisie en Telecommunicatie (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         30-Aug-99 16:48:04
  To: All                                               30-Aug-99 16:56:27
Subj: Re: multiple subnet support: Win9x vs OS/2

From: ad354@FreeNet.Carleton.CA (James Owens)

"Duane A. Chamblee" (duanec@eyebm.net) writes:
> On 27 Aug 1999 11:26:53 GMT, James Owens wrote:
> 
>>>>My OS/2 Warp 4 machine won't. It always uses the first address assigned
>>>>(in this case n.n.60.66) as the source address. This means that messages
>>>>sent to the other subnet contain the wrong source address and
>>>>communication fails (unless the receiving machine happens to be on
>>>>both subnets also.)
>>>>
>>>>Is there a fix?
> 
> this may be address in the MPTS updates and fixpaks. Are you running the
> basic installation from the Warp4 CD?
> 
> If so, You may want to update your stack to the latest 16-bit (MPTS 5.1x)
>   WR08423(install) + WR08424(fix)

OK, I instaled both, and LATEST40.EXE from the TCPIP\STACK directory for
good measure -- but the problem still exists.

Something has improved, because there used to be an ARP exchange involving
SOURCE n.n.60.0 and DEST n.n.99.1, and now it involves SOURCE n.n.99.2 and
DEST n.n.99.1.  However, when I ping n.n.99.1, the packet still goes out
with source address n.n.60.66, which the host at n.n.99.1 can't see.

I tried adding a route to n.n.60.0 from n.n.99.2 but it doesn't help. 
--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: fordp@southwestern.edu                            30-Aug-99 13:47:24
  To: All                                               30-Aug-99 16:56:28
Subj: DHCP Warp 3

From: Kacerdias <fordp@southwestern.edu>

I was recently given a copy of OS/2 Warp v3 (non-connect) to tinker
with. I can't for the life of me figure out how to get it to send and
receive TCP/IP packets via ethernet. I installed the FreeTCP package and
it all works fine except that I can't find a DHCP client for this
version of Warp anywhere. The NDIS drivers for this Intel EtherExpress
Pro 10+ are installed and working great, but I can't get an IP address
since everything on campus is dynamically based. I figure I should put
something in the lanstart.cmd file, but bootp doesn't do anything. Can
anyone help me out here? What do I need to get now?

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Posted via Supernews, http://www.supernews.com (1:109/42)

+----------------------------------------------------------------------------+

From: barrowcl@flash.net                                30-Aug-99 18:53:09
  To: All                                               30-Aug-99 16:56:28
Subj: SMTP Mail Server

From: "George Barrowcliff" <barrowcl@flash.net>

How can I find out the name of my SMTP server?
I want to use the SENDMAIL but need the name of the server

TIA GWB


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Bergen Brunswig (1:109/42)

+----------------------------------------------------------------------------+

From: bauman@ibm.net                                    30-Aug-99 20:33:08
  To: All                                               30-Aug-99 21:34:14
Subj: Socket Services-Fixed IRQ

From: bauman@ibm.net (Philip Bauman)

How does one disable PnP and specify I/O addresses and IRQ's for a Thinkpad
Laptop 390E?  Trying to get two ethernet cards running at one. 

TIA

Philip Bauman
bauman@ibm.net

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: frli@swipnet.se                                   30-Aug-99 22:44:16
  To: All                                               30-Aug-99 21:34:14
Subj: Create boot diskette with ethernet support

From: "FREDRIK LINDSTRM" <frli@swipnet.se>

Today I have all my backups in a external Hdd. I have managed with help from
others to create a single boot diskette so that I can restore my system
easy. Now I wounder if there are any knowledge in how to create a
diskette(s) with ethernet support so that I can store image files and so
forth on a another computer over a LAN. It does not need to be Warp4 as long
as I can create HPFS partitions and use ZIP then it is enough.

I have tried to find information on the net but have not so far been
succesfull.

If you know something about this subject please answer cause a would be very
greatfull.

Thanks for reading this
Fredrik Lindstrom




--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: A Customer of Tele2 (1:109/42)

+----------------------------------------------------------------------------+

From: donnelly@tampabay.rr.com                          30-Aug-99 21:19:01
  To: All                                               30-Aug-99 21:34:15
Subj: Re: "RoadRunner" cable service

From: donnelly@tampabay.rr.com (Buddy Donnelly)

On Mon, 30 Aug 1999 15:09:59, "Bradley G. Smith" 
<bgsmith/r6pnw_deschutes@fs.fed.xxus> a crit dans un message:

> On Mon, 30 Aug 1999 02:22:31 GMT, heloman@my-deja.com wrote:
> 
> >Is
> >anyone using RoadRunner and if so HOW did you get
> >around this lousy way of logging on? What lan card
> >would you recommend and where are the driver(s)
> >located? I look forward to your recommendations.....
> 
> Not all roadrunner systems use the log-on process. Mine does not for
example,
> and OS/2 works just fine. There is an OS/2 version of the log-on process,
> checkout: http://Larkin.NuclearWinter.com/rros2/

Roadrunner won't support OS/2 or Linux and hates supporting Macs, but 
there's lots of people using it with them.

Telephone Mical Solomon, Support Services Manager for RR Tampabay, at 
813-371-2214 and ask him lots of questions. He's very good at his job but 
probably unempowered by his boss suits. You should at least ask for a 
reduced fee for not needing the Windows install they're going to give you 
as part of the package.

If they don't reduce your fee, write me and I'll send you a "Sign up a 
friend" deal, and save a month's worth of fees in the bargain.



Good luck,

Buddy

Buddy Donnelly
donnelly@tampabay.rr.com


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: RoadRunner - TampaBay (1:109/42)

+----------------------------------------------------------------------------+

From: tomlins@cam.org                                   30-Aug-99 22:13:11
  To: All                                               30-Aug-99 21:34:15
Subj: Re: setting up connectivity between NT and OS/2

From: tomlins@cam.org

In <A0ny3.6120$2k6.77753@news1.rdc1.bc.home.com>, baden@unixg.ubc.ca   (Baden
Kudrenecky) writes:
>In <37C8DF53.1DACC59C@sonic.net>, Andrew Cannon <cannon@sonic.net> writes:   

Refer to these urls and read comments below:

http://huizen.dds.nl/~jacco2/samba/warp.html
Has the BEST explaination on how to configure OS2's peer support.

http://www.mit.edu/afs/athena/activity/o/os2/www/peer/WARPPEER.HTM
explains about OS2 vs NT/W95

>>My employer has installed an ISDN line to my home for remote access to
>>work.  This is the first time I have attempted to setup a network on
>>OS/2(always had used a modem to ISP).
>>
>>There doesn't seem to be any instructions on setting this up to an NT
>>network, and I have struggled with this for quite some time, so I
>>would appreciate any advice.  Here is what I have done so far:

>>3)MPTN -
>>    LAPS - removed No Network Adapter configuration
>>        added my LAN adapter
>>        added protocols
>>        0 - ... NETBIOS over TCP/IP
>>        0 TCP/IP
>
>   I have not done this, however, it is my understanding that
>you need to have both NETBIOS and TCP/IP on LAN0, and that
>NETBIOS over TCP/IP has to be on a separate interface, i.e.
>LAN1.

This is NOT correct.  I have my machine setup exactly like his
and it works correctly.

>>during bootup
>>ifconfig reports that the interface cannot be found (I believe
>>lan0 should be the default interface)

Do you see the nic driver load during boot?  Is it outputing any
error messages?  If its a PNP card you may have to fiddle with the
bios to ensure an iterupt is available.  To me it sounds like your
nic card is not being seen correctly...  Have you got the latest
drivers for it?

Also it would seem that where the nic drivers are in the config.sys
can make or break a system.  In my case I had to move the drivers to
just after:

DEVICE=D:\ibmcom\LANMSGDD.OS2 /I:D:\ibmcom /S
DEVICE=D:\ibmcom\PROTMAN.OS2 /I:D:\ibmcom
DEVICE=D:\IBMCOM\MACS\KTC100.OS2
DEVICE=D:\IBMCOM\MACS\DLKFET.OS2

If I leave them were MPTS puts them I get a hang, with PEER active, when
ifconfig tries to set the ip address (or dhcpcd).  If they are at the end of
the
config.sys I get traps during boot.  I also had to reorder these statements:

RUN=D:\ibmcom\PROTOCOL\NBTCP.EXE
DEVICE=D:\IBMCOM\PROTOCOL\TCPBEUI.OS2
DEVICE=D:\IBMCOM\PROTOCOL\NETBIOS.OS2
DEVICE=D:\IBMLAN\NETPROG\RDRHELP.200
IFS=D:\IBMLAN\NETPROG\NETWKSTA.200 /I:D:\IBMLAN /N

One other piece of advice I got was to add a 'dummy' .bat file containing
just 'exit' to my startup folder.  Apparently without this PEER can sometimes
hang the machine.  (I do really mean .bat it needs to start a dos session).

>>TME 10 NetFinity Network Interface reports
>>   Network Interface 'NETBIOS' atartup failed
>>
>>netstat -n reports no interfaces
>>
>>Seems like there is is fundamental problem here.  Any help or
>>suggestions would be great, Thanks
>>
>>--
>>Andy Cannon   http://www.sonic.net/~cannon
>>
>>
>>
>
>
>baden
>
>baden@unixg.ubc.ca
>http://baden.nu/
>OS/2, Solaris & Linux
>

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: me (1:109/42)

+----------------------------------------------------------------------------+

From: Jeff.davis.no@spam.ibm.net                        30-Aug-99 19:26:10
  To: All                                               31-Aug-99 03:52:24
Subj: Re: Problem with loopback and hostname

From: "Jeff Davis" <Jeff.davis.no@spam.ibm.net>

On Mon, 30 Aug 1999 01:40:53 GMT, Darin McBride wrote:
><slaps forehead> Duh.  Of course.  Try "host jeff".  Also, if you would,
>what's the output of running "sniffle /P" (the case is important on the
>parameter)?
host jeff gives 127.0.0.1  The same result for local host.
The sniffle output is....

The TCP/IP address specified for your primary interface does not match the
address returned by the name server or in your etc/hosts entry.  Check the
value of your HOSTNAME environment variable. Ensure that the IP address
associated with this name matches the IP address specified in the TCP/IP 
Configuration notebook. If this machine is configured as a localhost
(loopback interface is enabled), please remove the HOSTNAME environment
variable from your config.sys file.

HOSTNAME is also gone.  I got rid of the machine nae and TCP config did the
rest

>Work?  Where?  Not too many people have v6.1 yet, I would have imagined.  We
>only GA'd a bit ago...  If you're internal, reply via Notes.

I'm with the Government of Prince Edward Island.  I'm a systems manager.  I
was exposed to DB2 about 15 years 
ago and had a fondness for it then.  Still do.  The 6.1 single user edition
is available for download

>Java doesn't use any search engine.  Visual Age for Java, however, does. 
>Important question: which did you install first - VA Java, or DB2 PE?  Also,
>which version of VA Java?

Ah.. Your bang on.  I thought about my error on the way to work.  The README
says VA JAVA 2.0 Entry Edition
It was installed first. 
 
>
>What I'd really like to know is *what* exactly isn't working?  I assume both
>installed fine, but just the search in DB2 doesn't work, and the help (the
>ENTIRE help) in VA Java doesn't work.  I assume everything else works fine? 
>Of course, if you're developing, the help is somewhat important.  :-)
>
Your very close to being right on the money.  Both installs went Fine.  The
help works in BOTH products.  It just the search engine that is kaput.  I
suspect it is the same problem with both products.  VA JAVA isn't as clever
as DB2 so it doesn't check before starting the search server.  It starts the
server possibly when help is first loaded then times out on an error.  This
is only a guess.  The external symptoms are that I am merrily wandering
through help when this terrifying black screen appears from NetQuestion.
which says an error has occurred, hit enter to kill the server. Then I am
back
to the HTML help

I tried type netq.cfg  as per your instructions and all I got back was a
"beta"" character.  Could this be what is fried.

Anyway I will chug on.

Jeff

// ------------------------------------
// Jeff Davis
// Prince Edward Island, Canada



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: Rolf.Breuning@t-online.de                         31-Aug-99 00:13:19
  To: All                                               31-Aug-99 03:52:24
Subj: Re: Netscape API ?

From: Rolf Breuning <Rolf.Breuning@t-online.de>

Lamps stgt wrote:
> 
> Is there an C/C++ API for the Netscape Communicator? I want to send a
html-page
> to the Netscape Communicator from a C-application.
> Thomas

You might simply use the command line interface 
  (see help-command line options)
via a system call

regards 

  Rolf Breuning

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: T-Online (1:109/42)

+----------------------------------------------------------------------------+

From: Spammers@Bite.Me                                  30-Aug-99 23:00:09
  To: All                                               31-Aug-99 03:52:24
Subj: Apache 1.3.9 broke my REXX cgi

From: "Jaime A. Cruz, Jr." <Spammers@Bite.Me>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Okay, I'm getting frustrated here.  I don't know what happened, these things
worked with every other version of Apache 1.3 that I've tried.  Suddenly,
after upgrading to 1.3.9, whenever I try to access one of my many REXX CGI
scripts, my browser displays this cryptic message:

Internal Server Error

The server encountered an internal error or misconfiguration and was unable
to complete your request.

Please contact the server administrator, spammers@bite.me and inform them of
the time the error occurred, and anything
you might have done that may have caused the error.

More information about this error may be available in the server error log.


Apache/1.3.9 Server at Jaime Port 80

When I check the error_log file, I find this:
[Mon Aug 30 18:50:54 1999] [error] [client 10.0.0.1] Premature end of script
headers: H:\VIRTUAL_HTML\CGI-BIN\TESTCGI.CMD

This script has not changed in months.  Why is Apache 1.3.9 suddenly throwing
up over this (and ALL) of my cgi scripts??  What changed??  Help!!



Jaime A. Cruz, Jr.

o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o
o                                                 o
o  Visit the Nassau Wings Motorcycle Club at:     o
o  http://www.nassauwings.org/                    o
o  A Charter Member of the Motorcycle Web Ring!   o
o                                                 o
o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o



-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0 OS/2 for non-commercial use
Comment: PGP 5.0 for OS/2
Charset: cp850

wj8DBQE3yv6MgvzYfxgMc34RAt9VAJwPWQcRIygmyAROY1y0oKTFOPhycgCeL9t7
M4p89lhoZ8evJhlgLJfsfdU=
=4US/
-----END PGP SIGNATURE-----

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Nassau Wings Motorcycle Club (1:109/42)

+----------------------------------------------------------------------------+

From: drowelf@vnet.net.nospam                           30-Aug-99 18:40:07
  To: All                                               31-Aug-99 03:52:24
Subj: Re: DHCP Problem

From: "Eric Erickson" <drowelf@vnet.net.nospam>

On Sat, 28 Aug 1999 19:52:35 GMT, Frank Beythien wrote:

>On Sat, 28 Aug 1999 15:42:06, "Eric A. Erickson" 
><drowelf@nospamvnet.net> wrote:
>
>> My new notebook can not seem to receive the DHCP Offer from my Warp Server 
>> Advanced (TcpIp 4.1) Server. I can see the DHCP Server offering the address 
to
>> the client, but it never seems to accept it. I am running Warp 4 FP11 and
TcpIp V4.0 
>> with MPTS IP 8424 stack.  I know that DHCP is working properly since I can
get 
>> a DHCP Served address when I'm at a client site. 
>> 
>> So how can I debug this problem?
>
>run DHCPMON and activate logging in  DHCPcd.cfg.
>
>CU/2
>-- 
>Frank Beythien   fBeythien@gmx.de
Done, 

That, but I can't make heads or tails of the information in the 
log. DHCPMON just states that we are in discover mode. 




--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Elvish Software Foundry, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: drowelf@vnet.net.nospam                           30-Aug-99 18:44:13
  To: All                                               31-Aug-99 03:52:24
Subj: Re: warp 4 and two net cards

From: "Eric Erickson" <drowelf@vnet.net.nospam>

I have multiple systems with multiple cards with out a problem. 

1 Notebook with 2 PCMCIA Cards (Token Ring/Ethernet)
1 Server with 3 Cards (2 Ethernet. 1 Token Ring)
1 Server with 2 Cards (1 Ethernet, 1 Token Ring)

Its just a matter of using MPTS to configure all the cards so 
that there is no conflict. Depending on how flexible you 
driver software is it may require switching cards around
in PCI Slots to remove any IRQ conflicts.




--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Elvish Software Foundry, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: abstewa@ibm.net                                   30-Aug-99 20:43:09
  To: All                                               31-Aug-99 03:52:24
Subj: Re: "TNS-12560 Protocol Adapter Error" under OS2

From: Alon Stewart <abstewa@ibm.net>

You need to have the tnsnames.ora file in the Oracle directory
oraos2\network\admin.  You can run the config program that comes with Oracle
apps
or get the one that is on the server (network) to which you attach and copy it 
to
that directory.

EKadakal wrote:

> I am trying to access an Oracle database (NT) from a Warp4 workstation.
However
> I am getting the following error:
>
>             TNS-12560 Protocol Adapter Error
>
> Does anyone kn0ow anyting about this problem?
>
> Note: Regular IP ping is successful

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: abstewa@ibm.net                                   30-Aug-99 20:46:29
  To: All                                               31-Aug-99 03:52:24
Subj: Re: Can I telnet into OS/2 and auto start a program?

From: Alon Stewart <abstewa@ibm.net>

I would use rsh or rexec.  You will have better luck.  The dameon needs to be
started on the host before you can connect.

