Linux-Networking Digest #547, Volume #11 Tue, 15 Jun 99 21:13:41 EDT
Contents:
Samba: Had it; lost it. (Yuki Taga)
Re: Linux - Win98 via crossover (Yuki Taga)
Re: Linux - Win98 (Yuki Taga)
What do rc0.d, rc1.d, rc2.d do? (Dan Winchester)
Networking problems with RedHat 6.0 ([EMAIL PROTECTED])
Re: in.ftpd : login failed (M. Buchenrieder)
performance tuning (John Assalone)
Re: in.ftpd : login failed ([EMAIL PROTECTED])
Re: where the heck ARE the linux drivers @? (Bob)
Internet connection problem ("Francisco Noizeux")
Re: Samba: Had it; lost it. (Eric LEMAITRE)
Re: Red Hat 6.0 serving private intranet ([EMAIL PROTECTED])
Qmail can't forward to a POP box off of the server. (Ryan Hughes)
Re: Linux > NT Server ??? (Paul Carver)
Re: How to make lookups start with /etc/hosts? ("YouDontKnowWho")
Re: What do rc0.d, rc1.d, rc2.d do? (Bob Tennent)
Re: Does anyone know what ports 31789 and 31790 are for? ("George Georgakis")
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Yuki Taga)
Subject: Samba: Had it; lost it.
Reply-To: [EMAIL PROTECTED]
Date: Tue, 15 Jun 1999 13:31:35 GMT
Friends,
I had my Samba going both ways tonight (first time newbie). I could see my
Linux box in NT's Network Neighborhood and browse the shares and read files. I
could mount my NT box in Linux, as well. But after rebooting Linux, I was no
longer able to mount NT. The command was the same, and there is no possibility
that I mis-typed it, because I used the history file to bring up the command --
so I didn't type anything.
The command: smbmount \\\\akajishi_s\\C -c 'mount /mnt/tmp -u 123 -g 456'
Akajishi_s is the name of an NT BDC, and it has a valid share named C.
I would get a password prompt, enter the password, and get denied. Whatever
went wrong? (The only difference was that I had at first been logged in as a
different user on the BDC. But, it shouldn't make a bit of difference which
user is logged in on the BDC, should it? What counts is the Linux username,
which of course much be a valid account in the NT domain, and the password
given at the prompt, right? I mean, I *did* this once. I'm amazed it wouldn't
work again.)
Yuki ^_^
------------------------------
From: [EMAIL PROTECTED] (Yuki Taga)
Subject: Re: Linux - Win98 via crossover
Reply-To: [EMAIL PROTECTED]
Date: Tue, 15 Jun 1999 13:33:12 GMT
Is 255.255.255 a valid subnet mask for 161.68.1.0? I sure don't think so.
Yuki ^_^
On 15 Jun 1999 05:30:51 GMT, in article <7k4oeb$3p7$[EMAIL PROTECTED]>,
Nick <[EMAIL PROTECTED]> wrote:
>I have a linux machine up and running with IP of 161.68.1.1 Sub
>255.255.255.0 and win98 is 161.68.1.2 255.255.255.0
>They are connected via crossover, after I setup linux, i did the win box,
>rebooted it, and started pingin the windows box from linux during reboot,
>it did recieve a response, but as soon as i start up any application on
>the win machine, i no long recieve replys to my pings from the linux side.
>
># route
>Destination gateway Genmask Flags Metrics Ref Use Iface
>161.68.1.1 * 255.255.255.255 UH 0 0 0 eth0
>161.68.1.0 * 255.255.255.0 U 0 0 0 eth0
>127.0.0.0 * 255.0.0.0 U 0 0 0 lo
>
># ifconfig
>eth0 Link encap:Ethernet HW:xxxxxxxxxx
> inet addr:161.68.1.1 Bcast:161.68.1.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
>.......
>
>C:\WINDOWS\Desktop>route print
>
>Active Routes:
>
>Network Address Netmask Gateway Address Interface Metric
> 0.0.0.0 0.0.0.0 161.68.3.101 161.68.3.101 1
> 127.0.0.0 255.0.0.0 127.0.0.1 127.0.0.1 1
>161.68.0.0 255.255.0.0 161.68.3.101 161.68.3.101 1
>161.68.1.0 255.255.255.0 161.68.1.2 161.68.1.2 2
>161.68.1.0 255.255.255.0 161.68.3.101 161.68.3.101 1
>161.68.1.2 255.255.255.255 127.0.0.1 127.0.0.1 1
>161.68.255.255 255.255.255.255 161.68.1.2 161.68.1.2 1
>224.0.0.0 224.0.0.0 161.68.1.2 161.68.1.2 1
>
>
>Any suggestions? or is Windows determined not to work with Linux ;)
>
>Any help would be appreciated.
