Linux-Networking Digest #630, Volume #10 Thu, 25 Mar 99 16:13:36 EST
Contents:
Re: Sendmail Question ("R. Brooks")
Re: Diald question (Ronald Hovens)
Re: Can Linux browse NT? ("Lee Sharp")
Re: Samba and Win NT....I give up.....ack! (John McKee)
Re: Machine name themes - what do you use? ("- AJS")
Re: AOL and Linux (Malice)
LPD print jobs from AIX -> LINUX failing (Zing Zing Awungshi Shishak)
Re: ICQ Client and socks... (Matthew Mactyre)
SuSE 6.0 - PCMCIA Xircom CEM33 (Somf)
Re: sniffers ("Lee Sharp")
Re: pppd - demand dialing question (Kevin Martin)
Problems with IPX via PPP (Eric Rossing)
ftape with iomega ditto max pro extern ("ACE Alex")
----------------------------------------------------------------------------
From: "R. Brooks" <[EMAIL PROTECTED]>
Subject: Re: Sendmail Question
Date: Thu, 25 Mar 1999 09:28:41 -0500
You are right.
Send mail is trying to do a DNS look up and that made me
think it was locked up.
I used ntsysv to restart sendmail.
I rebooted and it worked fine.
then I did ps x and saw it was rejecting connections to port 25.
So I edited /etc/inetd.conf and uncommentd smtp.
I restarted sendmail and now ps x says:
it is running sendmail -bd -q1h
if I try to connect to port 25 it says it is connected
but I get rejected.
if I do a sendmail status it says
sendmail (pid 621) is running
I tried the connection again
and it says
sendmail dead but subsys locked
in my logs I see:
gateway inetd[602]: execv /usr/bin/smtpd no such file or directory.
I believe inetd should call the handling program but since
there is not smtpd things mess up.
I think my problem is in the inetd.conf file.
does anyone know how to fix it?
Thanks
Randall
Frank Hahn wrote:
> On Tue, 23 Mar 1999 15:57:17 -0500, R. Brooks <[EMAIL PROTECTED]> wrote:
> >When I first set up a computer with RH 5.1 I was
> >having problems with the Network card.
> >It seemed to lock up after sendmail was to start.
> >
> >so, I disabled sendmail and finally got the Network Card to work.
> >I have 2 questions.
> >1. Does anyone think sendmail/smtp will now work?
> >2. If I turn sendmail on and that is what locks it up, is there a way
> > to pass something to lilo to tell it not to load sendmail?
> >
> Sendmail is not causing your network card to lock up. When sendmail
> starts up, it is doing a DNS lookup. Once the lookup times out,
> your bootup most likely will continue.
>
> I start up sendmail with the following options:
>
> /usr/sbin/sendmail -bd -q
>
> I have a home network with diald running so I don't want diald
> bringing up the connection every 15 minutes or so.
>
> I know there are a couple of how-tos on setting up sendmail on
> Linux out there.
>
> I have a Slackware system. Redhat may be different.
>
> --
> Frank Hahn
--
_____________________________________________
Randall Brooks
[EMAIL PROTECTED]
http://www.glendinningprods.com
Senior Engineer
GMP (843) 399-6146 FAX (843) 399-5005
------------------------------
From: Ronald Hovens <[EMAIL PROTECTED]>
Subject: Re: Diald question
Date: Thu, 25 Mar 1999 14:44:39 +0100
Brian,
I read your question about diald and win'98 today. I don't have an answer to
your problem, butmaybe you can help me in setting up a similar infrastructure; I
myself tried to configure my linux server and win'98 clients for use with diald
but sofar I didn't succeed.
Can you send me some help, specifically I am interested in your following files:
/etc/ppp/options
diald.conf
Furthermore I am interested in the command you use to start the diald daemon
(which commandline options do you use). Many thanks in advance, your help will
save me a lot of time and headaches!
Ronald Hovens
Brian Smith wrote:
> I've got a couple of Win98 machines on ethernet, and a Linux box running IP
> masq and diald. I'm sure this is a pretty common set up, so maybe someone
> has come up with a good solution... is there a good way, from one of the
> windows machines, to tell diald "I'm done, disconnect *right now*" ?
