Linux-Networking Digest #951, Volume #10 Sat, 24 Apr 99 02:13:32 EDT
Contents:
high ping delay (Scott Wood)
Re: SMC1211 Driver -- unresolved dependency! ("Hoyt")
samba cant see win98 shares ("Ron")
Re: port forwarding (Greg Weeks)
linux computers for sale (warlock)
Re: how do you share a single email address? (Derek Shaw)
Re: POP & SMTP (Armand)
talk from prompt to xwindows (Ali Gonili)
How to get SNMP information from my Linux box ("Michael T. Spears")
Re: Two network cards ("J. S. Jensen")
Need help with DHCP on SuSE 6.1 ("glen")
Netscape and internet ("Curt")
Re: Anything wrong with using a 192.168.0.x private network? ("David K. Means")
ISP hookup (Cesar da Silva)
Logging onto Samba - not yet (John Hornblow)
Re: samba is running, but I get account not authorised to log on.. (John Hornblow)
ipchains under kernel 2.2.x (=?US-ASCII?B?ofOk363CxW+h8g==?=)
Re: Must replace Cisco with Linux in a hurry (Luca Filipozzi)
----------------------------------------------------------------------------
From: Scott Wood <[EMAIL PROTECTED]>
Subject: high ping delay
Date: Fri, 23 Apr 1999 18:38:45 -0700
I apologize for the length of this message,
I have seen this problem several times on Linux 2.2 hosts and routers...
# ping -c 10 192.168.19.2
PING 192.168.19.2 (192.168.19.2): 56 data bytes
64 bytes from 192.168.19.2: icmp_seq=0 ttl=255 time=3004.8 ms
64 bytes from 192.168.19.2: icmp_seq=1 ttl=255 time=3000.3 ms
64 bytes from 192.168.19.2: icmp_seq=2 ttl=255 time=3000.3 ms
64 bytes from 192.168.19.2: icmp_seq=3 ttl=255 time=3000.3 ms
64 bytes from 192.168.19.2: icmp_seq=4 ttl=255 time=3000.3 ms
64 bytes from 192.168.19.2: icmp_seq=5 ttl=255 time=3000.3 ms
64 bytes from 192.168.19.2: icmp_seq=6 ttl=255 time=3000.3 ms
--- 192.168.19.2 ping statistics ---
10 packets transmitted, 7 packets received, 30% packet loss
round-trip min/avg/max = 3000.3/3000.9/3004.8 ms
It looks like the first 3 packets are recieved, but the ping program doesn't see them
until 3 packets are in the queue, then it
looks like they took 3000 ms to return. I did a background tcpdump and pinged and saw
this:
PING 192.168.19.2 (192.168.19.2): 56 data bytes
18:10:25.307607 192.168.11.2 > 192.168.19.2: icmp: echo request
18:10:25.307833 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:26.302606 192.168.11.2 > 192.168.19.2: icmp: echo request
18:10:26.302817 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:27.302477 192.168.11.2 > 192.168.19.2: icmp: echo request
18:10:27.302686 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:28.302371 192.168.11.2 > 192.168.19.2: icmp: echo request
18:10:28.302579 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:29.302261 192.168.11.2 > 192.168.19.2: icmp: echo request
18:10:29.302465 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:30.302156 192.168.11.2 > 192.168.19.2: icmp: echo request
64 bytes from 192.168.19.2: icmp_seq=0 ttl=253 time=4994.8 ms
18:10:30.302359 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:31.302050 192.168.11.2 > 192.168.19.2: icmp: echo request
64 bytes from 192.168.19.2: icmp_seq=1 ttl=253 time=4999.6 ms
18:10:31.302265 192.168.19.2 > 192.168.11.2: icmp: echo reply
18:10:32.301943 192.168.11.2 > 192.168.19.2: icmp: echo request
64 bytes from 192.168.19.2: icmp_seq=2 ttl=253 time=4999.7 ms
18:10:32.302156 192.168.19.2 > 192.168.11.2: icmp: echo reply
--- 192.168.19.2 ping statistics ---
8 packets transmitted, 3 packets received, 62% packet loss
round-trip min/avg/max = 4994.8/4998.0/4999.7 ms
I had done several other ping tests and the response time rose to
about 5000ms. When I tcpdump from another host on the lan, it shows
the response immediately after the request (about 0.1ms on 100BaseT).
