Linux-Networking Digest #640, Volume #10         Fri, 26 Mar 99 14:13:47 EST

Contents:
  Re: Problems with Telnet (Pierre Bodart)
  Re: ypmake or ypinit for NIS setup ? (Thorsten Kukuk)
  Re: Linux and Cable Modem (Paul Tomblin)
  Re: Telnet connection over LAN takes for over (Pierre Bodart)
  Re: Linux Client on an NT network ("Mark Swope")
  Re: Help with an office internet solution ([EMAIL PROTECTED])
  How to Keep a Fixed IP ... (Desmond Coughlan)
  Re: Problem trying to start squid 2 in Linux 2.0.36 ([EMAIL PROTECTED])
  What multicast groups have I joined? (David S. Goldberg)
  netatalk:printing to non-PS printer on Mac from Linux (Jungshik Shin)
  Re: Telnet connection over LAN takes for over ("MattW")
  Re: IP watching between 2 Hosts (Terje Trane)
  msg from kernel about network card ... (Pierre Bodart)
  Re: ISA network card problem with RedHat 5.2 ([EMAIL PROTECTED])
  Re: Problems with FTP/Telnet (Pierre Bodart)
  modules net-pf-??? (Pierre Bodart)
  Re: modules net-pf-??? ("William Evans")
  Re: Viewing Win95 drive on Linux box... ("William Evans")
  Re: winmodems ("Rufus V. Smith")
  insmod for redhat 5.2 ([EMAIL PROTECTED])
  sendmail without a named on local net (root)
  Possible NFS problem.  Client: Linux 2.2.3, Server: Solaris 2.5 ("Roy Westfall")
  Re: Linux into 98 (Eric Rossing)
  WU FTP ("Stressed")
  Re: Viewing Win95 drive on Linux box... ("Greg Boes")

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

From: Pierre Bodart <[EMAIL PROTECTED]>
Subject: Re: Problems with Telnet
Date: Fri, 26 Mar 1999 17:01:00 +0100

Probably your Linux box is trying to log the originator of the
connection (your NT box) and it times out on the name resolution for the
IP address.
Try putting your NT IP address in the /etc/hosts file and check that the
order for name resolution is well first hosts file, and then DNS. This
is done with the line 'order hosts, bind' in the file /etc/host.conf

Hope that help :-)

Pierre

David P. Cook wrote:
> 
> I have so far successfully go the eth0 interface set up and I am able to
> ping both
> NT and linux machines. I also can telnet from linux over to the NT box. But
> when it
> comes to telneting into the Linux box I  have a problem. When I try to
> Telnet into
> my Linux box I get this
> 
> Red Hat Linux Release 5.2 (Apollo)
> Kernel 2.0.36 on an i486
> 
> Then after 60 seconds the I lost connection. No username prompt comes up..
> The only place I can think of where the problem is, is with the hosts.allow
> and hosts.deny files.. I have hosts.allow set up as follows
> 
> ALL:169.254.50.151
> 
> This is the IP address for my nt box.. The hosts.deny file is empty..
> 
> I also tried deleteing the hosts.allow file per a request of somebody at
> work and the problem still presists.
> 
> I hope I gave enough information. I would appreciate any help and
> suggestions.
> 
> Thanks,
> Dave

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

From: Thorsten Kukuk <[EMAIL PROTECTED]>
Subject: Re: ypmake or ypinit for NIS setup ?
Date: 26 Mar 1999 14:14:38 GMT

Hello,

Dominique Vorbrodt <[EMAIL PROTECTED]> wrote:
> Hello out there
> A hopefully simple question.
> According to standards I understand
> /usr/lib/yp/ypinit -m
> is the command to set-up a NIS database in /var/yp

> However on my Suse Linux Server 6.0 I can't find the command ypinit but only
> a command ypmake.
> Is this to be used ?

You could use ypmake. But you also could go to /var/yp and run make.

Or even better, you could read the SuSE NIS HOWTO at /usr/doc/packages/yp/HOWTO.
For this reason, it was written.

I have fixed the missing ypinit for the upcoming SuSE Linux 6.1 for some weeks.

