Linux-Networking Digest #391, Volume #10 Fri, 5 Mar 99 17:13:34 EST
Contents:
Re: samba 2.0.2 : login just as guest (Luc Taesch)
Re: Help with IP Routing Problem! (Michael23)
Re: NAT Support (Rick Onanian)
3c905b-tx nic (w00t)
Problems with Chat Script/pppd/diald on RedHat 5.1 ... (Eli White)
Re: Absolute PPP Frustration (Rick Onanian)
Re: decoding info in /proc/net (David Kirkpatrick)
Bringing linux into a Netware 5 environment? (Jude Travers-Frazier)
Pac and printer accounting ([EMAIL PROTECTED])
PPP not working with my ISDN TA (Magnus Sirwi�)
Re: Linux Networking Performance? (Michael John Erskine)
Re: SMTP problem ("Rocky Dean")
SENDMAIL MAIL GROUPS (coopers)
funny routing! ([EMAIL PROTECTED])
debugging IP_Masq and Portforwarding (Derek Ealy)
Ping works just fine.. But no http protocol?!? (Michael Kristensen)
Funky domains (Evan Wolenzik)
Re: Samba and NT4 (David Kirkpatrick)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Luc Taesch)
Crossposted-To: linux.samba,comp.protocols.smb
Subject: Re: samba 2.0.2 : login just as guest
Date: Fri, 05 Mar 1999 19:43:40 GMT
ok, i got it. a silly one, but i think its worthwhile for the FAQ.
i had ALL in hosts.deny, and that was denying access otherwise than
guest.
(can somebody forwrd it to the mail list, im not subsrcibed)
On Wed, 24 Feb 1999 17:39:58 GMT, [EMAIL PROTECTED] (Luc Taesch)
wrote:
>same problem since 1.9
>
>login works with 98
>but where guest is allowed, it works
>ie the status shows that guest is loggued, and not the ID i accessed
>
>funny enough, my home is still accessible ie i acces my luc directory,
>and my friends richard sees its richard one.
>
>ive tried with both encrypted, or without.
>
>same result, ie login says : wrong password (for the shares that deny
>guest)
>
>when i smbclient from the server, logged as luc, i can access the
>guest allowed shared, and status shows my id (luc), not guest.
>
>but even with smbclient , i cant access the shares that deny guest
>
>ive tried 1.9.18, 2.0, 2.0.
>
>hope u can help
>
>cheers
>
>luc
------------------------------
Date: Fri, 05 Mar 1999 12:16:41 -0700
From: Michael23 <[EMAIL PROTECTED]>
Subject: Re: Help with IP Routing Problem!
> How are you connected to the network? Do you still have one
> connection or got a second with the new address? When you say
> "host" are you running DNS and doing lookups or have nameserver
> addresses supplied by your ISP?
Hello David,
Thanks for your interest... I have one connection to the internet via DSL... but I
have 8 static IP's and I would like to put up another host on one of those. I
need to have two hosts up to run DNS (i.e. one host would be the primary
nameserver, and the other would be the secondary nameserver) BIND is running on
my primary machine and I'll get it running on the second once it can see the
outside world. So far I don't think that name service is an issue, because I have
been testing my connectivity using ping and telnet with IP addresses only.
One solution that I'm pretty sure will work is running the DSL into a hub and
running any machines that need to have real world IP's directly off the hub... but
I'd like to get it to work without buying new hardware if I can...
besides, It seems like the routing tables were designed to do specifically what
I'm trying to do... (i.e. route IP traffic)
I hope that helps... I've really been stuck on this one.
Thanks,
Michael23
------------------------------
From: Rick Onanian <[EMAIL PROTECTED]>
Subject: Re: NAT Support
Date: Fri, 05 Mar 1999 14:38:38 -0500
Allen wrote:
>
> That's good to know. And I suppose it's transparent to workstations behind
> it.
The workstations behind it have the linux box as their gateway. It is
pretty well transparent to the machines behind the gateway - some
special services (IRC, ICQ, gaming like Quake, etc.) can have some
trouble, resulting from the inability to make an incoming connection to
the client machine.
> However, my concern is for port-mapping from outside linux (from Internet)
> ... should I call it reverse hosting? Does it support reverse hosting?
I'm unsure of what you want.
Do you want, for example, to put up services (webserver, ftp server,
etc) behind the gateway? If so, you have to use one of many forwarding
schemes..I use ipportfw. Most of them will forward incoming connections
to a specified port on your gateway to an internal address. Options are
ipportfw, rinetd, ipautofw, and some others I can't remember right now.
