Linux-Networking Digest #279, Volume #11 Tue, 25 May 99 14:13:54 EDT
Contents:
knfsd doesn't follow mounts? ("Bart N. Locanthi")
Re: Samba and NT Domain ("Aaron Weiker")
Desktop ("v4cal")
Does one bad address spoil all the rest? ("Mark")
Re: I'm New in the linux world (Matt Starnes)
Re: IP Masquerading and pop3/ftp bigger then 30k (Duncan Simpson)
Re: Linksys Etherfast 10/100 setup problem (Jim Knoll)
Two 3c509B's --> problems.. (Ville Nummela)
Re: PPP with MSCHAP - hint (Clifford Kite)
sendmail question. ([EMAIL PROTECTED])
NAT timeouts ([EMAIL PROTECTED])
unwritable shared folder (bencecil)
Re: All I want for Christmas is a ppp internet connection ("Gene Heskett")
Re: Building a Virtual Comport (Henrik Carlqvist)
Firewall ("Henrik Krogh")
NIS sucks (Linux implemantion too) (Dirk Wetter)
How to get samba&win98 talking ("Jeremy Harrison")
Anyone using xinetd? Howto... ([EMAIL PROTECTED])
Re: Installing Intel Etherexpress Pro 10 ISA with RH 6.0 (J�n Gu�mundsson)
Re: What ports does SSH use? ([EMAIL PROTECTED])
----------------------------------------------------------------------------
From: "Bart N. Locanthi" <[EMAIL PROTECTED]>
Subject: knfsd doesn't follow mounts?
Date: Tue, 25 May 1999 09:05:03 -0700
the new, improved linux nfs isn't letting clients see mounted file systems
for me.
for example, if /usr is a mounted file system, then a client who mounts /
via nfs can't see anything in /usr.
this can't be right - it's either a bug or a config problem.
anyone got an answer?
------------------------------
From: "Aaron Weiker" <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc,linux.samba,comp.protocols.smb
Subject: Re: Samba and NT Domain
Date: Tue, 25 May 1999 12:10:26 -0400
I was able to get this working and the key to make it all work was that I
had to make sure that Samba use encryption. Any Windows NT machine that has
sp3 or greater on it can only connect with SMB when encryption first and
then plain text second, there is a registry key you can change to do this
but that would require going to every NT machine to do it. also you need to
make sure that the security=DOMAIN. As far as automation for creating a
matching set of passwords on the linux box I am not aware of any such thing.
I do know that you can create dumby accounts on the linux box and the
passwords don't need to match what is on the NT Domain as long as the
username matches. Also be sure that the smbusers file is set up for
encryption. There is a file that comes with Samba that can automatically
convert the passwd.conf file into the correct samba format. The directions
for that are in the doc's under encryption.
[EMAIL PROTECTED] wrote in message <7idctn$7a6$[EMAIL PROTECTED]>...
>I've read the MAN pages, and the samba howto, but I've still got some
>questions about how to make it work with an NT Domain. I've got an
>NT PDC, and two NT BDC's. I'm trying to set up my linux box so that
>when I try to connect to a SAMBA share, it checks the name and password
>against the NT password. I've tried setting security=server, but when I
>do that I can't even browse the shares on the samba box. I think
>this may be a problem with the guest account, but I don't know what
>it is. When I set it to security=domain, every user needs to have an
>account on the linux box, but I don't know how to automate that. I
>understand that there is a way to add something to smb.conf to make it
>work, but I didn't understand other posts on how to do that. I'll
>attach a copy of my smb.conf. Are there any SAMBA gurus who can help me
>start getting rid of NT on my network? Thanks,
> Greg.
>
>P.S. sorry for the cross post, but I need to get this figured out by the
>end of the week or the server will be made into NT...
>
>
>
>
>;********************section global*****************
>[global]
>workgroup = ntdom
>comment = greg's attemt at SAMBA
>strict locking = no
>share modes = yes
>password server = primus
>local master = no
>security = SERVER
>encrypt passwords = yes
>wins support = no
>os level = 0
>domain master = no
>prefered master = no
>preserve case = yes
>netbios name = linux
>case sensitive = no
>printing = bsd
>printcap name = /etc/printcap
>load printers = False
>print command /usr/bin/lpr -r -P %p %s
>create mode = 0755
>;********************section homes******************
>[homes]
>comment = Home Directories
>browsable = True
>writable = yes
>read only = no
>preserve case = yes
>short preserve case = yes
>;********************section userdata***************
>[userdata]
>comment = All userdata that you are allowed to see.
