Linux-Networking Digest #178, Volume #11 Sun, 16 May 99 19:13:42 EDT
Contents:
Re: samba troubles (password related ?) (Michael Abadjiev)
Re: File permission for group (Jon Skeet)
rlogin problems (root)
Losing default route (Sjoerd Venema)
simple ppp network ("Jean-Francois Joly")
A Real Puzzler, (try your hand) (Albert Schueller)
Network: Can I use Crossover Cable Instead of a Hub? (jhull)
Really Dumb Networking Question (TM) ("Johan Van Gompel")
Netgear EA201 NIC driver ("J. P.")
Re: A Real Puzzler, (try your hand) (Dave Platt)
Re: Null Cable For Ethernet (Kwan Lowe)
Diablo UDP packet problems behind IP Masquerading ("Justin Grochoski")
Apache compilation failes on RH6.0 (Michael Jenner)
PHP3 (3.0.7) compilation on RH6.0 fails (Michael Jenner)
my ISP times out ("marc*")
Re: Diald 0.16-5 problems ("Ian")
----------------------------------------------------------------------------
From: Michael Abadjiev <[EMAIL PROTECTED]>
Crossposted-To: linux.samba,comp.os.linux.setup
Subject: Re: samba troubles (password related ?)
Date: Sun, 16 May 1999 21:22:44 GMT
There is a registry file that forces Windows 98 to use plain text for
authentication you must run that file I think it comes with samba...
[EMAIL PROTECTED] wrote:
> If anyone could help me with this, I would sure appreciate it. I am
> trying to set up a Linux server w/ samba(v 2.0) to store some shared files on.
> I have everything installed, but Samba is giving me some problems. I
> have a smb.conf:
> [global]
> workgroup = UHC
> netbios name = THOTH
> server string = One Cool Server
> [homes]
> comment = home directory for the studs
> path = %H
> [tmp]
> comment = temporary files directory
> path = /tmp
> [testshar]
> comment = my test share
> path = /home/archive
> guest ok = Yes
>
> When I run 'smbclient -L thoth' on the linux machine, it works and lists
> the shares available, but from my windows(98) clients, it just says that
> thoth is inaccessible. When I try to do a 'net view \\thoth', I get
> this error message:
> Error 86: The specified network password is not correct. Type the
> correct
> password, or contact your network administrator for more information.
>
> I've gone through the DIAGNOSIS.txt file in the samba distribution, and
> get to test #8, which is the net view <servername> one. The things it
> says to do in there don't help. any other ideas?
>
> --
> Morgan Terry
> [EMAIL PROTECTED]
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Jon Skeet)
Subject: Re: File permission for group
Date: Fri, 14 May 1999 14:38:43 +0100
[EMAIL PROTECTED] wrote:
> Hey there...
>
> I have a Redhat 5.2 setup for server in a windows network.
> I got to windows users called User1 and User2.
> Both User1 and User2 got there home dir. on the LinuxServer without any
> problems.
> Then I created a folder called /home/public....
> created a Group called Gruppe1 with the members: User1 and User2
> I used chown -R .Gruppe1 . (in /home/public)
> then chmod -R go=rwx . (still in /home/public)
> And then i could create folders from User1 and User2
> Great .... :-)... but
> I cant create folders i folder created by the other user...... what do I do
> wrong ??????
The problem is that the folder created by the other user has the wrong
permissions. What you want is for the users to have umask 002 when
they're in /home/public, then everything they create will be
readable/writable by other members of the group.
Just type
umask 002
at the shell prompt (or put it in an RC file) to set this; alternatively
run chmod recursively on /home/public every so often.
--
Jon Skeet - [EMAIL PROTECTED]
http://www.pobox.com/~skeet/
------------------------------
From: root <[EMAIL PROTECTED]>
Subject: rlogin problems
Date: Mon, 17 May 1999 01:19:30 +0200
HI all,
I have some problems with my linux box.
I can=B4t do a rlogin or a slogin.
The Message is:
Connection reset by peer
Does anyone have an idea???
CU
Jan
------------------------------
From: Sjoerd Venema <[EMAIL PROTECTED]>
Subject: Losing default route
Date: Sun, 16 May 1999 23:37:01 +0200
Hi, I established a isp connection with dynamic ip. After disconnecting
I lose my default route. Does anyone know how to solve this problem??
THNX, Sjoerd
------------------------------
From: "Jean-Francois Joly" <[EMAIL PROTECTED]>
Subject: simple ppp network
Date: Sun, 16 May 1999 21:37:32 GMT
I`d like to setup my slackware linux computer as a server so other users can
access my machine by modem. (using ppp)
I thaught using the address 192.168.0.1 for my computer, the server. So what
I did is run netconfig and entered a host name, a domain name I invented,
net mask, etc.