>
> Telnetting to the linux box is not really an option - I need something I
> can make into a point-and-click icon on Windows for the wifey to be able to
> do it as well... Any good ideas? I did think of with replacing the internal
> modem with an external and just turning it off if I need the line
> disconnected, but that doesn't strike me as a very elegant solution.
>
> Thanks for any and all suggestions....
> Brian
------------------------------
From: "Lee Sharp" <[EMAIL PROTECTED]>
Subject: Re: Can Linux browse NT?
Date: Thu, 25 Mar 1999 13:02:08 -0600
Mike wrote in message <[EMAIL PROTECTED]>...
| I am fairly new to Linux.
Welcome!
|I am running it both at home & at work. My work computer is also on a
|network.
|So I have it setup dual boot with W95 & Linux2.2.1
|Can I browse the NT4.0 server at work?
Yes. Assuming you have NetBIOS over TCP/IP enabled on the NT Server, you
just need to run the Samba client. Samba comes with most distribution, but
the new version is out at www.samba.org and is much nicer. With Samba, you
will be able to join the domain, and brows, or be browsed by Windows
machines. If you will be interacting with NT 4.0 SP3 or later, or Win 98,
you will want to enable encrypted passwords in Samba.
|My sys admin says Im free to do so but only if I can figure it out as he
|has no idea how to do so.
Sounds like me. You may have to ask him to enable NetBIOS over TCP, but
that is easy.
|I have read many times how windows can hook up to a Linux server but can
|the reverse be done?
|Our server is netbui & tcp/ip
This is the only problem. If netbui is used for browsing, you are out of
luck. It is a nasty protocol, and a Windows default.
|I am able to ping all the machines at work so I know I have my eth0 card
|set up.
|Any help or a pointer to a URL that can help me would be appreciated.
|Thanks,
| Mike
Lee
--
SCSI is *NOT* magic. There are *fundamental technical reasons* why it is
necessary to sacrifice a young goat to your SCSI chain now and then. *
Black holes are where God divided by zero. - I am speaking as an individual,
not as a representative of any company, organization or other entity. I am
solely responsible for my words.
------------------------------
From: [EMAIL PROTECTED] (John McKee)
Subject: Re: Samba and Win NT....I give up.....ack!
Reply-To: [EMAIL PROTECTED]
Date: Thu, 25 Mar 1999 14:18:42 GMT
What did you use for the "os level" setting?
On Thu, 25 Mar 1999 05:53:48 GMT, [EMAIL PROTECTED] (vaclav vyvoda) wrote:
>I had a similar problem where the Samba server would not appear in
>Network Neighborhood. My problem was that I have set the "os level" to 0
>at some point in the past and forgot about it. You may want to check
>that particular settings.
>
>Good luck,
>
>Vas
>
>On JamesLay [EMAIL PROTECTED] wrote:
>: Ok......in a nutshell here is the deal:
>
>: I've got a NT PDC and a Libretto 70ct. I can see the \\Libretto\tmp
>: at home with the Win95 boxes and NT server there, but I just can't see
>: it here at work. I've added the domain name to the Libretto smb.conf
>: and I can ping everything on the network and the net in general. I've
>: added an account to the WinNT server here with the name of libretto.
>: And, I've even added the network to the hosts file. I also set the
>: securtiy option in smb.conf to server and to share. I still cannot
>: see it in Network Neighborhood...any help?
>
>: JamesLay
John McKee
[EMAIL PROTECTED]
------------------------------
From: "- AJS" <[EMAIL PROTECTED]>
Crossposted-To:
vmsnet.networks.misc,microsoft.public.windowsnt.domain,comp.unix.solaris,comp.os.os2.networking.server,comp.os.ms-windows.nt.admin.networking,comp.infosystems.www.servers.unix,comp.protocols.tcp-ip.domains
Subject: Re: Machine name themes - what do you use?
Date: Thu, 25 Mar 1999 12:03:06 -0800
> But I like that GIS company's solution of using a country
>name and then naming mount points for citys within the
>country! Talk about tracking down a problems being a cinch!