Mikael Wahlgren wrote:

> :>        It works. But, sadly, all the messages are shown in the window
which
> :>      telnet daeman runs. The remote window seems to be dead?!
> :>
> :>      How to correct such problem? Does the User ID, Group ID, or Shell
> :>      setup by passwd.exe have any effects on it?
>
> You can use OS2You as your telnet daemon instead...
>
> Mikael Wahlgren - mail@ridax.se
> Ridax programutveckling - PM2You/OS2You Remote Control for OS/2
> FTP 194.52.57.138 - http://welcome.to/ridax
> WIN2You - Remote Control for Windows 95/NT

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     31-Aug-99 01:07:19
  To: All                                               31-Aug-99 03:52:25
Subj: Re: Apache 1.3.9 broke my REXX cgi

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Mon, 30 Aug 1999 23:00:18 GMT, Jaime A. Cruz, Jr. wrote:

>-----BEGIN PGP SIGNED MESSAGE-----
>Hash: SHA1
>
>Okay, I'm getting frustrated here.  I don't know what happened, these things
>worked with every other version of Apache 1.3 that I've tried.  Suddenly,
>after upgrading to 1.3.9, whenever I try to access one of my many REXX CGI
>scripts, my browser displays this cryptic message:
>
>Internal Server Error
>
>The server encountered an internal error or misconfiguration and was unable
>to complete your request.
>
>Please contact the server administrator, spammers@bite.me and inform them of
>the time the error occurred, and anything
>you might have done that may have caused the error.
>
>More information about this error may be available in the server error log.
>
>
>Apache/1.3.9 Server at Jaime Port 80
>
>When I check the error_log file, I find this:
>[Mon Aug 30 18:50:54 1999] [error] [client 10.0.0.1] Premature end of script
>headers: H:\VIRTUAL_HTML\CGI-BIN\TESTCGI.CMD
>
>This script has not changed in months.  Why is Apache 1.3.9 suddenly throwing
>up over this (and ALL) of my cgi scripts??  What changed??  Help!!

Probably because the system more tightly checks the headers now?  Could you
post the headers your script provides?  Or perhaps the portion of the code
that produces the headers?

[I switched away from REXX to Perl a while back on my server...]

---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: dmcbride@no.tower.spam.to.org                     31-Aug-99 01:21:23
  To: All                                               31-Aug-99 03:52:25
Subj: Re: Problem with loopback and hostname

From: "Darin McBride" <dmcbride@no.tower.spam.to.org>

On Mon, 30 Aug 1999 19:26:20 -0300 (ADT), Jeff Davis wrote:

>On Mon, 30 Aug 1999 01:40:53 GMT, Darin McBride wrote:
>><slaps forehead> Duh.  Of course.  Try "host jeff".  Also, if you would,
>>what's the output of running "sniffle /P" (the case is important on the
>>parameter)?
>host jeff gives 127.0.0.1  The same result for local host.
>The sniffle output is....
>
>The TCP/IP address specified for your primary interface does not match the
>address returned by the name server or in your etc/hosts entry.  Check the
>value of your HOSTNAME environment variable. Ensure that the IP address
>associated with this name matches the IP address specified in the TCP/IP 
>Configuration notebook. If this machine is configured as a localhost
>(loopback interface is enabled), please remove the HOSTNAME environment
>variable from your config.sys file.
>
>HOSTNAME is also gone.  I got rid of the machine nae and TCP config did the
>rest

Hmmm... have you tried making jeff an alias for whatever IP address you *do*
have for your primary adapter?  I assume you have a network card setup?  This
would be LAN interface 0.  [Had to go bring up tcpcfg2 to find the name :-)]

For example, I have that adapter set up for 10.10.10.1, and have my hostname
as an alias for this address rather than localhost.  This way, when I type in
"host jeff", I get 10.10.10.1, or, at least, I would if that were my
hostname.

>>Work?  Where?  Not too many people have v6.1 yet, I would have imagined.  We
>>only GA'd a bit ago...  If you're internal, reply via Notes.
>
>I'm with the Government of Prince Edward Island.  I'm a systems manager.  I
>was exposed to DB2 about 15 years 
>ago and had a fondness for it then.  Still do.  The 6.1 single user edition

Fifteen years ago ... wow.  :-)

>is available for download

SU is a name that went out before I got to IBM.  :-)  Now called Personal
Edition (as you called it earlier).  I assume this means you have the try&buy
version?

>>Java doesn't use any search engine.  Visual Age for Java, however, does. 
>>Important question: which did you install first - VA Java, or DB2 PE?  Also,
>>which version of VA Java?
>
>Ah.. Your bang on.  I thought about my error on the way to work.  The README
>says VA JAVA 2.0 Entry Edition
>It was installed first. 

Oooo... Well, if it weren't for the fact that you're getting a serious
problem coming out of sniffle, I'd say this was your problem.  We have a
documented problem if VA Java 2.0 goes onto a system before DB2... and even
in the reverse, you need DB2's first fixpak (due out "RSN").

>>What I'd really like to know is *what* exactly isn't working?  I assume both
>>installed fine, but just the search in DB2 doesn't work, and the help (the
>>ENTIRE help) in VA Java doesn't work.  I assume everything else works fine? 
>>Of course, if you're developing, the help is somewhat important.  :-)
>>
>Your very close to being right on the money.  Both installs went Fine.  The
>help works in BOTH products.  It just the search engine that is kaput.  I
>suspect it is the same problem with both products.  VA JAVA isn't as clever
>as DB2 so it doesn't check before starting the search server.  It starts the
>server possibly when help is first loaded then times out on an error.  This
>is only a guess.  The external symptoms are that I am merrily wandering
>through help when this terrifying black screen appears from NetQuestion.
>which says an error has occurred, hit enter to kill the server. Then I am
>back to the HTML help

Hmmm... I heard rumours (obviously false) that NetQuestion was required for
all of Java's help...

>I tried type netq.cfg  as per your instructions and all I got back was a
>"beta"" character.  Could this be what is fried.

I can't find where I asked for the netq.cfg.  However, if you are feeling
adventurous, you may also want to play with the httpd.cnf in the netqos2
directory.  (I assume you know where this is.  If you don't, try typing
"db2netqd dir".  More undocumented commands.  Neat, huh?)  Don't play with
the port number... things won't work if you do.  This, however, won't fix the
sniffle problem.

>Anyway I will chug on.

I'm interested in getting this fixed even if just to update the NetQuestion
readme we have.  :-)
---
Disclaimer: unless explicitly mentioned otherwise, I do not speak,
nor have I ever spoken, for the company I work for.



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: Spammers@Bite.Me                                  31-Aug-99 01:50:00
  To: All                                               31-Aug-99 05:26:03
Subj: Re: Apache 1.3.9 broke my REXX cgi

From: "Jaime A. Cruz, Jr." <Spammers@Bite.Me>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

The code is the same as it's always been... here it is:

  Say 'Content-type: text/html'
  Say ''
  Say '<HTML>'


On Tue, 31 Aug 1999 01:07:39 GMT, Darin McBride wrote:

>On Mon, 30 Aug 1999 23:00:18 GMT, Jaime A. Cruz, Jr. wrote:
>
>>
>>Okay, I'm getting frustrated here.  I don't know what happened, these things
>>worked with every other version of Apache 1.3 that I've tried.  Suddenly,
>>after upgrading to 1.3.9, whenever I try to access one of my many REXX CGI
>>scripts, my browser displays this cryptic message:
>>
>>Internal Server Error
>>
>>The server encountered an internal error or misconfiguration and was unable
>>to complete your request.
>>
>>Please contact the server administrator, spammers@bite.me and inform them of
>>the time the error occurred, and anything
>>you might have done that may have caused the error.
>>
>>More information about this error may be available in the server error log.
>>
>>
>>Apache/1.3.9 Server at Jaime Port 80
>>
>>When I check the error_log file, I find this:
>>[Mon Aug 30 18:50:54 1999] [error] [client 10.0.0.1] Premature end of script
>>headers: H:\VIRTUAL_HTML\CGI-BIN\TESTCGI.CMD
>>
>>This script has not changed in months.  Why is Apache 1.3.9 suddenly
throwing
>>up over this (and ALL) of my cgi scripts??  What changed??  Help!!
>
>Probably because the system more tightly checks the headers now?  Could you
>post the headers your script provides?  Or perhaps the portion of the code
>that produces the headers?
>
>[I switched away from REXX to Perl a while back on my server...]
>
>---
>Disclaimer: unless explicitly mentioned otherwise, I do not speak,
>nor have I ever spoken, for the company I work for.

Jaime A. Cruz, Jr.

o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o
o                                                 o
o  Visit the Nassau Wings Motorcycle Club at:     o
o  http://www.nassauwings.org/                    o
o  A Charter Member of the Motorcycle Web Ring!   o
o                                                 o
o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o



-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0 OS/2 for non-commercial use
Comment: PGP 5.0 for OS/2
Charset: cp850

wj8DBQE3yya3gvzYfxgMc34RArVrAKDifd/BKT5g7Uzxv1vH7zkvpdhjagCcDt43
I3YBQi+ZbYoCPUImZ/4MGaQ=
=5mwp
-----END PGP SIGNATURE-----

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Nassau Wings Motorcycle Club (1:109/42)

+----------------------------------------------------------------------------+

From: tvoltagg@home.com                                 31-Aug-99 02:36:22
  To: All                                               31-Aug-99 05:26:03
Subj: NR/2 and Netscape for OS/2 through Proxy Server

From: Tom Voltaggio <tvoltagg@home.com>

I am running Warp 4 FP 11 in a computer connected as a "client" in a 2
computer Peer network.  The other computer is running win95 and is
connected to a cable modem through a second NIC.  I have Internet Gate
proxy server software on the "gateway" computer.  I am able to ping both
computers and the "client" runs PMMail/2 through the proxy server fine.
The problems:
1 - How can I configure NR/2 ver 2.0 to run through the proxy server to
the cable modem?  I tried putting the IP address of the "gateway" in the
news server box, but it will not connect.  Any ideas?  If not, does
anyone know of a news reader program which can run through a proxy
server (other than Netscape)?
2 - I cannot get Netscape for OS/2 4.61 or 4.04 to run on the "client"
I keep getting "broken pipe" messages.  I can run the "client" in Win 95
through the Boot Manager and Netscape works fine in Win 95, so I know
this is an OS/2 problem and not a TCP/IP problem.  I'm using Beta 2 of
Netscape 4.61 and Java 1.1.8.  I reinstalled 4.61 and 4.04 and also
deleted each one so only one was installed.  No luck.  The real kicker
is that when I run Netscape 4.03 for Windows 3.1 in Win-OS/2, it runs
fine!!!  Any ideas will be appreciated.  Thanks.


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network (1:109/42)

+----------------------------------------------------------------------------+

From: rickstarich@home.com                              31-Aug-99 03:05:05
  To: All                                               31-Aug-99 11:04:18
Subj: Re: LPD Ports

From: Rick Starich <rickstarich@home.com>


Hamed wrote:
> 
> I am using OS/2 Warp Server 4.0.  I am facing a problem configuring LPD
> ports.
> 
> I have enable 8 LPD ports in TCP/IP Printing Configuration and selected
> Autostart LPDPORTD.
> 
> During printer configuration (Printer template), when I view the printer
> ports I cannot see any LPD ports; all  other local ports (LPTx, COMx) are
> shown.
> 
> Any ideas?
Use the following REXX script to add (logical port objects)LPT4-LPT9 

/* Add LPT4 to LPT9 into OS2SYS.INI */
/* Title this file LPTX.CMD */
/* This script works by it is absolutely not supported by IBM */
call RxFuncAdd 'SysIni', 'RexxUtil', 'SysIni'
do i=4 to 9
        call SysIni 'SYSTEM','PM_SPOOLER_PORT','LPT'||i, ';'||'00'x
end
exit

if you change 9 to 64 you can have 64 (logical) LPT port objects

run this script before you configure lprportd THEN

Open the 'Programs' object on the desktop, 
then 'TCP/IP Internet (LAN)' folder 
then double click on the TCP/IP configuration icon
then select the 'Autostart' tab
then arrow down to 'lprportd' Service, and put a checkmark in 'Autostart
Service' checkbox.

Next select the 'Printing' tab
for the 'Remote Print Server' specify the host name of the print server
where you want your line printer request (LPR) request to be sent. <I
left this field blank>

for the 'Remote Print Server's Printer' specify the name of the printer
queue on the remote print server or a device name "155.203.12.33" where
you want your (LPR) request to be sent. If you specify a device name,
the LPD server determines the associated queue.

i.e. LPT4 is the name of an OS/2 printer and PIPELPD4 "155.203.12.33" is
the device name of the IP attached remote printer.

This info is saved as the LPR_PRINTER environment variable in your
config.sys file.

Now you need to reboot for these changes to take effect.

Now create a Folder on your DeskTop for your LPRportD printer objects.

Drag a Printer Object from the template folder into your newly created
LPRportD folder.

The Printer Object setup page appears.

Give the LPRportD printer a meaningful name <For 'Name' type in
"PipeLPD4">, select the type of printer it will use and then double
click on the printer object '\PIPE\LPD4' This will bring up the settings
screen for the LPD printer. The minimum info you need to enter is the
LPD Server "155.203.nn.nn" and for "LPD printer" the word "text"

Click OK then the CREATE Button. Repeat this process for as many
LPRportD printers you need to define. Remember you can change the
LPTX.CMD script loop counter from 9 to 64.

Next add the following line or as many as you need to the startup.cmd

SPOOL /D:LPT4 /0:PIPELPD4
SPOOL /D:LPT5 /0:PIPELPD5
SPOOL /D:LPT6 /0:PIPELPD6

The lines above uses the SPOOL command to redirect a print job destine
for an LPT port and sends it to the LPRportD printer object which is
identified with an IP address.

Note: You can not use LPRportD printer objects to redirect printer
output from a Windoze app running under WIN-OS2. WIN-OS2 only supports
Lpt 1-3

Hope this points you in the right direction. 

-- 
RickStarich@home.com
A Warp'ed mind is a terrible thing to waste. Get OS/2 Warp

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Micro Software Consultants Inc (1:109/42)

+----------------------------------------------------------------------------+

From: rickstarich@home.com                              31-Aug-99 04:00:20
  To: All                                               31-Aug-99 11:04:19
Subj: How do you connect NT wkst to Warp Server

From: Rick Starich <rickstarich@home.com>

Has anyone been sucessful using IBM's "Primary Logon Client for NT (June
99 version)" to access resources on a Warp Server? My IBM Primary Logon
Client for NT tells me it can't see DOMAIN01 my Warp Server domain.

I have a feeling it has something to do with my TCPIP settings on Warp
Server.

Under TCP/IP config 'Host Names' tab  I have
'The Computers Host Name' = msc01
'Local Domain Name' = domain01

'Name Resolution' tab
'Named Server Address' = 192.168.0.1 (Warp Server)
'Lan Domain Serach List' <No Entries>

'Hosts' tab
192.168.0.1    msc01         (Warp Server)  
192.168.0.3    c397631-a     (NT workstation)

Look thru Host list before going to Name Server (is checked)

NT workstation can ping Warp Server.
Warp Server can ping NT workstation.

Have some suggestions?
 
-- 
RickStarich@home.com
A Warp'ed mind is a terrible thing to waste. Get OS/2 Warp

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Micro Software Consultants Inc (1:109/42)

+----------------------------------------------------------------------------+

From: domi@kenavo.NOSPAM.fi                             30-Aug-99 21:59:14
  To: All                                               31-Aug-99 11:04:19
Subj: Re: SMTP Mail Server

From: domi@kenavo.NOSPAM.fi (Dominique Pivard)

On Mon, 30 Aug 1999 18:53:19, "George Barrowcliff" 
<barrowcl@flash.net> wrote:

> How can I find out the name of my SMTP server?
> I want to use the SENDMAIL but need the name of the server

They often are in the form smtp.ISP_domain or mail.ISP_domain

Both smtp.flash.net and mail.flash.net exist (same host, actually). 


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: None!! (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                31-Aug-99 06:45:14
  To: All                                               31-Aug-99 11:04:19
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37CAB209.C2B539FB@businessnet.net>, Sacha Prins <sprins@businessnet.net>
writes:
>
>
>Baden Kudrenecky wrote:
>
>> >each day. I might try to downlevel if IBM doesn't come up with any
solution soon. Are 8421+
>> >for SMP machines? Or do I need a different version for that? I'll look
around
>>
>>    No, if you go to
>> ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us,
>> you will see that there are separate packages for SMP:
>
>What does 'converged' mean? Stacks wr08423_conv and wr08424_convare that. I
thought that that
>means it can be installed on all OS/2 versions (including SMP). But I'm not
sure.

WR08210 ServicePak/Refresh services MPTS for WARP Connect, and
WARP Server LS 5.0 (Converged Stack) U.S. and U.K. versions.  For
LS 4.0 MPTS (Non-Converged Stack) use WR08150.

This ServicePak/Refresh provides service for IBM MPTS LAN Adapter and
Protocol Support that shipped with WARP Connect, and LAN Server 5.0
MPTS that shipped with WARP Server, (Converged-Stack).  Countries
that use U.S. and U.K. versions may apply this ServicePak/Refresh.


baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                31-Aug-99 07:13:11
  To: All                                               31-Aug-99 11:04:19
Subj: Re: DHCP Warp 3

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37CAD1D5.61159078@southwestern.edu>, Kacerdias <fordp@southwestern.edu>
writes:

>I was recently given a copy of OS/2 Warp v3 (non-connect) to tinker
>with. I can't for the life of me figure out how to get it to send and
>receive TCP/IP packets via ethernet. I installed the FreeTCP package and
>it all works fine except that I can't find a DHCP client for this
>version of Warp anywhere. The NDIS drivers for this Intel EtherExpress
>Pro 10+ are installed and working great, but I can't get an IP address
>since everything on campus is dynamically based. I figure I should put
>something in the lanstart.cmd file, but bootp doesn't do anything. Can
>anyone help me out here? What do I need to get now?
>

   I first installed an OS/2 LAN Requestor on OS/2 3.0 Red, and
then I added MPTS 8421 to get the DHCP support.

ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us/wr08421_conv

baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: Unknown                                           31-Aug-99 09:53:18
  To: All                                               31-Aug-99 11:04:19
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: Armin Pollak <pollak@informatik.unibw-muenchen.de

In <37C8BB6C.3CFC2787@dgraph.com>, Kris Kadela <kris@dgraph.com> writes:
>
>
>Baden Kudrenecky wrote:
>> 
>> In <37C64ECA.1755AFFC@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
>> >Baden Kudrenecky wrote:
>> >
>> >> In <37C599EB.87A3C267@businessnet.net>, Sacha Prins
<sprins@businessnet.net> writes:
>> >>
>> >>    AFAIK, the 86xx MPTS is really buggy, get an 8423 or 8424.
>> >
>> >How does 842x stack up against 86xx in performance? I'm afraid I need the
32 bit
>> >performance for the stuff I'm doing.
>> >
>> >Sacha
>> 
>>    AFAIK, 8423 and 8424 are 32 bit stacks.  I just looked on my
>> one computer with 8424, and it is definitely a 32 bit stack, and
>> the syslevel states that the previous version was a 8421, which
>> I believe also was a 32 bit stack, as I went 32 bit in spring
>> 1998, with the previous January MPTS (8421?), and it
>> significantly improved the TCP/IP performance.
>> 
>> baden
>> 
>> baden@unixg.ubc.ca
>> http://baden.nu/
>> OS/2, Solaris & Linux
>
>I did not know that 8421+ were 32 bit. Are you sure?
>Is this upgrade hell or what? I don't even know what level of MPTS I am
>at now.
>Does anyone have a stable MPTS 86xx install?? How?