  Thorsten

-- 
Thorsten Kukuk      http://www.suse.de/~kukuk/        [EMAIL PROTECTED]
SuSE GmbH           Schanzaeckerstr. 10             90443 Nuernberg
Linux is like a Vorlon.  It is incredibly powerful, gives terse,
cryptic answers and has a lot of things going on in the background.

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

From: [EMAIL PROTECTED] (Paul Tomblin)
Subject: Re: Linux and Cable Modem
Date: 26 Mar 1999 16:50:01 GMT
Reply-To: [EMAIL PROTECTED] (Paul Tomblin)

In a previous article, "Jahan K. Jamshidi" <[EMAIL PROTECTED]> said:
>I would like to have the linux system as gateway to internet using my cable
>modem.  That way I can have my win98 and win95 system use the linux box to
>get internet access.

Yeah, that works.  I'm doing the same thing, except I have three Linux boxes
(one's an Alpha, one's a 486 and the other is a Celery), two Macs, and two
Windows boxes.  One of the Linux boxes is *only* doing the cable modem router
thing, because it's just a crappy old 486 I had kicking around.

The Linux box needs two ethernet cards.  You have to compile the kernel to
support IP Masquarading.  Your first ethernet card, eth0, will connect to the
cable modem, and the second one will connect to the hub.  Configure the first
one to get its IP via dhcpd, and assign the second one 192.168.1.1.

You need a cable modem log-in client to run on the Linux box, and that will
depend on your cable modem company.  I've seen clients for @home and Time
Warner Road Runner.

Configure the other machines on your hub with ip numbers in the 192.168.1.x
range, and have them use 192.168.1.1 as their gateway.  Set their name servers
according to what the cable modem company tells you, or run a caching name
server on the linux box and point them all to that.

Enable IP Masquarade on the Linux box.  Get isinglass from tummy.com to keep
script kiddies from hacking your network.  Keep up to date with the latest
Linux rpms from RedHat (assuming you're using RedHat Linux).

If you don't understand anything, ask here.

-- 
Paul Tomblin, not speaking for anybody.
"Faced with the prospect of rereading this book, I would rather have my brains
ripped out by a plastic fork." - Charles Cooper reviews the new Gates book.
http://www.zdnet.com/zdnn/stories/news/0,4586,2230586,00.html

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

From: Pierre Bodart <[EMAIL PROTECTED]>
Subject: Re: Telnet connection over LAN takes for over
Date: Fri, 26 Mar 1999 17:09:24 +0100

Probably your Linux box is trying to log the originator of the
connection and it times out on the name resolution for the
IP address.
Try putting your IP addresses in the /etc/hosts file and check that the
order for name resolution is well first hosts file, and then DNS. This
is done with the line 'order hosts, bind' in the file /etc/host.conf

Hope that help :-)

Pierre
Matt G wrote:
> 
> When using Telnet or FTP to my Linux machine over my LAN I have to wait over
> a minute to get a connection. Onc connected it communicates instantly?
> 
> anyone have any ideas?
> 
> Realtive newbie...
> 
> Matt

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

From: "Mark Swope" <[EMAIL PROTECTED]>
Subject: Re: Linux Client on an NT network
Date: Thu, 25 Mar 1999 17:24:22 -0600

Not quite as terse (or concise):
http://www.samba.org

You need to run Samba on your laptop.  Its docs
look daunting, but it's really not too bad.  Just set it
up with the minimum of stuff in the smb.conf file
and add stuff after you get it working.

mas

Marc Hering wrote in message <[EMAIL PROTECTED]>...
>man smbmount :)
>
>
>
>Wayne Chunn wrote:
>>
>> Greetings,
>>
>> Is it possible to login to an NT network with my Linux laptop?
>>
>> Thanks in advance,
>>
>> Wayne
>>
>> ------------------  Posted via SearchLinux  ------------------
>>                   http://www.searchlinux.com



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

From: [EMAIL PROTECTED]
Subject: Re: Help with an office internet solution
Date: Fri, 26 Mar 1999 15:38:16 GMT

Is your default route of the linux box your router?