I am using a NetGear FA310TX with a real Dec Tulip chip on one link
and also the same NIC with the Lite-ON PNIC chip on another link and
both links experience the same problem. I am using kernel 2.2.5 with
iproute2. If I set both interfaces to down 'ip link set eth0 down'
and bring them back up and give them addresses again, the problem goes
away. I can reproduce this problem be running the NetPIPE test, found at:
http://www.scl.ameslab.gov/Projects/ClusterCookbook/nprun.html
I have run this test several times successfully on 2.2.4, but I had the
same problem occur under other circumstances.
Is this a kernel bug? Has anyone else had this problem? Does anyone know
what causes it?
--
+---------------------+--------------------------+
| Scott Wood | Test & Integration |
| [EMAIL PROTECTED] | Specialist |
|=====================+==========================|
| BroadLink Communications |
| -={ High Speed Wireless Networks }=- |
+------------------------------------------------+
------------------------------
From: "Hoyt" <[EMAIL PROTECTED]>
Subject: Re: SMC1211 Driver -- unresolved dependency!
Date: Fri, 23 Apr 1999 17:43:00 -0400
Jeffrey Rice wrote in message
news:[EMAIL PROTECTED]...
> Opps, I goofed a bit. The error I'm getting is "unresolved symbol", not
> dependency. I'm missing an header file, right? I'm lost...
I got this error message when I called my NE2000 clone driver "ne.o" instead
of just "ne" as an argument for insmod.
Hoyt
------------------------------
From: "Ron" <[EMAIL PROTECTED]>
Subject: samba cant see win98 shares
Date: Fri, 23 Apr 1999 13:44:14 -0700
new to linux
I have a small network of 1 linux redhat 5.2 kernel 2.0.36 running samba
1.9.18p10 and 1 win98 computer.
I recently changed my network security from share level to user level, and I
can't see my windows shares on the network anymore. I can see and access my
redhat shares from windows network neighborhood. Some of the symtoms which
seem to provide clues to the origin of the problem include the following:
When trying to set up the windows printer as a shared network printer I
recieve this message " you cannot view the list of users at this time.
Please try again later."
from windows DOS prompt running net view \\windowsHostname I get:
C:\>net view \\192.168.1.1
Error 1311: There are currently no logon servers available to service the
logon request.
in linux running smbclient -L windowshostname I get:
smbclient -L windowshostname
Added interface ip=192.168.1.254 bcast=192.168.1.255 nmask=255.255.255.0
SMBnegprot failed. myname=LINU destname=CERI - ERRSRV - 1311
and running testparm smb.conf I noticed 2 peculiar settings that might be a
problem, but I dont know how to change them.
protocol = NT1
announce as = NT
I see in Sery's book "Linux Network Toolkit" these are set as:
protocol: 8
announce as: 1
Can anyone identify the problem or point me to where to look for a solution?
TIA
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Greg Weeks)
Subject: Re: port forwarding
Date: Fri, 23 Apr 1999 19:17:20 -0500
In article <[EMAIL PROTECTED]>,
Raymond Doetjes <[EMAIL PROTECTED]> writes:
> I would advice you to forget the thing emailed by the guy before me.
> Since ipportfw is pretty unstable!!! and in some cases pretty hard to
> configure right.
It's very stable on the 2.0.36 kernels. I've heard of problems with
the 2.2.x kernels.
> I suggest you download the rinetd software. This is so very simple and
> so very good and stable.
> ALl you have to do is make a rinetd.conf file in the /etc dir with
> something like this in it.
>
>
> 0.0.0.0 80 192.168.1.1 80
> 0.0.0.0 2333 192.168.1.1 23
>
> This tells that a call to port 80 on one of the adapters will be
> forwarded to 192.168.1.1 on its port 80
> the other tells that a call on 2333 is being forwarded to 192.168.1.1 on
> port 23.
rinetd is very similar to redir. They're both very usable. They are a
little slower than ipportfw. The main problem I have with them is they
will show all accesses as from the firewall machine's internal IP
address in all logs on the internal server. If you don't care about
the logs then there isn't much of a reason to use ipportfw.