>
>------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: [EMAIL PROTECTED] (Yuki Taga)
Subject: Re: Linux - Win98
Reply-To: [EMAIL PROTECTED]
Date: Tue, 15 Jun 1999 13:34:36 GMT
The valid default subnet mask for that network ID is 255.255.0.0. Try it.
Yuki ^_^
On 15 Jun 1999 05:30:54 GMT, in article <7k4oee$3p7$[EMAIL PROTECTED]>,
Nick <[EMAIL PROTECTED]> wrote:
>I have a linux machine up and running with IP of 161.68.1.1 Sub
>255.255.255.0 and win98 is 161.68.1.2 255.255.255.0
>They are connected via crossover, after I setup linux, i did the win box,
>rebooted it, and started pingin the windows box from linux during reboot,
>it did recieve a response, but as soon as i start up any application on
>the win machine, i no long recieve replys to my pings from the linux side.
>
># route
>Dest
>
>
>------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: Dan Winchester <[EMAIL PROTECTED]>
Subject: What do rc0.d, rc1.d, rc2.d do?
Date: Tue, 15 Jun 1999 14:46:57 +0100
Hi
I'm running Redhat Linux 6.0, and I'm trying to get my firewall up
before the network is brought up.
In my /etc/rc.d directory I have lots of subdirectories named rc0.d,
rc1.d, rc2.d etc, all containing symlinks.
My question:
Do I simply put a symlink to my firewall script in one of these
directories? If so which one, and what to call the symlink? Why are
there so many seemingly identical symlinks in the various rcx.d
directories?
Many Thanks
Dan Winchester
------------------------------
From: [EMAIL PROTECTED]
Subject: Networking problems with RedHat 6.0
Date: Tue, 15 Jun 1999 22:48:06 GMT
After running RedHat Linux 6.0 for a couple of days, I rebooted the
machine to move it, however, when it came up again I constantly got the
error "Neighbour table overflow".
Not sure if this helps, but I can at least boot the unit, configure a
few things, and get it in a stable state again.
Steps:
1. Boot in single user mode. You can do that by typing 'linux s' at
the linux boot prompt.
2. Change directories to /etc/rc.d/rc2.d and create a subdirectory
called 'backup'. Move the file 'S10network' to backup:
mv S10network backup
3. Change directories to /etc/rc.d/rc3.d and create a subdirectory
called 'backup'. Move the same file 'S10network' to backup.
4. reboot.
When the system comes up, you won't get the error since networking is
now disabled. Once the system is up again, change directories to
'/etc/rc.d/init.d/' and type './network start'
On my system, I noticed that the loopback doesn't have an IP (typically
it's 127.0.0.1, however, it's not configured, so I typed:
ifconfig lo 127.0.0.1 mask 255.0.0.0 <press enter>
however, this command doesn't terminate on it's own, so I had to
press 'ctrl/c', but it does seem to take effect.
After poking around on the system, I found that my /etc/resolv.conf
file was corrupted (not sure why), and my default gateway would never
get set by default anymore. I reset the default gateway by entering:
$ route add default gw 192.168.1.1 <press enter>
This is obviously not the correct way to fix this, but I can't seem to
find anything else on the Internet about fixing this. The downside to
doing like this is that your web server and other network services
won't start. You have to manually start them.
RedHat should take a better look at this and come out with a patch
soon. This is a nasty problem.
Ed
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (M. Buchenrieder)
Subject: Re: in.ftpd : login failed
Date: Tue, 15 Jun 1999 07:42:35 GMT
[EMAIL PROTECTED] (peter) writes:
[...]
>the syslog-entry does not tell me a lot more:
>Jun 14 23:31:55 goldfisch ftpd[15120]: failed login from
>goldfisch.atat.at [192.168.1.1], root
^^^^^^^
[...]
root telnet logins are generally denied due to security risks.
If for some reasons you HAVE to do that, edit /etc/securetty
and /etc/ftpusers . It is not a very bright idea to do that.
Use SSH instead.
Michael
--
Michael Buchenrieder * [EMAIL PROTECTED] * http://www.muc.de/~mibu
Lumber Cartel Unit #456 (TINLC) & Official Netscum
Note: If you want me to send you email, don't munge your address.
