Linux-Networking Digest #848, Volume #10 Wed, 14 Apr 99 00:13:40 EDT
Contents:
RedHat 5.2 and Samba on a Server, And a Win/95 Box...Help (Michael Myers)
Re: TCP Stalls After ~100k (jedi)
Re: 3Com 3C900 Broke With Kernel 2.2.3 Upgrade (help please!) (Howard Au)
Re: linux/windows95 /98 (mist)
CallerID and mgetty ("Mike Redrobe")
Re: Need help upgrading sendmail (mist)
Re: Noticing a pattern: Red Hat + ethernet + 3c509 family = net (Andy Osborne)
Re: Port using by realaudio and irc (Joern Smock)
Re: really slow 100Mb ethernet ("David L. Courtney")
Plz help me!! pppd doesn't work. (Minjung Kim)
Re: Hooking a LAN onto the net using a linux box! (Jim Richardson)
TCP Stalls After ~100k (Eric Hancock)
Re: really slow 100Mb ethernet (Leslie Mikesell)
How to redirect external traffic to a server (Arash)
Re: linux version (Frank Sweetser)
Re: .shosts: Your host key cannot be verified: unknown or invalid host key? (Neil
Rickert)
Re: PPP connections problem with RedHat 5.2 / debugging problems ("Lee Howes")
Strange FTP problem ("Rich Simons")
----------------------------------------------------------------------------
From: Michael Myers <[EMAIL PROTECTED]>
Subject: RedHat 5.2 and Samba on a Server, And a Win/95 Box...Help
Date: Tue, 13 Apr 1999 14:43:55 -0400
I have a 486/100 HP NetServer LF with 56M ram, 4 gig HD, and 3com 10/100
Card.
hooked to a 3com superstack II Hub, along with a Win/95 Box, 64Meg, blah
blah blah,
My Problem, I can FTP, I can Telnet, I can Ping Both ways, and even see
the Apache Web server from my 95 Box, How or what do I need to do so I
can log in from the 95 box, (want to set up users) and have Drive
Mappings, ......Is this possiable?
I am an IS Manager in a NT Enviorment, and I am a Unix\Linux
Virgin.......I have Bought Samba Intergrating UNIX and
Windows.......reading it I was lost.
how do I edit the SMB.conf file in REdHat 5.2?
------------------------------
From: [EMAIL PROTECTED] (jedi)
Crossposted-To: comp.os.linux.setup
Subject: Re: TCP Stalls After ~100k
Date: Tue, 13 Apr 1999 11:42:49 -0700
On Tue, 13 Apr 1999 14:21:51 -0400, Eric Hancock <[EMAIL PROTECTED]> wrote:
>
>I'm having trouble with a Linux box and the network. The box is a P100,
>vanilla RedHat 5.2 / kernel 2.0.36.
>
>We can't seem to transfer files that are larger about 100k -- any transfer
>larger seems to stall. Eventually the client times out. We're having this
>trouble with FTP, SSH, SMTP, etc... I don't see anything in messages.
>
>I'm not sure where to look for this one. Let me know if this rings a bell
>-- web sites / FAQs are OK.
How does the effective thruput of the NIC under linux fare
next to the rest of the traffic on the network.
I had a cable modem + ISA NIC problem due to the amount
of pointless broadcast traffic going over my provider's
wires.
--
"I was not elected to watch my people suffer and die |||
while you discuss this invasion in committe." / | \
In search of sane PPP docs? Try http://penguin.lvcm.com
------------------------------
From: Howard Au <[EMAIL PROTECTED]>
Subject: Re: 3Com 3C900 Broke With Kernel 2.2.3 Upgrade (help please!)
Date: Tue, 13 Apr 1999 19:20:42 -0700
Thanks, I'll give it a try.
Chris Cantwell wrote:
> Perhaps this driver was compiled in the old kernel, and now it is configured
> as a module. Have you setup the conf.modules properly?
