Linux-Networking Digest #37, Volume #12 Wed, 28 Jul 99 17:13:39 EDT
Contents:
FW, WAN & MASQ (Michiel Dethmers)
Linux as a mailserver (EPrintingc)
SMB tunneling? IPX forwarding help? please? :) (Patrick Allison)
Re: diald goes up every 4 -5 min (rbinning)
help please: portfw invalid users from 80 to 8080 ([EMAIL PROTECTED])
Re: RH6.0 NFS: nfssvc -- function not implemented (A E Lawrence)
MS Access and Linux (Steve Porter)
Re: diald goes up every 4 -5 min (rbinning)
Re: ppp - ping OK but no telnet/ftp/netscape - answer. (Abdullah Ramazanoglu)
Free Long Distance (Kent Kling)
Re: MS Access and Linux (Jerry Walter)
Re: DHCPd ("John G. Drummond")
Re: e-mail problems (Matthew)
Linux Server is closing files opened by BDE or Microsoft-Word, etc ("Zorlu Yusuf")
Re: help! very slow ethernet IN ONE DIRECTION (William Calnan)
Re: PCI BIOS setup ? (Abdullah Ramazanoglu)
Re: DHCP questions (cable modem) (dead_grandmother)
Linux Server is closing files opened by BDE or Microsoft-Word, etc ("Zorlu Yusuf")
Re: Netscape scraps Limux! (John Thompson)
Re: root login problem with RH6.0 (Matthew)
----------------------------------------------------------------------------
Date: Wed, 28 Jul 1999 18:32:51 +0100
From: Michiel Dethmers <[EMAIL PROTECTED]>
Subject: FW, WAN & MASQ
Hi All,
I have a firewall installed (RH5.2) using masquerading for most of the
work. It's got three NICs, one inwards, one DMZ and one outwards.
Inwards is 192.168.2.2 as my network is 192.168.2.0/24
Now I want to add a WAN connection, which has a router at 192.168.2.1,
connecting to network 192.168.1.0. If I telnet into the router I can
ping the other side (router 192.168.1.1), but not from my network.
My network uses the FW as router, and the FW has a static router entry,
route add net 192.168.1.0 netmask 255.255.255.0 gw 192.168.2.1
(before the "default" which is my "outside" router).
Anyone who has more experiences with this setup? I can't figure out
what's wrong but I obviously haven't done the right thing.
Thanks
Michiel
------------------------------
From: [EMAIL PROTECTED] (EPrintingc)
Subject: Linux as a mailserver
Date: 28 Jul 1999 17:29:30 GMT
I am using Red Hat Linux 5.2 as a mailserver. The network seems to be
configured properly. All IP addresses are in place. This mailserver did work
for a little bit. I am using Microsoft Outlook as my Interoffice email
solution for now. I have setup an "Internet Email" account (only for in house
use).
The problem I am now getting is a "No transport provider available for
delivery". Please help. I have been stumbling on this for about 3 weeks now.
Reply to [EMAIL PROTECTED] or just reply to [EMAIL PROTECTED]
------------------------------
From: Patrick Allison <[EMAIL PROTECTED]>
Subject: SMB tunneling? IPX forwarding help? please? :)
Date: Wed, 28 Jul 1999 14:21:48 -0400
Hi -
I'm setting up a rather... curious network configuration, and though I'm
99.9%
sure that I can do everything I want to, I honestly can't quite figure
out
the *simplest* way to do things. Here's the basic setup:
At home -
Network with multiple PCs, and one Linux server (IP masq, PPP
connection)
- One IP address, and all the internal PCs have 192.168.1.x
addresses.
At school -
Network with multiple PCs (none of which I can affect in the
least)
One Linux server (mine)
- IP addresses for *everyone* - a normal Internet-connected
network.
At someone else's home -
Network exactly like mine at home (one IP address, linux masq.)
Now - what I want to be able to do:
I want an IPX network bridged across all three. For this, I should just
have to run
ipxtunnel, I think, but I've got a few questions about that. More later.
:)
(reason: games. duh.)
I want "Network Neighborhood" on the PCs at school to see *everything*
as well -
basically, I want to know if I can bridge the two SMB networks.
Apparently
a VPN would do this perfectly fine. (?) However, I'm not exactly sure
how to
set things up. Note that I can *not* simply make my linux server at home
a
WINS server, because I can't tell people at school to use my WINS
server.
