Linux-Networking Digest #101, Volume #10 Wed, 3 Feb 99 18:13:49 EST
Contents:
Re: How do I determine ppp dialup connect speed? (Ed Jones)
Mac to networked Linux print problems (Steve)
Re: UDP fwding in 2.0.36 for battlenet ([EMAIL PROTECTED])
Free X-Window Server & File System viewer for NT ([EMAIL PROTECTED])
Re: Help with simple manual network configuration (Robert Montgomery)
2.2.1 alpha/i386 teql and fsck (trojan or just buggy) ([EMAIL PROTECTED])
Re: Remote Access with root by telnet? (Thomas Zajic)
FTP .... H O W !!!! ("Morten Andersen")
POP/IMAP setup (Sean McEwan)
Re: Mac <--> Linux Networking questions (Bryan Scott)
NFS woes (Jay Thorne)
Re: Masquerading and Netmeeting (Bryan Scott)
Re: Romote "root" login (Paul Anderson)
Re: rpm question (Villy Kruse)
Pacific CommWare serial board (Keith Tucker)
Re: DHCP client (dhcpcd) for Caldera 1.3? (Ronny Bremer)
linux firewalling--HELP!! (Guido Dolci)
----------------------------------------------------------------------------
From: Ed Jones <[EMAIL PROTECTED]>
Subject: Re: How do I determine ppp dialup connect speed?
Date: Wed, 03 Feb 1999 06:14:11 +0000
Tom Gordon wrote:
>
> Hi,
>
> On windows when you connect using a ppp dialup account, it tells you the
> connect speed for the session. How can I determine that on Linux? I
> know about pppstats, but it doesn't tell you that info, only the current
> throughput.
>
> Thanks,
> Tom Gordon
Good question, I'm not sure what the best approach is, I just look at
the log file /var/log/messages . I have debug turned on in the ppp
options file with logs the ppp connect replies from the host. Maybe
there is a better way.
Ed
------------------------------
From: Steve <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.help
Subject: Mac to networked Linux print problems
Date: Tue, 02 Feb 1999 01:44:41 -0600
I'm a Linux newbie who's trying to setup a Linux print/file server
with a Mac but am slowly losing hair in trying to get the Linux printer
to respond. Still going through the various HOWTO's but could use some
help. :P
My setup:
Mac 8.5.1 / PowerCenter Pro
Debian GNU / Linux 2.1 on a Pentium 133
Lexmark Optra E+ Printer (with Postscript via Ghostscript)
I have the AppleTalk file sharing working with Netatalk, but I can't
get the printer to respond. The printer is recognized in the Mac
Chooser, but everytime that I try to print from my Mac, some data is
transmitted, and then I get a "Can't print because of a Postscript
error."
Not sure where I'm going wrong, and I was hoping that somebody could
steer me in the right direction.
My process so far....
1) Used MagicFilter to set my print filters (currently using defaults
of LaserJet4)
2) Downloaded Mac PPD files from Adobe's site for the HP printers and
placed them in both the Mac and Linux
3) Modified my netatalk papd.conf file to read
LaserPrinter:\
:op=me: pr=lp: pd=/usr/lib/ppd/ljet4ml.ppd:\
4) Selected LaserJet4ml on my Mac Laserwriter PPD config.
Tried futzing around with a few different .ppd configs, but I want to
be sure that I'm doing everything else right before going on the
printer driver goose chase. Also briefly tried using the Mac Desktop
Printer Utility to print from LPR instead of AppleTalk, but I just get
an error message there too.
Any help would be appreciated,
Steve
--
=====================================
To reply, remove SPMSX from the e-mail address ($%!^*! spammers...)
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: UDP fwding in 2.0.36 for battlenet
Date: Wed, 03 Feb 1999 08:43:38 GMT
In article <[EMAIL PROTECTED]>,
Mark <[EMAIL PROTECTED]> wrote:
> Is there any way to get the udp packets forwarded properly with
> this version?
>
I haven't tried it myself, but the command you need to use is ipautofw.
Specifically:
ipautofw -A -r tcp 6112 6112 -h <IP address of local machine>
ipautofw -A -r udp 6112 6112 -h <ditto>
If this doesn't work, do a search on DejaNews.
Mike
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED]
Subject: Free X-Window Server & File System viewer for NT
Date: Wed, 03 Feb 1999 09:19:09 GMT
Hi! Can anybody help me in finding: a) a free X-Window Graphic Server for
Windows NT (not XWin32, it's shareware) b) a File system viewer of my Linux
Server for a NT client, so I can locate the linux machine from Windows NT and
see my file system
Thank you very much for any help or reference !!!