> alias eth0 3c5x9 (I think this is right, you may want to check the driver
> name)
>
> did you rerun lilo?
> /sbin/lilo -v
>
> If all else fails, try recompiling the kernel with the 3c5x9 module, and
> setup conf.modules as above. Rerun lilo, and retry. See the Kernel-HOWTO.
>
> Chris Cantwell
>
> Howard Au wrote in message <[EMAIL PROTECTED]>...
> >I'm getting very desperate. I installed Mandrake Linux 5.3 (Kernel
> >2.0.36) for the first time a couple weeks ago and have been using it
> >problem free. But I recently installed the RPM's for the Kernel
> >2.2.3-ac4 upgrade using kpackage, updated the lilo.conf file, and the
> >new kernel booted just fine, except that my ethernet card has suddenly
> >stopped working. The "using DHCP for eth0" line found during the boot
> >process now says "Failed" after a few seconds. It's a PCI card, 3Com
> >3C900-TPO connected to a RoadRunner cable modem (uses DHCP server). When
> >I upgraded I installed all the RPM's available on the FTP site,
> >including the DHCP update, and as far as I can tell they all installed
> >with no problem.
> >
> >The card is detected properly during the boot and has no conflicts with
> >any other hardware. Here's some of the output from dmesg:
> >
> >-----------
> >3c59x.c:v0.99H 11/17/98 Donald Becker
> >http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> >eth0: 3Com 3c900 Boomerang 10baseT at 0xff00, 00:a0:24:d1:45:19, IRQ 10
> >
> >8K word-wide RAM 3:5 Rx:Tx split, 10baseT interface.
> >Enabling bus-master transmits and whole-frame receives.
> >-----------
> >
> >So I went to the /var/log/message file and had a look through it for any
> >info
> >that it might give about why it's not working... this is what I found
> >that was
> >related to my network card:
> >
> >-------------
> >Apr 9 20:47:05 COOLIE kernel: 3c59x.c:v0.99H 11/17/98 Donald Becker
> >http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> >Apr 9 20:47:05 COOLIE kernel: eth0: 3Com 3c900 Boomerang 10baseT at
> >0xff00, 00:a0:24:d1:45:19, IRQ 10
> >Apr 9 20:47:05 COOLIE kernel: 8K word-wide RAM 3:5 Rx:Tx split,
> >10baseT interface.
> >Apr 9 20:47:05 COOLIE kernel: Enabling bus-master transmits and
> >whole-frame receives.
> >Apr 9 20:46:32 COOLIE kerneld: Starting kerneld, version 2.1.121 (pid
> >141)
> >Apr 9 20:46:32 COOLIE kerneld: kerneld startup succeeded
> >Apr 9 20:46:32 COOLIE kerneld: started, pid=141, qid=0
> >Apr 9 20:46:32 COOLIE ifup: SIOCADDRT: Network is unreachable
> >Apr 9 20:46:34 COOLIE network: Bringing up interface lo succeeded
> >Apr 9 20:46:34 COOLIE ifup: Using DHCP for eth0...
> >Apr 9 20:46:34 COOLIE dhcpcd[233]: ioctl SIOCSIFBRDADDR (ifConfig):
> >Cannot assign requested address
> >Apr 9 20:47:04 COOLIE ifup: failed.
> >Apr 9 20:47:04 COOLIE network: Bringing up interface eth0 failed
> >--------------
> >
> >I can ping my own IP, but anything else just doesn't work. I've tried
> >specifying addresses manually (nameserver, gateway, etc) using ifconfig
> >and netcfg in X, as well as Linuxconf but nothing works. I usually get
> >the "cannot assign requested address" or some other such message. If
> >anyone has any ideas as to how to solve this, I'd be very grateful :)
> >
> >Howard
> >
------------------------------
From: mist <[EMAIL PROTECTED]>
Subject: Re: linux/windows95 /98
Date: Tue, 13 Apr 1999 18:41:39 +0100
Reply-To: mist <new$[EMAIL PROTECTED]>
Dave Goss <[EMAIL PROTECTED]> scribed to us that -
>
>
<snip>
> 192.168.0.7 is the linux machine. i have been unsucessful in
> accessing the internet thus far from the linux machine and hope
> ther is some one out ther that could help me.