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

From: Desmond Coughlan <[EMAIL PROTECTED]>
Subject: How to Keep a Fixed IP ...
Date: 26 Mar 1999 16:40:47 +0100

Hi,
I have a cable connection at home, but no fixed IP for the moment.
The ISP's DHCP server is supposed to check every 90 minutes, and if my
server is still using the IP (which, as it's up 24/24, it usually is),
it doesn't change it.

Last night, however, I was telnetting in from work, and at about
23:00, I lost the link.  It was 15:00 this afternoon, when I managed
to get another (different) IP.  This necessitated three reboots of the
server as "DHCPCD eth1" hung the server.

This is the third time this has happened, and frankly, it's becoming a
pain in the arse.  I can't connect, my users can't connect, and I have
to reboot the server to get another IP.

What I'd like you good people to tell me, is if there is a way to keep
my IP.  I've set up a cron job as follows:

0 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,21,22,23 * * * \
dhcpcd eth1

Will this work?  

If the worst comes to the worst, I really wouldn't mind paying an extra
100 FFr for a fixed IP, but my ISP doesn't plan to offer these for
another six weeks or so.  Would the above cron job keep my IP until
then? 

Your help would be appreciated.

-- 
Desmond Coughlan                |Restez zen ... Linux peut le faire
[EMAIL PROTECTED]
[www site under construction]
                                

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

From: [EMAIL PROTECTED]
Subject: Re: Problem trying to start squid 2 in Linux 2.0.36
Date: Fri, 26 Mar 1999 15:43:09 GMT

Try taking a look at the squid.out file and the syslog and the messages file

Squid is quite helpful in logging a lot of info if you look at these places.



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

From: [EMAIL PROTECTED] (David S. Goldberg)
Crossposted-To: comp.protocols.tcp-ip
Subject: What multicast groups have I joined?
Date: 26 Mar 1999 09:39:53 -0500

(I apologize if anyone sees an earlier version of this on c.o.l.n;
I've been told it may not have got off our local spool due to news
server problems)

I've got hosts running RH4.2 and RH5.2 (kernels 2.0.32 and 2.0.36
respectively) that are being used for a multicast demonstration.  Part
of the demo is explaining multicast concepts and I'd like to do that
visually.  On the Suns (sunos414 and 551) and SGIs I have, there's an
option to netstat (not consistent amongst them, of course :-) that
will display the groups that have been joined on the host.  I can't
seem to find any equivalent on the Linux boxen.  Any ideas on how to
dig out that information?  I've found /proc/net/igmp, but that output
isn't going to be clear to the people being shown the demo (yes, if
necessary, I can write a program to pretty it up, but I'd rather not
if it's already there).  The hosts are clearly working properly with
regard to multicast (they can join groups and properly send and
receive multicast traffic), btw, so there's no issue there.

Thanks,
-- 
Dave Goldberg
Post: The Mitre Corporation\MS B305\202 Burlington Rd.\Bedford, MA 01730
Phone: 781-271-3887
Email: [EMAIL PROTECTED]


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

From: Jungshik Shin <[EMAIL PROTECTED]>
Subject: netatalk:printing to non-PS printer on Mac from Linux
Date: 26 Mar 1999 15:19:30 GMT
Reply-To: [EMAIL PROTECTED]

  A friend of mine wishes to print from a Linux box in his lab with a HP
DJ 850c(attached to one of Macs in his lable) on Appletalk network.  You
may wonder why he doesn't attach the printer to his Linux box and
configure it to be visible from all the Macs in the network. I'm aware
that there's even a way(using ghostscript) to make HP DJ 850c look as if
it's a postscript printer in Mac Chooser. Problem with this approach is
his Linux box is not always running Linux, but sometimes it needs to run
Windows NT.

  So, my question is how to print out to a NON-postscript printer  on
appletalk network from a Linux box using netatalk. psf and its family
(symbolic links to psf) work only for PS printer, don't they?  I've been
using them to print to a number of public PS printers on my campus from
my Linux box and was successful in setting up my friend's RH 5.2 box the
same way(I wish RedHat supported appletalk printing in  its printtool.
In addition, its printer filter doesn't work well with psf/pap and I had
to resort to old magicfilter which have worked great for me). However,
I have no idea how to print with a non-PS printer.