Running fine here. MPTS WRG8600. TCPIP 4.1.
As already said in another posting, maybe you mixed up tcpip and mpts.
As far as I known you cant run all possible combinations.
Look at the readmes which combination is possible.

Hope this helps
	Armin

-----------Armin Pollak----Universitaet der Bundeswehr Muenchen
Telefon: 089 / 6004 3502     Werner-Heisenberg-Weg 39, 
FAX:      089 / 6004 3397       85577 Neubiberg, Deutschland
email:    pollak@informatik.unibw-muenchen.de
     
This OS/2 system uptime is 1d 0h 23m 05s 250ms (en).

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Institut fuer Informatik 3.3 Universitaet der Bun
(1:109/42)

+----------------------------------------------------------------------------+

From: jknott@ibm.net                                    31-Aug-99 05:52:03
  To: All                                               31-Aug-99 11:04:20
Subj: Re: How do I Change IP Address info

From: jknott@ibm.net (James Knott)

Curious.  I have changed the IP address hundreds of times and never 
had a problem.  What sort of problem do you get?  How is it confused?


In article <W4xy3.226$1E3.19455@news.uswest.net>,
"Justin Greywolf" <diego@cryogen.com> wrote:
>That may be what it says there...I am only going from experience with all of
>the machines that we use here.  If we do not change the IP in MPTCONFG.INI,
>the workstation will be confused.
>
>
>James Knott wrote in message <+k8x3odSRE4J089yn@ibm.net>...
>>In article <HGTw3.547$H.32280@news.uswest.net>,
>>"Justin Greywolf" <diego@cryogen.com> wrote:
>>>
>>>James Knott wrote in message <9vmw3odSRYfO089yn@ibm.net>...
>>>
>>>>>It could not hurt to also change the IP statement in the MPTCONFG.ini,
>>>>>located in the same directory.
>>>>
>>>>Why would you want to do that?  Changing setup.cmd, either manually or
>>>>via tcpcfg does what's required.
>>>
>>>
>>>Not if you are changing the IP.  You need to change it in both files or it
>>>will lead to problems.
>>
>>What problems.  According to the "Network Adapters and Protocol
>>Services Guide" (quoted below), if TCP/IP is installed, most of that
>>file, including the part containing the IP address is ignored.
>>
>>
>>
>>" The Sockets MPTS section contains the TCP/IP configuration
>>parameters.  When used, this section generates an  MPTCONFIG.INI file
>>with the values specified.  During MPTS installation/configuration (if
>>there is no IBM TCP/IP  product on the workstation), the values
>>specified in the Sockets MPTS section are used to generate the
>>SETUP.CMD file.
>>
>>The MPTS section contains several segments that are [CONTROL],
>>[IFCONFIG], [ROUTE], [DHCP], and [NETBIOS].  If  the workstation
>>currently has an IBM TCP/IP Base Kit product on it, only the [CONTROL]
>>section needs to be  specified (if required) to configure Local IPC
>>sockets, NetBIOS Sockets, or TCP/IP Sockets access.  If the
>>Workstation does not have the IBM TCP/IP product on the system, a
>>complete Sockets MPTS section (with all  segments properly completed)
>>must be specified to enable TCP/IP sockets access.
>>
>>You can specify a valid MPTCONFG.INI file within the Sockets MPTS
>>section.  The information specified in this  section overwrites an
>>existing MPTCONFG.INI file.  To update a currently configured
>>MPTCONFG.INI file, use the  MPTS Update Section.  The Sockets MPTS
>>section should be placed after any NetBIOS sections."
>>
>>
>>
>>As shown above, only the control part of the file is used when TCP/IP
>>is installed.  The IP address is in the ipconfig section, and
>>therefore ignored.
>>
>>
>>--
>>E-mail jknott@ca.ibm.com
>>_________________________________________________________________________
>>The above opinions are my own and not those of ISM Corp., a subsidiary of
>>IBM Canada Ltd.
>
>

-- 
E-mail jknott@ca.ibm.com
_________________________________________________________________________
The above opinions are my own and not those of ISM Corp., a subsidiary of
IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: jknott@ibm.net                                    31-Aug-99 05:56:28
  To: All                                               31-Aug-99 11:04:20
Subj: Re: What does INETWAIT.EXE do?

From: jknott@ibm.net (James Knott)

Inetwait is simply a delay.  Enter "inetwait /?" at the command prompt
for details.  I don't see anything in mptstart.cmd or setup.cmd that 
would respond to an error code.



In article <cimumabqoootu.pminews@biggins.fe.psu.edu>,
"Larry Belan" <IHateSPAM@montypython.org> wrote:
>Can anyone point me to the answer of what INETWAIT does?  It's called in the
>\mptn\bin\mptstart.cmd file, and if there's an errorcode, it passes up the
>setup.cmd which init's the lan interfaces.
>
>I ask, since I'm trying to use my laptop with two networks (Ether pcmcia and
>ISA  TR in a dock) and if I boot without the dock, the ethenet card never
>starts.
>
>The ethernet is lan0 and the TR is lan1.  I never use BOTH at the same time.
>
>Thanks for any pointers!
>--
>Larry Belan
>PennState - Fayette
>www.fe.psu.edu
>
>

-- 
E-mail jknott@ca.ibm.com
_________________________________________________________________________
The above opinions are my own and not those of ISM Corp., a subsidiary of
IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: nospam_madbrain@thetaband.com                     31-Aug-99 03:44:11
  To: All                                               31-Aug-99 11:04:20
Subj: Theta Band Software releases WarpCharge payment software

From: nospam_madbrain@thetaband.com

Santa Clara, Calif. - Theta Band Software LLC released WarpCharge, a new
product that enables web sites to securely process credit card orders directly
over the Internet.

WarpCharge is the first credit card processing software available for the OS/2
platform.  WarpCharge provides the missing piece in your OS/2 e-commerce
solution : the payment system.

"Until now, merchants running secure web sites on IBM OS/2 Warp Server for
E-business had no means to process payments online. They could accept online
orders, but had to process the payment manually." said Julien Pierre,
President of Theta Band Software. "Shoppers like to get their purchase
immediately, and by processing credit card payments automatically and in
real-time, WarpCharge enables that sort of instant gratification."

WarpCharge is not just for e-commerce - it can be used for nearly all types of
businesses : whether you are taking credit card orders by mail, over the phone
or real-time over the Internet, WarpCharge is the perfect solution.

WarpCharge comes with sample CGI scripts for use on any secure OS/2 web
server, so that you can start taking Internet orders immediately.

WarpCharge also comes with an extensive REXX interface that lets developers
integrate credit card processing facility into any REXX-enabled OS/2
application, or into custom applications.

There are two editions of WarpCharge :

WarpCharge Business

WarpCharge Business is for any business that needs to process credit cards
over the phone, web, or in custom REXX applications. 

WarpCharge for Internet Service Providers

WarpCharge for ISPs lets Internet Service Providers offer credit card
processing capability to all their customers.

AVAILABILITY

WarpCharge is available for purchase from the Theta Band Software web site at
http://www.thetaband.com . WarpCharge Business requires a PC running IBM OS/2
Warp 4, OS/2 Warp Server or OS/2 Warp Server for E-Business. In addition,
WarpCharge for ISPs requires an OS/2 secure web server program.

ABOUT THETA BAND SOFTWARE LLC

Theta Band Software, headquartered in Santa Clara, California, was founded in
1997 and develops Internet and multimedia software products for IBM OS/2 Warp
that are marketed and sold on the world wide web.

-- 
--------------------------------------------------------------------
Julien Pierre               http://www.madbrain.com
Theta Band Software LLC     http://www.thetaband.com
--------------------------------------------------------------------

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Posted via Supernews, http://www.supernews.com (1:109/42)

+----------------------------------------------------------------------------+

From: sperber@airmail.net                               31-Aug-99 04:11:20
  To: All                                               31-Aug-99 14:56:01
Subj: Re: DHCP Warp 3

From: Darryl Sperber <sperber@airmail.net>

On Mon, 30 Aug 1999 13:47:49 -0500, Kacerdias <fordp@southwestern.edu> wrote:

>  Can anyone help me out here? What do I need to get now?

DHCP was added into Warp 3 with UN00959.  However I believe this required
TCP/IP version 3.0, which is what came with Warp Connect.

And it also required the underlying MPTS fixes... WR08210, or the much newer
WR08423 and WR08424.  These latter brought TCP/IP up to version 4.0.

And then you'd want IC17248, the latest TCP/IP 4.0 stack, DOSBOX, etc. to
bring your whole environment most-current.

But to be honest, I'm not sure this maintenance is applicable to plain Warp 3.


I know it applies to Warp Connect (since that's what I use... including DHCP
to support my cable modem), but I'd hate to steer you wrong regarding Warp 3.
I just haven't used plain Warp 3 for many years.

But the answer to the basic question is that DHCP (which is inherent in Warp
4) was added to Warp Connect with UN00959.


--
//
//   Darryl Sperber  (sperber@airmail.net)
//

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Airnews.net! at Internet America (1:109/42)

+----------------------------------------------------------------------------+

From: hg@uni.de                                         31-Aug-99 14:38:15
  To: All                                               31-Aug-99 14:56:01
Subj: SSH startup problems...

From: Heiko Glueckher <hg@uni.de>

After the successfull installation of the actual os/2 ssh client I get
the error message "Could not create directory '//ssh'" each time when
opening a ssh-session. The start of ssh is incredibly slow, maybe
because it tries several times to create this folder. After some 2
minutes of waiting I get the connection and everything works fine and
fast. Is there a possibility to force the ssh-client to create or use
another directory?

Thanks,

Heiko

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Rechenzentrum der Universitaet Freiburg, Germany (1:109/42)

+----------------------------------------------------------------------------+

From: engel003@mc.duke.edu                              31-Aug-99 09:46:15
  To: All                                               31-Aug-99 14:56:01
Subj: Re: How do you connect NT wkst to Warp Server

From: Leigh Engelhart <engel003@mc.duke.edu>

You need to run the Browser service on the server and have Netbios on both
the workstation and the server.  TCP-IP shouldn't have anything to do with
it.

Rick Starich wrote:

> Has anyone been sucessful using IBM's "Primary Logon Client for NT (June
> 99 version)" to access resources on a Warp Server? My IBM Primary Logon
> Client for NT tells me it can't see DOMAIN01 my Warp Server domain.
>
> I have a feeling it has something to do with my TCPIP settings on Warp
> Server.
>
>
> Have some suggestions?
>
> --
> RickStarich@home.com
> A Warp'ed mind is a terrible thing to waste. Get OS/2 Warp

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Duke University (1:109/42)

+----------------------------------------------------------------------------+

From: Unknown                                           31-Aug-99 16:00:26
  To: All                                               31-Aug-99 14:56:01
Subj: Re: SSH startup problems...

From: Armin Pollak <pollak@informatik.unibw-muenchen.de

In <37CBCCC6.A9F7426F@uni.de>, Heiko Glueckher <hg@uni.de> writes:
>After the successfull installation of the actual os/2 ssh client I get
>the error message "Could not create directory '//ssh'" each time when
>opening a ssh-session. The start of ssh is incredibly slow, maybe
>because it tries several times to create this folder. After some 2
>minutes of waiting I get the connection and everything works fine and
>fast. Is there a possibility to force the ssh-client to create or use
>another directory?
>

Startuptime is need to gemerate session key.....
Did you set environement variabel user and home
and points home to an existing directory??

Hope this helps
	Armin

-----------Armin Pollak----Universitaet der Bundeswehr Muenchen
Telefon: 089 / 6004 3502     Werner-Heisenberg-Weg 39, 
FAX:      089 / 6004 3397       85577 Neubiberg, Deutschland
email:    pollak@informatik.unibw-muenchen.de
     
This OS/2 system uptime is 1d 6h 28m 32s 687ms (en).

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Institut fuer Informatik 3.3 Universitaet der Bun
(1:109/42)

+----------------------------------------------------------------------------+

From: REMOVE_THISjbradley@msc.net                       31-Aug-99 08:42:12
  To: All                                               31-Aug-99 14:56:01
Subj: Connectivity

From: REMOVE_THISjbradley@msc.net

I have 4 machines in my home network, as follows:

A- PPro 200 with both 10 Mbps and 100 Mbps ethernet. The 100 Mbps card is
setup as 192.168.0.1 Netmask 255.255.255.0 and the 10 Mbps card is setup
as 192.168.1.1 Netmask 255.255.0.0 . This machine currently has the modem
(USR Sportster internal). IP forwarding is turned on.

B- PPro 200 with 100 Mbps ethernet. 192.168.0.3 Netmask 255.255.255.0

C- P133 with 100 Mbps ethernet. 192.168.0.5 Netmask 255.255.255.0

D- 486 with ISA bus with 10 Mbps ethernet. (If it had a PCI bus, or the
ISA fast ethernet cards that I've found had OS/2 drivers, I'd change the
card.)

I'd like to setup D as my internet server, using InJoy for IP network
address translation with dial on demand (or linux with diald). The problem
that I have, is that I can't get B or C to recognize D, nor D to recognize
B or C (I can ping anything from A, A from anything, but I can't ping D
from B or C, nor B and C from D). Both B, C, and D have A setup as their
gateway.

What do I need to do to get B and C to recognize D?

Jim Bradley -- Maryville, MO USA

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         31-Aug-99 14:08:08
  To: All                                               31-Aug-99 14:56:02
Subj: simple test for possible bug

From: ad354@FreeNet.Carleton.CA (James Owens)

I think I am up against a bug in the way OS/2 assigns an alias to a
network card (up to and including TCPIP 4.02w). The card can receive using the
alias, but cannot send using the alias. 

If someone could try this simple test, it would help me eliminate hardware
or setup problems. (If there is something obviously wrong with the setup
or my expectations, I would greatly appreciate more information about that.)

You need three hosts on an ethernet. 

- One has address a.b.c.1
- One has address a.b.d.1
   (where a, b, c, and d are at your convenience) -
 One is OS/2 machine to be tested, with TCP/IP on one LAN adapter. 

On the machine to be tested, type

route -c
ifconfig lan0 a.b.c.2 netmask 255.255.255.0
ifconfig lan0 a.b.d.2 netmask 255.255.255.0 alias

Both hosts are able to ping OS/2 machine, but OS/2 machine is able to
ping only c.1

Now type

route -c
ifconfig lan0 a.b.d.2 netmask 255.255.255.0
ifconfig lan0 a.b.c.2 netmask 255.255.255.0 alias

Again, both hosts are able to ping OS/2 machine, but OS/2 machine
is able to ping only d.1.

Can anyone confirm this?

--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: lsunley@mb.sympatico.ca                           31-Aug-99 14:40:25
  To: All                                               31-Aug-99 14:56:02
Subj: Re: Connectivity

From: lsunley@mb.sympatico.ca (Lorne Sunley)

On Tue, 31 Aug 1999 13:42:24, REMOVE_THISjbradley@msc.net wrote:

> I have 4 machines in my home network, as follows:
> 
> A- PPro 200 with both 10 Mbps and 100 Mbps ethernet. The 100 Mbps card is
> setup as 192.168.0.1 Netmask 255.255.255.0 and the 10 Mbps card is setup
> as 192.168.1.1 Netmask 255.255.0.0 . This machine currently has the modem
> (USR Sportster internal). IP forwarding is turned on.
> 
> B- PPro 200 with 100 Mbps ethernet. 192.168.0.3 Netmask 255.255.255.0
> 
> C- P133 with 100 Mbps ethernet. 192.168.0.5 Netmask 255.255.255.0
> 
> D- 486 with ISA bus with 10 Mbps ethernet. (If it had a PCI bus, or the
> ISA fast ethernet cards that I've found had OS/2 drivers, I'd change the
> card.)
> 
> I'd like to setup D as my internet server, using InJoy for IP network
> address translation with dial on demand (or linux with diald). The problem
> that I have, is that I can't get B or C to recognize D, nor D to recognize
> B or C (I can ping anything from A, A from anything, but I can't ping D
> from B or C, nor B and C from D). Both B, C, and D have A setup as their
> gateway.
> 
> What do I need to do to get B and C to recognize D?
> 

You didn't say what the IP address of D is. Given the
two IP addresses on A I assume that the IP address of
D is something like 192.168.1.2 or something like that.

What you have here is two separate subnets. 

The netmask settings on B and C preclude them seeing
the 192.168.1.X series of addresses

You can change the netmask settings to 255.255.0.0 on all
machines or you can change the IP addresses.

You should change the IP address to all be the same subnet
(192.168.0.X) This will give machine A two addresses
192.168.0.1 and 192.168.0.10 both with a netmask of 255.255.255.0

Machine D IP address of 192.168.0.10 netmask 255.255.255.0
and a router/gateway of 192.168.0.10 (10 mbps card address)

Machine B and C would have a router of 192.168.0.1 
(100 mbps card address)

When you set up machine D as the dial on demand server
the default route for the A, B, C  machines will have to point to 
192.168.0.10 (D address) to route IP addresses that are not
part of the 192.168.x.x subnet to the D machine so it can pass
them on to the external internet. The D machines default route
will be the ISP router/gateway value.