>
I think you'd be better off if you gave a bit more information about
what you're having trouble with.
>From what I can see, you'll be best off connecting using the linux box,
setting all the windows machines' default gateways to the linux box's IP
addy, and running IP masquerading on the linux box to connect the
network to the outside world. Read the ISP howtos and the firewalling
and masquerading howtos.
>
>[ A MIME image / gif part was included here. ]
>
... and don't post gifs to non-binary groups. Not even small blank
gifs.
--
Mist.
------------------------------
From: "Mike Redrobe" <[EMAIL PROTECTED]>
Subject: CallerID and mgetty
Date: 13 Apr 99 23:15:24 +0000
I'm in the process of moving all my netowrking related stuff over onto a
linux box, and have done the samba,ip-masq,diald and email stuff...
...but can't seem to get CallerID logged using mgetty.
I've put the init string (AT#CLS=8#CID=1) in
/etc/mgetty+sendfax/mgetty.config
It happily shows me CID in minicom, but never in mgetty ?
Surely the mgetty log is just echoing the modem responses
(I do get RING RING..)
I'm sure this is something stupid...
--
Mike
------------------------------
From: mist <[EMAIL PROTECTED]>
Subject: Re: Need help upgrading sendmail
Date: Tue, 13 Apr 1999 18:44:21 +0100
Reply-To: mist <new$[EMAIL PROTECTED]>
Ham Radio Op <[EMAIL PROTECTED]> scribed to us that -
>I am currently running 8.8.4 sendmail and would like to goto 8.9.3
>sendmail. I am running slackware 3.1 I believe. Is there a faq or
>can someone give me hints on how to do this??? Thanks.
There is substantial documentation included in the .tar.gz that you
download. Go to <URL:http://www.sendmail.org> and get hold of the
latest source. You might need to play around with the configuration
files but eventually everything should be fine. There's a large book of
operating instructions included in postscript format, which is nice.
--
Mist.
------------------------------
From: Andy Osborne <[EMAIL PROTECTED]>
Subject: Re: Noticing a pattern: Red Hat + ethernet + 3c509 family = net
Date: Tue, 13 Apr 1999 19:15:58 GMT
Eric Caldwell wrote:
>
I have noticed several problems with the 509 family in the past (I work
for a UK based Network Reseller). I find it best to ALLWAYS use the
3c5x9cfg.exe (I can never remember the correct file name) file from the
driver disks under dos to disable plug and prey and to set an IRQ and IO
manually.
This tends to give better results even when using Microtoss :-) NOS'.
Andy.
> I had the same problem recently and decided to take the easy way out. I went
> down to CompUSA and purchased a (PC) Linksys Etherfast 10/100 card and it's
> working beautifully and only cost $29.00.
>
> I think that my 3c509 wasn't totally PNP compliant which what was causing
> the problem. It work great until I upgraded my PC to an ABIT BX 2.0 MB and
> overclocked celery.
>
> Hope this helps.
>
> P.S. Linksys has Linux on it OS compatibility list on the box! Just use the
> Tulip driver and things work great.
>
> Nonya Bidness <[EMAIL PROTECTED]> wrote in message
> news:7egvdc$9m3$[EMAIL PROTECTED]...
> > Read some of the posts lately. I have been having a problem connected to
> > the internet. I dual boot and the net works fine with Win98. Linux has
> the
> > card recognized, it is activated. I can ping myself. I can NOT get out
> to
> > the net. I have the 3c509 card family and can't reach the nameservers.
> > unreachable???
> > what is the deal? I am headed to check Red Hat's page for any help...just
> > wanted to point this out if anyone is noticing PLEASE help us out...we are
> > becoming a mass epidemic.