>path = /home
>writable = yes
>create mode = 0770
>;********************section gregsplace*************
>[gregsplace]
>browsable = yes
>comment = This is a test share
>path = /root
>writable = no
>
>
>
>
>
>--== Sent via Deja.com http://www.deja.com/ ==--
>---Share what you know. Learn what you don't.---
------------------------------
From: "v4cal" <[EMAIL PROTECTED]>
Subject: Desktop
Date: Tue, 25 May 1999 09:16:57 -0700
i want to know if there is a way that i can have my desktop on windows
follow me using linux
like if i log on to a system on my network i get my desktop and if i go to a
differnt system on my net work it will show up when i log on
------------------------------
From: "Mark" <[EMAIL PROTECTED]>
Subject: Does one bad address spoil all the rest?
Date: Tue, 25 May 1999 09:23:52 -0700
I'm running Sendmail 8.9.3-1 on a RedHat 5.2 machine with about 250 users.
Most of the user's have Outlook or Outlook Express as a mail client. The
'problem' is that when a local user sends an e-mail to several addresses,
one bad address causes all the other addresses to fail as well. For
example, someone sending a message to 25 good addresses and 1 mistyped local
address will get an "undeliverable error" for all the addresses, not just
the incorrect one.
Is this a function of Sendmail that can be changed?
Thanks,
Mark
------------------------------
From: Matt Starnes <[EMAIL PROTECTED]>
Subject: Re: I'm New in the linux world
Date: Tue, 25 May 1999 11:56:17 -0500
Wow, can I have one?
Brandon wrote:
> Khaled Al-Thukair wrote:
> >
> > Hi
> > I'm new to the linux world
> > If Some one Can direct me to the starting point in it
> > what I have right now is:
> > PII 450 with 128 ram and 10 gb hd
> > PII 350 with 128 ram and 4.3 gb hd
> > AMD 300 with 6.4 gb hd and 64 ram
> > and A 486 DX 66 with 400 MB HD and 16 RAM
> And what exactly are we supposed to use this info for?
>
> --
>
>
> "Bill Gates?, I dont know any Bill Gates. Oh, you mean 'by putting
> every conceivable
> feature into an OPERATING SYSTEM, whether you want it or not, is
> innovation' Bill
> Gates? Yeah, I know the monopolizer"
>
> http://web.mountain.net/~brandon/main.htm
> For Beginners in Linux, Emulation, Midis, Playstation Info, and
> Virii.
------------------------------
From: [EMAIL PROTECTED] (Duncan Simpson)
Subject: Re: IP Masquerading and pop3/ftp bigger then 30k
Date: 25 May 1999 15:11:25 GMT
In <7ibvms$ns6$[EMAIL PROTECTED]> "Ryan Yetter" <[EMAIL PROTECTED]> writes:
>Ok, new problem that no one I know who's running it has. My ftp isn't
>working right on files larger then 30k....and my email client (Outlook and
>others I've tried) fails to download messages larger then 30 k. I'm running
>linux 2.2.9 with ip masquerading (everything else with the masquerading
>works fine, ie telnet, http, etc) and running the internet connection across
>a dynamic ip PPP connection. Now, like I said, everything but ftp and pop3
>seem to not mind about the size of files. Any help appreciated.
I *think* this is a server side issue and therefore not something you
can fix. I hit trouble with one week of bugtraq in a POP
mailbox. After my ISP moved to a new domain name and mail server I
managed >1 month of bugtraq with *no* trouble (it did take a while
though). If you are in realy freindly terms with the admin he should
be able to put it somewhere you can ftp your mailbox and the rest is
easy enough with a minimal perl script (just split the messages apart
using the From without a colon lines and feed them to port 25, or
whatever).
30k is a little small but speak POP 3 manaually to the server and see
what happens. The ftp issue is odd unless you are using alleged ftp
servers with bandwidth limits that cut your connection and only
support the rest command with offset 0 (which makes it rather
pointlesss). ftp.microsoft.com features such a ftp server (and many
other sites using NT). The only fix known is using a faster network
connection or uppgrading the remote software to something decent
(wu-ftpd, as used by various major archives, is a good quality ftp
daemon that is a few tads cheaper than IIS).