Greg Weeks
--
http://durendal.tzo.com/greg/
------------------------------
Subject: linux computers for sale
From: [EMAIL PROTECTED] (warlock)
Date: Sat, 24 Apr 1999 01:35:43 GMT
I bought 2 for my office servers... great systems.
http://homestead.com/linux2000
------------------------------
From: Derek Shaw <[EMAIL PROTECTED]>
Subject: Re: how do you share a single email address?
Date: Sun, 18 Apr 1999 09:36:19 GMT
fetchmail can be configured to do a credible job of this. I saw
instructions for doing this in the IDG book "The Liunux Network" (ISBN
1-55828-589-X). Here's a really brief summary.
1 - you must have your own domain name.
2 - You must arrange with your ISP that all e-mail to your domain is
directed into one psuedouser account.
3 - .fetchmailrc must be edited
poll myisp.com
protocol POP3
localdomains mydomain.org
no dns
no envelope
user pseudouser with password pseudouserpassword to
fred
peter
here
make a script that invokes fetchmail (e.g. --
fetchmail -f /usr/local/etc/.fetchmailrc -syslog )
configure syslog to handle error reporting how you want
run the script with cron as often as you see fit.
Peter Marks wrote:
> I'm interested in finding a way to share a single email account
> amongst a group of users who have accounts on a linux box.
>
> Basically as the email needs to go to a single address [EMAIL PROTECTED]
> there needs to be some clue as to who it should get forwarded to.
>
> I've seen a scheme that uses a line either in the subject or the first
> line of the mail in the form "Attn: peter" or there was a trick that
> had the real name along side the <email@address> "Fred Smith" that is
> used.
>
> Can someone point me to a way to do the mail sorting with sendmail (or
> perhaps fetchmail can do this).
>
> Many thanks,
>
> peter
--
Derek Shaw
Business Information Systems
Victoria, BC.
voice: 250-885-2021 fax: 250-386-4060
------------------------------
From: [EMAIL PROTECTED] (Armand)
Subject: Re: POP & SMTP
Date: 24 Apr 1999 01:40:15 GMT
Reply-To: [EMAIL PROTECTED]
It is the Mail-HOWTO that explains quit efficiently how to set up qmail.
In article <[EMAIL PROTECTED]>,
Seb Bacon <[EMAIL PROTECTED]> writes:
> try fetchmail & qmail (start at www.qmail.org). you need two
------------------------------
From: Ali Gonili <[EMAIL PROTECTED]>
Subject: talk from prompt to xwindows
Date: 24 Apr 1999 02:32:21 GMT
I tried to do a talk request from one machine to another through a LAN we
have setup... We can successfully talk, ftp, and mount various components
of the machines, but when I do a talk request from my prompt to the other
machine, it only works if the other machine is not in an xwindows
environment. Even if I have a xterm or a xiterm open, then talk request is
not seen and I only get a beep. I only see the talk request when I exit my
xsession. Any suggestions?
================== Posted via SearchLinux ==================
http://www.searchlinux.com
------------------------------
From: "Michael T. Spears" <[EMAIL PROTECTED]>
Subject: How to get SNMP information from my Linux box
Date: Fri, 23 Apr 1999 23:07:06 -0400
I use MRTG running on my Linux box to pull stats from a couple switches and
a Sun box. When I tried to pull from my Linux box, I think I got a
connection rejected.
I have not configured anything on my box in the way of server side SNMP
stuff (i.e. community names, who can/can't pull from it, etc). This is a
Debian 2.x install, so maybe some of this is pre-configured.
First, is SNMP reporting a function of the kernel or a separate daemon.
Second, is there a good resource on this. I did not see a How-To on it.
Thanks,
Mike
------------------------------
From: "J. S. Jensen" <[EMAIL PROTECTED]>
Subject: Re: Two network cards
Date: Fri, 23 Apr 1999 17:04:18 -0600
Zoltan Pittner wrote:
> 200.200.200.0 * 255.255.255.0 U eth1
> 200.200.200.0 * 255.255.255.0 U eth0
> 127.0.0.0 * 255.255.255.255 lp
> in this case the eth0 will never work.