> > Thanks!
> >
> >
------------------------------
From: Joern Smock <[EMAIL PROTECTED]>
Subject: Re: Port using by realaudio and irc
Date: 14 Apr 1999 01:00:35 GMT
"Spirou" <[EMAIL PROTECTED]> writes:
> Hy everybody !!!!
> What is the number of the port used by the irc and realaudio ????
Spirou,
for realaudio, it's TCP port 7070 and UDP ports 6970-7170. I don't
use irc.
Joern
------------------------------
From: "David L. Courtney" <[EMAIL PROTECTED]>
Subject: Re: really slow 100Mb ethernet
Date: Wed, 14 Apr 1999 02:27:44 GMT
dude wrote:
>
> Hello, everybody. I'm having yet another problem with my network, and I hope
> someone can give me some hint's here. The problem being: My lan is composed
> of several win9x, NT, IRIX's and a Linux, acting as a fileserver (plus other
> insignificant thingies). The setup is pure 100Mb, i.e. no switches. Now, the
> transfer rates -------
I am not an expert, but you said send any suggestions at all. So, what
about disabling the built-in ethernet business on the Compaq and
installing a card like the other cards on the network? At least that
might eliminate the card/driver as a problems source.
David C.
------------------------------
From: Minjung Kim <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Plz help me!! pppd doesn't work.
Date: Wed, 14 Apr 1999 11:56:29 +0900
Hi!
I'm newbie in linux.
I tried to connect to my ISP with ppp.
--- #>/usr/sbin/pppd /dev/modem 57600 debug connect "/usr/sbin/chat -v
`` AT OK ATD01432 CONNECT '\d\c'"
But the modem didn't dial and log message is below.
---
Apr 13 12:22:35 localhost pppd[341]: pppd 2.3.5 started by root, uid 0
Apr 13 12:22:36 localhost chat[342]: expect (AT)
Apr 13 12:23:21 localhost chat[342]: alarm
Apr 13 12:23:21 localhost chat[342]: Failed
Apr 13 12:23:21 localhost pppd[341]: Connect script failed
Apr 13 12:23:22 localhost pppd[341]: Exit.
Apr 13 12:24:35 localhost kernel: PPP: ppp line discipline successfully
unregistered
---
How can I do ??? Plz help me!!
My sytem is RedHat 5.2.
------------------------------
From: [EMAIL PROTECTED] (Jim Richardson)
Subject: Re: Hooking a LAN onto the net using a linux box!
Date: 14 Apr 1999 01:48:22 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 11 Apr 1999 21:13:10 GMT,
[EMAIL PROTECTED], in the persona of <[EMAIL PROTECTED]>,
brought forth the following words...:
>Hello!
>
>I wrote here once before conserning this, but it don't seem to be posted, so
>I'll try it again:
>
>Me and some friends have connected our 3 PCs (Windows 98) into a little LAN
>(Only using the standars TCP/IPX and IP stuff that comes with win98).
>
>Now we have been offered a Cable modem so we are planning to use a linux box
>as a router. the problem is:
>
>Exactly WHAT do we need to be able to get all 3 win98 machines connected to
>the Internet. I haven't got that much experience with linux, but I now some
>stuff.
>
>As far as I know, i need a machine with 2 NICs, and put that between the modem
>and the LAN:
>
> |------| |------| |-------|
> |Modem | <-----------> |Linux | <-----------> | LAN |
> |------| |------| |-------|
>
>The thing is: WHAT kinda stuff is needed on the linux box?
>
>Ip-masquerading?
>
>I'll get right to the case:
>
> I need HELP! :-)
>
>Tell me ALL kinda stuff that is needed to be installed, and are there any
>other HARDWARE needed (I can't possibly think of any)?
>
>All protocols, all programs, just letmeknow! :-)
>
>In advance, thanks ALOT!