>
> Maybe I'll name my machines after Baseball teams and the
>mount point names can be well known players from those teams.
>Now, I wonder if I can rename 'root' to manager names from
>those teams?
Well, you could always name your machines after trees and leave "Root" alone
;^).
Personally, I like mythology... stick with a period/culture and do a little
research as the network grows... (right now I need the name of an
English/Celtic witch or 'being' associated with illusions - for a Video
Presentation server).
- AJS
ps. I like the cities idea too.
------------------------------
From: Malice <[EMAIL PROTECTED]>
Subject: Re: AOL and Linux
Date: Thu, 25 Mar 1999 12:21:13 -0700
Unfortuanatly it can not be done, aol installs an exe file on you HDD to
check mail via a web browser, it is not supported under linux nor NT.
messed up huh?
Malice
P.S. actually the file it installs is in conjunction with a browser plug
in that will only work with 95 or 98.
------------------------------
From: [EMAIL PROTECTED] (Zing Zing Awungshi Shishak)
Crossposted-To: comp.unix.aix
Subject: LPD print jobs from AIX -> LINUX failing
Date: 25 Mar 1999 15:36:58 -0500
Hello all,
I have a problem LPD printing from AIX (4.3.2.0) to a remote printer on a
Linux box(red hat 5.0).
If I send two print jobs (both postscript btw) from AIX in quick succession
to the remote queue (setup raw w/ no filter) on the linux box only the
first job prints and the other disappears.
An lpq -P<rp>, on the linux box showed that only the first job was received,
but I then used a filter to redirect the job to a file and the file
contained both jobs (interestingly, cat - > /dev/lp0 in the filter did not
work... only the first job came through, so much for a quick fix.)
Well, I have a feeling either LINUX or AIX has incorrectly supported the
lpd protocol and was wondering if anyone has heard of this or if there is
a fix somewhere.
TTTTTTIA
zing
--
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Matthew Mactyre)
Subject: Re: ICQ Client and socks...
Reply-To: [EMAIL PROTECTED]
Date: Thu, 25 Mar 1999 14:43:36 GMT
On Wed, 24 Mar 1999 20:13:06 -0500, "Curt" <[EMAIL PROTECTED]> wrote:
>ICQ will work thourgh a socks proxy server if you use a socks wrapper like
>sockscap32
>or another one from hummingbird. Take a look at www.socks.nec.com .
Thanks for the advice, but I didn't see a Sockscap client for Linux on
www.socks.nec.com. I would still like to be able to compile with the
--enable-socks5 switch turned on. There are lot of clients that I'm
going to need socks support with...
I have made some progress; however, I'm still attempting to compile
Licq, and have gotten a step closer. I downloaded socks5-
1.0r9-0.src.rpm from
http://rufus.w3.org/linux/RPM/contrib/libc6/i386/socks5-
s5watch-devel-1.0r9-0.i386.html and installed it and was able to
compile much farther then I have before, but it still fails.
However, I'm still having trouble with the make, during the complile I
get the follow set of errors, see below, from the socket.cpp. I am a
complete Newbie, but if someone would be kind enough to point me in
the right direction to get this problem solved I would appreciate it.