Lorne Sunley

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MBnet Networking Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: diego@cryogen.com                                 31-Aug-99 07:36:14
  To: All                                               31-Aug-99 14:56:02
Subj: Re: How do I Change IP Address info

From: "Justin Greywolf" <diego@cryogen.com>

James Knott wrote in message ...
>
>Curious.  I have changed the IP address hundreds of times and never
>had a problem.  What sort of problem do you get?  How is it confused?
>


Usually will just not know which IP to use.  If the Router Statement and
NetName are set up correctly, they will usually go out to the server, and
find out their correct IP, change and reboot.  They are fine then.  But
sometimes they will just get stuck, and not connect to the network.  Change
the IP in that file, reboot, and they are almost always fine

Justin Greywolf
----------------------------------------------------------------------------
--------------------------------------------------
Xcentric(dot)net - bringing communication to the new millenium
Email, Web site design, hosting, programming, network consulting
http://www.xcentric.net

----------------------------------------------------------------------------
--------------------------------------------------------------------------


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: U S WEST Interprise (1:109/42)

+----------------------------------------------------------------------------+

From: nospam@nospam.noway.com                           31-Aug-99 11:50:24
  To: All                                               31-Aug-99 14:56:02
Subj: NFS Client

From: "Roberto F. Salomon" <nospam@nospam.noway.com>

Is there a command line utility to mount NFS drives on the client?


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: nospam (1:109/42)

+----------------------------------------------------------------------------+

From: rmcculloch@rocketmail.com                         31-Aug-99 12:33:08
  To: All                                               31-Aug-99 20:08:13
Subj: Re: HELP!  Cannot get network logon screen!

From: "Randy McCulloch" <rmcculloch@rocketmail.com>

Microsoft gotcha, if you first tell your computer you don't want logon
screens by not giving a username and password and click OK, on
initialization then a registry entry will prevent you from getting a network
login after installing a card, protocol and client.  The Fix!  Run regedit,
search for autologon or autologin, I can't remember, when found delete the
key, close regedit reboot the computer.  Voila.  Should do the trick.



Jimmy Mac <jmacd@svn.net> wrote in message news:37AC3EC1.FDC43676@svn.net...
> Control panel, network properties, Client Services for MS, check the box
for
> "Log into NT Domain"
>
> J..
>
>
> fugaZi wrote:
>
> > Hello,
> >
> > I installed a 3com MHz PCMCIA network card and I'm having a big problem.
I
> > cant get a network logon screen.  I installed the card with no problems.
> > There are no hardware conflicts or anything like that.  I can ping other
PC
> > on the network, so that means TCP/IP is working.  I get a link light on
the
> > card and I can browes the Internet.  I'm primarly using a static IP
address
> > but tried hooking it up to a DHCP server based network.  So, I
reinstalled
> > the network configuration, still nothing.
> >
> > I have a......
> > Toshiba 510CDT
> > Windows 95b
> > 3com MHz 3CCFE574BT 3Com Megahertz 10/100 LAN PC Card
> >
> > Thanks,
> >
> > Corey
> > Work charbaugh@bma.com
> > Home charbaugh@kc.rr.com
> > ICQ# 44847658
>



--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: EarthLink Network, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         31-Aug-99 19:41:15
  To: All                                               31-Aug-99 20:08:13
Subj: OS/2 Requester -- for Warp Server only?

From: ad354@FreeNet.Carleton.CA (James Owens)

I'm running OS/2 Warp 4 (not Warp Server) and I want to map to some drives
on an NT server (TCPIP and ethernet). Do I need OS/2 requester? At the
moment I do not have OS/2 Requester installed. Is it included with the
Warp client, or available for the client? 

--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: rickstarich@home.com                              31-Aug-99 20:43:16
  To: All                                               31-Aug-99 20:08:13
Subj: Re: How do you connect NT wkst to Warp Server

From: Rick Starich <rickstarich@home.com>

Thanks for the reply

I am checking all components again. Noticed the readme.1st said to
uninstall logon client before applying a service pack. I need to
uninstall and reinstall the client logon and see if this fixes the
problem.

Also the following is from the readme.1st 

    IBM Networks Coordinated Logon Client for Windows NT         Page 1
of 5
   
    -----------------
    2.0 Prerequisites
    -----------------

         You must have the following prerequisite components installed
on
         your system before you can install IBM Networks Client:

           *   The Microsoft Workstation service
           *   An appropriate network adapter device driver
           *   A supported protocol (either NetBEUI or TCP/IP)

NetBEUI is different the netbios right? My NT workstation is not using
NetBEUI but does have NetBIOS in the Services tab of the Network
section. TCPIP is an installed protocol.

Are you using NetBEUI?

I keep fiddling with it, and let you know the outcome.

Thanks again. RickStarich@home.com

Leigh Engelhart wrote:
> 
> You need to run the Browser service on the server and have Netbios on both
> the workstation and the server.  TCP-IP shouldn't have anything to do with
> it.
> 
> Rick Starich wrote:
> 
> > Has anyone been sucessful using IBM's "Primary Logon Client for NT (June
> > 99 version)" to access resources on a Warp Server? My IBM Primary Logon
> > Client for NT tells me it can't see DOMAIN01 my Warp Server domain.
> >
> > I have a feeling it has something to do with my TCPIP settings on Warp
> > Server.
> >
> >
> > Have some suggestions?
> >
> > --
> > RickStarich@home.com
> > A Warp'ed mind is a terrible thing to waste. Get OS/2 Warp

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Micro Software Consultants Inc (1:109/42)

+----------------------------------------------------------------------------+

From: engel003@mc.duke.edu                              31-Aug-99 17:20:26
  To: All                                               31-Aug-99 21:19:21
Subj: Re: How do you connect NT wkst to Warp Server

From: Leigh Engelhart <engel003@mc.duke.edu>

I've forgotten the precise relationship between NetBios and NetBEUI--they're
closely related somehow--but NetBios is what you want on the workstation.
Double check that NetBios is also installed on the server and is bound to the
correct adapter, etc.

Re: the IBM Client having to be installed after applying an NT SP.  I've hit
that gotcha quite a few times.  Absolutely, whenever you apply a service pack,
first uninstall the Client by deleting it from the Services page of the
Network
configuration notebook.  Then re-install it after the service pack has been
applied.  Otherwise you may not be able to log in at all on reboot.  I've
actually had to re-install the operating system due to forgetting this.  Even
with an Emergency Repair Disk, to get back to the regular NT logon, you may
have
to overwrite so much of the Registry that you have to re-install much of your
software.

BTW, for performance sake, you may want to use regedit to create a DWORD
value,
~Winlogon|'SlowLinkDetectEnabled' (with value 0).  You can also change the
value
of ~LogoffPolicy|DeleteUser from 1 to 0.  Not deleting the profile speeds up
logon some.

To be able to see the Warp Server in the Network Neighborhood, you also need
to
be running the Browser service on the server.

Rick Starich wrote:

> Thanks for the reply
>
> I am checking all components again. Noticed the readme.1st said to
> uninstall logon client before applying a service pack. I need to
> uninstall and reinstall the client logon and see if this fixes the
> problem.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Duke University (1:109/42)

+----------------------------------------------------------------------------+

From: lsunley@mb.sympatico.ca                           31-Aug-99 21:54:05
  To: All                                               31-Aug-99 21:19:21
Subj: Re: OS/2 Requester -- for Warp Server only?

From: lsunley@mb.sympatico.ca (Lorne Sunley)

On Tue, 31 Aug 1999 19:41:30, ad354@FreeNet.Carleton.CA (James Owens) 
wrote:

> 
> I'm running OS/2 Warp 4 (not Warp Server) and I want to map to some drives
> on an NT server (TCPIP and ethernet). Do I need OS/2 requester? At the
> moment I do not have OS/2 Requester installed. Is it included with the
> Warp client, or available for the client? 
> 

It's included. The install procedures reference it as 
"File and Print Services Client". You can install it by executing the
"INSTALL.CMD" file in the root directory of the Warp 4 CD

You will also have to add OS/2 NETBIOS and/or NETBIOS
over TCP/IP in the MTPS setup in order to talk to the NT
server.

WARNING - this will fail if the length of the PATH in your PATH
statement exceeds 250 bytes or something like that. reduce it
before you do the network install and save yourself some 
headaches......

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MBnet Networking Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: iallen@fox.nospam.nstn.ca                         31-Aug-99 22:21:26
  To: All                                               01-Sep-99 10:43:22
Subj: Re: Connectivity

From: iallen@fox.nospam.nstn.ca (Iain Allen)

On Tue, 31 Aug 1999 14:40:50, lsunley@mb.sympatico.ca (Lorne Sunley) 
wrote:

> You didn't say what the IP address of D is. Given the
> two IP addresses on A I assume that the IP address of
> D is something like 192.168.1.2 or something like that.
> 
> What you have here is two separate subnets. 
> 
> The netmask settings on B and C preclude them seeing
> the 192.168.1.X series of addresses
> 
> You can change the netmask settings to 255.255.0.0 on all
> machines or you can change the IP addresses.
> 
> You should change the IP address to all be the same subnet
> (192.168.0.X) This will give machine A two addresses
> 192.168.0.1 and 192.168.0.10 both with a netmask of 255.255.255.0
> 
> Machine D IP address of 192.168.0.10 netmask 255.255.255.0
> and a router/gateway of 192.168.0.10 (10 mbps card address)

So now you have a card on A and a card on D both with 192.168.0.10  ? 
This is a mistake, right?  If I'm missing something, please 
explain...not meaning to criticize, just wanting to see if I can 
figure it out... :)

> 
> Machine B and C would have a router of 192.168.0.1 
> (100 mbps card address)
> 
> When you set up machine D as the dial on demand server
> the default route for the A, B, C  machines will have to point to 
> 192.168.0.10 (D address) to route IP addresses that are not
> part of the 192.168.x.x subnet to the D machine so it can pass
> them on to the external internet. The D machines default route
> will be the ISP router/gateway value.
> 
> Lorne Sunley

--
-----------------------------------------------------------
Iain allen <iallen@fox.nstnDOTca>
-----------------------------------------------------------


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: PSINet (CA-TORONTO-FEED) (1:109/42)

+----------------------------------------------------------------------------+

From: webster@lexmark.com@                              31-Aug-99 17:16:23
  To: All                                               01-Sep-99 10:43:23
Subj: NFS Server 'execute' bits

From: webster@lexmark.com@   (Gary R. Webster)

Hey, OS/2 people !!   Here's a good one for you.

I'm using the NFS server included with Aurora to do file sharing to some
Solaris & Linux clients.  It is working fine except for one thing.
It appears that the NFS server is 'guessing' at the NFS 'execute' bits
(user, group, & other) according to file extension.  This leaves some
extensions, like '.sh' un-executable by these clients, which is bad.

In the IBM 'book' "IBM TCP/IP v2.0 for OS/2: NFS Guide", I found the
following in section 2.4:
	"The user-execute, group-execute, and other-execute bits are set
	depending on the file's extension.  If the file ends with .exe, .EXE,
   .com, .COM, .cmd, .CMD, .out, .OUT, .nfs, or has no extension,
   the three execute bits are set."
So, this appears to be a problem in the NFS server itself, not an O/S
'association' issue.  This 'method' also appears to have been
carried-over to the Aurora NFS server (NFSD.exe).

Judging from the behavior of my Aurora NFS server, the NFS 'executable'
extensions have been 'updated' to include '.bat', & now omit '.nfs' .

Anywho, this seems like an awfully short-sighted mechanism, & I can't
believe that there isn't some way to configure this 'list' .   Does anyone
have an idea on this one?  I have read through all 'redbooks' & the like,
which I can find, that are even remotely related to OS/2 NFS.

To make matters worse, I have a Windows NT4 Server, using Windows Services
for Unix (SFU) NFS Server, which doesn't exhibit this problem.  It shows all
files as executable to the NFS clients.  This may not be correct,
but it offers the functionality that we need.

Thanks for any advice.


Gary R. Webster
webster@lexmark.com
Lexington, Ky.  USA


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Lexmark International, Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: ekadakal@aol.com                                  31-Aug-99 23:25:08
  To: All                                               01-Sep-99 10:43:23
Subj: -<ERROR>- unable to open transport

From: ekadakal@aol.com (EKadakal)

Hi Everyone:

I get the follwing error (in tracing) when I try to access an Oracle database
from a Warp 4 client machine. Any idea?

-<ERROR>- soc -1 error - operation=3, ntresnt[0]=530, ntresnt[1]=-3,
ntresnt[2]=0
-<ERROR>- nsres: id=0, op=65, ns=12560, ns2=0; nt[0]=530, nt[1]=-3, nt[2]=0
-<ERROR>- unable to open transport

Thanks

Ercan

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: AOL http://www.aol.com (1:109/42)

+----------------------------------------------------------------------------+

From: ekadakal@aol.com                                  31-Aug-99 23:31:12
  To: All                                               01-Sep-99 10:43:23
Subj: -<ERROR>- unable to open transport

From: ekadakal@aol.com (EKadakal)

Hi Everyone:

I get the follwing error (in tracing) when I try to access an Oracle database
from a Warp 4 client machine. Any idea?

-<ERROR>- soc -1 error - operation=3, ntresnt[0]=530,
ntresnt[1]=-3,ntresnt[2]=0
-<ERROR>- nsres: id=0, op=65, ns=12560, ns2=0; nt[0]=530, nt[1]=-3, nt[2]=0
-<ERROR>- unable to open transport

Thanks

Ercan

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: AOL http://www.aol.com (1:109/42)

+----------------------------------------------------------------------------+

From: nospamnclarke@bellatlantic.net                    01-Sep-99 00:22:12
  To: All                                               01-Sep-99 10:43:24
Subj: Re: Configuring BellAtlanticNet ADSL

From: "Nigel Clarke" <nospamnclarke@bellatlantic.net>

On Wed, 25 Aug 1999 23:37:44 -0400 (EDT), Philip R. Mann wrote:

:>On Thu, 26 Aug 1999 02:55:55 GMT, Nigel Clarke wrote:
:>
:>:>The address they give you is a class B and most TCP implementations
:>:>recognise it as such therefore they put in 255.255.0.0. 
:>
:>Evidently not Win 9x.

I meant to imply correctly implemented TCP/IP implementations when I said
most. Sorry 'bout that.

:>
:>What does this mean in real life?
:>
:>And thanks for the earlier private message; it pointed me in the right
direction.
:>
:>----- PRM -----

The original grouping gave out class A addresses to big companies using
the lower 24 bits to denote individual IP addresses. Class B's went to
smaller companies using the lower 16 bits for addresses and Class C 
went to those sites needing just 253 addresses (.0 and .255 being
reserved). To mask out the address bits you use the netmask to OR
(or is it AND :-) out the address to identify your subnet. For example
you can have subnets AAA.BBB.100.DDD through AAA.BBB.104.DDD
and the corresponding netmask would be 255.255.252.0 - check the bits
to see exactly how it works. Your machine knows that with the 255.255.252.0
mask that all addresses from 100 to 104 are in the same subnet.

So in this case Bell Atlantic's Class C has just over 65 thousand addresses
that they can hand out. You machine knows that all addresses from AAA.BBB.0.1
to AAA.BBB.255.254 are on the same subnet.

It doesn't actually matter much in this case unless you have it set wrongly.

Nigel


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: nospamnclarke@bellatlantic.net                    01-Sep-99 00:24:18
  To: All                                               01-Sep-99 10:43:24
Subj: Re: Configuring BellAtlanticNet ADSL

From: "Nigel Clarke" <nospamnclarke@bellatlantic.net>

On Wed, 25 Aug 1999 22:56:50 -0400 (EDT), Philip R. Mann wrote:

:>On Wed, 25 Aug 1999 16:28:07 -0400, Stephen Eickhoff wrote:
:>
:>:>Even though they say the netmask should be 255.255.255.0, I believe it
should be 
255.255.0.0. I don't know why..
:>
:>I received a message from Nigel Clarke who uses BellAtlantic.Net down South
and made this 
discovery by comparing with his setup.  Thanks to both of you.  This leads to
a question --- what 
does using subnet 255.255.0.0 rather than 255.255.255.0 affect (other than
letting my system 
work<g>?  Should I let my BA representative know so that they can do something 
on their end?
:>
:>>using a hostname as your DNS is a catch-22
:>
:>I'm not doing that.  I listed the results from the host command in case that 
would help in some 
way.
:>
:>As it now stands, I have the connection but it has a tendency to drop when
using PMINews or 
Netscape; I am able to get it back by using the Route command to delete one of 
the lines that 
somehow gets added --- another trick I learned from Nigel --- but it makes it
hard to let programs 
run and autodownload work.
:>
:>

Make a shadow of the setup.cmd file and keep it on your desktop. If I see that
the line has choked up due to the routing table getting lots of extra entries
I just
run it to clear the table back to the original. Works a treat - a fix from IBM 
would 
work better but we'll have to wait and see.