>
>Odd Arne Beck
>informatics student
>
>-----------== Posted via Deja News, The Discussion Network ==----------
>http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Go straight to www.linuxrouter.org
They have all the info you need.
--
Jim Richardson
www.eskimo.com/~warlock
All hail Eris
"Linux, because a cpu is a terrible thing to waste."
------------------------------
From: Eric Hancock <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: TCP Stalls After ~100k
Date: Tue, 13 Apr 1999 14:21:51 -0400
Reply-To: [EMAIL PROTECTED]
I'm having trouble with a Linux box and the network. The box is a P100,
vanilla RedHat 5.2 / kernel 2.0.36.
We can't seem to transfer files that are larger about 100k -- any transfer
larger seems to stall. Eventually the client times out. We're having this
trouble with FTP, SSH, SMTP, etc... I don't see anything in messages.
I'm not sure where to look for this one. Let me know if this rings a bell
-- web sites / FAQs are OK.
Eric -
--
"Our aim is to make our standards ubiquitous..." Alec Sanders, Microsoft
Eric D. Hancock ................................... [EMAIL PROTECTED]
Web Home ................................ http://www2.cybernex.net/~edh/
Technical volunteering opportunities ......... http://www.voluntech.org/
------------------------------
From: [EMAIL PROTECTED] (Leslie Mikesell)
Subject: Re: really slow 100Mb ethernet
Date: 13 Apr 1999 21:57:24 -0500
In article <7f0dug$oh7$[EMAIL PROTECTED]>,
dude <[EMAIL PROTECTED]> wrote:
>Hello, everybody. I'm having yet another problem with my network, and I hope
>someone can give me some hint's here. The problem being: My lan is composed
>of several win9x, NT, IRIX's and a Linux, acting as a fileserver (plus other
>insignificant thingies). The setup is pure 100Mb, i.e. no switches. Now, the
>transfer rates between any 2 machines on the lan, excluding the Linux box, is
>in the range of 1-5 MB/sec, but all transfer to and from the Linux box is ca.
>200-300 KB/sec.
I saw about that speed once when someone accidentally set one end of
the interconnect between two switches to half duplex and the other
to full. Setting both to full put things back at 8+MB/sec.
>* Double checked/changed all hubs and cables to and from the patient. *
>Made sure neither routed nor gated are running * Forced full duplex / 100 Mb
>on the NIC (just in case) *
If you are plugged into a hub instead of a switch, you need to be
set to half duplex, but it should autodetect that.
>One thing which I don't understand (well, there are more then one, but who's
>counting :) is the frame errors I'm seeing when I ifconfig the nic:
>
>UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
>RX packets:73162 errors:387 dropped:0 overruns:0 frame:387
>TX packets:60522 errors:0 dropped:0 overruns:0 carrier:0
>collisions:7962
>Interrupt:9 Base address:0x6480
>
>What do these mean? Is there cause for alarm? (Note, these numbers are after
>ca. 4 hrs of very little activity)
You most likely have several cards that think they should be in full
duplex when the use of a hub actually forces half duplex. Or you
have cables that aren't up to cat 5 specs, or both.
Les Mikesell
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Arash)
Crossposted-To: de.comp.os.linux.networking,de.comp.os.linux.misc
Subject: How to redirect external traffic to a server
Date: Tue, 13 Apr 1999 22:25:59 GMT
Hi,
i would like to send my entire tcp/udp traffic to one server
in internet. This is because my ISP wants me to use a proxy server.
Otherwise i would have to pay by MB each month. (pls don't ask
for details on this)
I am using a linux server and one windows pc. Masquerading works
fine. The problem is that not every application uses a proxy server.
Using ICQ or RealAUdio/Video, i can see that the traffic goes
directly to the servers. And i want it to go through the proxy server.
The best way for me would be some kind of redirector just like
the way ipfwadm or ipchains work for internal redirection.