Ludwig
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Robert Montgomery <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.install,comp.os.linux.setup,comp.os.linux.redhat
Subject: Re: Help with simple manual network configuration
Date: Wed, 03 Feb 1999 21:14:34 GMT
When I do a netstat -nr I get this:
Kernel IP routing table
Destination Gateway Genmask Flags MSS Window irtt
Iface
24.65.228.0 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 24.65.228.1 0.0.0.0 UG 1500 0 0
eth0
Also, I noticed that I can ping my own IP, but not the gateway or either
nameservers, so I'm not even getting out at all... Should I see the
nameservers listed with a netstat -nr? If so how do I add them?
Another thing I noticed is that when my system attempts to do a network
reload (/etc/rc.d/rc5.d/s10network), I get an error saying "netmask
doesnt match route address". Does this make any sense?
I'll take anymore ideas you've got!
Thanks
Thanks...
Rob
> Robert, I'm not an expert in these matters, and everything looks
> familiar in your posting, and identical to mine expect for one
> detail.. your name servers are on a different network.. 24.64.0.0 from
> your gateway/ip 24.65.0.0 .
>
> You probably checked your routing using the command "netstat -nr" or
> "route" and you know that you should see something like
>
> 0.0.0.0 24.65.228.1 0.0.0.0 UG ... etc.
>
> at the bottom of that display.. is there any routing to your nameservers
> in this display? Of course an "arp" command should show the ethernet
> address for your gateway, yourself and any other ethernet computers on
> your subnet.
>
> Just a thought.. Ed
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: fa.linux.kernel
Subject: 2.2.1 alpha/i386 teql and fsck (trojan or just buggy)
Date: Wed, 03 Feb 1999 16:30:50 GMT
I am attempting to "bond" to 2 NICs together to form a 200Mbps link between
two linux boxes using teql. Bad things have happened!!
I downloaded iprouter2 from an ftp site listed in a post from Eric Kluft on
1/14/99 to fa.linux.net.
I took a brief look before I compiled, but I admit not fully checking it.
It died (make) and took ld out with it. (any action at the terminal resulted
in a non-alpha character being printed and ld saying it could not load
anything).
Upon rebooting, fsck(in the intel) seg faulted (see below).
Feb 2 22:12:48 PPRO kernel: swap_free: Trying to free nonexistent swap-page
Feb 2 22:12:48 PPRO last message repeated 3 times Feb 2 22:12:48 PPRO
kernel: Unable to handle kernel paging request at virtual address 002d0039
Feb 2 22:12:48 PPRO kernel: current->tss.cr3 = 00101000, `r3 = 00101000 Feb
2 22:12:48 PPRO kernel: *pde = 00000000 Feb 2 22:12:48 PPRO kernel: Oops:
0000 Feb 2 22:12:48 PPRO kernel: CPU: 0 Feb 2 22:12:48 PPRO kernel: EIP:
0010:[<c011d4c0>] Feb 2 22:12:48 PPRO kernel: EFLAGS: 00010216 Feb 2
22:12:48 PPRO kernel: eax: caf2875c ebx: 002d002d ecx: caf28fe0 edx:
caf2881c Feb 2 22:12:48 PPRO kernel: esi: 002d002d edi: c0088c60 ebp:
00000000 esp: cb28df7c Feb 2 22:12:48 PPRO kernel: ds: 0018 es: 0018 ss:
0018 Feb 2 22:12:48 PPRO kernel: Process netscape (pid: 700, process nr: 30,
stackpage=cb28d000) Feb 2 22:12:48 PPRO kernel: Stack: fffffffc 00000000
002d002d c01138a7 c0088c60 c0088c60 c0088c60 c01182d2 Feb 2 22:12:48 PPRO
kernel: c0088c60 cb28c000 ffffffff fffffffc 0000000 4ffffffff fffffffc
00000000 Feb 2 22:12:48 PPRO kernel: Call Trace: [<c01138a7>] [<c01182d2>]
[<c0118466>] [<c0108c00>] [<c010002b>] Feb 2 22:12:48 PPRO kernel: Code: 8b
4b 0c 89 4c 24 10 8b 73 04 8b 6b 08 29 f5 8b 43 28 85 c0
********************************** I could not boot this machine as this
happened at boot each time. I moved this disk to my alpha and ran fsck
successfuly, fixing numerous errors.
Then I rebooted the intel machine (one with the now fixed drive). It worked
for a little while and then hung. When it went down, it took the alpha with
it.