> Thanks.
> Allen
>
> Eldir Tomassen wrote in message <[EMAIL PROTECTED]>...
> >Ye sit does. Buy on linux it's often called
> >masquerading instead.
> >
> >
> >Eldir
> >[EMAIL PROTECTED]
--
rick - a guy in search of raw (ISO) cd images of SuSE and Slackware
===============
My opinions don't exist, and as such, are not anyone elses. I do not
represent anyone, not even myself, and especially not my employer.
---
Looking for a 1968 Camaro SS convertible, black interior,
beat-up rustbucket that is in need lots of restoration and TLC.
---
To email me, take out the papers and the trash
[EMAIL PROTECTED]
------------------------------
From: w00t <[EMAIL PROTECTED]>
Subject: 3c905b-tx nic
Date: Thu, 04 Mar 1999 14:05:04 -0800
I have a prob with the 3c905b nic card. Let say I'm in win98 and I
reboot into slack. When it his my nic card I get a ff:ff:ff:**** irq11
error. Now if I disable the damn nic card in win98 reboot back into
win98 then reboot once more back into slack my nic seems to work.
Everyone tells me it is plug and prey but all of that is turned off from
what I can tell.
Has anyone solved this problem. It seems that there is about 3 people
here at work that is having this problem..
Out.
Kevin
------------------------------
From: Eli White <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.questions,comp.os.linux.setup,comp.os.linux.help,comp.os.linux.admin,comp.os.linux.misc
Subject: Problems with Chat Script/pppd/diald on RedHat 5.1 ...
Date: Thu, 04 Mar 1999 17:06:57 -0500
I've been having a problem with my chat script I am trying to use to
automate connecting to the net through diald.
I have everything set up ok and can dial in through minicom, and run
pppd by hand ... Everything works fine (as long as I am REALLY QUICK on
the draw - bellatlantic.net likes to drop you quickly if you don't start
talking ppp)
But when I try to do it automatically through diald/chat ... it breaks
... in particular, the chat script breaks ...
diald automatically starts running the chat script just like it should,
however in the chat, it never sees the 'CONNECT 48000/ARQ' string, and
timeouts instead ...
I see the CONNECT line just fine when going thorugh minicom ...
Anyway please have an answer for me?
Included below are my chat script, and the log file of the attempt ...
Thanks,
Eli
[EMAIL PROTECTED]
chat-script:
ABORT BUSY ABORT 'NO CARRIER' '' ATZ OK ATDT4103395344
CONNECT
(NOTE: Using PAP, so no login stuff there)
/var/log/messages:
Mar 4 20:05:29 shield diald[326]: Running connect (pid = 401).
Mar 4 15:05:29 shield chat[401]: abort on (BUSY)
Mar 4 15:05:29 shield chat[401]: abort on (NO CARRIER)
Mar 4 15:05:29 shield chat[401]: send (ATZ^M)
Mar 4 15:05:29 shield chat[401]: expect (OK)
Mar 4 15:05:29 shield chat[401]: ATZ^M^M
Mar 4 15:05:29 shield chat[401]: OK
Mar 4 15:05:29 shield chat[401]: -- got it
Mar 4 15:05:29 shield chat[401]: send (ATDT4103395344^M)
Mar 4 15:05:30 shield chat[401]: expect (CONNECT)
Mar 4 15:05:30 shield chat[401]: ^M
Mar 4 15:06:15 shield chat[401]: alarm
Mar 4 20:06:15 shield diald[326]: Connect script failed.
Mar 4 15:06:15 shield chat[401]: Failed
Mar 4 20:06:17 shield diald[326]: Delaying 30 seconds before clear to
dial.
------------------------------
From: Rick Onanian <[EMAIL PROTECTED]>
Subject: Re: Absolute PPP Frustration
Date: Fri, 05 Mar 1999 14:59:11 -0500
A rule of thumb for this sort of thing:
If you can manually set the IO and IRQ that the modem uses by changing
something physical on the modem itself, such as jumpers or DIP switches,
than it will work without any trouble in Linux. Else, it probably will
not work.
Look to see if you can. Also, follow this link and click on the link it
has for modem problems: http://dingo.mcrnet.net
[EMAIL PROTECTED] wrote:
>
> I'm a new linux user, trying to install and congigure my modem, PPP style.
> I've been trying with two non-winmodem PnP 28.8s, and to no avail thus far.