   TIA, 

       Jungshik Shin


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

Reply-To: "MattW" <[EMAIL PROTECTED]>
From: "MattW" <[EMAIL PROTECTED]>
Subject: Re: Telnet connection over LAN takes for over
Date: Fri, 26 Mar 1999 08:35:06 -0800

If your Lan is using DNS then get DNS to resolve properly.  That way you
sholdn't have to add the hosts entry.  DNS needs to resolve both forward and
reverse and return appropriate info.

If your not using DNS then to the /etc/hosts file.

MattW

Pierre Bodart <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Probably your Linux box is trying to log the originator of the
> connection and it times out on the name resolution for the
> IP address.
> Try putting your IP addresses in the /etc/hosts file and check that the
> order for name resolution is well first hosts file, and then DNS. This
> is done with the line 'order hosts, bind' in the file /etc/host.conf
>
> Hope that help :-)
>
> Pierre
> Matt G wrote:
> >
> > When using Telnet or FTP to my Linux machine over my LAN I have to wait
over
> > a minute to get a connection. Onc connected it communicates instantly?
> >
> > anyone have any ideas?
> >
> > Realtive newbie...
> >
> > Matt



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

From: Terje Trane <[EMAIL PROTECTED]>
Crossposted-To: 
comp.os.linux.admin,comp.protocols.tcp-ip.domains,de.comm.protocols.tcp-ip
Subject: Re: IP watching between 2 Hosts
Date: Fri, 26 Mar 1999 18:27:42 +0100

Barry Margolin wrote:
> 
> In article <[EMAIL PROTECTED]>,
> Clemens Heise <[EMAIL PROTECTED]> wrote:
> >Hi !
> >I just need an idea, how to monitor IP between 2 hosts. I tried tcpdump,
> >but it does not have the right affect. Maybe
> >the handling is wrong.
> >E.g.:
> >tcpdump is installed on Host 192.168.100.5 and should listen in IP
> >between Host 192.168.100.10 AND 192.168.100.15
> >I tried command:  tcpdump -pqni eth0 host 192.168.100.10 and host 192.168.100.15
> >It takes no affect. In my opinion tcpdump listens only on IP on eth0.
> 
> The -p option specifies that it shouldn't go into Promiscuous mode, so it
> will only see traffic to or from itself.  Don't use that option if you want
> to see traffic of other hosts.
> 

You also have to make shure the interface is in promiscuous mode.
I don't know what operating system and version you are using. 
Here's from man ifconfig on a Linux-system:

       [-]allmulti
              Enable  or  disable  the  promiscuous  mode  of the
              interface.  This means that all incoming frames get
              sent  to  the  network  layer of the system kernel,
              allowing for networking monitoring.

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

From: Pierre Bodart <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,linux.admin.isp
Subject: msg from kernel about network card ...
Date: Fri, 26 Mar 1999 16:51:27 +0100

Hi,

I'm running a box with RH Linux kernel 2.0.34 and three network cards.
I'm using it as a router. It's running fine except that sometimes I can
find in the log this line :

kernel: eth1 : XMT status = 0xffff8060

The number can also be 0xffff80a0 or 0xffff80e0.
This is only reported for eth1, never for eth0 or eth2.
I can have up to 60 lines like that in a week.

Does someone have an idea of what this means ?
Or can someone point me to where to start looking for the explanation of
such error message ?

All ideas (even silly ?) are welcome.

Thanks in advance for your time.

Pierre

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

From: [EMAIL PROTECTED]
Subject: Re: ISA network card problem with RedHat 5.2
Crossposted-To: comp.os.linux.hardware,comp.os.linux.setup
Date: Fri, 26 Mar 1999 15:34:57 GMT

This would indicate that the irq/base address aer not the same in the card as your 
config.

Try manually setting the resources for your card.