At various points throughout the evening I got seg faults of various types
involving interupt handlers as well as the above (those were lost to the log
files so are not reproduced here)
No message on the alpha, in fact I could restart xdm with ctrl-alt-bkspc, but
it would never get back into the window manager after the log in.
I rebooted the intel machine and it again seg faulted on fsck. (in the mean
time the alpha came back up on a reboot).
2 more iterations of fsck seg faulting, moving the disk to the alpha and
running fsck and the intel machine appears to be running, but the disk is
very corrupted as my dmesg shows that it is trying to do some very strange
things on boot.
I see kernel messages with large portions of my file-system (as seen in an
ls) included in a non-sense, but non-fatal error. I will assume that the
fscks on a currupted disk have left me with a disk that boots and runs, but
will fail soon.
Now the Alpha has joined in and my milo/lilo appear to have been corrupted as
I get PYXIS (164Lx chipset) PAL errors on boot that kill the kernel shortly
after it unzips.
SO,... The question is - Is teql really this buggy, or have I downloaded a
trojan which has been remarkably effective at trashing my hard drives
(seemingly accross a network as well - although I did move the "infected"
disk to the alpha so I guess the net is out of it)? Of course, I always
acknowledge full blame for such things in trying to do strange things.
I have found precious little documentation on teql, could someone point me to
any?
This is where I got my fs killer version of iprouter (use at your own risk!!
and then let me know if I have some other problem and this worked fine for you
:))
ftp://ftp.inr.ac.ru/ip-routing/iproute2-current.tar.gz
Also, I am a little confused as to the relationship between equal cost
multipath and teql (kernel compile options).
Are these mutually exclusive methods to do the same thing, or different levels
of hardware/software "channel-bonding"?
Am I taking the hard route here and I could just have these 4 NICS between two
computers with only 2 IPs between them and equal cost multipath routing would
take care of it?
Thanks,
/Mike
(or in deference to my limping linux mini-cluster -> "/!i^.@" ["/Mike" as it
would have appeared if typed on my trashed xterm after running make on this
software])
SYSTEM info:
Alpha 164LX 533 RH 5.1 2.2.1 (no patches)
egcs 1.1.1
2 NICS (Tulip)
P-PRO 200 196MB RAM RH 5.2 2.2.1 (no patches)
egcs 1.1.1
4 NICS (2 eepro100, 1 tulip, 1 eexpress16)
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Thomas Zajic <[EMAIL PROTECTED]>
Subject: Re: Remote Access with root by telnet?
Date: Wed, 03 Feb 1999 21:15:09 GMT
Robert Barnette wrote:
> Look at the "securetty" file in the /etc/ directory and either remove or
> comment out the lines that list the secure TTY locations. Those are listed
> as places where root can't telnet in from. When you remove them, root can
> come in from anywhere.
Huh? I think you�re mixing this up with /etc/ftpusers (which indeed
lists all users that are _not_ allowed to ftp in).
If /etc/securetty does not exist, everybody may telnet in as root. If
it does exist, root may _only_ login on the ttys that _are_ listed.
Thomas
--
=---------------------------------------------------------------------=
- Thomas Zajic aka ZlatkO ThE GoDFatheR, Vienna/Austria -
- Spam-proof e-mail: thomas(DOT)zajic(AT)teleweb(DOT)at -
=---------------------------------------------------------------------=
------------------------------
From: "Morten Andersen" <[EMAIL PROTECTED]>
Subject: FTP .... H O W !!!!
Date: Wed, 3 Feb 1999 20:56:07 +0100
I need to disable my FTP server witch is build in Redhat 5.1....
I'm trying to set up an FTP-forwarder..... it says that the address is
allready used....
Thanx.. .:o)
------------------------------
From: Sean McEwan <[EMAIL PROTECTED]>
Subject: POP/IMAP setup
Date: Wed, 03 Feb 1999 16:09:34 -0600
I got PPP access to my linux box, now I am trying POP/IMAP services, and
they half-work. I can retrieve mail using Netscape or Eudora, but cannot
send mail. From Eudora, I get 'Connection Refused (10061)'. From
Netscape, I get connection refused. I don't have DNS working, so I am
trying to mail to [EMAIL PROTECTED] Will someone help me please? My
mail is ==>[EMAIL PROTECTED] Thanks much.
------------------------------
Date: Tue, 02 Feb 1999 17:53:22 -0700
From: Bryan Scott <[EMAIL PROTECTED]>
Subject: Re: Mac <--> Linux Networking questions
I got it working on three Macintoshes and two windows boxes and two
Linux boxes..