Nigel


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: lsunley@mb.sympatico.ca                           01-Sep-99 00:23:18
  To: All                                               01-Sep-99 10:43:24
Subj: Re: Connectivity

From: lsunley@mb.sympatico.ca (Lorne Sunley)

OOPS  <typo mode off>

Sorry about that the IP address of D should be something like
192.168.0.11

Each IP address assignment has to be unique.
Unfortunately my eyes are failing (obviously :-)

Lorne Sunley

On Tue, 31 Aug 1999 22:21:53, iallen@fox.nospam.nstn.ca (Iain Allen) 
wrote:

> On Tue, 31 Aug 1999 14:40:50, lsunley@mb.sympatico.ca (Lorne Sunley) 
> wrote:
> 
> > You didn't say what the IP address of D is. Given the
> > two IP addresses on A I assume that the IP address of
> > D is something like 192.168.1.2 or something like that.
> > 
> > What you have here is two separate subnets. 
> > 
> > The netmask settings on B and C preclude them seeing
> > the 192.168.1.X series of addresses
> > 
> > You can change the netmask settings to 255.255.0.0 on all
> > machines or you can change the IP addresses.
> > 
> > You should change the IP address to all be the same subnet
> > (192.168.0.X) This will give machine A two addresses
> > 192.168.0.1 and 192.168.0.10 both with a netmask of 255.255.255.0
> > 
> > Machine D IP address of 192.168.0.10 netmask 255.255.255.0

TYPO TYPO - S/B Machine D IP address 192.168.0.11

> > and a router/gateway of 192.168.0.10 (10 mbps card address)
> 
> So now you have a card on A and a card on D both with 192.168.0.10  ? 
> This is a mistake, right?  If I'm missing something, please 
> explain...not meaning to criticize, just wanting to see if I can 
> figure it out... :)
> 
> > 
> > Machine B and C would have a router of 192.168.0.1 
> > (100 mbps card address)
> > 
> > When you set up machine D as the dial on demand server
> > the default route for the A, B, C  machines will have to point to 
> > 192.168.0.10 (D address) to route IP addresses that are not

TYPO TYPO - S/B D address of 192.168.0.11

> > part of the 192.168.x.x subnet to the D machine so it can pass
> > them on to the external internet. The D machines default route
> > will be the ISP router/gateway value.
> > 
> > Lorne Sunley
> 
> --
> -----------------------------------------------------------
> Iain allen <iallen@fox.nstnDOTca>
> -----------------------------------------------------------
> 
> 


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: MBnet Networking Inc. (1:109/42)

+----------------------------------------------------------------------------+

From: Spammers@Bite.Me                                  01-Sep-99 00:24:06
  To: All                                               01-Sep-99 10:43:24
Subj: Apache 1.3.9 and cgi scripts

From: "Jaime A. Cruz, Jr." <Spammers@Bite.Me>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I did some experimentation and discovered the problem was related to using
Object REXX as my default environment.  When I use SWITCHRX to switch to
Classic REXX, my cgi scripts work fine.  I relayed this information to Brian
Havard and this was his response:

  Some CGI security was tightened in this version. It appears that Object
Rexx
  requires the DPATH environment variable to exist but it's no longer
included
  in a CGI's environment by default (it used to include everything which is
  considered insecure). To make it work just add

  PassEnv DPATH

  to your httpd.conf

I just thought I'd pass this information along.
Jaime A. Cruz, Jr.

o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o
o                                                 o
o  Visit the Nassau Wings Motorcycle Club at:     o
o  http://www.nassauwings.org/                    o
o  A Charter Member of the Motorcycle Web Ring!   o
o                                                 o
o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o_o&o



-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0 OS/2 for non-commercial use
Comment: PGP 5.0 for OS/2
Charset: cp850

wj8DBQE3zGQZgvzYfxgMc34RAjX0AJ9dwU2LFi+iapA8pJL7rVC8dSeIKwCg/x68
5sW7RlW8aCUteg67Ux4N/Y8=
=fItV
-----END PGP SIGNATURE-----

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Nassau Wings Motorcycle Club (1:109/42)

+----------------------------------------------------------------------------+

From: chall_1@my-deja.com                               01-Sep-99 00:24:06
  To: All                                               01-Sep-99 10:43:24
Subj: Re: PPP Over Ethernet

From: chall_1@my-deja.com

iVasion (WinPoETs designers ) have an experimental one.  see
www.iVasion.com

In article <7qai2e$k23$1@holly.prod.itd.earthlink.net>,
  bklein@earthlink.net wrote:
> I just got my DSL line installed via Earthlink and Pac-Bell.
Earthlink's
> connection requires WinPoET (PPP over ethernet) so at the moment I
have to
> use Win 9x. Does anyone know of a PPPOE driver for OS/2? TIA
>
>   Brian
>
>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Deja.com - Share what you know. Learn what you do
(1:109/42)

+----------------------------------------------------------------------------+

From: jhickey@gcfn.org                                  31-Aug-99 20:47:19
  To: All                                               01-Sep-99 10:43:24
Subj: Re: NFS Client

From: jhickey@gcfn.org (John F Hickey)

Roberto F. Salomon (nospam@nospam.noway.com) wrote:
: Is there a command line utility to mount NFS drives on the client?

IBM has an NFS Server/Client package for OS/2 TCP/IP v2.1 which works
pretty well for us on Warp4.  We use Woolongong on DOS and OmniNSF on
WIN98 and everything works out pretty well.  

All of that is separate purchase packages - not included with the OS.  

- John

-- 
                  John F. Hickey
jhickey@gcfn.org                   www.cyberpro.com/jhickey

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The Greater Columbus FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: iallen@fox.nospam.nstn.ca                         01-Sep-99 01:43:21
  To: All                                               01-Sep-99 10:43:24
Subj: Re: Connectivity

From: iallen@fox.nospam.nstn.ca (Iain Allen)

On Wed, 1 Sep 1999 00:23:36, lsunley@mb.sympatico.ca (Lorne Sunley) 
wrote:

> OOPS  <typo mode off>
> 
> Sorry about that the IP address of D should be something like
> 192.168.0.11
> 
> Each IP address assignment has to be unique.
> Unfortunately my eyes are failing (obviously :-)
> 
> Lorne Sunley
> 

Thanks for confirming that.  I had to squint at it fairly hard myself 
before I was convinced they were the same address... :)


> On Tue, 31 Aug 1999 22:21:53, iallen@fox.nospam.nstn.ca (Iain Allen) 
> wrote:
> 
> > On Tue, 31 Aug 1999 14:40:50, lsunley@mb.sympatico.ca (Lorne Sunley) 
> > wrote:
> > 
> > > You didn't say what the IP address of D is. Given the
> > > two IP addresses on A I assume that the IP address of
> > > D is something like 192.168.1.2 or something like that.
> > > 
> > > What you have here is two separate subnets. 
> > > 
> > > The netmask settings on B and C preclude them seeing
> > > the 192.168.1.X series of addresses
> > > 
> > > You can change the netmask settings to 255.255.0.0 on all
> > > machines or you can change the IP addresses.
> > > 
> > > You should change the IP address to all be the same subnet
> > > (192.168.0.X) This will give machine A two addresses
> > > 192.168.0.1 and 192.168.0.10 both with a netmask of 255.255.255.0
> > > 
> > > Machine D IP address of 192.168.0.10 netmask 255.255.255.0
> 
> TYPO TYPO - S/B Machine D IP address 192.168.0.11
> 
> > > and a router/gateway of 192.168.0.10 (10 mbps card address)
> > 
> > So now you have a card on A and a card on D both with 192.168.0.10  ? 
> > This is a mistake, right?  If I'm missing something, please 
> > explain...not meaning to criticize, just wanting to see if I can 
> > figure it out... :)
> > 
> > > 
> > > Machine B and C would have a router of 192.168.0.1 
> > > (100 mbps card address)
> > > 
> > > When you set up machine D as the dial on demand server
> > > the default route for the A, B, C  machines will have to point to 
> > > 192.168.0.10 (D address) to route IP addresses that are not
> 
> TYPO TYPO - S/B D address of 192.168.0.11
> 
> > > part of the 192.168.x.x subnet to the D machine so it can pass
> > > them on to the external internet. The D machines default route
> > > will be the ISP router/gateway value.
> > > 
> > > Lorne Sunley
> > 
> > --
> > -----------------------------------------------------------
> > Iain allen <iallen@fox.nstnDOTca>
> > -----------------------------------------------------------
> > 
> > 
> 
> 

--
-----------------------------------------------------------
Iain allen <iallen@fox.nstnDOTca>
-----------------------------------------------------------


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: PSINet (CA-TORONTO-FEED) (1:109/42)

+----------------------------------------------------------------------------+

From: sbogom@voicenet.com                               31-Aug-99 22:06:21
  To: All                                               01-Sep-99 10:43:24
Subj: TCP/IP perfomance tuning

From: Stuart Bogom <sbogom@voicenet.com>

Can anyone point me to performance tuning information for OS/2's TCP/IP?
I am running the 8610 update on a Warp Server box  whose primary
function is a Lotus Notes server. I never see more than 130 or so
sockets in use, although more than this number of Notes sessions are
active. Is there a way to tune this and other IP parameters?

Stuart Bogom

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Origin Line 1 Goes Here (1:109/42)

+----------------------------------------------------------------------------+

From: chadpaul@netexpress.net                           01-Sep-99 03:57:12
  To: All                                               01-Sep-99 14:27:05
Subj: two netcards update - no go

From: chadpaul@netexpress.net

Hello,

I have tried a few things. I have one 3com 3c905 card and a 10 base t linksys
pci.

I put the linksys in pci slot 1, and mpts reports it as adapter nic 1
the 3com is nic 0 in mpts.

under mpts, I edit the 3com card so that I tell its
MAC address and the pci slot it is in.  I left pci bus # blank.
I even tried it with the MAC address blank.

Two people wrote that they managed to have things going.
Can someone post the MPTS settings they used for their two cards.
What brand or brands of cards were used and the settings for each in MPTS?

I am drawing a few blanks.
Any more ideas or suggestions? 
I am willing to copy someone's configuration to make it happen.
Even buying the same cards and everything. 

Maybe in the long run I am better buying the server type net cards with dual
ports.
Intel and Adaptec make them. Pricy but might be a easier solution.

Any specific configuration file to look closely at?
If I need to add in a clip of a config file to the posting, what file or files 
should I get and put into the posting?


Any ideas most welcomed,

chadpaul@netexpress.net

Chad Pauli

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network (1:109/42)

+----------------------------------------------------------------------------+

From: chadpaul@netexpress.net                           01-Sep-99 04:03:10
  To: All                                               01-Sep-99 14:27:06
Subj: dual port server nic for os/2?

From: chadpaul@netexpress.net

Hello,

I am considering to buy maybe an Intel or Adaptec
dual port ethernet card.

I am wanting to use it to route between my LAN and the internet via TCI@home's
cable modem.

Has anyone used such network cards?
Any tried such cards under OS/2?

Maybe I am better off just buying a small ethernet to ethernet router for
about $500.00.

Any ideas or comments welcomed.


chadpaul@netexpress.net

Chad Pauli

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network (1:109/42)

+----------------------------------------------------------------------------+

From: domi@kenavo.NOSPAM.fi                             31-Aug-99 19:38:27
  To: All                                               01-Sep-99 14:27:06
Subj: Re: NFS Client

From: domi@kenavo.NOSPAM.fi (Dominique Pivard)

On Tue, 31 Aug 1999 14:50:48, "Roberto F. Salomon" 
<nospam@nospam.noway.com> wrote:

> Is there a command line utility to mount NFS drives on the client?

How about mount.exe?

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: None!! (1:109/42)

+----------------------------------------------------------------------------+

From: fordp@southwestern.edu                            01-Sep-99 01:12:21
  To: All                                               01-Sep-99 17:47:19
Subj: Re: DHCP Warp 3

From: Kacerdias <fordp@southwestern.edu>

>
>    I first installed an OS/2 LAN Requestor on OS/2 3.0 Red, and
> then I added MPTS 8421 to get the DHCP support.
>
> ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us/wr08421_conv
>
> baden

Ok, pardon my ignorance, but where can I get this "LAN requestor"? I got the
disk
images of 8421 and have those ready to go, I just need this other thing. I
guess all
this works with FreeTCP and IBM's TCP/IP stack, right? Gees. I've gotten DOS,
Windows
3-98, and Linux down pat. This is the final frontier for me. :)

Thanks for the help!

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Posted via Supernews, http://www.supernews.com (1:109/42)

+----------------------------------------------------------------------------+

From: kris@dgraph.com                                   01-Sep-99 00:55:02
  To: All                                               01-Sep-99 17:47:19
Subj: Re: two netcards update - no go

From: Kris Kadela <kris@dgraph.com>

I have a D-Link and a Linksys both working fine under Warp 4. I did not
do anything other than install the drivers, select the NICs in MPTS and
assign protocols to each one. Both are PCI NICs. I use similar setup at
home with a 3Com and a Linksys. 3Com is ISA and Linksys is PCI.

chadpaul@netexpress.net wrote:
> 
> Hello,
> 
> I have tried a few things. I have one 3com 3c905 card and a 10 base t
linksys pci.
> 
> I put the linksys in pci slot 1, and mpts reports it as adapter nic 1
> the 3com is nic 0 in mpts.
> 
> under mpts, I edit the 3com card so that I tell its
> MAC address and the pci slot it is in.  I left pci bus # blank.
> I even tried it with the MAC address blank.
> 
> Two people wrote that they managed to have things going.
> Can someone post the MPTS settings they used for their two cards.
> What brand or brands of cards were used and the settings for each in MPTS?
> 
> I am drawing a few blanks.
> Any more ideas or suggestions?
> I am willing to copy someone's configuration to make it happen.
> Even buying the same cards and everything.
> 
> Maybe in the long run I am better buying the server type net cards with dual 
ports.
> Intel and Adaptec make them. Pricy but might be a easier solution.
> 
> Any specific configuration file to look closely at?
> If I need to add in a clip of a config file to the posting, what file or
files should I get and put into the posting?
> 
> Any ideas most welcomed,
> 
> chadpaul@netexpress.net
> 
> Chad Pauli

-- 

**********************
DigiGraph Technical
http://www.dgraph.com
**********************

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: DigiGraph Technical (1:109/42)

+----------------------------------------------------------------------------+

From: msmanju@mindless.com                              01-Sep-99 14:52:00
  To: All                                               01-Sep-99 17:47:19
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: manju <msmanju@mindless.com>

Hi
  Can you give the netstat -m output ?

Manju

Sacha Prins wrote:

>  Dear people,
>
> I'm desperate. I'm experiencing TCPIP bugs for over 4 months now and
> no-one seems te be able to help me. I've purchased an IBM supportline
> contract to no avail (yet, after 4 months). The problems are as
> folows:
>
>    * MPTS8610 : I'm experiencing some kind of memory leak problem
>      where MBUFs allocated from free clusters (see 'netstat -m')
>      aren't freed anymore. This leads to an unusable system after so
>      many days (depending on the load on the machine). I run a Java
>      HTTPD in OS/2's JVM 1.1.7. I can't remember having these problems
>      with Lotus Go, but I did have similar problems with Xitami and
>      Apache. This problem is killing me softly since my servers stop
>      functioning properly every 3 to 4 days. This is not helping my
>      business at all.
>    * MPTS8620 : Same memory leak problem as 8610, but a new DNS
>      resolution problem is introduced here. Every now and then
>      applications will stop resolving host names (i.e. DNS queries
>      fail) until those programs are restarted. I've experienced the
>      problem with InetMail and Netscape so far. Netscape I don't care
>      about, but InetMail I do. When InetMail cannot query DNS anymore
>      all mail that is relayed through InetMail is queued indefinitely,
>      or until I restart InetMail.
>
> Please, please, pretty please with sugar on top. If anyone from IBM is
> reading this, or if anyone else that can help me is reading this I'll
> name my first born after you if you actually do help me.
>
> I'll scan these newsgroups every hour, but you can also mail me at
> sprins@businessnet.net
>
> Thanks in advance.
>
> Sacha Prins
>

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: XXXXXXXXXXX (1:109/42)

+----------------------------------------------------------------------------+

From: evsi@naverex.kiev.ua                              01-Sep-99 09:16:09
  To: All                                               01-Sep-99 17:47:19
Subj: Re: Configuring BellAtlanticNet ADSL

From: evsi@naverex.kiev.ua (Sergey I. Yevtushenko)

On Wed, 1 Sep 1999 00:24:36, "Nigel Clarke" 
<nospamnclarke@bellatlantic.net> wrote:

> :>:>Even though they say the netmask should be 255.255.255.0, I believe it
should be 
> 255.255.0.0. I don't know why..
> :>
> :>I received a message from Nigel Clarke who uses BellAtlantic.Net down
South and made this 
> discovery by comparing with his setup.  Thanks to both of you.  This leads
to a question --- what 
> does using subnet 255.255.0.0 rather than 255.255.255.0 affect (other than
letting my system 
> work<g>?

Try to run netstat -r and look at output. I guess default router is 
set to IP reachable only with
mask 255.255.0.0. Anyway, using such mask for large network simplifies
setup although introduces 
some drawbacks.

Regards,
	Sergey.

*--------------------------------------
ES@Home

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: ES@Nome (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                01-Sep-99 10:20:04
  To: All                                               01-Sep-99 17:47:19
Subj: Re: DHCP Warp 3

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <37CCC3DA.54801283@southwestern.edu>, Kacerdias <fordp@southwestern.edu>
writes:
>>
>>    I first installed an OS/2 LAN Requestor on OS/2 3.0 Red, and
>> then I added MPTS 8421 to get the DHCP support.
>>
>> ftp://ftp.software.ibm.com/ps/products/mpts/fixes/english-us/wr08421_conv
>>
>> baden
>
>Ok, pardon my ignorance, but where can I get this "LAN requestor"? I got the
disk
>images of 8421 and have those ready to go, I just need this other thing. I
guess all
>this works with FreeTCP and IBM's TCP/IP stack, right? Gees. I've gotten DOS, 
Windows
>3-98, and Linux down pat. This is the final frontier for me. :)

   I am not sure if you even initially need Lan Requestor
installed, or where to obtain it.  It comes packaged with Lan
and Warp Servers, and you might be able to find it somewhere if
8421 does not install by itself, which it may just do.

baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: msmanju@mindless.com                              01-Sep-99 17:57:05
  To: All                                               01-Sep-99 17:47:19
Subj: Re: Begging for help with either MPTS8610 or MPTS8620

From: manju <msmanju@mindless.com>

Also netstat -r output

manju wrote:

> Hi
>   Can you give the netstat -m output ?
>
> Manju
>
> Sacha Prins wrote:
>
> >  Dear people,
> >
> > I'm desperate. I'm experiencing TCPIP bugs for over 4 months now and
> > no-one seems te be able to help me. I've purchased an IBM supportline
> > contract to no avail (yet, after 4 months). The problems are as
> > folows:
> >
> >    * MPTS8610 : I'm experiencing some kind of memory leak problem
> >      where MBUFs allocated from free clusters (see 'netstat -m')
> >      aren't freed anymore. This leads to an unusable system after so
> >      many days (depending on the load on the machine). I run a Java
> >      HTTPD in OS/2's JVM 1.1.7. I can't remember having these problems
> >      with Lotus Go, but I did have similar problems with Xitami and
> >      Apache. This problem is killing me softly since my servers stop
> >      functioning properly every 3 to 4 days. This is not helping my
> >      business at all.
> >    * MPTS8620 : Same memory leak problem as 8610, but a new DNS
> >      resolution problem is introduced here. Every now and then
> >      applications will stop resolving host names (i.e. DNS queries
> >      fail) until those programs are restarted. I've experienced the
> >      problem with InetMail and Netscape so far. Netscape I don't care
> >      about, but InetMail I do. When InetMail cannot query DNS anymore
> >      all mail that is relayed through InetMail is queued indefinitely,
> >      or until I restart InetMail.
> >
> > Please, please, pretty please with sugar on top. If anyone from IBM is
> > reading this, or if anyone else that can help me is reading this I'll
> > name my first born after you if you actually do help me.
> >
> > I'll scan these newsgroups every hour, but you can also mail me at
> > sprins@businessnet.net
> >
> > Thanks in advance.
> >
> > Sacha Prins
> >

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: XXXXXXXXXXX (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   01-Sep-99 12:00:04
  To: All                                               01-Sep-99 17:47:20
Subj: Re: DHCP Warp 3

From: {R}@sunshine.tm (Richard Ashton)

In article <s57z3.10063$2k6.148762@news1.rdc1.bc.home.com>
           baden@unixg.ubc.ca "Baden Kudrenecky" writes:

>   I am not sure if you even initially need Lan Requestor
>installed, or where to obtain it.  It comes packaged with Lan
>and Warp Servers, and you might be able to find it somewhere if
>8421 does not install by itself, which it may just do.