>
> To this point, I've tried using Minicom just to establish communication with
> the modem, but this has been unsuccessful. Also, I've tried CHAT, with proper
> PPP parameters given at the command line. It attempts to ATZ, and never
> recives an OK.
>
> I'm running turboLinux, and their probing utilities have found the modem, but
> pretend it's not there when it's time to dial out.
>
> Maybe I'm going about this completely backwards, but I know there is a simple
> process out there I need to follow, and I'd be undescribably indebted to
> anyone who could detail that process from A - Z. I'm a fairly adept guy,
> just treading on unfamiliar ground. No need to write a book, just tell me
> where to go, and what to do when I get there! THANKYOU!! -- in advance.
>
> Chris Beams-->
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
--
rick - a guy in search of raw (ISO) cd images of SuSE and Slackware
===============
My opinions don't exist, and as such, are not anyone elses. I do not
represent anyone, not even myself, and especially not my employer.
---
Looking for a 1968 Camaro SS convertible, black interior,
beat-up rustbucket that is in need lots of restoration and TLC.
---
To email me, take out the papers and the trash
[EMAIL PROTECTED]
------------------------------
From: David Kirkpatrick <[EMAIL PROTECTED]>
Subject: Re: decoding info in /proc/net
Date: Fri, 05 Mar 1999 15:38:20 +0000
Reply-To: [EMAIL PROTECTED]
Not exactly but they can be divined from source. snmp_get_info
in proc.c sets up this info which is tracked in variables defined
in snmp.h. If you look at the code you can see what each one
tracks and where its incremented.
d
G Agdesteen wrote:
>
> Hello,
> can anyone point me to a how-to or a descrition or a book about decoding
> the information contained in the /proc/net, and especially
> /proc/net/snmp??
>
> A class project here at NMSU is to analyze a small beowulf cluster, and
> I'm digging
> for alternate ways to extract networking information.
>
> Thanks,
> geir
> (remove NOSPAM if reply by mail)
--
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Jude Travers-Frazier)
Subject: Bringing linux into a Netware 5 environment?
Date: Thu, 04 Mar 1999 22:10:57 GMT
I am currently running a Netware 5 pure IP network, serving about 700
users (all of who use Win95 machines).
I have an NT server that provides web-based services to all users
(POP3/SMTP and web-pages).
What I would like to do is replace the NT server with a linux machine,
and have those web services provided by linux.
What I am concerned about is the following: I do not want to have to
create 700 accounts on linux. I would like to be able to configure
linux to provide POP3 services for all my users (and hopefully
personal web-page services) without having to do much on the linux
machine, and keeping the linux machine invisible to the users (so, for
example, they can continue to change their netware passwords without
having to worry about changing/syncing linux passwords).
If anyone can offer ideas or point me to resources on this one, I'd be
very appreciative.
Many thanks!
jtf
------------------------------
From: [EMAIL PROTECTED]
Subject: Pac and printer accounting
Date: Fri, 05 Mar 1999 17:57:43 GMT
We found a utility already in RH5.2 called pac that is supposed to control
printer accounting. Read the man pages, etc. but haven't been able to get it
working so far. Anyone know anything about pac?
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Magnus Sirwi� <[EMAIL PROTECTED]>
Subject: PPP not working with my ISDN TA
Date: Thu, 4 Mar 1999 23:23:58 +0100
I can't ping my domain server either by name or by ip address when I make a
dial-up connection using my isdn TA with kppp.
If i however change the device to my analog modem everything works fine...
although slow :-(
Suggestions and hints anybody? My kernel is 2.2.2 and I'm using kde 1.1. The TA
is a ZyXEL omni.net plus.
\Magnus
------------------------------
Date: Fri, 05 Mar 1999 20:48:27 +0000
From: Michael John Erskine <[EMAIL PROTECTED]>
Subject: Re: Linux Networking Performance?
Bill;
I am sorry that I can't speak to the questions you have asked here. I too am
frustrated by the Holy wars this sort of conversation can generate. I have no
desire to become part of that.
The BSD code running on Sun workstations never seemed to perform as well as the
socket libraries running on the SGI's but both appeared to work well enough to
keep me happy. Personally I liked the SGI's better.
What I have seen is a P120 with 64M of ram switching packets for 100+ W95 boxes
and showing a MAXIMUM system load of 12%. That was a Linux box and it was the
best performance I have ever seen on any machine which was not a vendor built
router. This machine ran day after day, month after month and just never went
down.
I don't know why they used a variable length buffer but I believe the high
priests who do Linux networking code are listed in the source files and you can
mail them directly for an answer.