The Macintoshes need to have TCP/IP manually set. Get MacTCP for System
6/7, and set up:
* Manual
* Class C address
* 192.168.2.<blah> will work for in-home networks
* Gateway IP is the IP address of your linux box. This has to be the
same first three numbers (192.168.2.x) and the netmask should be 255.255.255.0.
For MacOS 8:
Connect via: Ethernet
Configure: Manually
IP Address: 192.168.2.something
Subnet Mask: 255.255.255.0
Router Address: your.linux.box's.ip (same first three numbers)
Name Server: your.linux.box.ip (make sure you have named working)
Additional Search domains:
domain.of.your.linux.box/home.network
I've set up all my linux boxes to be 192.168.2.1 and then any of the
Macintoshes to be 192.168.2.2 to 192.168.2.255 (anything between 2 and
255 is valid).
netatalk is available as well for doing file sharing and printer
sharing. Really nice package. I've had these three Macs up with the
linux and Windows boxes for two years now, at home, and have another
hybrid network here at work.
-- Bryan Scott
-- CTR Online Systems Administration
"Thomas S. Martinson" wrote:
>
> I have been working on this same problem. What I have tried so for is to
> use differnt software on the MAC's that support "tcp/ip". I have used the
> software "Dave" and it works ok. It makes your mac's work like Windows
> Machines. I am also going to try "ShareWay IP" from www.opendoor.com. It
> is supose to give IP support for the Mac's.
>
> togtog wrote:
>
> > Hi,
> >
> > I just asked this question over in comp.os.linux.questions before I saw
> > this group so I'm reposting this here.
> >
> > I've asked this question for what seems like a year now in many places,
> > mostly IRC, without any luck. I hope you fine people can lend a hand.
> > :-)
> >
> > I'm trying to network a Macintosh and my Linux (RedHat) box via TCP/IP
> > without any luck. For the past year I have had them connected via
> > Appletalk (netatalk) through my ethernet card without problem. I just
> > can't seem to get the TCP/IP working. Neither computer will see the
> > other.
> >
> > How am I supposed to do this?
> >
> > I figure my cables, hub, and hardware are all fine because Appletalk
> > works.
> > I'll be honest, I don't know a whole lot about routing and TCP/IP. I'm
> > sure its just some stupid little number set wrong someplace, or
> > something I forgot to install. I don't really want to go out any buy a
> > $50 Networking book just to find out how to do this one task.
> >
> > Thank you for any help, have a good day!
> >
> > -Paul
> >
> > PS- Software stats:
> > Linux Box:
> > RedHat 5.2, Kernel 2.0.36
> >
> > Macintosh:
> > MacOS 7.6.0
> > Open Transport 1.1.2 <-- 802.3 compatible
------------------------------
From: Jay Thorne <[EMAIL PROTECTED]>
Subject: NFS woes
Date: Wed, 03 Feb 1999 22:13:49 GMT
I'm trying to mound an nfs export from one machine to another.
When I call rpcinfo -p servermachine from the root account of
clientmachine, it never returns. Strangely enough, it does work perfectly
if you call it from a user account. Worse, every once in a while it will
work.
Strace follows:
close(3) = 0
munmap(0x400c8000, 4096) = 0
socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 3
getpid() = 1128
bind(3, {sin_family=AF_INET, sin_port=htons(880),
sin_addr=inet_addr("0.0.0.0")}, 16) = 0
connect(3, {sin_family=AF_INET, sin_port=htons(111),
sin_addr=inet_addr("x.x.x.x")}, 16
<hangs here>
information:
client machine:
Kernel 2.0.35
libc5.4.46
mount from util-linux 2.7e
rpcinfo from slack 3.2 dist
Server machine
RH 5.0 w/all current patches
w/ portmap updates
nfs updates
glibc updates
etc.
--
Jay Thorne [EMAIL PROTECTED] KE Software
http://www.kesoftware.com
------------------------------
Date: Tue, 02 Feb 1999 17:44:48 -0700
From: Bryan Scott <[EMAIL PROTECTED]>
Subject: Re: Masquerading and Netmeeting
I've also noticed that programs like CUSeeMe, NetMeeting, and VIRC pull
up specific IP addresses. If you're behind the firewall/masquerading
server, you can't get to it from outside. I have a linux box serving
five other machines and good luck trying to get other machines
connecting to services on them.