So, simple question on that is, how do I set things up so that the Linux
servers
pass SMB packets back and forth?
(reason: mp3s. do you have any idea how many mp3s are at universities?
:) )
I also want the network at "someone else's home" to see each individual
PC at my
home. I should simply be able to run IP-IP tunneling between the two
Linux servers
(one at my home, one at theirs) and the two would see each other
transparently.
But, again, how do I ensure that SMB packets are passed?
(reason... games, file sharing, even possibly printer sharing. ooh.
cool.)
Augh. Concerns.
Any help is appreciated!
Patrick
------------------------------
From: rbinning <[EMAIL PROTECTED]>
Crossposted-To: redhat.networking.general
Subject: Re: diald goes up every 4 -5 min
Date: Wed, 28 Jul 1999 18:31:59 GMT
[EMAIL PROTECTED] wrote:
> Just tell diald to ignore winblows netbios requests!
>
> In diald conf... filter
> ignore tcp tcp.source=tcp.netbios-ns,tcp.dest=tcp.netbios-ns
> ignore tcp tcp.source=tcp.netbios-dgm,tcp.dest=tcp.netbios-dgm
> ignore tcp tcp.source=tcp.netbios-ssn,tcp.dest=tcp.netbios-ssn
>
> ignore udp udp.source=udp.netbios-ns,udp.dest=udp.netbios-ns
> ignore udp udp.source=udp.netbios-dgm,udp.dest=udp.netbios-dgm
> ignore udp
> udp.source=udp.netbios-ssn,udp.dest=udp.netbios-ssn
>
> Then diald will not dial outside for winblows lookups
Thanks for the advice,
I changed my filter to read as advised, and still had the same problem. I then
tried to narrow it down even further using tcpdump on sl0 and found that indeed
it was a Samba request as Mike mentioned in reply.
Now that I have got it narrowed to this what are the pro's and cons of running
a caching -nameserver and if someone could direct me to the proper HOWTO'S to
get me headed in the right direction.
If I misunderstood any of the advice given please let me know.
Thanks for all the help...
Rob
rob wrote:
> >
> > Hello all,
> >
> > Thanks for the advice so far recieved. I now have it down to every 15 min.
> >
> > that it dials out!
> >
> > here is the culprit accurding to tcpdump -i eth0:
> >
> > arp who has abby.chelsea.com tell barbie.chelsea.com
> >
> > arp reply abby.chelsea.com is at 0:0:b4:85:22:7c
> >
> > 127.0.0.2.61076 > 204.209.196.5.domain:1251+ (46)
> >
> > 127.0.0.2.61076 > 204.209.196.5.domain:1251+ (46)
> >
> > 127.0.0.2.61076 > 204.209.196.5.domain:1251+ (46)
> >
> > abby.chelsea.com.137 > 192.168.1.255.137: udp 50
> >
> > barbie.chelsea.com.137 > abby.chelsea.com.137:udp 62
> >
> > abby.chelsea.com.139 > barbie.chelsea.com.1111: . ack 1 win 32736
> >
> > barbi.chelsea.com.1111 > abby.chelsea.com.139: . ack 1 win 8508 (DF)
> >
> >
> >
>
>
------------------------------
From: [EMAIL PROTECTED]
Subject: help please: portfw invalid users from 80 to 8080
Date: 28 Jul 1999 17:29:13 GMT
Reply-To: [EMAIL PROTECTED]
I have a Slackware 4.0 machine masquerading for the network. Only certain
users are allowed access to the internet. To get that access they go to a web
page which calls IPChains to add them to the masq tables. They browse, yadda,
yadda, they're happy. If they are NOT allowed to access the net and try and
access an external site, say www.slashdot.org, the masq tables are set to forward
ANY request they make to port 8080 on the IPMasq server. The webpage they
come to tells them how they are trying to access things they are not allowed to.
Basically I redirect all "unauthorized" TCP/UDP stuff to port 8080 on the gateway.
This scenerio only works if I use the "redir" command available from the IP Masq
resource page. I have to forward port 8080 to 8081, apache running on 80 and 8081.
If I setup Apache to run on 80 and 8080 the redirected user always ends up at
the server running on port 80, not the one on 8080.
Anybody have any thoughts as to why this is happening? I like to not have to use
the "redir" command.