++summary++
hostname: joey
domain: fisherman.org
host address: 192.168.0.1
net mask: 255.255.255.000
But, when I reboot and try to run telnet (telnet joey), it says it can`t
find the network or something like that. What's the problem? Why can't I
telnet to my own computer? Is there more to setting up my own dial-in ppp
network?
Also, I installed mgetty and it works. (As far as I know, it can answer the
modem and the user can access the shell) From there, what do I have to do
use PPP now? If I want to have an static IP for each individual user, what
do I have to do?
Thanks!
Jean-Francois
------------------------------
From: [EMAIL PROTECTED] (Albert Schueller)
Crossposted-To:
comp.os.linux.setup,comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux.setup
Subject: A Real Puzzler, (try your hand)
Date: Sun, 16 May 1999 21:41:12 GMT
Reply-To: [EMAIL PROTECTED]
Please read on, you've probably never run into this before!
As you may have noted from the subject heading I'm experiencing a
very strange problem with my newly upgraded Redhat 6.0 box. It's a Pentium
133 on an ethernet, kernel version 2.2.9.
The problem is that when I connect to it from home (over a 14.4 modem, ppp
connection) from either windows or linux (rh6.0), everything runs slowly
and by slowly, I mean at the processing level. For instance, I did the
following timings of a clean mutt source tree compilation:
These are tcsh time results (default output). I took the trouble of
labeling some of the relevant fields, so you needn't run off to the tcsh
man page.
>From office:
> time make
CPU (s) Ker (s) Wall Clock CPU%
290.830u 12.270s 5:36.29 90.1% 0+0k 0+0io 76724pf+0w
>From home (linux):
> time make
CPU (s) Ker (s) Wall Clock CPU%
2235.410u 87.380s 40:22.81 95.8% 0+0k 0+0io 76562pf+0w
>From home (windows):
> time make
CPU (s) Ker (s) Wall Clock CPU%
2115.200u 83.350s 36:59.63 99.0% 0+0k 0+0io 74436pf+0w
As you can see from home the compilation takes roughly 8 times longer. I
can think of no reason for this whatsoever. The connections from home are
made using ssh or ttssh, but I don't think ssh is the problem.
Interestingly, I tried logging in from a few other boxes on the same
subnet at the problem box and got differing results. From one machine,
again over ssh, the timings are normal, on another the timings are slow.
They are both rh5.1 boxes. All connections via ssh.
I've been administrating linux boxes for a few years now and cannot think
of any explanations. Any ideas anyone?
thanks,
Albert
--
Albert Schueller Department of Mathematics
Office Phone: 509-522-4432 Whitman College
Public Key: http://carrot.whitman.edu/pgp.key Walla Walla, WA USA 99362
------------------------------
From: jhull <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.questions,comp.os.linux.setup,comp.os.linux.misc
Subject: Network: Can I use Crossover Cable Instead of a Hub?
Date: Sun, 16 May 1999 16:58:23 -0400
Reply-To: [EMAIL PROTECTED]
Hi there,
Here is the question: Can I use crossover cable instead of a Hub?
If I don't have to use a hub, then could someone please tell me what I
could have done wrong?
I am trying to set up a local network connecting two PCs, one running NT
server, and the other running Linux RH 5.2/Windows 98.
Before I started to configure the NT-Linux network, I succefully
configured the NT-Windows98 network using a crossover cable (without
using a hub).
Then I ran netcfg to configure my network on the Linux box. I
configured two interfaces, i.e., lo-->127.0.0.1 and
eth0-->200.50.50.49. It seemed to be OK, until I ran ifconfig which
only reported my local loopback, but did not recongnize my network card.
Thanks in advance.
James
------------------------------
From: "Johan Van Gompel" <[EMAIL PROTECTED]>
Subject: Really Dumb Networking Question (TM)
Date: 16 May 1999 20:45:01 GMT
Hello,
Suppose there are two networks connected through a leased line. There is a
server and a router on each side of the leased line. Employees of the same
department are located on each side of the leased line. Each employee has
his/her own computer with a 192.168.x.y (x == entity/department number) IP
address and 255.255.0.0 netmask. The computers in network 1 can talk to
each other and so can those in network 2. But how does a computer from
network 1 talk to a computer in network 2?
Do I have to set up a routing table on each of the servers or is it as easy
as setting those 2 servers up as a gateway? Am I talking nonsense here?
Sorry folks, but this is my first leased line experience. Pitty me, but
don't ostracize me. ;-)
Please forward replies to [EMAIL PROTECTED] as well as posting to this
newsgroup.
--
Johan Van Gompel
------------------------------
Date: Sun, 16 May 1999 13:09:14 -0700
From: "J. P." <[EMAIL PROTECTED]>
Subject: Netgear EA201 NIC driver
I've been reading that the tulip driver supports the Netgear NICs,
but so far I could only ascertain that all these posts were referring
to the Netgear FA 310 TX PCI card. But what about the 10baseT ISA
card from Netgear, the EA201? Is the tulip driver supporting this NIC,
too?