Another thing you might do is just write some sockets code on linux and check it
for yourself perhaps you will figure it out. You could also read their source.
I wish you luck in your quest for this most holy of grail.....
:)
[EMAIL PROTECTED] wrote:
> At the risk of violating religious boundaries and inciting a flame war, I'd
> like to ask a few questions about Linux's network architecture.
>
> First and foremost, is there a distribution or patch of Linux that utilizes
> the actual 4.4BSD-Lite(R2) source, rather than the standard Linux networking
> code? Is anybody working on such a project?
>
> Secondly, I'd like to ask about some design decisions regarding some
> important structures in Linux's network implementation. In particular, could
> anyone tell me why the sk_buff structure apparently has a variable-length
> data segment (or provide a pointer to better documentation if this is not the
> case)? mbuf clusters (the BSD equivalent of Linux's sk_buff structure) were
> designed to be fixed length for several very important reasons, and I can't
> understand why somebody would have designed network structures differently.
> I'm by no means the leading expert on networking, and if there are good
> reasons why these differences exist I'd be glad to hear them.
>
> I've toyed around with Linux since '94, but have used FreeBSD exclusively on
> any production machines I've ever set up, mostly because of the remarkable
> efficiency of the BSD networking code. Until recently (the past year or so),
> Linux's networking capability under heavy load has... well... it's sucked, to
> be perfectly honest. More recent changes to the networking code (for
> example, allowing data to be stored in a separate sk_buff structure in a way
> that more closely resembles how mbufs and mbuf clusters work) have made me
> reconsider my loyalties in this religous holy war. My development philosophy
> has always leaned more towards the bazaar than the cathederal, but I'm pretty
> much unwilling to give up my faith in the BSD networking code. It would
> really make my day if somebody could tell me that I can have both. :)
>
> -Bill Clark
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
--
Michael Erskine, Simply Computers!, Urbanna, Virginia, 804-758-3798
[EMAIL PROTECTED]
============= Posted via Newsfeeds.Com, Uncensored Usenet News ============
http://www.newsfeeds.com/ The Largest Usenet Servers in the World!
============= Over 66,000 Groups, Plus a Dedicated Binaries Server ============
------------------------------
From: "Rocky Dean" <[EMAIL PROTECTED]>
Subject: Re: SMTP problem
Date: Fri, 5 Mar 1999 15:49:09 -0500
edit the file /etc/mail/relay_allow and allow your local hosts to relay
email.
Alejandro Miguel wrote in message <7bp7u5$ngm$[EMAIL PROTECTED]>...
>
>nico wrote:
>> I've recently obtained a domain name.
>> I've a static ip adress and a linux with
>> a sendmail working.
>> The PC states behind a DNS on which our
>> domain is declared.
>> Before I obtained the domain name I've never
>> had a problem associated with sendmail.
>> Now when I test my new e-mail I'm receiving such message :
>>
>> Remote-MTA: DNS; dns-name
>> Diagnostic-Code: SMTP; 551 <e-mail> ... we do not relay
>>
>> Does it come from my sendmail configuration, or from an error
>> in the DNS domain name's declaration ?
>> Can anyone help me, thank's for advance.
>> nico
>>
>
>You problem is related with relaying, You should check your sendmail.cf if
>it contains something like FEATURE(promiscuous_relay)dnl,
>if not, add it, just for a test, after that you can investigate the
>implications of it.
>
>
>Alejandro Miguel
>
>------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: coopers <[EMAIL PROTECTED]>
Subject: SENDMAIL MAIL GROUPS
Date: Thu, 04 Mar 1999 14:39:14 -0800
I am looking for a way to send mail to all the users on my mail server
without having to write out manually all the user names in the
/etc/aliases file. This way, mail to [EMAIL PROTECTED] would result
in mail to all my POP users on the server.
-Coop
------------------------------
From: [EMAIL PROTECTED]
Subject: funny routing!
Date: Thu, 04 Mar 1999 01:46:59 GMT
I have linux setup as PPP router. I am using netscape from a Windows 95
client. I am using apache web server as http proxy. The problem is that I
can browse the web but cannot read mail or news from my ISP!
My routing table looks like this:
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
209.191.13.101 0.0.0.0 255.255.255.255 UH 0 0 0 ppp0
127.0.0.1 0.0.0.0 255.255.255.255 UH 0 0 0 lo
192.168.100.0 0.0.0.0 255.255.255.0 U 0 0 22 eth0
0.0.0.0 209.191.13.101 0.0.0.0 UG
I have no idea how it got 209.191.13.101 but it works fine!