Here's what my IP Masq tables look like:
spacebridge:/# ipchains -L
Chain input (policy ACCEPT):
target prot opt source destination ports
ACCEPT all ------ 192.168.3.100 anywhere n/a
ACCEPT all ------ localnet/16 localnet/16 n/a
ACCEPT all ------ anywhere anywhere n/a
REDIRECT tcp ------ anywhere anywhere any -> any =>
8080
REDIRECT udp ------ anywhere anywhere any -> any =>
8080
Chain forward (policy DENY):
target prot opt source destination ports
ACCEPT all ------ 192.168.5.18 anywhere n/a
MASQ all ------ anywhere anywhere n/a
Chain output (policy ACCEPT):
Spacebridge is my Linux gateway, IP Addr 192.168.5.18.
Thanks very much,
Devon P. Hubner
"Miserable Glitches!"
-- Shockwave
[EMAIL PROTECTED]
------------------------------
From: A E Lawrence <[EMAIL PROTECTED]>
Crossposted-To: ox.os.linux,uk.comp.os.linux
Subject: Re: RH6.0 NFS: nfssvc -- function not implemented
Date: Wed, 28 Jul 1999 20:01:32 +0100
Yu Qian Zhou wrote:
>
> After I compiled the new kernel of my RH6.0, the NFS server can on longer
> be started, it reports
>
> % /etc/rc.d/init.d/nfs start
>
> ...
> NFS: nfssvc -- function not implemented
> ...
>
> I suspect I may have missed some modules in the "make xconfig". Can
> anyone help?
>
The problem is that your kernel was not configured to support the new
kernel-based nfs server. Recompile with something like:
CONFIG_EXPERIMENTAL=y
CONFIG_NFSD=m
And yes this is just one of the many little problems that arise when
upgrading to RH6.0. :-)
Another one I discovered yesterday is:-
rpm -qf /usr/include/scsi/scsi.h
glibc-devel-2.1.1-6
and
$ ls -l /usr/include/scsi/
lrwxrwxrwx 1 root root 25 Apr 18 1998 /usr/include/scsi
-> ../src/linux/include/scsi
Thus upgrading glibc-devel-2.2.2.6 corrupts your kernel if you happen to
be on a different version from that current when the rpm was built. I
was on 2.2.10, and found that the recompilation failed for that reason.
I just reinstalled the kernel from the tar ball.
Adrian
--
Dr A E Lawrence (from home)
------------------------------
From: Steve Porter <[EMAIL PROTECTED]>
Subject: MS Access and Linux
Date: Wed, 28 Jul 1999 18:40:08 GMT
HELP! I have a client who has a small (10 user) peer-to-peer MS Win98 network,
with each node having MS Access97 installed. They keep a database on one of the
Win98 w/s, and multiple people can access the db file at the same time. So here
I come along and sell them a Linux server. I set up Samba, set up user names,
logons, p/w, permissions, etc. I created a Public deirectory on the Linux
(Mandrake 6.0) server and copied the Access database file over to the Linux
Public folder. Tested each users connectivity to the server and specifically to
the Access db file. But here's the catch...only one user can access the file at
a time. When a second user tries to access the file, that user gets a message
similar to "the number of allowable users has been exceeded...". Again this did
not happen when the same file was residing on a directory on the Win98 station.
I'm clueless.....any ideas?
Steve
------------------------------
From: rbinning <[EMAIL PROTECTED]>
Crossposted-To: redhat.networking.general
Subject: Re: diald goes up every 4 -5 min
Date: Wed, 28 Jul 1999 18:47:10 GMT
Mike Jagdis wrote:
> In article <rpqdgl$0$37nsp8d$[EMAIL PROTECTED]>, rob wrote:
> >Thanks for the advice so far recieved. I now have it down to every 15 min.
> >that it dials out!