------------------------------
From: John Assalone <[EMAIL PROTECTED]>
Subject: performance tuning
Date: Tue, 15 Jun 1999 09:55:08 -0400
Hi all. I just set up a VAResearch server (VArServer 500). It's a PII400
1GB RAM, 9GB SCSI drive. It came pre-installed with RedHat 5.2, 2.0.36
kernel. I recompiled and installed a pretty bare 2.2.10 kernel. I'm
running apache 1.3.6 with most of the modules turned off (i'm only
serving static html pages - i'm using apache's mod_mmap_static module).
I've been doing some load testing using ZiffDavis's WebBench, running on
about 30 Windows workstations, each workstation generating about 100
child processes.
I ran WebBench on the out-of-the-box config and got about a 3.6M/s
transfer rate (that's before I did the tweaks mentioned above). After
that i've been able to pull out around 5.8M/s. Some people above me (i'm
the lowly linux minion in my company) are pushing for a transfer rate of
near 10M/s. We are building a site with projected peak transfer of
41M/s. (i didn't come up with these numbers, so if they're crazy don't
yell at me).
Is there any other perf tuning i can do to up these numbers?
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: in.ftpd : login failed
Date: Tue, 15 Jun 1999 23:58:19 +1000
>
> and the use I tried to logon with (root) is also listed in /etc/ftpusers
>
> what the hell is wrong here.
> (tcp/wrapper ist also not the problem, cause I come to the login ...)
/etc/ftpusers is a list of users which are not allowed to use the ftp
service. You couldnt' login as root as root is mentioned in /etc/ftpusers.
Solution: don't ftp as root (good idea!) or just delete root from
/etc/ftpusers.
Good luck,
Joe
------------------------------
From: [EMAIL PROTECTED] (Bob)
Crossposted-To: alt.os.linux,comp.os.linux.help,linux.help
Subject: Re: where the heck ARE the linux drivers @?
Date: Tue, 15 Jun 1999 13:52:48 GMT
[EMAIL PROTECTED] (Mike Ching) wrote:
>If the card is an Allied Telesis AT1500, you can use the LANCE driver. I
>didn't recognize the ati1500 reference but if it has DMA, maybe you mistyped
>the name.
>
TX very much! - I was surprised to find direct at1500 support in a
file w/o 1500 in the name (considering the 1700 is named
specifically)!!
appreciate the assist...
________________________________________________
Definition of Windows 95:
A 32 bit upgrade to 16 bit extensions for an 8 bit operating system
designed to run on a 4 bit processor by a 2 bit company that
doesn't like 1 bit of competition.
------------------------------
From: "Francisco Noizeux" <[EMAIL PROTECTED]>
Subject: Internet connection problem
Date: Wed, 16 Jun 1999 02:02:11 +0200
I have Red Hat 6.0 but when I try to connect to internet with de kppd it
connects to my server, "Loggin into network..." but when my computer sends
the pppd commands the server reports "connection timed out".
What can I do to resolve it.
------------------------------
From: Eric LEMAITRE <[EMAIL PROTECTED]>
Subject: Re: Samba: Had it; lost it.
Date: Tue, 15 Jun 1999 16:00:03 +0200
Yuki Taga a �crit :
> The command: smbmount \\\\akajishi_s\\C -c 'mount /mnt/tmp -u 123 -g 456'
> Akajishi_s is the name of an NT BDC, and it has a valid share named C.
> I would get a password prompt, enter the password, and get denied. Whatever
> went wrong? (The only difference was that I had at first been logged in as a
> different user on the BDC. But, it shouldn't make a bit of difference which
> user is logged in on the BDC, should it? What counts is the Linux username,
> which of course much be a valid account in the NT domain, and the password
> given at the prompt, right? I mean, I *did* this once. I'm amazed it wouldn't
> work again.)
Hi !
Perhaps the login name is not good (-n station_name), because the host name will be
used by default. Verify your accounts and use the "-n" parameter with smbmount.
Bye !
--
Responsable de formation pour les fili�res Internet et Linux
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Red Hat 6.0 serving private intranet
Date: Wed, 16 Jun 1999 00:02:03 +1000
Have you edited your /etc/hosts file? It should contain a list of all the
hosts on your local network , e.g.
#IP name nickname
192.168.1.1 Linux
192.168.1.2 Mac
Good luck,
Joe
Bruce Fletcher wrote:
> Short 'n sweet:
>
> Linux box --- crossover 10BT cable --- Mac, neither can ping the other.