Thanks,
Matthew
socket.o: In function `INetSocket::SetAddrsFromSocket(unsigned
short)':
/root/licq-0.61/src/socket.cpp:155: undefined reference to
`SOCKSgethostbyname'
/root/licq-0.61/src/socket.cpp:165: undefined reference to
`SOCKSgetpeername'
socket.o: In function `INetSocket::GetIpByName(char *)':
/root/licq-0.61/src/socket.cpp:187: undefined reference to
`SOCKSgethostbyname'
socket.o: In function `INetSocket::OpenConnection(void)':
/root/licq-0.61/src/socket.cpp:216: undefined reference to
`SOCKSconnect'
socket.o: In function `INetSocket::StartServer(unsigned int)':
/root/licq-0.61/src/socket.cpp:245: undefined reference to `SOCKSbind'
/root/licq-0.61/src/socket.cpp:250: undefined reference to
`SOCKSlisten'
socket.o: In function `INetSocket::CloseConnection(void)':
/root/licq-0.61/src/socket.cpp:269: undefined reference to
`SOCKSclose'
socket.o: In function `INetSocket::SendRaw(CBuffer &)':
/root/licq-0.61/src/socket.cpp:284: undefined reference to `SOCKSsend'
socket.o: In function `INetSocket::RecvRaw(void)':
/root/licq-0.61/src/socket.cpp:307: undefined reference to `SOCKSrecv'
socket.o: In function `TCPSocket::RecvConnection(TCPSocket &)':
/root/licq-0.61/src/socket.cpp:335: undefined reference to
`SOCKSaccept'
socket.o: In function `TCPSocket::SendPacket(CBuffer &)':
/root/licq-0.61/src/socket.cpp:371: undefined reference to `SOCKSsend'
/root/licq-0.61/src/socket.cpp:386: undefined reference to `SOCKSsend'
socket.o: In function `TCPSocket::RecvPacket(void)':
/root/licq-0.61/src/socket.cpp:433: undefined reference to `SOCKSrecv'
/root/licq-0.61/src/socket.cpp:463: undefined reference to `SOCKSrecv'
main.o: In function `main':
/root/licq-0.61/src/main.cpp:22: undefined reference to `SOCKSinit'
collect2: ld returned 1 exit status
make[1]: *** [licq] Error 1
make[1]: Leaving directory `/root/licq-0.61/src'
make: *** [all_recursive] Error 2
------------------------------
From: Somf <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.hardware
Subject: SuSE 6.0 - PCMCIA Xircom CEM33
Date: 19 Mar 1999 08:00:24 GMT
Can anyone help. I know there are others with the same problem:
The System see's the card just fine. Got Link, Got traffic light
blinking, but can't get system to talk to the network. I can ping the
localhost, but nothing on the network, not even the default router. Network
configs look okay to me!
Using: SuSE 6.0
System: Toshiba Tecra 8000
Ethernet Card: PCMCIA - Xircom CEM33
Per SuSE Documentation, I have:
Disabled the Network configs for the Ethernet in YAST,
Created a Scheme and used lilo to setup the network (it appears to be fine).
in rc.config set: PCMCIA_PCIC_OPTS="do_scan=0 irq_mask=0xefff"
ALL HELP IS GREATLY APPRECIATED IN ADVANCED!
Here are my current configurations. Maybe someone see's something I don't:
# cardctl ident
cardctl.txt
Socket 0:
product info: "Xircom", "CreditCard Ethernet+Modem 33.6", "CEM33", "1.00"
manfid: 0x0105, 0x110d
function: 2 (serial)
Socket 1:
no product info available
# ifconfig eth0
eth0 Link encap:Ethernet HWaddr 00:80:C7:57:F8:4C
inet addr:139.103.100.188 Bcast:139.103.100.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0
Interrupt:3 Base address:0x2d0
# netstat -rn
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt Iface
139.103.100.0 0.0.0.0 255.255.255.0 U 1500 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 3584 0 0 lo
0.0.0.0 139.103.100.254 0.0.0.0 UG 1500 0 0 eth0
FROM dmesg:
....
xirc2ps_cs.c 1.31 1998/12/09 19:32:55 (dd9jn+kvh)
eth0: Xircom: port 0x2d0, irq 3, hwaddr 00:80:C7:57:F8:4C
ttyS03 at 0x02e8 (irq = 3) is a 16450
eth0: media 10BaseT, silicon revision 1
....
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "Lee Sharp" <[EMAIL PROTECTED]>
Subject: Re: sniffers
Date: Thu, 25 Mar 1999 14:45:50 -0600
[EMAIL PROTECTED] wrote in message
<7dcm0i$nlp$[EMAIL PROTECTED]>...
|Are there any good sniffers for linux>>> any that are GUI's
I have not actually run these yet, but here is what I have found so
far...
http://ethereal.zing.org/ Designed to be a clone of Network Associates
Sniffer Pro. It is a beta...
http://www.et-inf.fho-emden.de/~tobias/epan/ It doesn't look like anyone's
but it is GUI. Binaries and packages for Deban only...
http://mojo.calyx.net/~btx/karpski.html Another GUI Sniffer.
http://www-serra.unipi.it/~ntop/ This is not really GUI based, or a sniffer,
but it is a nice overview, and can be browsed HTML.