> >here is the culprit accurding to tcpdump -i eth0:
> >arp who has abby.chelsea.com tell barbie.chelsea.com
> >arp reply abby.chelsea.com is at 0:0:b4:85:22:7c
> >127.0.0.2.61076 > 204.209.196.5.domain:1251+ (46)
>
> It's a domain request (i.e. DNS look up). It comes from 127.0.0.2
> which, presumably, is what you using for a local IP address on the
> link when it is down so it is something on the local machine doing
> the look up. Every 10-15 mins is typical of the frequency that
> Windows machines generate queries for their workgroup name. Probably
> this query is getting picked up by samba which is then trying to
> satisfy it by querying DNS. The simplest way of "fixing" it is
> to use the samba config option that turns off forwarding queries
> to DNS (it's in the man page, dns something-or-other). The best
> way of fixing it is to run a name server that is authoritative
> for the local domain so that you can fail bogus queries locally
> rather than bouncing yet another query for "WORKGROUP" off the,
> already overloaded, root nameservers.
>
> Mike
> Hi Mike,
> Thanks for the info! indeed I believe that you hit the nail on the head.
> After doing a tcpdump on sl0, I found that it was indeed a Samba request to
> find WORKGROUP on my ISP 's DNS. Now that I have it narrowed to this How do
> I go about configuring and using a caching name server? I have the O'rielly
> NAGuide, but some of the examples seem a little out dated. Any direction you
> cou give me our point me to would be greatly appreciated.
Thanks again...
Rob
------------------------------
From: Abdullah Ramazanoglu <[EMAIL PROTECTED]>
Subject: Re: ppp - ping OK but no telnet/ftp/netscape - answer.
Date: Wed, 28 Jul 1999 22:52:54 +0300
[EMAIL PROTECTED] wrote:
>
> I all,
>
> For the past few days I have struggled with the problem that I have
> been able to dial in to my ISP, start ppp successfully, and ping to any
> other machine, but I have been unable to telnet, ftp or use netscape.
> I'm using RH6.0 and ppp 2.3.7.
>
> When I attempt to telnet I get :
>
> telnet fred.net.nl
> Trying xxx.xxx.xx.xx...
> Connected to fred.net.nl.
> Escape character is '^]'.
>
> and then it stops there - no login prompt.
>
> Same for ftp, and using netscape I get
>
> Connect: Host www.his.com contacted. Waiting for reply...
>
> and we wait forever.
>
> I was unable to solve the problem even after reading the PPP-HOWTO,
> scanning the network administrators guide, and reading the many
> hundreds of postings on this and other linux newsgroups.
>
> I have stumbled onto the solution for my case:
>
> To solve my problem I needed to change the following line in my
> /etc/ppp/ppp-on script:
>
> Original (as from the PPP-HOWTO):
> REMOTE_IP=0.0.0.0 # Remote IP address if desired. Normally 0.0.0.0
>
> New:
> REMOTE_IP=xxx.xxx.xx.xx # where xxx.xxx.xx.xx is the IP address of my
> ISP.
>
> I know this is trivial, and I should have spotted it before, but it
[snip]
I think this must be something deeper than ppp, related to TCP code. I'm
having exactly the same problem, not on the ppp but on ethernet. I have
two 2.2.5 machines (RH6.0) with Cnet-Pro120c NICs (tulip) connected
directly via cross UTP. I can ping, traceroute, tftp, but can not
rlogin, telnet, nfs and http. I don't think that this is a reverse DNS
lookup phenomenon because http is not working either. ICMP and UDP
apparently works, but TCP gets stuck. Receiving side gets request (e.g.
login), authorizes and opens a session, but requesting side doesn't get
an answer. Machines are quite a standard RH6.0 installation, which I
also run at work without this problem. I had suspected the NICs that I
use, and tried my NIC on the machine at work-site: It runs without
problems. The remaining difference between home and work-site machines :
- Home : Direct UTP connection / Work : Using hub.
- Home : One of machines has Internet connection (ppp defined) / Work :
No ppp
- Home : Machines are Celeron (o/c) & 486 / Work : Pentium-MMX
No other difference I can think of. I'm continuing to struggle, and if I
get any achievements I will post it here. Any recommendations?
Thanks
--
Abdullah Ramazanoglu [ aramazanoglu AT demirbank DOT com DOT tr ]
------------------------------
From: [EMAIL PROTECTED] (Kent Kling)
Subject: Free Long Distance
Date: Wed, 28 Jul 1999 19:10:35 GMT
Reply-To: [EMAIL PROTECTED]
Check out the editor's page @ http://www.economite.com/editor.html for
more information
------------------------------
From: Jerry Walter <[EMAIL PROTECTED]>
Subject: Re: MS Access and Linux
Date: Wed, 28 Jul 1999 15:55:12 -0400
Though I am not sure, but I seem to remember that there was a
administrator out there for this purpose. I believe it was called the
Access Workgroup Administrator. This created a file that specified what
users and domains could attache to a access db.