>
> Details:
> I have a new Linux box, running Red Hat 6.0. It also has a small Win95
> test partition, with which I verified that all the hardware (including
> the ethernet card) is functional. After much frustration, I have
> managed to get the ethernet card to be recognized by linux. ifconfig
> reports eth0 is alive and well, and the box can ping itself both with
> localhost and eth0's IP address, but the first packet never goes
> anywhere when pinging my Mac (an old G3 running MacOS 8.5) and the Mac
> can't ping the Linux box either.
>
> I understand the basics of UNIX, but I am no sysadmin. How do I get
> there from here? Is there some lower-level diagnostic than ping that I
> should be using?
>
> - Bruce
>
> PS; as a service to anyone searching usenet for ethernet adapter advice,
> I would like to state that my ethernet card with 'EN5038' on the main
> chip is aparently a Realtek 8139 clone, and the rtl8139 driver can find
> the card and even query it for irq and i/o address.
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
------------------------------
From: Ryan Hughes <[EMAIL PROTECTED]>
Subject: Qmail can't forward to a POP box off of the server.
Date: 15 Jun 1999 21:30:47 GMT
Hello,
If anyone out there could answer this, I would be most appreciative.
I recently installed RedHat 5.2 on an old 486 in plans of it replacing our
current mail server, running on NT. I dumped sendmail and installed qmail.
Everything's cool, except I can not get the .qmail files to forward to
email address. The will forward to local mailboxes, but not to boxes that
are off of the server.
Say we have the virtual domain, client-domain.com. I listed this domain
name in /var/qmail/control/rcpthosts and /var/qmail/control/locals. I
added a /var/qmail/control/virtualdomains file, with this in it:
client-domain.com:client1138
The user client1138 already has a home directory and maildir all setup.
Here is how that looks:
root# pwd
/home
root# ls -al
total 3
drwxr-xr-x 12 root root 1024 Jun 15 11:10 .
drwxr-xr-x 16 root root 1024 Jun 11 23:57 ..
drwx------ 3 client1138 client1138 1024 Jun 15 11:44 client1138
root# cd client1138/
root# ls -al
total 5
drwx------ 3 client1138 client1138 1024 Jun 15 11:44 .
drwxr-xr-x 12 root root 1024 Jun 15 11:10 ..
-rw-r--r-- 1 client1138 client1138 11 Jun 15 11:10 .qmail
-rw-r--r-- 1 root root 21 Jun 15 12:05 .qmail-test
drwx------ 5 client1138 client1138 1024 Jun 15 11:10 Maildir
The .qmail-test file contains the line:
&[EMAIL PROTECTED]
The actual email address is of course a valid one. Now, when I send a test
message to [EMAIL PROTECTED], I don't get a MAILER-DAEMON error, or a
message in my maillog about it dying, nothing. But- when I change the cile
to say:
&anotherlocaluser
The mail delivers just fine. I do eventually get this message mailed to
root: (forwarded from root to another box, of course)
<[EMAIL PROTECTED]>:
Sorry, no mailbox here by that name. (#5.1.1)
Any ideas are greatly appreciated. Thanks.
-Ryan Hughes
[EMAIL PROTECTED]
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: [EMAIL PROTECTED] (Paul Carver)
Subject: Re: Linux > NT Server ???
Date: Wed, 16 Jun 1999 03:53:17 GMT
On Mon, 07 Jun 1999 08:28:15 +0000, J�n Gu�mundsson
<[EMAIL PROTECTED]> wrote:
>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)
This quote is particularly interesting in light of the adds I've seen
recently in Forbes magazine. Microsoft is running adds touting Windows
NT as "3 times more reliable that Windows 95". Gee, who was it that
was selling Windows 95? And what's the sales pitch for Windows 98,
"one third as reliable as Windows NT"?
------------------------------
From: "YouDontKnowWho" <[EMAIL PROTECTED]>
Subject: Re: How to make lookups start with /etc/hosts?
Date: Tue, 15 Jun 1999 14:06:43 GMT
I didn't quite understand your post, so maybe this will be wrong.
I think that the "order hosts,bind" line belongs in /etc/resolv.conf,
not /etc/hosts.
--
And now we return to our regularly scheduled,
uncommonly entertaining thread...
Duncan McIntyre wrote in message <[EMAIL PROTECTED]>...