--
Duncan (-:
"software industry, the: unique industry where selling substandard goods is
legal and you can charge extra for fixing the problems."
------------------------------
Subject: Re: Linksys Etherfast 10/100 setup problem
From: [EMAIL PROTECTED] (Jim Knoll)
Date: Tue, 25 May 1999 12:16:54 GMT
[EMAIL PROTECTED] writes:
>I'm trying to set up my Linksys Etherfast 10/100 to run on a Linux box
>but can't seem to get Linux to recognize it. When the machine boots up
>I get these messages:
>% Setting up Network Device eht0
>% SCIOSIFADDR: No such device
>% SCIOSIFBRDADDR: No such device
>% SCIOSIFNETMASK: No such device
>[Using SuSE 5.1 kernel 2.0.33]
>/proc/pci gives the following:
>Bus 0 device 9 function 0:
>Ethernet controller: Unknown vendor unknown device (rev 32)
>Vendor id=11ad. Device id=2 Medium devsel. Fast back-to-back capable.
>Irq 11. Master Capable. Latency=64 I/O at 0xe800 Non-prefetchable
>32-bit memory at 0xe5000000
>I compiled the tulip driver and when I do "insmod tulip", I get this
>message:
>% /lib/modules/2.0.33/net/tulip.o: couldn't find the kernel version the
>module was compiled for.
>Any ideas about what's causing the problem? I'd really appreciate some
>help. Thanks in advance.
>Jay Oram
>--== Sent via Deja.com http://www.deja.com/ ==--
>---Share what you know. Learn what you don't.---
I just installed RH 5.2 and had a LinkSys Etherfast 10/100
card installed in the PCI slot. I installed the server selection
of RH and when it promped me for NIC I selected Digital (Tulip).
RH installed the NIC and everything seems to work okay so far.
Don't know if this helps or not...
Jim
[EMAIL PROTECTED]
------------------------------
From: Ville Nummela <[EMAIL PROTECTED]>
Subject: Two 3c509B's --> problems..
Date: Tue, 25 May 1999 19:02:38 +0300
I have a problem with two 3c509B Cyclones; If I plug in just one card,
everything works fine. If I insert another card, linux says it has found
one card but it doesn't work.. any ideas..?
--
| ViGe / gasp inc. | http://www.lut.fi/~vnummela | [EMAIL PROTECTED] |
| It Has To Work. - RFC 1925 |
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: PPP with MSCHAP - hint
Date: 25 May 1999 11:09:58 -0500
Richard Birchall ([EMAIL PROTECTED]) wrote:
: The remotename option is needed because the NT RAS server does not give
: its name in the MS-CHAP challenge.
Thanks. I didn't know about that NT-ism.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* Better is the enemy of good enough. */
------------------------------
From: [EMAIL PROTECTED]
Subject: sendmail question.
Date: 25 May 1999 10:02:57 -0700
------------------------------
From: [EMAIL PROTECTED]
Subject: NAT timeouts
Date: Tue, 25 May 1999 16:17:28 GMT
I am using NAT over a ppp device and if a connection
behind the NAT is idle for a short while (not sure on
exact timeout), that connection goes dead. Note,
the ppp session is still open, and connections that
are not translated (ie connections from my linux
machine to the world) do not go dead when idle.
Anyone know how to set the timeouts via ipchains, or
is it set some other way? Thanks, Bill
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
From: bencecil <[EMAIL PROTECTED]>
Subject: unwritable shared folder
Date: Tue, 25 May 1999 23:28:40 +1000
Hi there,
I have set up a Linux (Redhat 5.2) file and print server at home
attached to which are two windoze95 machines. I am trying to establish
a shared directory to add to the current home directories that the win95
machines map to when they boot up.
No matter how hard I try, I cannot seem to get write access to the
shared folder... my smb.conf file looks like this...
----[snip]----
[shared]
comment = common file space
path = /shared
valid users = bob ted
writable = yes
----[snip]----
I have tried adding...
public = yes
browsable = yes
only guest = yes
create mask = 0765
... all with no change.
Can someone help me create a *writeable* shared directory please?