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

From: Pierre Bodart <[EMAIL PROTECTED]>
Subject: Re: Problems with FTP/Telnet
Date: Fri, 26 Mar 1999 17:38:18 +0100

Probably your Linux box is trying to log the originator of the
connection (your NT box) and it times out on the name resolution for the
IP address.
Try putting your NT IP address in the /etc/hosts file and check that the
order for name resolution is well first hosts file, and then DNS. This
is done with the line 'order hosts, bind' in the file /etc/host.conf

Hope that help :-)

Pierre

Andrew wrote:
> 
> I am having trouble connecting to my linux machine (redhat 5.2) using telnet
> and ftp.  I can ping the machine with no problems from my nt box and it can
> ping the nt box as well (responce times are as they should be, no timeouts
> either)  When i try to by ftp or telnet it takes around 60 seconds or so to
> connect.  I will ftp over to my linux box and it will say connected to
> 10.10.10.2 but then it will take 30-60 seconds to bring up the login screen.
> Same for telnet.  Also, once logged in under ftp or telnet it is quick (no
> delays).  The part i am confused about is when i ftp from my linux machince
> to itself (localhost) or to my winnt box (nt server 4.0) it connects very
> quickly without any problem.  The only way i seem to be able to remedy the
> problem is to stop the dns server sevice on my nt server box and connect to
> the ip of the linux machine.  If anyone has any ideas as to how i could fix
> this problem i would greatly appreciate a responce
> 
> Andrew
> [EMAIL PROTECTED]

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

From: Pierre Bodart <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,fr.comp.os.linux.configuration
Subject: modules net-pf-???
Date: 26 Mar 1999 16:35:07 GMT

Hi,

I've recompiled my kernel 2.0.34 (RedHat distribution) and since then
I'm getting those lines almost each time I'm using some network services
(ftp, login, ...) :

modprobe: can't locate module net-pf-4
modprobe: can't locate module net-pf-5

I've checked again my kernel options and I was unable to find something
related to 'net-pf' ... I've also search in some HowTo and on the web
unsuccesfully :-(

Can someone help me ?
What are those modules ? How do I enable them (If needed ?) ?

All ideas are welcome !!!

Thanks in advance for your time,

Pierre

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

From: "William Evans" <[EMAIL PROTECTED]>
Subject: Re: modules net-pf-???
Date: 26 Mar 1999 13:03:38 -0500

>>>>> "Pierre" == Pierre Bodart <[EMAIL PROTECTED]> writes:

    Pierre> Hi,
    Pierre> I've recompiled my kernel 2.0.34 (RedHat distribution) and since then
    Pierre> I'm getting those lines almost each time I'm using some network services
    Pierre> (ftp, login, ...) :

    Pierre> modprobe: can't locate module net-pf-4
    Pierre> modprobe: can't locate module net-pf-5

These refer to ipx and appletalk protocols, respectively.  If you
aren't using either of these, don't worry ... it doesn't mean
something is misconfigured.

To disable these messages, add "alias net-pf-4 off" (and one for 5) to
/etc/conf.modules (or /etc/modules.conf, whichever you have) and
they'll go away.

    Pierre> I've checked again my kernel options and I was unable to find something
    Pierre> related to 'net-pf' ... I've also search in some HowTo and on the web
    Pierre> unsuccesfully :-(

/linux/Documentation/modules.txt


-- 
William Evans                 < william . evans @ computer . org >

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

From: "William Evans" <[EMAIL PROTECTED]>
Subject: Re: Viewing Win95 drive on Linux box...
Date: 26 Mar 1999 13:07:40 -0500

>>>>> "Jon" == Jon Slater <[EMAIL PROTECTED]> writes:

    Jon> Can I create a "mount point" on my Linux box that containes the Win95 C:
    Jon> drive?

samba should contain a binary called smbmount.  Use it to mount it.
Check out the man page for more information (such as restrictions,
command-line arguments, etc).

If you don't have one (mine is /usr/bin/smbmount), check where you got
the package from and make sure you didn't forget anything.  (It came
in samba-2.0.2.)

You also have to have smbfs available to the kernel, either compiled
in or (preferably) as a module.  Since you're using it to mount a
win95 box, make sure you also enable the Win95 workaround.

HTH

-bill

-- 
William Evans                 < william . evans @ computer . org >

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

From: "Rufus V. Smith" <[EMAIL PROTECTED]>
Subject: Re: winmodems
Date: Fri, 26 Mar 1999 13:04:57 -0500


That just feeds the spiral of needing more and more
powerful machines if more and more devices are leaching
computing power!