Any succesful users of this NIC here?
Thanks for any feedback,
Joe
------------------------------
From: [EMAIL PROTECTED] (Dave Platt)
Crossposted-To:
comp.os.linux.setup,comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux.setup
Subject: Re: A Real Puzzler, (try your hand)
Date: Sun, 16 May 1999 22:06:23 GMT
>As you can see from home the compilation takes roughly 8 times longer. I
>can think of no reason for this whatsoever. The connections from home are
>made using ssh or ttssh, but I don't think ssh is the problem.
>
>Interestingly, I tried logging in from a few other boxes on the same
>subnet at the problem box and got differing results. From one machine,
>again over ssh, the timings are normal, on another the timings are slow.
>They are both rh5.1 boxes. All connections via ssh.
>
>I've been administrating linux boxes for a few years now and cannot think
>of any explanations. Any ideas anyone?
I'd guess that the machine which is suffering from the inconsistent
results, has APM power-management features enabled. The CMOS/BIOS
settings are probably set to put the system into a lower-power mode of
operation after some number of minutes of inactivity. The APM
low-power mode in question is probably reducing the CPU clock speed
(either by actually slowing down the clock, or by running the clock in
short bursts and stopping it entirely in between bursts).
On a lot of these systems, "inactivity" means "no keyboard or mouse
use".
If you compile locally (from the console) the occasional keyboard
activity is enough to keep the system "awake". If you dial in from
home, though, it'll probably have been an hour or more since you used
the keyboard; the APM portions of your motherboard chipset won't have
detected any "activity" and the chipset/BIOS will have put the system
into low-power mode. Your over-the-modem or over-the-network accesses
to the machine don't count as "activity" (based on your APM CMOS
settings) and so the system runs at about the speed of an 8 MHZ 386.
When you did the test via other machines on the same subnet, and got
inconsistent result, I'd suspect that you probably had touched the
host's keyboard or mouse sometime during or before the tests you
ran... and as a result, some of the tests were done in APM low-power
mode and some in full-power mode.
Best fix is probably to disable the APM "sleep" or "standby" mode via
your CMOS setup screen. Linux already gives you most of the
advantages of this mode, because it idles the CPU during periods of
inactivity (unlike Win3.1 and DOS).
--
Dave Platt [EMAIL PROTECTED]
Visit the Jade Warrior home page: http://www.radagast.org/jade-warrior/
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
------------------------------
From: Kwan Lowe <[EMAIL PROTECTED]>
Subject: Re: Null Cable For Ethernet
Date: Fri, 14 May 1999 07:01:43 -0400
This info is in the Ethernet Howto near the end of the file. You need
to take a normal patch (straight through) cable, cut off one end, swap
the transmit and receive wires and recrimp. This creates a null cable.
mike wrote:
>
> I have heard that one can modify a standard twisted pair ethernet
> cable to allow two computers to connect with each other without
> using a hub. Can one purchase such a cable. Is a " patch cable "
> such a cable?
> Mike
------------------------------
From: "Justin Grochoski" <[EMAIL PROTECTED]>
Subject: Diablo UDP packet problems behind IP Masquerading
Date: Sun, 16 May 1999 17:22:39 -0400
I have IP Masquerading set up on my linux box and my internal machines can
interact via numerous ip protocols (HTTP, FTP, ...) but UDP packets do not
seem to be masqueraded. I could not find a module to load to perform this
function either.
I am running RH5.2 (kernel 2.0.36).
Can anyone point me to a way to masquerade all UDP packets or generically
all ip packets to a specific host if it is not masquerading on that port for
another host?
example:
192.168.0.1 is my linux "router" using ip masquerading with external
(internet) address 200.200.200.1
192.168.0.2 is my local machine running Diablo
Now, when a packet originates from 192.168.0.2 port 1400, it goes through
the gateway at 192.168.0.1 and is masqueraded to originate from
200.200.200.1 port 69005. A reply then comes back to 200.200.200.1 port
69005 and is masqueraded to 192.168.0.2 port 1400. This works great.
What I need to do is say that if a port on 200.200.200.1 is not being
masqueraded, forward the packet to the exact same port on 192.168.0.2.
Or perhaps set up some rules on how to do this. Since 192.168.0.2 is not an
internet ip address, I still need to masquerade instead of just forwarding,
but I want all unknown traffic (or traffic on specific ports) to be
masqueraded to that machine (192.168.0.2).