-- Bryan Scott
-- CTROnline Systems Administration
waco wrote:
>
> Brian Watson wrote:
>
> > I just got a linux server set up to deal with my internal household
> > network (2 machines). The only fault I can find, which I know is
> > microsofts, is that netmeeting will not receive any audio or video. I
> > know this is something to do with internal packet structures for
> > netmeeting but there is a loadable module for cuseeme, is there anything
> > equivalent for netmeeting?
> >
> > What other solutions are there? Can I make the linux server essentially
> > act as a router to one machine temporarily while I use netmeeting?
> >
> > --
> > ---------------------------------
> > ICQ UIN #804161
> > Web Page http://home.austin.rr.com/biscuitshouse
> > ---------------------------------
>
> I believe I read somewhere that netmeeting uses a proprietary socket not
> socks5 etc., so linux does not support it (i think)
------------------------------
From: [EMAIL PROTECTED] (Paul Anderson)
Crossposted-To: comp.os.linux,comp.os.linux.admin,comp.os.linux.help,nl.comp.os.linux
Subject: Re: Romote "root" login
Date: 1 Feb 1999 14:38:33 -0500
pazuzu <[EMAIL PROTECTED]> writes:
>I'm sorry, and I'm a bit new at this. Why is it more a security problem
>that people from outside can directly login as root (password guessing?)
>then people login in as guest and do a su -?
>
Instead of only knowing one password, they need to know two passwords, and a
valid user on your system.
------------------------------
From: [EMAIL PROTECTED] (Villy Kruse)
Crossposted-To: comp.os.linux.setup,comp.os.linux.admin,comp.os.linux.questions
Subject: Re: rpm question
Date: 3 Feb 1999 10:34:08 +0100
In article <[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote:
>mv traffic-vis-0_23-1_i386.rpm traffic-vis-0.23-1.i386.rpm
>
>because rpm looks for the i386 part.
>
Actually it doesn't. You could rename the file to 'my-old-hat' if you
wish and rpm wouldn't care. What is more important is that you download
with ftp in binary mode.
Villy
------------------------------
From: Keith Tucker <[EMAIL PROTECTED]>
Subject: Pacific CommWare serial board
Date: Wed, 03 Feb 1999 07:43:43 -0600
Hello All,
I have a pacific CommWare high speed serial board that I want to run
with my Linux box. I had Kernel 2.0.30 that I have upgraded to 2.2.1.
My question is: is there any special option that I need to look for
when I recompile the Kernel or does 2.2.1 just support the board (or
more importantly, the UART) without any special configuration. Thanks
in advance.
Keith Tucker
[EMAIL PROTECTED]
------------------------------
From: Ronny Bremer <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.caldera
Subject: Re: DHCP client (dhcpcd) for Caldera 1.3?
Date: Wed, 03 Feb 1999 22:47:44 +0100
Hi,
I can't tell you why Caldera still ignores the presence of DHCP (hint hint hint
:-))).
I do have both DHCPCD for LIBC5 and GLIBC. If anybody is interested, please send
me an email. I'll make sure you'll get it.There is only one exception: it does
not support Token Ring yet. I hope to implement it into the latest sources soon.
Regards,
Ronny Bremer
Wadels wrote:
> Help?
>
> DHCP is a common reality, but Caldera doesn't seem to support its client or
> be interested. "Our OpenLinux distribution currently does not include a DHCP
> client. So it isn't something that we are familure with to even give you
> some pointers." (End of knowledge base entry)
>
> So, I searched for a general use DHCP client, to find dhcpcd v. 1.3 from
> Phystech (or something like that). It looks great, except that its readme
> warns that it will not even compile without Glibc2. I know Caldera has
> glibc2 _runtime_ libraries, so my question is, will this suffice? Has anyone
> tried it?
>
> If it cannot be made to work, does anyone have a dhcpcd ver. 0.70 or better
> that was made for libc5 systems (Caldera or RedHat 4.2)?
>
> Thanks.
------------------------------
From: Guido Dolci <[EMAIL PROTECTED]>
Subject: linux firewalling--HELP!!
Date: Wed, 03 Feb 1999 15:09:37 +0100
I need a firewall, so I'm setting up a linux box with two eth cards.
As I need my LAN still have the old public IP addresses, I've been
suggested to do:
LAN ------------ eth1 linux eth0 ----------- eth0 cisco serial
========internet
195.xx 195.xx.xx.2 10.0.1.2 10.0.1.1
195.xx.xx.1
that is a static route to 10.0.1.1 on the cisco and then a static route
from 10.0.1.2 to 195.xx.xx.2 on the linux.
Good Idea.... but in terms of routing tables...how do I config the linux
box??
Thanks
Guido
------------------------------
** 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
******************************