Incidentally, USB devices are also going to be doing
this (leaching) as I understand a lot of the control of USB resides
in the Master computer.  Anyone know more about that?



Paul Carver wrote in message <[EMAIL PROTECTED]>...
>On Sat, 20 Mar 1999 16:37:26 -0500, "batags"
><[EMAIL PROTECTED]> wrote:
>
>>I have installed RH 5.2 and have set up ppp0 through linuxconf, But I
still
>>can't get connected to my isp. I have all the info from my isp <like
>>gateway-primary dns-sec dns> But I still can't hear the modem dialing up
or
>>even get connected. I have a UsRobotics 56k Winmodem, Could that be my
>>problem? And if it is,  is there a way to solve it without getting a new
>>modem? Any help will be appreciated
>>Thanks
>>Bu||
>>
>
>I know the Linux community thinks poorly of Winmodems and many people
>believe that winmodems shouldn't be supported on Linux even if the
>manufacturers released the information necessary to write Linux
>drivers. I read an article that suggested that software modems may be
>the wave of the future however. The author based this assertion on the
>increasing power of CPUs in single user machines and there may be some
>sense in his position. The processing power to "modem" at 56K is
>relatively constant. As CPU continue to increase it will take a
>smaller and smaller percentage of the CPU's power to do the "modem"
>work. Isn't there a point where it becomes silly to use a dedicated
>modem processor to do work that the CPU can do without adding
>noticeable load? Might we see a day when you can't buy a hardware
>modem?
>



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

From: [EMAIL PROTECTED]
Subject: insmod for redhat 5.2
Date: Fri, 26 Mar 1999 18:10:50 GMT

Perhaps someone else has had this problem:

  I have a red hat 5.2 system up and running (build date 10/98) on a HP
Vectra XM 5/133. The Vectra comes with the enhanced HP 10/100 ethernet card.
When I issued the commands "insmod hp" or "insmod hp-plus", the command
returns an error of "device or resource busy". When I attempted to re-install
the whole OS and include support for this card, it told me it could not find
the device on my system. The card works fine off of my windows partition.
When I thought the card or PCI slot might be at fault, I installed an Intel
Ether Express pro card. Again, same problem. Is there a patch for this?

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

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

From: root <[EMAIL PROTECTED]>
Subject: sendmail without a named on local net
Date: Fri, 26 Mar 1999 18:54:35 +0100

Hello, perhaps somebody can help me ... and please don�t worry about my english.

I�ve got a linux-box running, connected to a local net. There is one mailserver 
(let�s say �mailserver�) for the hole net without any nameserver running. All
the mail produced on my box has to be forwarded to this mailserver, including mail
to root. To make it easier, there is only one account (root) on the box, because
it�s a firewall.

So I configured sendmail...
I am running Suse 6.0 (never trie it) but that�s not the problem. I do configure
sendmail over this wunderpretty m4 preprocessor and the important parts look like
this:

        # sendmail soll alles weiterreichen an den Mailserver mailserver
        # der mu� nat�rlich in /etc/hosts drinnenstehen
        define(`LOCAL_RELAY',`relay:mailserver')
        define(`MAIL_HUB',`relay:mailserver')

        # wir m�chten nicht, da� sendmail DNS benutzt, wir haben n�mlich kein named
        FEATURE(`nodns')
        FEATURE(`nocanonify')

I allready tried configurations where i defined the mailserver as a SMART_HOST,
tried other mailers and so on. The results always look like this:

        Mar 26 14:18:16 ... sendmail[688]: OAA00649: to=root, ctladdr=root (0/0), 
delay=00:04:03,
        xdelay=00:00:00, mailer=relay, relay=mailserver, stat=Deferred: Name server: 
mailserver:
        host name lookup failure

Please don�t think, I am too stupid. Of course there is an entry for the mailserver
in /etc/hosts und my bindings in /etc/host.conf are all right. I can do a
  ping mailserver
and all other software has no problems with my mailserver, only sendmail...

As I can see, sendmail uses DNS-Calls to resolve the name and does not read any lines
in my /etc/hosts. Also sendmail does not use any system-calls like gethostbyname() to
resolve the name (I wrote a litte C-Programm to test it, my programm got the right IP
for the name �mailserver�).