Lee
--
SCSI is *NOT* magic. There are *fundamental technical reasons* why it is
necessary to sacrifice a young goat to your SCSI chain now and then. *
Black holes are where God divided by zero. - I am speaking as an individual,
not as a representative of any company, organization or other entity. I am
solely responsible for my words.
------------------------------
From: [EMAIL PROTECTED] (Kevin Martin)
Subject: Re: pppd - demand dialing question
Date: Thu, 25 Mar 1999 15:21:16 GMT
In article <[EMAIL PROTECTED]>, it says [EMAIL PROTECTED]
(Bob) wrote:
>I would greatly appreciate some gentle nudging as to how to invoke pppd
>to demand dial. Been around the man page, etc but have not found any
>pseudo HOW-TOs or web pages to help explain this part [...].
I could never diald get it to compile, let alone run. So I made pppd
demand-dialing work, with much help from this very group, and documented the
process for future former newbies. See
http://www.brasscannon.com/Linux/
It's a "Hands-on How-to" on two subjects that take up a lot of space here:
demand-dialing with pppd, and IP Masquerade.
As for your immediate question, once you add the "demand" parameter to your
pppd options file, you invoke it like so:
% pppd :10.0.0.2
The IP address on the right of the colon is a dummy address that is NOT
LOCAL to your LAN. (I'm assuming your ISP assigns you a dynamic IP address.
If you have a static IP address, use that instead.) You can put that into
/etc/rc.d/rc.local as the last line, if you want it available all the time.
Or make it an alias in your .profile and run it when you want to enable
pppd:
alias ppp-start="pppd :10.0.0.2"
Calling it "ppp-start" makes it a nice partner to the ppp-stop script from
the distribution, which I use to force a hangup (or disable ppp if I'm
expecting a fax on my modem line).
When configured this way, pppd won't dial out right away; it'll just sit
there until someone on your Linux box (or your LAN, if Linux is configured
as a gateway to do IP Masq, the next logical step) does something non-local,
like "ping my.isp.com" -- at which point it will happily "reach out, reach
out and touch someone." Set it up to drop after x minutes of idle time and
you've got a nice intelligent gateway that doesn't tie up your phone all the
time.
------------------------------
From: [EMAIL PROTECTED] (Eric Rossing)
Subject: Problems with IPX via PPP
Reply-To: [EMAIL PROTECTED]
Date: Thu, 25 Mar 1999 15:22:34 GMT
I've been trying to configure my Linux machine as an IPX/PPP server.
According to the IPX HOWTO (I have version 2.3, 6-May-1998), if I'm running
the ipxd daemon, an IPX route to my ppp0 device should be automatically
added when I establish the connection.
My problem is that this not happen.
After making my PPP connection from my Win95 client, I am able to telnet
into the Linux box, so I can see that the ppp0 device is NOT added as an IPX
interface. Furthermore, after I manually add it, if I try to find my Novell
server from the Win95 client (routing IPX through the Linux server), the
Linux computer completely freezes up. My telnet session stops responding,
and I'm not able to do anything (even log in) at the Linux server's console.
My only way out of this is to cold-restart the Linux server.
Any ideas on how I can get around this problem would be greatly appreciated!
Eric Rossing
[EMAIL PROTECTED]
------------------------------
From: "ACE Alex" <[EMAIL PROTECTED]>
Subject: ftape with iomega ditto max pro extern
Date: Thu, 25 Mar 1999 21:55:03 +0100
Hi, has anybody got a iomega ditto max pro extern backup drive to work in
linux? Ftape should have support for it but when i try to use it my disc
drive makes some noise! How do i configure that i have an extern drive that
uses the lpt port and not a intern one using the disk controller?
please msg me
/alex
[EMAIL PROTECTED]
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.networking) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Networking Digest
******************************