8421 contains no lan requester, only MPTS fixes.

8423 is better then 8421.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: bauman@ibm.net                                    01-Sep-99 19:18:11
  To: All                                               01-Sep-99 19:58:09
Subj: Re: two netcards update - no go

From: bauman@ibm.net (Philip Bauman)

In message <Eu1z3.14351$ok3.4090@news.rdc1.il.home.com> -
chadpaul@netexpress.net writes:
:>
:>Hello,
:>
:>I have tried a few things. I have one 3com 3c905 card and a 10 base t
linksys pci.
:>
:>I put the linksys in pci slot 1, and mpts reports it as adapter nic 1
:>the 3com is nic 0 in mpts.
:>
:>under mpts, I edit the 3com card so that I tell its
:>MAC address and the pci slot it is in.  I left pci bus # blank.
:>I even tried it with the MAC address blank.
:>
:>Two people wrote that they managed to have things going.
:>Can someone post the MPTS settings they used for their two cards.
:>What brand or brands of cards were used and the settings for each in MPTS?
:>
:>I am drawing a few blanks.
:>Any more ideas or suggestions? 
:>I am willing to copy someone's configuration to make it happen.
:>Even buying the same cards and everything. 
:>
:>Maybe in the long run I am better buying the server type net cards with dual 
ports.
:>Intel and Adaptec make them. Pricy but might be a easier solution.
:>
:>Any specific configuration file to look closely at?
:>If I need to add in a clip of a config file to the posting, what file or
files should I get and put into the posting?
:>
:>
:>Any ideas most welcomed,
:>
:>chadpaul@netexpress.net
:>
:>Chad Pauli
I've still not resolved a similar problem on my laptop TP 390E.

Using two PCMCIA Cards (same make and model) MPTS has them configured as Lan0
and Lan1 but, Slot 2 will not power on when there is a card in Slot 1.  Slot
two alone is fine.  Both cards are set for different I/O's and different
IRQ's.  Trying to run a cable modem off of the lan1 card using DCHP.

This has been very frustrating.  What's more frustrating is it works in
WINbows!

HELP! Anyone?


Philip Bauman
bauman@ibm.net

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: morgannalefey@my-deja.com                         01-Sep-99 19:17:17
  To: All                                               01-Sep-99 19:58:09
Subj: Re: two netcards update - no go

From: morgannalefey@my-deja.com

In article <Eu1z3.14351$ok3.4090@news.rdc1.il.home.com>,
  chadpaul@netexpress.net wrote:

> I have tried a few things. I have one 3com 3c905 card and a 10 base t
linksys pci.
>
> I put the linksys in pci slot 1, and mpts reports it as adapter nic 1
> the 3com is nic 0 in mpts.

I have two 3c905s.  I put one in PCI slot 3 and one in PCI slot 4.  The
slot numbers that OS/2 sees, however, are *different*.  In my computer,
PCI slot 3 is noted in MPTS driver options as PCI slot uhm... 10, I
think.  I think it reports 11 for slot 4.  I don't know WHY it does
this, I only know that it did.  So I have to put the higher slot
numbers into MPTS.


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Deja.com - Share what you know. Learn what you do
(1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         01-Sep-99 20:44:28
  To: All                                               01-Sep-99 21:47:11
Subj: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

After updating MPTS and TCPIP, and then installing File and Client
Services (which required a work-around using PEERRMT.EXE from the
Installation CD, to accept the newer MPTS.EXE), I could connect to some NT
peers but not others -- specifically the server I want to map.  When I
tried I would get "SYS0234: More data is available".

From Dejanews I learned that this was supposedly fixed with IP8402. But I
have installed IP8402, and I still have the same problem.

NOTES: First I tried to install it using the CSF disk CSFTOOL1.DSK from
IP8402, but after inserting Disk 1 I got an error message about an
out-of-date component (I didn't note the exact wording, but I could
reproduce it if necessary.) Then I used the CSF disk FIXT139, which came
with the MPTS fix IP8424, and this seemed to work OK. The level is now
8402. 

Also during the installation the CONFIG.SYS PATH did not include
C:\IBMLAN\NETPROG or C:\MUGLIB, which had been added by the File and
Client Services installation. (I had hidden an older, longer CONFIG.SYS
and restored it after.)

The bottom line is that I still can't connect to the server, and the error
message is exactly the one IP8402 was supposed to correct.  Should I just
start over with the original TCPIP, MPTS, and File and Client from the Install
CD, and then add fixes in numerical order?
--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: jknott@ibm.net                                    01-Sep-99 05:47:15
  To: All                                               02-Sep-99 10:42:05
Subj: Re: How do I Change IP Address info

From: jknott@ibm.net (James Knott)

In article <5QRy3.271$yQ4.31078@news.uswest.net>,
"Justin Greywolf" <diego@cryogen.com> wrote:
>
>James Knott wrote in message ...
>>
>>Curious.  I have changed the IP address hundreds of times and never
>>had a problem.  What sort of problem do you get?  How is it confused?
>>
>
>
>Usually will just not know which IP to use.  If the Router Statement and
>NetName are set up correctly, they will usually go out to the server, and
>find out their correct IP, change and reboot.  They are fine then.  But
>sometimes they will just get stuck, and not connect to the network.  Change
>the IP in that file, reboot, and they are almost always fine

What happens if you remove that part of the file?


-- 
E-mail jknott@ca.ibm.com
_________________________________________________________________________
The above opinions are my own and not those of ISM Corp., a subsidiary of
IBM Canada Ltd.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: mjain@impetus.co.in                               02-Sep-99 15:06:24
  To: All                                               02-Sep-99 10:42:05
Subj: RPC on Solaris

From: "Manish Jain" <mjain@impetus.co.in>

Hi,
Does any one of you know of any website or new group having information on
Sun Solaris RPC .
Also, is the  source code of Sun RPC available at any ftp site ?

Manish


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: VSNL (1:109/42)

+----------------------------------------------------------------------------+

From: gerben@kabelnoord.nl                              02-Sep-99 13:42:03
  To: All                                               02-Sep-99 15:03:08
Subj: Somebody must know about this NT TCP/IP  issue? RIGHT???

From: "Gerben @ Work" <gerben@kabelnoord.nl>

Hello Experts,

I have a big problem.
I have a computer (workstation, win98) who is connnected in two different
ways.

Either it is connected:
1. to a NT network through TCP/IP.
or
2. stand alone to a cable modem (the other place I live)

Now every time I am at a different location I have to reconfigure my TCP/IP
settings for this one NIC!. Is there a way to switch TCP/IP settings at
startup. So I dont have to reconfigure my NIC manually every time?

Thanx in advanced,
(plz reply by email! :)

Digital Performer

Hello Experts,

I have a big problem.
I have a computer who is connnected in two different ways.

Either it is connected:
1. to a network through TCP/IP.
or
2. stand alone to a cable modem (the other place I live)

Now every time I am at a different location I have to reconfigure my TCP/IP
settings for this one NIC!. Is there a way to switch TCP/IP settings at
startup. So I dont have to reconfigure my NIC manually every time?

Thanx in advanced,
(plz reply by email! :)

Digital Performer




--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Atriserv (1:109/42)

+----------------------------------------------------------------------------+

From: ulvb@online.no                                    02-Sep-99 15:23:15
  To: All                                               02-Sep-99 15:03:09
Subj: Re: two netcards update - no go

From: ulvb@online.no (Ulv Bjerkan)

On Wed, 1 Sep 1999 03:57:24, chadpaul@netexpress.net wrote:

> Any ideas most welcomed,

You have do edit d: \ibmcom\protocol.ini manually with OADDRESS for 
each card:  With two 3com509B you get:

Protocol.ini:

EL3IBMO2_nif]

   DriverName = ELNK3$
   MaxTransmits = 20
   NetAddress = "onemacadress"
   IoAddress = 0x300

[EL3IBMO2_nif2]

   DriverName = ELNK32$
   MaxTransmits = 20
   NetAddress = "another macaddress"
   IoAddress = 0x210

--
Ulv Bjerkan 
http://home.sol.no/~ulvb/os2dl.html
 ftp://ftp.sol.no/user/ulvb
---------------------------------------

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Telenor Nextel (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   02-Sep-99 15:45:06
  To: All                                               02-Sep-99 15:03:09
Subj: Re: connect to NT server: IP08402 didn't work

From: {R}@sunshine.tm (Richard Ashton)

In article <37CE005B.AF2E4BB3@ibm.net> abstewa@ibm.net "Alon Stewart" writes:

>Of all the suggestions I have seen on this site as well as many web pages on
the>subject concerning hosts, resolv, protocol.ini, etc. files, the one common 
thing>that I have seen is: if the NT box has the NETBEUI protocol adapter
installed,
> you
>can see it from OS/2 if it does not you can't.  You should check that on the
NT
>box you can't see.

Done properly by someone that know what they are doing NT can
communicate with Warp Server or Warp 4 using:

1) NetBUEI/NetBIOS
2) TCPBUEI/NetBIOS
3) TCP/IP  (there is no native file system for TCP/IP in NT)

Any combination of these, is possible, for Warp Server you need.

XRnW040 (or better), WR08423 and IP08267.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         02-Sep-99 15:54:08
  To: All                                               02-Sep-99 16:41:13
Subj: Re: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

Alon Stewart (abstewa@ibm.net) writes:
> Of all the suggestions I have seen on this site as well as many web pages on 
the
> subject concerning hosts, resolv, protocol.ini, etc. files, the one common
thing
> that I have seen is: if the NT box has the NETBEUI protocol adapter
installed, you
> can see it from OS/2 if it does not you can't.  You should check that on the 
NT
> box you can't see.

It doesn't.  The administrator says he'll add it when he gets the chance. 
Thanks.
--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: mark.parent@sympatico.ca                          02-Sep-99 18:23:17
  To: All                                               03-Sep-99 03:37:15
Subj: Problem with ftpget (FTPAPI) in VisualAge C++ 

From: mark.parent@sympatico.ca (Mark Parent)

I'm having a consistent problem with ftpget, where I receive an access
violation each time I attempt to execute the routine.  I've even
hardcoded the function in my code, matching the syntax and example
shown in the documentation to no effect.  (I'm not having any problems
with other ftp API calls, such as ftprename or ftpappend, which I
assume means that I've compiled and linked correctly)

Here's my current (hardcoded) function call:

rc = ftpget ( "10.2.1.11", "PARENT", "MARK", NULL, "edp.xxx",
"edp.xxx", "w", T_BINARY );

I'm using the parameters in this order:  address, userid, password,
account, local filename, remote filename, mode, type

Does anyone have any similar experience with ftpget ??


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Sympatico (1:109/42)

+----------------------------------------------------------------------------+

From: baden@unixg.ubc.ca                                02-Sep-99 19:32:18
  To: All                                               03-Sep-99 06:09:27
Subj: Re: Somebody must know about this NT TCP/IP  issue? RIGHT???

From: baden@unixg.ubc.ca   (Baden Kudrenecky)

In <936272658.888298@morotai>, "Gerben @ Work" <gerben@kabelnoord.nl> writes:
>Hello Experts,
>
>I have a big problem.
>I have a computer (workstation, win98) who is connnected in two different
>ways.
>
>Either it is connected:
>1. to a NT network through TCP/IP.
>or
>2. stand alone to a cable modem (the other place I live)
>
>Now every time I am at a different location I have to reconfigure my TCP/IP
>settings for this one NIC!. Is there a way to switch TCP/IP settings at
>startup. So I dont have to reconfigure my NIC manually every time?

I have two batch files:

ipup.cmd
@ifconfig lan0 24.113.29.27 up

ipdown.cmd
@ifconfig lan0 24.113.29.27 down

   You can modify these for your situation.  Just 'ipdown' the
cable modem ip and 'ipup' the NT ip, et vice versa, or make an 
inclusive NT.cmd or cable.cmd.



baden

baden@unixg.ubc.ca
http://baden.nu/
OS/2, Solaris & Linux

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   02-Sep-99 20:13:01
  To: All                                               03-Sep-99 06:09:28
Subj: Re: connect to NT server: IP08402 didn't work

From: {R}@sunshine.tm (Richard Ashton)

In article <7qmgif$v0@freenet-news.carleton.ca>
           ad354@FreeNet.Carleton.CA "James Owens" writes:

>
>Richard Ashton ({R}@sunshine.tm) writes:
>
>> Done properly by someone that know what they are doing NT can
>> communicate with Warp Server or Warp 4 using:
>>
>> 1) NetBUEI/NetBIOS
>> 2) TCPBUEI/NetBIOS
>> 3) TCP/IP  (there is no native file system for TCP/IP in NT)
>>
>> Any combination of these, is possible, for Warp Server you need.
>>
>> XRnW040 (or better), WR08423 and IP08267.
>
>Thanks. I'm running Warp 4 (client) FP 6, WR08424, IP08402.  We do have
>TCP/IP communications (ping is ok and FTP gets a login prompt at least),
>but for File and Client Services I have to use NetBIOS over TCP/IP, and my
>guess is that NT has to use NetBUEI at the other end.

There are two ways for you to proceed.

1) as above needs NetBEUI (I always spell it wrong) installed on the
   target NT machine,

2) TCPBEUI installed on the WARP 4 box this is done by selecting
   NETBIOS OVER TCPIP in the MPTS configuration.

BOTH NEED additional work. You must have a userid and password defined
on the NT DOMAIN ( if ther is one) or on the NT Workstation if that is
the target.

You must have the WORKGROUP name or the NT Domain name in the Domain
part of IBMLAN.INI.

You must have the SAME user id and password defined to the LOCAL
UPMACCTS for which the usual starting place is userid of userid and
password of password. (it's in \MUGLIB)

Then you will be able to do

net use z: \\ntserver\ntresource, provided the ntresource is shared.
another password can be imposed by NT's share level security.

Welcome to the world of interoperability.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: bhk@dsl.co.uk                                     02-Sep-99 18:31:06
  To: All                                               03-Sep-99 06:09:28
Subj: Re: connect to NT server: IP08402 didn't work

From: bhk@dsl.co.uk (Brian {Hamilton Kelly})

On 2 Sep, in article <7qm6j9$q6t@freenet-news.carleton.ca>
     ad354@FreeNet.Carleton.CA "James Owens" wrote:

> Alon Stewart (abstewa@ibm.net) writes:
> > Of all the suggestions I have seen on this site as well as many web pages
on
>  the
> > subject concerning hosts, resolv, protocol.ini, etc. files, the one common
>  thing
> > that I have seen is: if the NT box has the NETBEUI protocol adapter
installed,
>  you
> > can see it from OS/2 if it does not you can't.  You should check that on
the
>  NT
> > box you can't see.
> 
> It doesn't.  The administrator says he'll add it when he gets the chance. 
> Thanks.

I wouldn't bother; you've been fed a red herring here, I suspect (because
Alon has misunderstood your problem).  Whilst NetBEUI is one possible
transport between NT and OS/2, it is NOT a REQUIREMENT.

An NT server system is perfectly visible on OS/2 even when only using
NetBIOS over TCP provided you're on the same sub-net.  If you're on a
different one, there are workarounds using the RFCHOSTS and RFCBCST
files: there have been recent threads here that mention these (wish I
could lay my hands on the URL of the site with the information on using
these: it's already helped me, and someone else to whom I passed on the
information).

With regard to your ACTUAL problem (which Alon seemed to miss/ignore) of
getting the "More data available" from the NT server.  I too suffer from
this problem, and after investigating at Deja, found the IP08402 fix that
you mention.  However, unlike yourself, I didn't try applying this as it
stood, because I knew my system had already had other fix packs applied
which would have meant (as you found) that the 8402 fix wouldn't be able
to be applied.

On the assumption that every fix pack for an OS/2 component (whether a
mainstream FP, or one of these "offshoot" fixes such as IP08402)
incorporates all APARs distributed in a previous fixpack, I applied (like
yourself) the latest IP084xx fix that I could find at Boulder; like
yourself, I found that it *DIDN'T* cure the "More data is available"
problem.

I suspect that when the later fixes in the IP084xx range were created,
IBM inadvertently omitted the particular APAR fix for the SYS3502
problem.  I had been intending to post a query just like yours (although
I would have hoped to have received more helpful replies than you seem to
have got to date).

Question: how can one determine just which fixes are in a particular pack
like this, and, more importantly, if it is indeed the case that the
particular fix has been "lost" in later packs, how does one go about
getting IBM to put it back in?

-- 
Brian {Hamilton Kelly}                                         bhk@dsl.co.uk
    "But we're a university.  We /have/ to have a library!..."said Ridcully,
         "What sort of people would we be if we didn't go into the library?"
    "Students", said the Senior Wrangler, morosely. [TP: The Last Continent]

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Dragonhill Systems Ltd (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         02-Sep-99 21:41:04
  To: All                                               03-Sep-99 06:09:28
Subj: Re: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

Richard Ashton ({R}@sunshine.tm) writes:

>>Thanks. I'm running Warp 4 (client) FP 6, WR08424, IP08402.  We do have
>>TCP/IP communications (ping is ok and FTP gets a login prompt at least),
>>but for File and Client Services I have to use NetBIOS over TCP/IP, and my
>>guess is that NT has to use NetBUEI at the other end.
> 
> There are two ways for you to proceed.
> 
> 1) as above needs NetBEUI (I always spell it wrong) installed on the
>    target NT machine,
> 
> 2) TCPBEUI installed on the WARP 4 box this is done by selecting
>    NETBIOS OVER TCPIP in the MPTS configuration.