>I have just installed Netscape 4.6 on my Redhat 6.0-ish system, and I
>now have a need to finally sort out a problem which has been bugging
me
>for a while.
>
>My setup is a number of linux and Windows boxes on an ethernet LAN. I
>use one of the Linux boxes to dial-up my ISP and masquerade for the
>other boxes. I also have Squid running on the box since masquerading
>doesn't seem to work 100%.
>
>Anyway, the issue is that I would like my Linux Workstation to use
its
>/etc/hosts file for name resolution before trying any nameservers. I
>have
>
>/etc/hosts:
>order hosts,bind
>.
>.
>
>/etc/nsswitch.conf
>.
>.
>hosts: files dns
>.
>.
>
>But nslookup never goes to /etc/hosts. Apparently Netscape uses the
same
>mechanism as nslookup to resolve names, because when I start it it
hangs
>for several minutes while it tries to resolve a bunch of names. If I
>start ppp on the server before I start Netscape there is no problem.
>However the mechanism I have to start ppp uses a web-page on the
server
>to bring it up and down and give statistics. Which means I'm screwed,
>since Netscape can't find the address of the server unless ppp is
>already running.
>
>What am I missing? I'm sure this must be simple.
>
>Other info. kernel is 2.2.2, glibc-2.1-0.990222
>
>=========================
>Duncan McIntyre
>
>
------------------------------
From: [EMAIL PROTECTED] (Bob Tennent)
Subject: Re: What do rc0.d, rc1.d, rc2.d do?
Date: 15 Jun 1999 13:58:01 GMT
Reply-To: rdt(a)cs.queensu.ca
On Tue, 15 Jun 1999 14:46:57 +0100, Dan Winchester wrote:
>
>In my /etc/rc.d directory I have lots of subdirectories named rc0.d,
>rc1.d, rc2.d etc, all containing symlinks.
>
>Do I simply put a symlink to my firewall script in one of these
>directories? If so which one, and what to call the symlink? Why are
>there so many seemingly identical symlinks in the various rcx.d
>directories?
>
rcn.d is processed when entering run-level n, where n=3 for
console multi-user and n=5 for X. You've got the right idea
about putting in the symlink. It doesn't much matter what you
call it, but the first letter should be S (start) and the next two
digits are to put it in the right place in the start-up sequence.
You probably want it just before the network link.
Bob T.
------------------------------
From: "George Georgakis" <[EMAIL PROTECTED]>
Subject: Re: Does anyone know what ports 31789 and 31790 are for?
Date: Wed, 16 Jun 1999 00:16:23 GMT
I wouldn't worry too much about that.
a) any port over 1024 is non-privileged. It's VERY unlikely to be an
exploit.
b) how much damage can 1 byte of malicious data every 15 minutes do?
My guess is that it's just a stray bit of data just floating around,
occasionally hitting those ports.
George
===========================================================================
I never reply by email as a) I don't give out my real email address freely,
and b) it stops other NG users from reading the solutions to problems
If necessary, however, I can be contacted thru geegs (a) linuxstart DOT com
==========================================================================
Thomas Zajic <[EMAIL PROTECTED]> wrote in article
<3766c231.213367@news>...
> On Tue, 15 Jun 1999 20:02:26 GMT, David Kennedy wrote:
>
> > Hmm, I see you have the same thing?
>
> Yup, seems to be some kind of new sport these days, somebody else does
> the same thing obviously:
>
> Jun 15 21:41:25 sphere udplog: dgram to port 31789 from
212.17.74.136:31790
> (1 bytes)
> Jun 15 22:04:11 sphere udplog: dgram to port 31789 from
212.17.74.136:31790
> (1 bytes)
> Jun 15 22:18:09 sphere udplog: dgram to port 31789 from
212.17.74.136:31790
> (1 bytes)
>
> <paranoia>
> Was there some new exploit/vulnerability discovered recently? Are the
> script kiddies at work already?
> </paranoia>
>
> > I will let you know the solution if I find one. Please do the same.
>
> I will, sure!
>
> Thomas
> --
> =--- Thomas Zajic aka ZlatkO ThE GoDFatheR, Vienna/Austria
---=
> =-- "It is not easy to cut through a human head with a hacksaw." M.C.
--=
> =-- Posted with Free Agent 1.11/32 running on Linux 2.0.36/Wine-990226
--=
> =--- Spam-proof e-mail: thomas(DOT)zajic(AT)teleweb(DOT)at
---=
>
------------------------------
** 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
******************************