HTH
Jerry Walter
Steve Porter wrote:
>
> HELP! I have a client who has a small (10 user) peer-to-peer MS Win98 network,
> with each node having MS Access97 installed. They keep a database on one of the
> Win98 w/s, and multiple people can access the db file at the same time. So here
> I come along and sell them a Linux server. I set up Samba, set up user names,
> logons, p/w, permissions, etc. I created a Public deirectory on the Linux
> (Mandrake 6.0) server and copied the Access database file over to the Linux
> Public folder. Tested each users connectivity to the server and specifically to
> the Access db file. But here's the catch...only one user can access the file at
> a time. When a second user tries to access the file, that user gets a message
> similar to "the number of allowable users has been exceeded...". Again this did
> not happen when the same file was residing on a directory on the Win98 station.
> I'm clueless.....any ideas?
>
> Steve
--
================================================
Jerry Walter
Bentley Systems, Inc.
Geoengineering Business Group
[EMAIL PROTECTED]
(937) 332-0053
================================================
------------------------------
From: "John G. Drummond" <[EMAIL PROTECTED]>
Subject: Re: DHCPd
Date: 28 Jul 1999 17:39:44 GMT
Peter Buelow <[EMAIL PROTECTED]> wrote:
> Get a new copy from www.isc.org and rebuild it. It kinda looks like
> you have either a broken version or one that is looking in the wrong
> places for stuff.
I have tried this since my post. And a new problem raised its ugly head:
the compile dies, searching for ethernet.h.
Specifically:
======
blacksburg:/usr/src/dhcp/dhcpcd-1.3.17-pl9# make
gcc -O2 -fomit-frame-pointer -Wall -malign-loops=2 -malign-jumps=2
-malign-fun
ctions=2 -I. -c buildmsg.c
In file included from buildmsg.c:26:
client.h:26: net/ethernet.h: No such file or directory
make: *** [buildmsg.o] Error 1
=======
I am running glibc2.0.7pre6. If I copy ethernet.h from
/usr/src/linux/include/net/ to /usr/include, it bombs thus:
=============
In file included from buildmsg.c:26:
client.h:132: `ETHER_ADDR_LEN' undeclared here (not in a function)
client.h:158: field `ethhdr' has incomplete type
buildmsg.c: In function `buildDhcpDiscover':
buildmsg.c:51: `ETHER_ADDR_LEN' undeclared (first use this function)
buildmsg.c:51: (Each undeclared identifier is reported only once
buildmsg.c:51: for each function it appears in.)
buildmsg.c:53: `ETHERTYPE_IP' undeclared (first use this function)
buildmsg.c: In function `buildDhcpRequest':
buildmsg.c:126: `ETHER_ADDR_LEN' undeclared (first use this function)
buildmsg.c:128: `ETHERTYPE_IP' undeclared (first use this function)
buildmsg.c: In function `buildDhcpRenew':
buildmsg.c:203: `ETHER_ADDR_LEN' undeclared (first use this function)
buildmsg.c:205:e this function)
make: *** [buildmsg.o] Error 1
========
So, I guess I have the wrong ethernet.h. I am not a developer, so
I'm not sure what else to check. Any suggestions?
Thanks,
-John Drummond
--
-=-=-=-
<[EMAIL PROTECTED]> http://falcon.jmu.edu/~drummojg/
"It is better to be wrong than to be vague."
--Freeman Dyson
------------------------------
From: Matthew <[EMAIL PROTECTED]>
Subject: Re: e-mail problems
Date: Wed, 28 Jul 1999 21:03:56 +0100
Hi,
I can't help you with the programs you are using, but I have done a
similar thing myself. I gave up with sendmail - it is a bummer of a beast
to set up: try exim from www.exim.org - works wonderfully with practically
no configuration to do (at least it's understandable!). Fetchmail was the
same. For the Pop-3 server I use qpopper - you just install and run it and
it does the rest. I can't remember it's web site, but I think it comes
from a company called Qualcomm.
If what you say is that you can telnet to port 110 on the server and
access the mail (i.e. the Linux software is working fine), then it could
be several things...