No, because the networking code will automatically match the packet as a
200.200.200.0 packet through the first routing entry, then shove it off to
eth1
--
J. S. Jensen
mailto:[EMAIL PROTECTED]
http://www.Paramin.COM
------------------------------
From: "glen" <[EMAIL PROTECTED]>
Subject: Need help with DHCP on SuSE 6.1
Date: Sat, 24 Apr 1999 23:19:29 -0000
So I have SuSe 6.1 running, and sometimes I get a valid IP address, but I
never get a hostname from my server. I've tried following the instructions
to run dhclient and dhcpd, both which came with the CD, and I have also
tried running other dhcp's which I downloaded. I cannot get any of them to
work. I also try to get the dhcp setup running in YAST, but it is
highlighted so is not an option. I really don't know what other info to
give, but I have tried everthing I can think of. Any suggestions welcome,
Glen
------------------------------
From: "Curt" <[EMAIL PROTECTED]>
Subject: Netscape and internet
Date: Fri, 23 Apr 1999 19:01:07 -0500
i have two machines connecting through a hub to my cable modem...the problem
arises when i use netscape...i can ping my other win 98 machine and the
gateway without a hitch..but in netscape i get two errors...netscape is
unable to locate the server:blah blah the server does not have a dns
entry...check the server name in the location and try again..the other
message is...warning : the folowing hosts are unkown
home.netscape.com
home6.netscape.com
internic.net
this means that some or all hosts will be unreachable.
perhaps ther is a problem with your name server?
if you site must use a non-root name server, you will need to set the
$socks_ns environment variable to point at the apporpriate name server. it
may (or may not ) be nesessary to set this variable, or the socks host
preference, to the ip address of the host in question rather than its name.
this is my nework information on the linux box
Names
Hostnames: eaglecom.net
Domain: eaglecom.net
Search for hostnames in additonal domains:
Nothing
Nameservers: 216.96.32.1
Hosts
IP Name Nicknames
127.0.0.1 localhost localhost.localdomain
216.96.32.XX .eaglecom.net cvonlintel
Interfaces
Interface IP proto atboot
active
lo 127.0.0.1 none yes
active
eth0 216.96.32.XX none yes
active
Routing
Network Packet Forwarding (IPv4)
Default Gateway: 216.96.32.1
Default Gateway Device: eth0
interface network address netmask
gateway
nothing in these intries
win98 machine
IP address: 216.96.32.XX
Subnet Mask: 255.255.255.0
WINS Configuration: Disabled
Gateway: 216.96.32.1
DNS Configuration
Enabled
Host : cvonlintel Domain: eaglecom.net
SNS Server Search Order
207.1.165.3
207.182.160.2
any help would be appreciated
thanks
------------------------------
From: "David K. Means" <[EMAIL PROTECTED]>
Subject: Re: Anything wrong with using a 192.168.0.x private network?
Date: Fri, 23 Apr 1999 17:19:18 -0700
Tim Gibson <[EMAIL PROTECTED]> wrote in message
news:bt6U2.55$[EMAIL PROTECTED]...
> Am I ever going to run into any problems if I use a 192.168.0.x private
> network? I am afraid that the zero will be thought of as a network like
> when you have 192.168.0.0. Should I just play it safe and start with
> 192.168.1.x? I need multiple subnets (192.168.2.x and 192.168.3.x) so I
> will be using a 255.255.255.0 subnet mask. Thanks.
The folklore about not using all-zeroes and all-ones has gotten blown
further and further out of proportion. The actual situation is that you
should
not use HOST number 0 or all-ones on any subnet, since all-ones is the
broadcast address, and some brain-dead software from the distant past
uses all-zeroes as the broadcast address. But this does not, so far as I
know
extend into the network number part. Especially in the case you cited, you
are in the Class C network space, and so 192.168.0.x and 192.168.1.x
are actually different networks, not just different subnets.
So, to make a long flame short, feel free to use 192.168.0.x, but avoid
the host addresses 192.168.0.0 and 192.168.0.255
------------------------------
From: [EMAIL PROTECTED] (Cesar da Silva)
Crossposted-To: linux.redhat.misc
Subject: ISP hookup
Date: Sat, 24 Apr 1999 00:05:06 GMT
Reply-To: [EMAIL PROTECTED]
Hi!
I'm running Red hat v5.2 and I have a problem with my host name.
Before I dial-up to my ISP provider (with 'ifup ppp0') I have my host
name as it should (localhost).