--
Thanks,
Justin
------------------------------
From: Michael Jenner <[EMAIL PROTECTED]>
Subject: Apache compilation failes on RH6.0
Date: Sun, 16 May 1999 22:13:56 GMT
Apache compilation failes on RH6.0
./configure --sysconfdir=/etc/httpd \
--datadir=/home/httpd \
--logfiledir=/var/log/httpd \
--enable-module=most \
--enable-shared=max \
--disable-rule=WANTHSREGEX
Had to disable modules (--disable-module=) "rewrite" and
"auth_dbm"
otherwice a ndbm.h header file is missing.
--disable-module=rewrite --disable-module=auth_dbm
to make it compile. However auth_dbm seems to be a must as Apache
starts to complain on a lot of normally ok configurations in the
config-file.
/etc/rc.d/init.d/httpd start
Invalid command 'UserDir', perhaps mis-spelled or defined by a
module
not included in the server configuration
/usr/local/apache/bin/apachectl start: httpd could not be started
What am I doing wrong ?
Thanks in advance
/Michael
------------------------------
From: Michael Jenner <[EMAIL PROTECTED]>
Subject: PHP3 (3.0.7) compilation on RH6.0 fails
Date: Sun, 16 May 1999 22:16:34 GMT
PHP3 (3.0.7) compilation on RH6.0 fails
Running:
./configure --with-apxs=/usr/sbin/apxs \
--with-config-file-path=/etc/httpd \
--with-mysql \
--with-zlib \
--with-system-regex
make
gives:
functions/gdttf.c:20: freetype.h: No such file or directory
make: *** [functions/gdttf.o] Error 1
Hints, suggestions etc are very welcome
Regards,
Michael
------------------------------
From: "marc*" <[EMAIL PROTECTED]>
Subject: my ISP times out
Date: Mon, 17 May 1999 00:29:32 +0200
hi,
i set up a connection through gazel isdn card.
i'm using kppp.
everything seems to work fine at first: it says "modem ready", then it dials
my isp number, then it says "connecting to network..."
....but it doesn't go beyond that point.
ANd it finally times out.
To what is this timing out related bec i think that my isp is effectively
reached.
I have entered the correct DNS and Gateway settings of my isp in kppp.
So ?
tks.
marc.
------------------------------
From: "Ian" <[EMAIL PROTECTED]>
Subject: Re: Diald 0.16-5 problems
Date: Mon, 17 May 1999 10:37:55 +1200
It may be the old problem with the Win machines sending out their "hello!
I'm here!" type messages. They do this about every 20mins or so.
If it is, you can just exclude all tcp and udp port 137 and 139 requests
with ipfwadm. I can never remember the command but it is fairly straight
forward.
Ian
Paul Massey wrote in message <[EMAIL PROTECTED]>...
>Hi all
>
>I have RH 5.2 and trying to use diald to automatically dial out as
>necessary.
>
>diald.conf is
># /etc/diald.conf
>mode ppp
>fifo /etc/diald.ctl
>accounting-log /var/log/diald/account.log
>crtscts
>local 193.237.58.184
>remote 158.152.1.200
>netmask 255.255.255.0
>defaultroute
>modem
>device /dev/modem
>path-route /sbin/route
>path-ifconfig /sbin/ifconfig
>path-pppd /usr/sbin/pppd
>buffer-packets
>buffer_size 65536
>buffer-fifo-dispose
>buffer-timeout 600
>connect "/usr/sbin/chat -f /etc/ppp/demon/chat.demon"
>lock
>speed 115200
>restrict 00:00:00 07:44:59 1-5 * *
>or-restrict 18:15:00 23:59:59 1-5 * *
>up
>restrict * * 0,6 * *
>up
>restrict 07:45:00 18:14:59 1-5 * *
>down
>include /usr/lib/diald/standard.filter
>
>I have set it only connects automatically during cheap rate times on the
>phone bill it's okay during the standard rate time when it is down
>however as soon as cheap rate is allowed the phone connects with a
>masqurade request.
>
>this is checked by ysing the following command
>
>the firewall is set up thus
>
>ipfwadm -F-p deny
>ipfwadm -F -m a -S 192.168.0.0/16 -D 0.0.0.0/0
>
>Yes I know it should be -S 192.168.1.0/24 but I might want to add a
>second ethernet card as a 192.168.2.1 and allow the other hosts to use
>the same connection to the internet.
>
>ipfwadm -Ml
>List of Masqurade requests is blank. but the phone sta#ill dials out.
>The standard.filter is unchanged from the inital installation.
>
>I have two Windoze 95 machines connected as 192.168.1.2 and 192.168.1.3
>I know but the ankle biters want to use those "popular games". Although
>I think I am winning slowly. I look forward to using wine and then they
>can use the games want to on an OS that works. I have considered that
>the samba might be producing extra requests but they should show in the
>ipfwadm list
>
>What am I doing wrong ?
>--
>Paul Massey
------------------------------
** 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
******************************