Thank you in advance
Ben Cecil
------------------------------
Date: 25 May 99 09:03:07 -0500
From: "Gene Heskett" <[EMAIL PROTECTED]>
Subject: Re: All I want for Christmas is a ppp internet connection
Gene Heskett sends Greetings to Sandwalker ;
> Actually, I want it now.
Here, I've cleared out all entries in the chat window of that option in
linuxconf, and I believe it would show the same in netcfg. Do have the
proper entries in the pap area.
Then you will probably have to add in the probably blank ppp0 options
window, 'asyncmap 0x000a0000', without the quotes. This will cause ppp
to negotiate a link with the xon/xoff characters 'escaped' and they will
not then cause un-explained lockups.
Be sure you have their DNS machines numeric address in resolv.conf too.
Cheers, Gene
--
Gene Heskett, CET, UHK |Amiga A2k Zeus040 50 megs fast/2 megs chip
Ch. Eng. @ WDTV-5 |A2091,GuruRom,1g Seagate,CDROM,Multiface III
<[EMAIL PROTECTED]> or |Buddha + 4 gig WDC drive, 525 meg tape
<[EMAIL PROTECTED]>|Stylus Pro, EnPrint, Picasso-II, 17" vga
RC5-Moo! 22kkeys/sec isn't much, but it all helps
--
------------------------------
From: Henrik Carlqvist <[EMAIL PROTECTED]>
Subject: Re: Building a Virtual Comport
Date: Mon, 24 May 1999 21:39:22 +0200
"Jim D." wrote:
> According to me, a great solution would be to create a virtual com
> port, with which pppd will talk instead of the real one. A program
> would redirect all data recieved on the virtual com to the real one,
> Does anybody find good this idea, think it's possible, and can
> giving me advises to build this application ? Does anybody have an
> other (better or not) solution ?
It would be possible to patch the kernel to get such a
/dev/virtual-comport, but I think it would be easier to insert an extra
computer between the modem and computer. Then you could run your program
in this extra computer listening to both com-ports and sending input
from one com-port to the other com-port. That is something like this:
Without debugging computer With extra debugging computer
========================== =============================
============ ========= ============ ============= =========
I Computer I I Modem I I Computer I I Debugger I I Modem I
I I I I I I I I I I
I com2:I----I I I com2:I---Icom1: com2:I--I I
I I I I I I I I I I
============ ========= ============ ============= =========
regards Henrik
--
spammer strikeback:
root@localhost [EMAIL PROTECTED] [EMAIL PROTECTED]
[EMAIL PROTECTED] [EMAIL PROTECTED] [EMAIL PROTECTED]
------------------------------
From: "Henrik Krogh" <[EMAIL PROTECTED]>
Subject: Firewall
Date: Tue, 25 May 1999 18:52:10 +0200
I am installing af linux-server (RH 5.2) as a firewall/proxy, according to
the How-to.
I have installed 2 nic's in the server, one with an IP that is on the
internet and one with IP that is on our LAN.
The linux-server can see the internet (I am able to ping various
internet-adresses) and the LAN (I am able to ping local machines).
The computers in the LAN can ping both nic's in the linux-server, but not
internet-adresses.
I have compiled the kernel as decribed in the How-to and Ipforwarding is
turned on!
Hope you have an answer
Henrik Krogh
------------------------------
From: Dirk Wetter <[EMAIL PROTECTED]>
Subject: NIS sucks (Linux implemantion too)
Date: Tue, 25 May 1999 15:32:16 +0200
hi folks,
NIS is somehow a bad thing, under Linux this is more true. I've got a
problem where I need some help.
What I am looking for is a NIS client implementation which is
comparable to the one of Slowlaris. This means one of the given server
goes down, the next one takes over the buisiness. When the last of the
specified servers is not available and the others are up again, it
should bind again to one of those.
ypbind 3.3 of Linux however offers two possiblities (correct me, if I
am wrong,but I couldn't find any docu on that): The failover mechanism
is the same (it only pings the server all the time), but if the
last of the given servers is down (the others are reachable again...)
it brodcasts in the subnet. The one who answer "hey, I can serve you
with a nice (maybe faked) passwd.byname map (and other maps...)" is
becoming the server. This is something I cannot install for security
reasons. Another idea is to start the client with -no-ping, the
broadcast mechanism doesn't work, but the failover doesn't work too!