But as soon as I connect to my ISP-provider I get something else
(t0985645). And the new host name stays even after I've logged off.
What I want and what I think is supposed to be is that my host name
remain the same when I log in, and after I've disconnected, to my
ISP-provider.
How do I fix this? What am I doing wrong?
Thanks,
Cesar da Silva
------------------------------
From: [EMAIL PROTECTED] (John Hornblow)
Subject: Logging onto Samba - not yet
Date: Sat, 24 Apr 1999 04:56:02 GMT
Reply-To: [EMAIL PROTECTED]
Hi,
I can see my lynix server from my NT workstation...
I can ping it... and via versa, Ive enables public access in the
config file etc but when I click on the icon in network nebourhood I
get prompted for username and password
I get: "you are not authorised to log on from this station"
the "station" is in the HOSTS file...
does my workstation need an account similar to NT workstations
requiring accounts on NT server as well as users?
thanks
John
=====================================================
John Hornblow
homepage http://homepage.ihug.co.nz/~johnhb/
gliding page http://www.soar.co.nz/
=====================================================
------------------------------
From: [EMAIL PROTECTED] (John Hornblow)
Subject: Re: samba is running, but I get account not authorised to log on..
Date: Sat, 24 Apr 1999 05:02:01 GMT
Reply-To: [EMAIL PROTECTED]
just an update, I ran sbmclient -L server
the client has a browser list...
it shows all the shares Ive enabled...
but still "this workstation is not authorized to log on"
what step have I missed?
double thanks!
John
=====================================================
John Hornblow
homepage http://homepage.ihug.co.nz/~johnhb/
gliding page http://www.soar.co.nz/
=====================================================
------------------------------
From: =?US-ASCII?B?ofOk363CxW+h8g==?= <[EMAIL PROTECTED]>
Subject: ipchains under kernel 2.2.x
Date: 24 Apr 1999 05:33:14 GMT
Something will go wrong,
such as ftp site that uses non-standard port(21),
connections of some web sites will fail...etc
Is there any way to fix these problem?
Any document?
thanks.
--
---=============--- PENG Chung-Feng ---====================---
*** Department of Computer Science and Information Engineering ***
*** National Chiao Tung University ***
*** [EMAIL PROTECTED] \ [EMAIL PROTECTED] ***
---=============--- Clavin at Life ---====================---
------------------------------
From: [EMAIL PROTECTED] (Luca Filipozzi)
Crossposted-To: comp.os.linux,alt.os.linux
Subject: Re: Must replace Cisco with Linux in a hurry
Date: Fri, 23 Apr 1999 20:50:34 -0700
In article <[EMAIL PROTECTED]>,
[EMAIL PROTECTED] says...
> In article <7fq9qn$6nm$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
> says...
> > X-Article-Creation-Date: Fri Apr 23 17:15:07 1999 GMT
> > X-Http-User-Agent: Mozilla/3.04 (Win95; I; 16bit)
> > X-Http-Proxy: 1.0 x1.dejanews.com:80 (Squid/1.1.22) for client 209.113.193.216
> >
> > In short: Can I hook a 56k DSU that seems to have a serial interface to a
> > Linux box and use the Linux box as a router to replace my dead Cisco?
> >
>
> Basically, you need to buy a synchronous serial card that is supported by
> Linux. You should also take a look at: http://www.etinc.com/
>
> Also, take a look at this Linux Journal article:
> http://www.linuxjournal.com/issue50/2928.html
>
> which talks about a router card from this company:
> http://www.sangoma.com/
>
> This web page has links to a lot of serial card vendors:
> http://www.ntlug.org/~cbbrowne/lincomhard.html
>
> You could also look at www.blackbox.com to see if they have any "dumb"
> synchronous serial cards.
>
>
I found more stuff.
ImageStream sell an all-in-one-box solutions (running Linux):
http://www.imagestream-is.com/
They also resell (and use) the synchronous cards from SDL Communications:
http://www.sdlcomm.com/
The RISCom/N2s is what you're after.
The pricing at ImageStream for this card is $655. ImageStream indicates
Linux support. I didn't surf the SDL Communications site for price or
support.
The price seems high to me... let me know if you find cheaper stuff.
--
Luca Filipozzi <[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
******************************