Is it possible to tell sendmail really to not use DNS vor Name-Resolving ?

If anyone knows any good ideas (please no answers like: did you start sendmail ?)

Thanks, thanks, and ... nice weekend for Linux (it�s raining again here in Stuttgart)


Snuggles

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

From: "Roy Westfall" <[EMAIL PROTECTED]>
Subject: Possible NFS problem.  Client: Linux 2.2.3, Server: Solaris 2.5
Date: Fri, 26 Mar 1999 12:24:27 -0600

I am having the following problem which I assume is an NFS issue.  I am
mounting an NFS drive using autofs,
from a Solaris 2.5 server on my x86 Linux box with the 2.2.3 kernel with NFS
builtin.

Everything works fine, except:

Software builds normally on my local drives just fine, but if I move the
sourcecode to the NFS mounted drive and rebuild, ar, ranlib, and /or ld
complain that the object files are bad (INVALID STRING OFFSET).  I am
assuming
that this is due to some subtle error or interaction in NFS.

Any assistance will be greatly appreciated.


    Roy Westfall -- ZEH Graphics Systems
    [EMAIL PROTECTED]



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

From: [EMAIL PROTECTED] (Eric Rossing)
Subject: Re: Linux into 98
Reply-To: [EMAIL PROTECTED]
Date: Tue, 23 Mar 1999 14:01:15 GMT

On Tue, 23 Mar 1999 08:29:19 -0500, "Matt Anderson" <[EMAIL PROTECTED]> wrote:

: Ok, I guess I will have to tone down my earlier enthusiasm a little bit.
: After playing around a little bit, I discovered that windows will not accept
: a connection from linux, even though I can ping it just fine.  I know this
: should be simple, but again I need just a little bit of assistance, even
: point me to a HOWTO or a man page.  Thanks a lot
: 
: Matt Anderson
: BTW--I want to ultimately winproxy my internet connection into linux (i'm a
: poor college student and for some GOD-AWFUL reason I'm stuck with a
: winmodem) .

You need to have a program that will accept a connection on a particular
port running on your Win9x box before you can connect.  Win98, I _think_
comes with "Personal Web Server", so you could set that up to make HTTP
connections, but doesn't come with any other TCP/IP server software.  

In addition, by itself, Win9x is unable to forward packets.  You need to get
additional software before you can access the Internet from your Linux box
through your Win9x box.  I know such software exists, but I don't remember
any specifics....

Eric Rossing
[EMAIL PROTECTED]

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

From: "Stressed" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,linux.redhat.install,linux.redhat.misc
Subject: WU FTP
Date: Fri, 26 Mar 1999 13:27:18 -0500

I've got a RHL 5.2 server going. Can someone give me specifics on CHROOTing
real users to specific directories upon login, (or other ways of limiting
users to specific areas that don't involve re-vamping my system). Specific.
Meaning, answers like, "see the man page for chroot", or "it's on the
cd-rom", or "check the HOWTO" aren't going to help. I've already read these,
as well as many examples and I have been to the WU site(s). Apparently, I'm
just missing the whole picture here. People are still able to log in and
"see" things I don't want them to. Please help. Thanks.

JM

[EMAIL PROTECTED]

P.S. I also would like to know about enabling telnet in a secure manner.
Right now, I have it disabled, since any valid user can apparently login via
telnet, cd to my /etc folder, download /etc/passwd and have their way with
it along with shadow and anything else thay can "see".

What I want to do is, eventually, be able to securely telnet to my Linux
server (from a remote location), and transfer files via other FTP servers
running on machines locally attached to the Linux server. I can then FTP to
the Linux machine and DL the files to my remote location. Does this make
sense?

Thanks again...





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

From: "Greg Boes" <[EMAIL PROTECTED]>
Subject: Re: Viewing Win95 drive on Linux box...
Date: Fri, 26 Mar 1999 12:36:11 -0600


Jon Slater wrote in message <[EMAIL PROTECTED]>...
>For example, I have a mount point (called C_Drive) that allows me access
>to my Win95 partition.


You need smbfs for this.  It enables use to mount 'shares' .It's distributed
with Samba these days I believe.

later



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


** 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
******************************

Reply via email to