Now I'm confused. All other things being equal, is it enough for me to
have NetBIOS over TCP/IP (and of course TCP/IP) drivers on my NIC, or is
it _also_ necessary for the NT server to have a NETBEUI driver to
interface with my NetBIOS?

Note that I have the userid, password, and domain.  I can perform a LAN
Server logo successfully using the icon in the OS/2 Peer group.  Once
logged on, I can see many other machines on the network, and I can access
the shared resources on them (well -- I can see drives, I haven't tried
printing.) The problem comes when I attempt to connect to the shared
resources of the LAN server (to map drives) -- then I get SYS0234: More
data is available. 

According to the APAR for this, the server is trying to send more than 4K
of something -- so I figure I'm connected.  However, the APAR fix has not
helped. I'm now waiting for the admin to add NETBOOEY (whatever) -- but if
there has been a misunderstanding and this will not be helpful, please let
me know.



 > 
> BOTH NEED additional work. You must have a userid and password defined
> on the NT DOMAIN ( if ther is one) or on the NT Workstation if that is
> the target.
> 
> You must have the WORKGROUP name or the NT Domain name in the Domain
> part of IBMLAN.INI.
> 
> You must have the SAME user id and password defined to the LOCAL
> UPMACCTS for which the usual starting place is userid of userid and
> password of password. (it's in \MUGLIB)
> 
> Then you will be able to do
> 
> net use z: \\ntserver\ntresource, provided the ntresource is shared.
> another password can be imposed by NT's share level security.
> 
> Welcome to the world of interoperability.
> 
> {R}
> 


--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: heloman@my-deja.com                               02-Sep-99 22:00:13
  To: All                                               03-Sep-99 06:09:28
Subj: RoadRunner and card setup?

From: heloman@my-deja.com

I am moving closer to getting RoadRunner service. I
purchased a Linksys 10/100 Lan Card. NOW that I have
it installed in the bus slot what next? The drivers
give several choices: OS/2 Lan, DOS Requester, etc. I
was thinking of the OS/2 Lan. Also what do I need to
do regarding OS/2? Since I am already leaving msgs
via the internet via my modem the TCPIP is already up
and running. Do I need to make adjustments to this
for the new land card? If not is there something that
MPTS needs to know? Since I have not subscribed as of
yet what numbers, netmask, io, address, etc. do I
need to enter in the tcpip/mpts setups? Is there a
very nice and easy step by step webpage that might
answer the bulk of these questions. Thanks in advance
for any answers given.  <<ah the simplicity of just
being a standalone desktop model>>


Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Deja.com - Share what you know. Learn what you do
(1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         02-Sep-99 21:55:26
  To: All                                               03-Sep-99 06:09:28
Subj: Re: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

Brian {Hamilton Kelly} (bhk@dsl.co.uk) writes:

>> It doesn't.  The administrator says he'll add it when he gets the chance. 
>> Thanks.
> 
> I wouldn't bother; you've been fed a red herring here, I suspect (because
> Alon has misunderstood your problem).  Whilst NetBEUI is one possible
> transport between NT and OS/2, it is NOT a REQUIREMENT.

The admin hasn't done anything yet, so it's now on hold while I figure
this out. . .

> An NT server system is perfectly visible on OS/2 even when only using
> NetBIOS over TCP provided you're on the same sub-net. 

Happily, I'm on the same subnet. I'd like to be on two subnets, which is a
different problem I can't get solved: when I define an alias for the same
NIC using ifconfig, all packets go out with the source address for the
first subnet.  As a result, machines only on the second subnet cant find
me to reply.  

But I digress.  Let's say I'm on one subnet, the same one as the NT
server. 

> With regard to your ACTUAL problem (which Alon seemed to miss/ignore) of
> getting the "More data available" from the NT server.  I too suffer from
> this problem, and after investigating at Deja, found the IP08402 fix that
> you mention.  However, unlike yourself, I didn't try applying this as it
> stood, because I knew my system had already had other fix packs applied
> which would have meant (as you found) that the 8402 fix wouldn't be able
> to be applied.

Well, I applied it, but using a CSF for a different fix (8424). Here also,
I'm confused.  The 8424 CSF was 1.39, which warned that 1.37 et al would
never work again after I used it.  However, the 8402 CSF identified itself
as 2B. And 1.37 for FP 6 came on 2 disks, while 1.39 came on one. This is not
the group for it, but if anyone can explain something about the CSF disk
in these fixpaks I'd be grateful.

> On the assumption that every fix pack for an OS/2 component (whether a
> mainstream FP, or one of these "offshoot" fixes such as IP08402)
> incorporates all APARs distributed in a previous fixpack, I applied (like
> yourself) the latest IP084xx fix that I could find at Boulder; like
> yourself, I found that it *DIDN'T* cure the "More data is available"
> problem.
> 
> I suspect that when the later fixes in the IP084xx range were created,
> IBM inadvertently omitted the particular APAR fix for the SYS3502
> problem.  I had been intending to post a query just like yours (although
> I would have hoped to have received more helpful replies than you seem to
> have got to date).

But I used 8402, superseded by 8401, which was the first one referenced
for the problem.  Surely it wasn't lost that fast?  Or do you mean it
might have been undone in advance, as it were, by the 8424 I had already
applied?

> Question: how can one determine just which fixes are in a particular pack
> like this, and, more importantly, if it is indeed the case that the
> particular fix has been "lost" in later packs, how does one go about
> getting IBM to put it back in?

My question is whether I can safely unwind this mess by removing MPTS, TCPIP,
and File and Client, and starting from the beginning -- or is that risky,
or futile?
--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: xrstad@azstarnet.com                              03-Sep-99 01:09:09
  To: All                                               03-Sep-99 06:09:28
Subj: Re: both static and dynamic on same adapter?

From: xrstad@azstarnet.com (Robert Stadler)

On Mon, 23 Aug 1999 15:36:49, ad354@FreeNet.Carleton.CA (James 
Owens) wrote:

This has worked for me: (from mptn\bin\setup.cmd)

route -fh
arp -f
ifconfig lan0 192.168.1.2 netmask 255.255.255.0
ipgate on
sleep 5

route -fh
arp -f
ifconfig lo 127.0.0.1
DHCPSTRT -i lan0
ifconfig lan0 192.168.1.2 netmask 255.255.255.0 alias
ipgate on
------------------------
Notes:
1:  192.168.x.x is one RFC-defined address space for private 
LAN's
2:  The loopback interface (127.0.0.1) was needed for StarOffice 
5.1
3.  ipgate is turned on as per InJoy Firewall documentation
4.  InJoy Firewall need only be used if your machine is to route 
traffic
    from one subnet to the other.
5.  This SETUP.CMD file gets creamed when TCPCFG is run.
6.  The DHCP-provided address uses a different netmask 
(255.255.252.0),
    but all seems to work anyway.
7.  It would appear that without the initial ifconfig something
    is not initialized properly somewhere.
   
Good Luck.


> 
> Currently (if I have this right) I am connected to two networks through a
> single LAN adapter. One uses IPX and the other uses TCP/IP. The TCP/IP
> connection is to a Unix server using a static address. The IPX connection
> is to a Novell Netware server.  (I can also connect to the internet via a
> dial-up adapter.)
> 
> Our sysadmin is phasing out the Novell server in favour of another one
> that uses dynamic IP addressing. Most of our site is now Windows-based, and
> the sysadmin has succeeded in connecting Windows machines dynamically to
> the new server, and statically to the UNIX system, all on the same
> physical network.
> 
> The sysadmin does not know how to set this up for OS/2, or even if it is
> possible, and is not interested in researching it. So, here I am. Is it
> possible, and if so, how?  (Simply setting up LAN interface 1 for DHCP in
> the TCP/IP configuration and booting didn't work.) 
> 
> I have OS/2 Warp 4, Fixpak 6. I'm not sure of my TCP/IP version, but AFAIK
> it has not been upgraded since installation. 
> 
> --
> 
>                                      James Owens  ad354@Freenet.carleton.ca
>                                      Ottawa, Ontario, Canada

--------------------------------------------------------
This late bulletin just in:  OS/2 is still dead.
--------------------------------------------------------
SpamGuard: Remove the initial "x" from my e-mail address

xrstad@azstarnet.com
Robert Stadler

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Posted via Supernews, http://www.supernews.com (1:109/42)

+----------------------------------------------------------------------------+

From: bhubleybhubley@home.com                           03-Sep-99 02:00:13
  To: All                                               03-Sep-99 06:09:28
Subj: Want to add your Cable Modem and OS/2 Warp link to my website?

From: "Brad Hubley" <bhubleybhubley@home.com>

If anyone who has created a Cable modem and OS/2 Warp website and would like
to have me place a link to their site from mine please email me with the URL.


Brad

____________________________________________________
  Windoze 95 @ work 'cause I have to,
  OS/2 Warp 4 @ home 'cause I want to!

  Visit my Cable Modems and OS/2 Warp 4 website @
  http://members.home.net/bhubley/cableintro.html
____________________________________________________

  Please remove 1 bhubley from my address before replying


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: @Home Network Canada (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         02-Sep-99 18:44:15
  To: All                                               03-Sep-99 06:09:29
Subj: Re: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

Richard Ashton ({R}@sunshine.tm) writes:

> Done properly by someone that know what they are doing NT can
> communicate with Warp Server or Warp 4 using:
> 
> 1) NetBUEI/NetBIOS
> 2) TCPBUEI/NetBIOS
> 3) TCP/IP  (there is no native file system for TCP/IP in NT)
> 
> Any combination of these, is possible, for Warp Server you need.
> 
> XRnW040 (or better), WR08423 and IP08267.

Thanks. I'm running Warp 4 (client) FP 6, WR08424, IP08402.  We do have
TCP/IP communications (ping is ok and FTP gets a login prompt at least),
but for File and Client Services I have to use NetBIOS over TCP/IP, and my
guess is that NT has to use NetBUEI at the other end. 

Admin still hasn't got around to installing it (has to take the system
down). 



--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: Baschti@Gmx.de                                    02-Sep-99 20:53:21
  To: All                                               03-Sep-99 06:09:29
Subj: Trap 000e after pinging in Dos-Window

From: "Sebastian Lotz" <Baschti@Gmx.de>

Hi,

we have a big problem with some of our Warp 4 - PC in our firm.

After I installed the DOS/WINDOWS support of TCPIP 4.00 from the original
Warp 4.0 CD, I reinstalled Fixpack 9 (german)
On some of our PC occurs now a trap 000e when I started command.com an
execute Ping with a valid adress. (We use Citrix Metaframe 16Bit-Client and
it crashes, too)
Some older Models seems to have not such a problem, everything is running
fine.

Some Ideas?
If not, where should I search for differences of the running and the failing
systems?

greetings from germany

Sebastian Lotz


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: T-Online (1:109/42)

+----------------------------------------------------------------------------+

From: abstewa@ibm.net                                   02-Sep-99 23:32:11
  To: All                                               03-Sep-99 10:34:22
Subj: Re: connect to NT server: IP08402 didn't work

From: Alon Stewart <abstewa@ibm.net>

Sorry folks, but I did not misunderstand the problem and I stand by my
statement
of NETBEUI on the NT boxes, it has to be installed.  I've had to install it
myself on several PC's I wanted to connect to.  For the "more data available"
problem, make sure that the share name on the NT box is C  not C$ and that
none
of the folders are shared.  You then can net use the \\ntbox\C or use the File 
&
Print Client to see all the NT's resources from OS/2.

James Owens wrote:

> Brian {Hamilton Kelly} (bhk@dsl.co.uk) writes:
>
> >> It doesn't.  The administrator says he'll add it when he gets the chance.
> >> Thanks.
> >
> > I wouldn't bother; you've been fed a red herring here, I suspect (because
> > Alon has misunderstood your problem).  Whilst NetBEUI is one possible
> > transport between NT and OS/2, it is NOT a REQUIREMENT.
>
> The admin hasn't done anything yet, so it's now on hold while I figure
> this out. . .
>
> > An NT server system is perfectly visible on OS/2 even when only using
> > NetBIOS over TCP provided you're on the same sub-net.
>
> Happily, I'm on the same subnet. I'd like to be on two subnets, which is a
> different problem I can't get solved: when I define an alias for the same
> NIC using ifconfig, all packets go out with the source address for the
> first subnet.  As a result, machines only on the second subnet cant find
> me to reply.
>
> But I digress.  Let's say I'm on one subnet, the same one as the NT
> server.
>
> > With regard to your ACTUAL problem (which Alon seemed to miss/ignore) of
> > getting the "More data available" from the NT server.  I too suffer from
> > this problem, and after investigating at Deja, found the IP08402 fix that
> > you mention.  However, unlike yourself, I didn't try applying this as it
> > stood, because I knew my system had already had other fix packs applied
> > which would have meant (as you found) that the 8402 fix wouldn't be able
> > to be applied.
>
> Well, I applied it, but using a CSF for a different fix (8424). Here also,
> I'm confused.  The 8424 CSF was 1.39, which warned that 1.37 et al would
> never work again after I used it.  However, the 8402 CSF identified itself
> as 2B. And 1.37 for FP 6 came on 2 disks, while 1.39 came on one. This is
not
> the group for it, but if anyone can explain something about the CSF disk
> in these fixpaks I'd be grateful.
>
> > On the assumption that every fix pack for an OS/2 component (whether a
> > mainstream FP, or one of these "offshoot" fixes such as IP08402)
> > incorporates all APARs distributed in a previous fixpack, I applied (like
> > yourself) the latest IP084xx fix that I could find at Boulder; like
> > yourself, I found that it *DIDN'T* cure the "More data is available"
> > problem.
> >
> > I suspect that when the later fixes in the IP084xx range were created,
> > IBM inadvertently omitted the particular APAR fix for the SYS3502
> > problem.  I had been intending to post a query just like yours (although
> > I would have hoped to have received more helpful replies than you seem to
> > have got to date).
>
> But I used 8402, superseded by 8401, which was the first one referenced
> for the problem.  Surely it wasn't lost that fast?  Or do you mean it
> might have been undone in advance, as it were, by the 8424 I had already
> applied?
>
> > Question: how can one determine just which fixes are in a particular pack
> > like this, and, more importantly, if it is indeed the case that the
> > particular fix has been "lost" in later packs, how does one go about
> > getting IBM to put it back in?
>
> My question is whether I can safely unwind this mess by removing MPTS,
TCPIP,
> and File and Client, and starting from the beginning -- or is that risky,
> or futile?
> --
>
>                                      James Owens  ad354@Freenet.carleton.ca
>                                      Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Global Network Services - Remote Access Mail & Ne
(1:109/42)

+----------------------------------------------------------------------------+

From: cannon@sonic.net                                  02-Sep-99 22:46:08
  To: All                                               03-Sep-99 11:24:27
Subj: Re: setting up connectivity between NT and OS/2

From: Andrew Cannon <cannon@sonic.net>

Thanks for the information.  It really helped.  Unfortunately, I have
two problems:

    1) NETBEUI does not load at boot up, i.e. SYS1201 occurs, which later
    causes NET2 and NetWksta.200 to fail to load.

    2) DNS does not work when I call out through the modem.

tomlins@cam.org wrote:

> In <A0ny3.6120$2k6.77753@news1.rdc1.bc.home.com>, baden@unixg.ubc.ca  
(Baden Kudrenecky) writes:
> >In <37C8DF53.1DACC59C@sonic.net>, Andrew Cannon <cannon@sonic.net> writes:
>
> Refer to these urls and read comments below:
>
> http://huizen.dds.nl/~jacco2/samba/warp.html
> Has the BEST explaination on how to configure OS2's peer support.
>
> http://www.mit.edu/afs/athena/activity/o/os2/www/peer/WARPPEER.HTM
> explains about OS2 vs NT/W95
>
> >>My employer has installed an ISDN line to my home for remote access to
> >>work.  This is the first time I have attempted to setup a network on
> >>OS/2(always had used a modem to ISP).
> >>
> >>There doesn't seem to be any instructions on setting this up to an NT
> >>network, and I have struggled with this for quite some time, so I
> >>would appreciate any advice.  Here is what I have done so far:
>
> >>3)MPTN -
> >>    LAPS - removed No Network Adapter configuration
> >>        added my LAN adapter
> >>        added protocols
> >>        0 - ... NETBIOS over TCP/IP
> >>        0 TCP/IP
> >
> >   I have not done this, however, it is my understanding that
> >you need to have both NETBIOS and TCP/IP on LAN0, and that
> >NETBIOS over TCP/IP has to be on a separate interface, i.e.
> >LAN1.
>
> This is NOT correct.  I have my machine setup exactly like his
> and it works correctly.
>
> >>during bootup
> >>ifconfig reports that the interface cannot be found (I believe
> >>lan0 should be the default interface)
>
> Do you see the nic driver load during boot?  Is it outputing any
> error messages?  If its a PNP card you may have to fiddle with the
> bios to ensure an iterupt is available.  To me it sounds like your
> nic card is not being seen correctly...  Have you got the latest
> drivers for it?
>
> Also it would seem that where the nic drivers are in the config.sys
> can make or break a system.  In my case I had to move the drivers to
> just after:
>
> DEVICE=D:\ibmcom\LANMSGDD.OS2 /I:D:\ibmcom /S
> DEVICE=D:\ibmcom\PROTMAN.OS2 /I:D:\ibmcom
> DEVICE=D:\IBMCOM\MACS\KTC100.OS2
> DEVICE=D:\IBMCOM\MACS\DLKFET.OS2
>
> If I leave them were MPTS puts them I get a hang, with PEER active, when
> ifconfig tries to set the ip address (or dhcpcd).  If they are at the end of 
the
> config.sys I get traps during boot.  I also had to reorder these statements:
>
> RUN=D:\ibmcom\PROTOCOL\NBTCP.EXE
> DEVICE=D:\IBMCOM\PROTOCOL\TCPBEUI.OS2
> DEVICE=D:\IBMCOM\PROTOCOL\NETBIOS.OS2
> DEVICE=D:\IBMLAN\NETPROG\RDRHELP.200
> IFS=D:\IBMLAN\NETPROG\NETWKSTA.200 /I:D:\IBMLAN /N
>
> One other piece of advice I got was to add a 'dummy' .bat file containing
> just 'exit' to my startup folder.  Apparently without this PEER can
sometimes
> hang the machine.  (I do really mean .bat it needs to start a dos session).
>
> >>TME 10 NetFinity Network Interface reports
> >>   Network Interface 'NETBIOS' atartup failed
> >>
> >>netstat -n reports no interfaces
> >>
> >>Seems like there is is fundamental problem here.  Any help or
> >>suggestions would be great, Thanks
> >>
> >>--
> >>Andy Cannon   http://www.sonic.net/~cannon
> >>
> >>
> >>
> >
> >
> >baden
> >
> >baden@unixg.ubc.ca
> >http://baden.nu/
> >OS/2, Solaris & Linux
> >



--
Andy Cannon   http://www.sonic.net/~cannon


--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Sonic,Santa Rosa CA,http://www.sonic.net (1:109/42)

+----------------------------------------------------------------------------+

From: uliw@erdw.ethz.ch                                 03-Sep-99 13:15:16
  To: All                                               03-Sep-99 17:08:13
Subj: Re: connect to NT server: IP08402 didn't work

From: Uli Wortmann <uliw@erdw.ethz.ch>

Alon Stewart <abstewa@ibm.net> writes:

> 
> Sorry folks, but I did not misunderstand the problem and I stand by my
statement
> of NETBEUI on the NT boxes, it has to be installed.  I've had to install it
> myself on several PC's I wanted to connect to.  For the "more data
available"
> problem, make sure that the share name on the NT box is C  not C$ and that
none
> of the folders are shared.  You then can net use the \\ntbox\C or use the
File &
> Print Client to see all the NT's resources from OS/2.