Is TCP/IP up on the machine that runs Outlook? can you ping the Linux box?
Are you using Outlook or Outlook Express? Outlook Express is relatively
easy to set up for internet e-mail, Outlook is not. If you're using
Outlook then you need to add the internet protocol support to it. It is in
the help somewhere, but I can't remember where.
Do any other e-mail clients work (i.e. Pegasus mail or Pine on another
Linux box after setting up multiple inboxes)?
Hope this helps.
Regards,
Matthew.
On Fri, 23 Jul 1999, C. Andrew Zook wrote:
>I am trying to set up an e-mail system on our 20 user network. I have
>installed Sendmail and ids-pop3d and Fetchmail. I can send mail to the
>server without a problem, I can send mail to users without a problem. The
>mail is definately on the server because I can manually get into ids-pop3d
>and see that there are messages awaiting delivery. I just can't get my
>client e-mail application (outlook) to connect!
>
>I don't know which program I should be configuring here! I have Fetchmail,
>Sendmail, and ids-pop! My question is..... what programs should I have! I
>have been unable to find any how-tos to tell me exactly what I need to meet
>my requirements.
>
>Thanks!
>
------------------------------
From: "Zorlu Yusuf" <[EMAIL PROTECTED]>
Subject: Linux Server is closing files opened by BDE or Microsoft-Word, etc
Date: Wed, 28 Jul 1999 22:04:11 +0200
Since I have a linux-Server with Samba, i'am havin some exeptions by some
applications like
app. working together with bde or app. like winword. I think, that linux is
closing the files, opened
by that application.
What is to do, in this case?
thanks!
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (William Calnan)
Subject: Re: help! very slow ethernet IN ONE DIRECTION
Date: Wed, 28 Jul 1999 19:05:15 GMT
Hi, I had a friend that had a similar problem, as it turned out, he had
installed one of those maxmtu type programs on his windows machine.
After uninstalling the defaults it fixed his problem.
Hope this helps.
Simeon Cran ([EMAIL PROTECTED]) wrote:
: Further to my previous post... I am getting very slow transfers
: (~7kbytes/sec) from linux to windows, but normal (>500kbytes/sec) from
: windows to linux. This is surely a config problem.. right? But where?
: One more hint: dmesg is reporting "ARP: arp called for own address" over and
: over again.
: OK, so what do I need to do to fix all this???? It all works, just slowly in
: one direction.
: HELP!
------------------------------
From: Abdullah Ramazanoglu <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.setup
Subject: Re: PCI BIOS setup ?
Date: Wed, 28 Jul 1999 23:42:51 +0300
[EMAIL PROTECTED] wrote:
>
> In my 'dmesg' there's a line just above the 'eth0' setup that
> indicates 'PCI BIOS has not enabled this device'
> If that's correct, why does boot indicate eth0 is up OK ?? Pinging
> my net address works, but appears to be a local function (?) in that
> hub led's never indicate activity.
> Card address. node interrupt, and irq are recorded correctly
> Getting RH6 to recognize this SOHOWARE FastEthernet card is a real
> pain in the ass !! The tulip 0.89 driver appears ok, would upgrading
> the kernel be of any help in fixing the bios ?? Current = 2.2.5-15
> How/where can I reconfigure the bios?
> At this point don't know if I'm more stubborn or stupid as it'd been
> much cheaper and easier to upgrade Windoz . . . . .
For BIOS setting, interrupt your machine's PowerOnSelfTest procedure
(mine is interrupted by pressing DEL) to enter BIOS setting menu of the
PC. In some menu there should be an option "PnP aware OS" --> disable
this. (Linux IS PnP aware but the name of this option is misleading. It
should have read "You have ancient 16bit DOS apps")
You have not indicated your NIC. But those newly introduced $15 cards
usually need latest tulip driver. Here is what can be done:
1. If you haven't installed kernel-sources yet, you will find
"kernel-sources-2.2.5-15.i386.rpm" in CD. Install it:
- mount /mnt/cdrom
- cd /mnt/cdrom/RedHat/RPMS
- rpm -Uvv kernel-sources-2.2.5-15.i386.rpm (or use GnoRPM graphical
tool)
2. Download http://cesdis.gsfc.nasa.gov/linux/drivers/test/tulip.c
copy it as root to "/usr/src/linux/drivers/net" directory, and recompile
kernel. If you haven't done it before, read Kernel-HOWTO and related
chapter in RedHat Installation Guide. They are on the CD --> Open GnoRPM
and install all the docs you can find on CD)
3. From linuxconf --> Basic host information : Set your IP and ether
card config data. For a PCI card leave I/O port and Irq fields blank.