Now when I try traceroute on mail or nntp server, I get the following strange
message:
# traceroute 205.231.236.9
traceroute: Warning: Multiple interfaces found; using 192.168.100.1 @ eth0
traceroute to 205.231.236.9 (205.231.236.9), 30 hops max, 40 byte packets
1 * * *
why is it using eth0 when it should use ppp0?
thanks,
kal
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Derek Ealy)
Subject: debugging IP_Masq and Portforwarding
Date: Thu, 04 Mar 1999 22:45:48 GMT
Hi,
I've been running IP Masquerading on my RH Linux 5.1 machine for a few
months now. Had everything working FTP and even was able to establish
VPN connections from NT machines through the Linux gateway out to
other NT machines. Last weekend I decided to try to bring my web site
to my internal network and would like to run in on NT and IIS4. This
means that I need to route http traffic through my Linux gateway to
the IIS box on my internal network.
Ever since I began this endeavor I've had tons of problems.
Masquerading seems to work ok, but I can no longer FTP from my
internal Windows machines to other machines on the Internet. My VPN no
longer connects to one of my clients sites. And I've never been able
to successfully get my http traffic routed to the IIS machine. I've
tried individually installing rinetd, ipportfw and ippfvs (Linux
Virtual Server), on my upgraded 2.0.36 kernel. None of these have done
the trick. I assume that this is just a bit of configuration ignorance
on my part. I need suggestions on how to track down where the problem
really lies.
What tools are there available that will allow me to trace where
packets are going on my network? I need to know if things are actually
making through the firewall coming, or if they are being stopped going
out.
Any help would be greatly appreciated.
Thanks, Derek
http://www.grandprixsw.com
Anti spam notification:
Please don't just hit reply to send me email. Remove the despaminator (nospam.)
from my email address. My correct addr is: [EMAIL PROTECTED]
------------------------------
From: Michael Kristensen <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux
Subject: Ping works just fine.. But no http protocol?!?
Date: Fri, 05 Mar 1999 22:12:49 +0100
I have a problem with my newly installed RH5.2.
I've only just installed it and got PPP up and running, and it seems to
work fine; I can ping, telnet and ftp.
But I can't http to any servers, neither using Lynx or Netscape. And
this is even though I can ping the web servers, using IP-no. or DNS
name, with no problem what so even.
Any help would be appreciated.
Best regards,
Michael
------------------------------
From: Evan Wolenzik <[EMAIL PROTECTED]>
Subject: Funky domains
Date: 4 Mar 1999 22:52:41 GMT
Hi,
There's a radio station in St. Louis that's promoting its new website,
www.97.fm . Now, .fm is Micronesia, but a search at apnic.net didn't
turn up anything, and InterNIC doesn't recognize them either. And yet
this domain WORKS. How is this possible?
I realize you can program anything you want into your own named files,
but how do you get people outside of your domain to see this?
Or, could they really by registered in Micronesia, and I just missed
it somehow?
Thanks,
-Evan
------------------------------
From: David Kirkpatrick <[EMAIL PROTECTED]>
Subject: Re: Samba and NT4
Date: Thu, 04 Mar 1999 18:06:20 +0000
Reply-To: [EMAIL PROTECTED]
David,
I can't exactly remember but I think I could see a linux puter
on NT but could not login until the things below. I think I
could see the linux puter in neighborhood before I got samba
going though. I have some suggestions below for smb and getting
the icon.
I setup smbpasswd, smbusers and the shares simmilar to Fred but
with my settings. Are you on the same network? Some checks:
see man smbpasswd
encryption = yes in smb.conf
username map = /etc/smbusers
/etc/networks =
xxx.xxx.xxx.xxx domain.com where its the same as workgroup
in smb.conf
/etc/hosts name.domain.com
xxx.xxx.xxx.xxx
/etc/lmhosts same as hosts.
Setup the same files on NT.
David Kingston wrote:
>
> Greetings,
>
> I and having trouble with Samba and NT4.
>
> I am using RH5.2, and I am able to Telnet, FTP and Ping into the Linux
> box, but for some reason, I cannot see the Linux box on my network.
>
> I have checked all of the "obvious" (to me at least) information in
> smb.conf, but to no avail.....
>
> Thanks in advance
>
> David Kingston
--
[EMAIL PROTECTED]
[EMAIL PROTECTED]
[EMAIL PROTECTED]
------------------------------
** 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
******************************