The more data is avail is caused by resource names (or comments) which
are longer than 8 characters. Try a 
	
	net view \\maschine_name

I'll bet youl'll see long names or names with spaces or special
characters.

BTW, have you searched the apar database that this problem was indeed
fixed. I consider this still a myth.

	Cheers

		Uli

-- 
	Uli Wortmann           Fax (Switzerland) (1) 632  1080
	Dept. of Geology       Fon                        3694
	ETH-Zuerich    http://www.erdw.ethz.ch/~bonk/bonk.html
	Visit the SPOC-team at http://www.spoc.ethz.ch

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Dept. of Geology, ETH-Zuerich (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   03-Sep-99 10:50:26
  To: All                                               03-Sep-99 17:08:13
Subj: Re: connect to NT server: IP08402 didn't work

From: {R}@sunshine.tm (Richard Ashton)

In article <7qmqtk$4sh@freenet-news.carleton.ca>
           ad354@FreeNet.Carleton.CA "James Owens" writes:
>Richard Ashton ({R}@sunshine.tm) writes:
>
>>>Thanks. I'm running Warp 4 (client) FP 6, WR08424, IP08402.  We do have
>>>TCP/IP communications (ping is ok and FTP gets a login prompt at least),
>>>but for File and Client Services I have to use NetBIOS over TCP/IP, and my
>>>guess is that NT has to use NetBUEI at the other end.
>>
>> There are two ways for you to proceed.
>>
>> 1) as above needs NetBEUI (I always spell it wrong) installed on the
>>    target NT machine,
>>
>> 2) TCPBEUI installed on the WARP 4 box this is done by selecting
>>    NETBIOS OVER TCPIP in the MPTS configuration.
>
>Now I'm confused. All other things being equal, is it enough for me to
>have NetBIOS over TCP/IP (and of course TCP/IP) drivers on my NIC, or is
>it _also_ necessary for the NT server to have a NETBEUI driver to
>interface with my NetBIOS?

It is not necessary for the NT server (or workstation) to have
the NetBUEI protocol installed in order to communicate with any box
that has NetBIOS over TCP/IP installed.

You do have to tell IBMLAN.INI about TCPBUEI

; OS/2 Peer initialization file
[networks]

  net1 = TCPBEUI$,0,LM10,100,150,14 ; This information ....

[requester]

  COMPUTERNAME = BOFH
  DOMAIN = DOMH     <---- this is the NT domain or workgroup name


>Note that I have the userid, password, and domain.  I can perform a LAN
>Server logo successfully using the icon in the OS/2 Peer group.  Once
>logged on, I can see many other machines on the network, and I can access
>the shared resources on them (well -- I can see drives, I haven't tried
>printing.) The problem comes when I attempt to connect to the shared
>resources of the LAN server (to map drives) -- then I get SYS0234: More
>data is available.
>
>According to the APAR for this, the server is trying to send more than 4K

64K it is a netbios remote call limit.

>of something -- so I figure I'm connected.  However, the APAR fix has not
>helped. I'm now waiting for the admin to add NETBOOEY (whatever) -- but if
>there has been a misunderstanding and this will not be helpful, please let
>me know.

Adding NetBEUI to the NT server is one solution other posts here have
others.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   03-Sep-99 11:07:24
  To: All                                               03-Sep-99 17:08:13
Subj: Re: connect to NT server: IP08402 didn't work

From: {R}@sunshine.tm (Richard Ashton)

In article <19990902.1731.28010snz@dsl.co.uk>
           bhk@dsl.co.uk "Brian {Hamilton Kelly}" writes:

>Question: how can one determine just which fixes are in a particular pack
>like this, and, more importantly, if it is indeed the case that the
>particular fix has been "lost" in later packs, how does one go about
>getting IBM to put it back in?

The readme on the first disk contains the APAR list.

If you need IBM help, Brian, Mail me.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: {R}@sunshine.tm                                   03-Sep-99 11:06:07
  To: All                                               03-Sep-99 17:08:13
Subj: Re: connect to NT server: IP08402 didn't work

From: {R}@sunshine.tm (Richard Ashton)

In article <7qmrp8$563@freenet-news.carleton.ca>
           ad354@FreeNet.Carleton.CA "James Owens" writes:

>Brian {Hamilton Kelly} (bhk@dsl.co.uk) writes:
>
>>> It doesn't.  The administrator says he'll add it when he gets the chance.
>>> Thanks.
>>
>> I wouldn't bother; you've been fed a red herring here, I suspect (because
>> Alon has misunderstood your problem).  Whilst NetBEUI is one possible
>> transport between NT and OS/2, it is NOT a REQUIREMENT.
>
>The admin hasn't done anything yet, so it's now on hold while I figure
>this out. . .
>
>> An NT server system is perfectly visible on OS/2 even when only using
>> NetBIOS over TCP provided you're on the same sub-net.

[...]

>
>But I digress.  Let's say I'm on one subnet, the same one as the NT
>server.
>
>> With regard to your ACTUAL problem (which Alon seemed to miss/ignore) of
>> getting the "More data available" from the NT server.  I too suffer from
>> this problem, and after investigating at Deja, found the IP08402 fix that
>> you mention.  However, unlike yourself, I didn't try applying this as it
>> stood, because I knew my system had already had other fix packs applied
>> which would have meant (as you found) that the 8402 fix wouldn't be able
>> to be applied.
>
>Well, I applied it, but using a CSF for a different fix (8424). Here also,
>I'm confused.  The 8424 CSF was 1.39, which warned that 1.37 et al would
>never work again after I used it.  However, the 8402 CSF identified itself
>as 2B. And 1.37 for FP 6 came on 2 disks, while 1.39 came on one. This is not
>the group for it, but if anyone can explain something about the CSF disk
>in these fixpaks I'd be grateful.

IBM keep messing with the CSF disks they are up to 141 now. You are
confused, as are many who havent done this before.

The IBM.PEER fixes necessary for WARP 4 are in order of appearance

8401
8402
8406
8407
8410
8412


Most of these are cululative and should include all previous fixes
You should be running with 8412 applied, I am and I have not seen this
problem.

The MPTS fixes are in order of appearance

8415 (I think, a long time back)
8420
8421
8423
8424

These are cumulative as above except 8424, I run with only 8423. 8423
is not applied with CSF just run MPTS off the first disk.

>But I used 8402, superseded by 8401, which was the first one referenced
>for the problem.  Surely it wasn't lost that fast?  Or do you mean it
>might have been undone in advance, as it were, by the 8424 I had already
>applied?

8402 superceded 8401 and now you should be on 8412, this fixes LAN
requester and Peer things and is nothing to do with MPTS.

8424 is an MPTS fix which need 8423 applied first.

>My question is whether I can safely unwind this mess by removing MPTS, TCPIP,
>and File and Client, and starting from the beginning -- or is that risky,
>or futile?

It can be dangerous :( if you have 8412 and 8423 there should not be a
problem, as someone else posted the NT resource must not be C$ (unless
you are an admin on the NT domain) the nt resource should be shared
with a name such as CDRIVE or whatever.

{R}

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: D4 (1:109/42)

+----------------------------------------------------------------------------+

From: bhk@dsl.co.uk                                     03-Sep-99 14:49:04
  To: All                                               03-Sep-99 17:08:13
Subj: Re: connect to NT server: IP08402 didn't work

From: bhk@dsl.co.uk (Brian {Hamilton Kelly})

On 03 Sep, in article <uli12hflcuoei.fsf@erdw.ethz.ch>
     uliw@erdw.ethz.ch "Uli Wortmann" wrote:

> The more data is avail is caused by resource names (or comments) which
> are longer than 8 characters. Try a 
>         
>         net view \\maschine_name
> 
> I'll bet youl'll see long names or names with spaces or special
> characters.

No; it's caused by the complete list of shared resources available being
longer then 4kB in toto.  I can use net view of certain servers (at work)
and see resources with names well in excess of eight characters, but I
still get the "more data available" after about twenty lines of such.
(perhaps the server shouldn't be proferring quite so many shares;-)

> BTW, have you searched the apar database that this problem was indeed
> fixed. I consider this still a myth.

It claims to have been fixed in one particular IP084xx fix-pack; as
suggested by me earlier, I *suspect* that the fix was somehow or other
dropped from later versions of that fix-pack.  (I base my concept of its
having been fixed upon Deja-archived postings from people who said that
applying the fix had cured the problem for them; I base my conception
that the fix was lost/dropped/stolen/strayed upon my empirical evidence
that applying a later FP did NOT cure the problem.  I have never tried
applying the one version of the FP that claimed to have fixed that
particular APAR.)

-- 
Brian {Hamilton Kelly}                                         bhk@dsl.co.uk
    "But we're a university.  We /have/ to have a library!..."said Ridcully,
         "What sort of people would we be if we didn't go into the library?"
    "Students", said the Senior Wrangler, morosely. [TP: The Last Continent]

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: Dragonhill Systems Ltd (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         03-Sep-99 13:58:29
  To: All                                               03-Sep-99 17:08:13
Subj: Re: both static and dynamic on same adapter?

From: ad354@FreeNet.Carleton.CA (James Owens)

Robert Stadler (xrstad@azstarnet.com) writes:
> On Mon, 23 Aug 1999 15:36:49, ad354@FreeNet.Carleton.CA (James 
> Owens) wrote:
> 
> This has worked for me: (from mptn\bin\setup.cmd)

> route -fh
> arp -f
> ifconfig lan0 192.168.1.2 netmask 255.255.255.0
> ipgate on
> sleep 5
> 
> route -fh
> arp -f
> ifconfig lo 127.0.0.1
> DHCPSTRT -i lan0
> ifconfig lan0 192.168.1.2 netmask 255.255.255.0 alias
> ipgate on

Thanks, I'll give this a try -- but I'm not optimistic.  I've found that
these commands are effective in real time from the command line, and by
now I must have tried the equivalent.  

My problem is not getting two addresses on the same card -- I can do that.
My problem is getting OS/2 to use the aliased address properly, by
incorporating it as the source address when sending to the secondary
subnet. (Am I right in thinking it's supposed to do that?)

Fortunately I have an indefinite lease on the DHCP network, so in practice
I can switch between networks from the command line by issuing an ifconfig
for whichever subnet I want to be on. (Sometimes I have to clear existing
routes first.) This workaround will do, since I don't really need to be on
both subnets simultaneously. 


--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

From: sprins@businessnet.net                            03-Sep-99 16:34:28
  To: All                                               03-Sep-99 17:08:13
Subj: INETCFG docs for 86xx?

From: Sacha Prins <sprins@businessnet.net>

Hi,

Does anybody know a location where the different INETCFG options are
discussed (in detail if possible)?

Much obliged,

Sacha

P.S. These are the params:

keepalive       :KeepAlive (sec)
tcpswinsize     :TCP SendWindow Size
tcprwinsize     :TCP RecvWindow Size
udpswinsize     :UDP SendWindow Size
udprwinsize     :UDP RecvWindow Size
lingertime      :Linger Time (sec)
probecnt        :Number Of KeepAlive Probes
ipforward       :IP Forwarding flag ON/OFF
mtudiscover     :TCP Path MTU Discovery ON/OFF
arpkillc        :ARP Complete TimeOut (sec)
arpkilli        :ARP InComplete TimeOut (sec)
lipcstswinsize  :LIPC Stream SendWindow Size
lipcstrwinsize  :LIPC Stream RecvWindow Size
lipcdgswinsize  :LIPC Dgram SendWindow Size
lipcdgrwinsize  :LIPC Dgram RecvWindow Size
synattack       :SYN Attack Flag ON/OFF
syncookie       :SYN Cookie Flag ON/OFF
firewall        :FIREWALL ON/OFF
multidefrt      :Multiple Default Routes ON/OFF
winscale        :TCP Window Scale ON/OFF
timestmp        :TCP TimeStamps ON/OFF
realslow        :TCP TW Q Slow Timeout ticks
perfhttp        :Fast Path HTTP ON/OFF
tcpttl          :TCP  TTL (hops)
udpttl          :UDP  TTL (hops)
icmpttl         :ICMP TTL (hops)
fragttl         :Fragment TTL (sec)
reusetw         :Reuse Time Wait ON/OFF
msl             :TCP MaxSegLifeTime (sec)
cc              :TTCP function ON/OFF

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: A2000 Kabeltelevisie en Telecommunicatie (1:109/42)

+----------------------------------------------------------------------------+

From: ad354@FreeNet.Carleton.CA                         03-Sep-99 15:35:03
  To: All                                               03-Sep-99 19:57:18
Subj: Re: connect to NT server: IP08402 didn't work

From: ad354@FreeNet.Carleton.CA (James Owens)

Brian {Hamilton Kelly} (bhk@dsl.co.uk) writes:
> On 03 Sep, in article <uli12hflcuoei.fsf@erdw.ethz.ch>

> It claims to have been fixed in one particular IP084xx fix-pack; as
> suggested by me earlier, I *suspect* that the fix was somehow or other
> dropped from later versions of that fix-pack.  (I base my concept of its
> having been fixed upon Deja-archived postings from people who said that
> applying the fix had cured the problem for them; I base my conception
> that the fix was lost/dropped/stolen/strayed upon my empirical evidence
> that applying a later FP did NOT cure the problem.  I have never tried
> applying the one version of the FP that claimed to have fixed that
> particular APAR.)
> 

Thanks to everyone who is helping me with this. A word of explanation may
be in order. 

I'm the last OS/2 client in the company.  The administrator, who was
originally hired to administer a Novell office network and help with an
OS/2 token ring product, has lost all patience with OS/2 -- "Good OS, but
no drivers and no support."  We are migrating the product to NT, and to
simplify his life he is also migrating the office network to NT. 

People in our company are encouraged to choose their own tools, management
being essentially libertarian, and I adopted OS/2 for my workstation long
ago while everyone else was waiting for Chicago.  Since then I have
invested in it, and for me to migrate to NT would involve considerable
effort (learning new apps and rejigging REXX calls, for example.) I do
have Win95 on another partition, and increasingly I am forced to use it --
for example, to run some Win32s 1.30 apps, or to work with the latest WP
and Word documents.

There is some quiet, practical pressure for me to abandon OS/2 whenever
something like this comes up.  For years I have been finding solutions and
demonstrating that, at the system level anyway, OS/2 is every bit as
capable as Win32 if not more so.  Luckily, with the help of people such as
you, I am a fast study.

Under these circumstances, getting the network changed on my behalf is out
of the question. It is enough to ask the admin to add a protocol driver
for NetBEUI -- and I gather and deeply hope that this alone can solve the
SYS0234 error.  If the NT resource is C$, offers shared folders, or has
other inconvenient properties, that's just tough for me -- and OS/2.

At Uli Wortmann's suggestion I ran NET VIEW.  Here is the output:

--
Shared resources at \\jsi-server

Netname      Type         Used as  Comment
-------------------------------------------------------------------------------

Add-ins      Disk                  "Access to EDK objects"
Address      Disk                  "Access to address objects"
D_Drive      Disk                  
E_Drive      Disk                  
FujitsuP     Print                 Fujitsu PrintPartner 10V
HP-1         Print                 HP 4000
HP5P         Print                 Hewlett Packard LaserJet 5P
mailbox      Disk                  
NETLOGON     Disk                  Logon server share 
Resources    Disk                  "Event logging files"
ROOT         Disk                  
SYS          Disk                  
tracking.log Disk                  "Exchange message tracking logs"
NET3513: More data is available than can be returned.
--

Sure enough, there are long comments - but it seems to accept them, as far
as it goes.  Unfortunately at the end there is that "more data" message.

The IBMLAN.INI file suggestions offered by Richard Ashton are already in
place (though I can only aspire to BOFH!).

There is an APAR for SYS0234:

--
APAR#: IC15963
. . .
ERROR DESCRIPTION

Shared Resources and Network Connections GUI displays sys0234 
on connections page when connected to a server that has more 
than 4K of data returned on the NetShareEnum call. 
Click on Shared Resources
Click on Connections page
Click on create connection
Click on a server that has numerous shares on it (>4K of data) 
Error pops up sys0234:More data is available
--

It occurs to me that SYS0234 might be returned for more than one reason,
and that this APAR fixes only one of them. (Another APAR fixes a false
SYS0234 from WRPRINT, for example.)  Perhaps there are other causes with
no fix yet. 

Meanwhile, I have advised the admin that the NetBEUI install is back on
after all, and I'll let you know what happens when he gets around to it.

Once again, thanks to all.

PS The Dejanews reports said 8402 worked -- but that's what I installed. 
Maybe you can't use CSF 1.39 with it though. I'm downloading 8412 as I
write -- I see it includes CSF 1.39.



--

                                     James Owens  ad354@Freenet.carleton.ca
                                     Ottawa, Ontario, Canada

--- WtrGate+ v0.93.p7 sn 165
 * Origin: Usenet: The National Capital FreeNet (1:109/42)

+----------------------------------------------------------------------------+

+============================================================================+