Thanx,
Arash :-)
------------------------------
From: Frank Sweetser <[EMAIL PROTECTED]>
Subject: Re: linux version
Date: 13 Apr 1999 17:14:21 -0400
"calvyn du toit" <[EMAIL PROTECTED]> writes:
> im looking for a version of windows that will run on a 486 with 4mb or ram,
> ive tried redhat 5.2 & debian but they hanging during the installation, any
> ideas
IIRC, slackware can be convinced to install in 4M. also the (fairly old -
ie, 1.2 kernel) Bogus distro should work.
--
Frank Sweetser rasmusin at wpi.edu fsweetser at blee.net | PGP key available
paramount.ind.wpi.edu RedHat 5.2 kernel 2.2.5 i586 | at public servers
Sam: What will you have, Norm?
Norm: Well, I'm in a gambling mood, Sammy. I'll take a glass of whatever
comes out of that tap.
Sam: Oh, looks like beer, Norm.
Norm: Call me Mister Lucky.
-- Cheers, The Executive's Executioner
------------------------------
From: [EMAIL PROTECTED] (Neil Rickert)
Crossposted-To: comp.security.ssh
Subject: Re: .shosts: Your host key cannot be verified: unknown or invalid host key?
Date: 13 Apr 1999 12:01:13 -0500
[EMAIL PROTECTED] (Georg Schwarz) writes:
>now when trying to log in from the one (helena) to the other (luthien),
>with an appropriate .shosts file being in place, I get:
>...
>helena.physik.tu-berlin.de: Remote: Your host key cannot be verified:
>unknown or invalid host key.
>helena.physik.tu-berlin.de: Remote: The host name used to check the key
>was 'helena.physik.tu-berlin.de'.
>helena.physik.tu-berlin.de: Remote: Try logging back from the server
>machine with the canonical host name using ssh, and then try again.
>...
>what exactly is the remote server complaining about?
>I'm not using an /etc/ssh/ssh_known_hosts on any machine. Might this be
>the problem?
It is looking for information which could be in either the system
'ssh_known_hosts' file, or in your $HOME/.ssh/known_hosts file. Each
system should have known_hosts information about the other. When you
login to system B from system A, that puts the information about B
into the 'known_hosts' file on A. Then login the other way, to put
the information needed into the 'known_hosts' file on B. Thereafter
you should be set for using '.shosts' validation.
------------------------------
From: "Lee Howes" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: PPP connections problem with RedHat 5.2 / debugging problems
Date: 13 Apr 1999 17:16:08 GMT
There should only be two double quotes, the first and last. All others
should be singles.
piezas <[EMAIL PROTECTED]> wrote in article <[EMAIL PROTECTED]>...
> Jean-Claude wrote:
>
> > I am trying to debug my ppp connection and I wonder why I can't get
> > connected from this line :
> >
> > /usr/sbin/pppd /dev/ttyS1 57600 debug connect "/usr/sbin/chat -v ''
> > 'AT&F0' OK ATD5555555 CONNECT '\d\c' "
> >
> > I am not sure where to place single or double quotes. That may be the
> > problem...
> >
> > Thanks
> >
> > Jean-Claude
>
> just try quickppp u will find it at >>>>linux.box.sk
> el will work perfect!!!!!!
>
>
------------------------------
From: "Rich Simons" <[EMAIL PROTECTED]>
Subject: Strange FTP problem
Date: Tue, 13 Apr 1999 23:46:58 GMT
I can't figure this one out. I can get files, large and small, just fine.
I can send small files, but I can't send large files (>~20k). Sending
large files is very erratic, that is, quick bursts and then long periods
of no activity. I suspect I am having the same problem with the small
files but they finish before the problem occurs.
It seems like a timing problem, packet size problem or possibly a
configuration problem. My nic (generic ne2000) seems to work fine
(http works great). I have tried different FTP servers with the same
results.
Any clues would be greatly appreciated. Also, is there any easy way
to "monitor" FTP activity to actually see who is waiting on whom?
Thanks,
Rich Simons
------------------------------
** 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
******************************