Also if the ypbind should realize that he cannot connect to the server
(without pinging).
What I need is an easy to use solution for >200 PCs (most of them are
still runnning a libc5 based distribution).
Any ideas ?
bye,
~dirkw
_________________________________________________________
Dirk Wetter @ DESY (IT division/UNIX), Bldg. 2B/214
Notkestr. 85, D-22607 Hamburg, GERMANY
+49-(40)-8998-4667,fax:-4429 http://www.desy.de/~dirkw
PGP pub key available mailto:[EMAIL PROTECTED]
------------------------------
From: "Jeremy Harrison" <[EMAIL PROTECTED]>
Subject: How to get samba&win98 talking
Date: Tue, 25 May 1999 23:37:21 +1000
I have a Rh5.2 box and a win98 box. I now have it so that I can see the RH
box under network neighbourhood, but when i try to connect I get
'\\linuxserver is not accessable. the specified computer did not recieve
your request. Try again later'
I havent tried connecting the linux box to win98 shares yet.
is this anything to do with encrytped passwords?
i am a bit lost here, am surfing the net looking for clues...
Also which file controls the number of root telnet sessions allowed? I can
login as anything except root from my win98 box....
thanks in advance
Jeremy
------------------------------
From: [EMAIL PROTECTED]
Subject: Anyone using xinetd? Howto...
Date: Tue, 25 May 1999 16:57:05 GMT
Please email reply!!!
Ok, I'm new to xinetd. I got the source, compiled, installed and now
what??? I think I read to convert inetd.conf to xinetd.conf. OK. Now how
do I get xinetd to start at boot?
Vitals:
RedHat 5.1
Kernel 2.2.1
xinetd-2.1.8.6b5
I copied /etc/rc.d/init.d/inet to <same>/inet.o and edited inet. Changed
all instances of inet to xinet. Now on boot I get something like:
Starting XINET: xinetd nice: no such file or directory
This message is incomplete since I can't write that fast...
I can go back to inetd, but I would like to lick this problem. any
suggestions? directions. What would, if you will, would the entry in
/etc/rc.d/init.d/<entry> be?
Thanks a bunch!!!
-Wes Yates
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
Date: Tue, 25 May 1999 08:08:13 +0000
From: J�n Gu�mundsson <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: Installing Intel Etherexpress Pro 10 ISA with RH 6.0
Jack Richins wrote:
>
> I'm installing RH 6.0 and I'm having trouble getting my Intell EtherExpress
> Pro 10 detected. Autodetection hangs. I copied IRQ and I/O port information
> from Windows, which detects it fine. However, the RH manual shows a MEM
I had the same problem with Slackware;
my problem was that although my ethernetcard did show up on the
right IO port and IRQ it didn't work. I usually tried to put it in with
the command
#insmod eepro io=0x300 irq=10 (it's loaded as a module)
but nothing worked
Now I can trick it in by doing these commands
#modprobe eepro io=0x200 irq=10
(device or resource busy)
#modprobe eepro io=0x300 irq=10
#rmmod eepro
#modprobe eepro
#modprobe eepro io=0x300
#/etc/rc.d/rc.inet1 start
It looks like I have to try first to put the card into wrong IO-port
then the right one, pull it out and put it in again without giving up
the IRQ- adress
It's a tricky thing to do but I put the whole procedure into
a script and it works now...
--
The idea that Bill Gates has appeared like a knight in shining armour
to lead all customers out of a mire of technological chaos neatly
ignores the fact that it was he who, by peddling second-rate
technology, led them into it in the first place. (Douglas Adams)
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: What ports does SSH use?
Date: Tue, 25 May 1999 17:37:55 GMT
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (mike dombrowski) wrote:
> As the above says what ports does ssh use? Is it just 23? I thought I
> read somewhere that there were others. I need my sysadmin to open them
> on the firewall. Thanks
There are at least two ways to figure this out:
-do a `netstat -a` and look for open ports
-look in /etc/services
Both methods will tell you that sshd listens to port 22 [NOT 23], and
only to port 22.
HTH
--
-Bill Clark
Systems Architect
ISP Channel
http://locale.ispchannel.com/
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
** 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
******************************