Good luck,
--
Abdullah Ramazanoglu [ aramazanoglu AT demirbank DOT com DOT tr ]
------------------------------
From: [EMAIL PROTECTED] (dead_grandmother)
Subject: Re: DHCP questions (cable modem)
Date: Wed, 28 Jul 1999 21:02:18 GMT
On 28 Jul 1999 20:17:48 GMT, [EMAIL PROTECTED] (Patrick Hulst)
wrote:
>Good day all!
>
>I'm using the @Home network. For whatever reason, my provider gives me a new
>IP address every day (or every other day or whatever). My Linux box (RH 6.0,
>2 NICs, 64mb ram, AMD K6-2 350) is acting as a IP Masq/forwarding box for my
>house and when they change the IP it no longer forwards any traffic. I can
>ping the new address from another PC on the network, but I can't get past it.
>
>Is there anything I should be doing? Is there another way to refresh my route
>tables automatically? (All the other computers inside my network use the
>internal NIC as their gateway).
>
>Anyone with ideas?
>
>Thanks in advance.....
hey... i'm setting up the same thing... wierd... same type of
computer, ram, processor, and version of RH...
having some problems... if you had some time... could you let me know
of of the tings you did to get it up and running???
it would be a great help...
------------------------------
From: "Zorlu Yusuf" <[EMAIL PROTECTED]>
Subject: Linux Server is closing files opened by BDE or Microsoft-Word, etc
Date: Wed, 28 Jul 1999 22:03:32 +0200
Since I have a linux-Server with Samba, i'am havin some exeptions by some
applications like
app. working together with bde or app. like winword. I think, that linux is
closing the files, opened
by that application.
What is to do, in this case?
thanks!
[EMAIL PROTECTED]
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Crossposted-To: comp.protocols.smb,linux.redhat.ppp,linux.redhat.misc
Subject: Re: Netscape scraps Limux!
Date: Wed, 28 Jul 1999 13:20:46 -0600
Piotr Kaminsky wrote:
> Monte Phillips <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
> > Well Andrew, as a matter fact it does. AAnd if you'd give it some
> > thought you would see: That one netscape is the primary browser in the
> > *nix world. Being such it is also the primary source by which many
> > access SAMBA files.
> > If Netscape screws linux then samba will suffer because many potential
> > users of *nix will say to hell with it and stay with windows. Andrew
> > I know you are a longtime unix man, to bad really since it gives you a
> > myopic view on the world.
> > I realize unix has been a cozy little clique for decades, but the
> > future requires recruits, good, smart recruits. Anything which may
> > hinder that effort is or should be a concern of all aspects of this
> > effort.
> > You may think little of linux, but it is the talent pool of the future
> Interesting... Same thing was said about OS/2.
Indeed it was, but remember at the time OS/2 was first
released it had very high hardware requirements and was
quite costly to purchase. Linux has neither of these
shotcomings. By the time reasonably-priced hardware could
run OS/2 effectively, Microsoft had already entrenched
Windows with their now-illegal exclusionary preload
licensing pacts.
> Which will prevail, in your opinion?
Only time will tell, but linux appears to have the momentum
to make an end-run around Microsoft in the next few years.
--
-John ([EMAIL PROTECTED])
------------------------------
From: Matthew <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: root login problem with RH6.0
Date: Wed, 28 Jul 1999 21:12:35 +0100
Hello,
I haven't tried it with RH6, but have you tried removing the
/etc/securetty file? Just move it to securetty.old or something and then
try the telnet again. As for security...
Regards,
Matthew
On Fri, 23 Jul 1999, CNelson wrote:
>Hi,
>
>Any body know how to enable root login from a telnet session in RH6.0?
>
>Yes, I knew you can add /dev/ttyp? or something similiar to
>/etc/securetty. It works for most case, likes RH5.2 or other
>distribution but just not work for RH6.0.
>
>Does anybody using RH6.0 and know how to get it?
>
>Thanks
>
>Nelson
>
>
>
------------------------------
** 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
******************************
