Linux-Networking Digest #231, Volume #11 Fri, 21 May 99 17:13:44 EDT
Contents:
Linux as router for 2 NT subnets ? (Christian Motschke)
Re: Modem and Sound Card Problem - RedHat 6 on Presario 5630 (TS Stahl)
Re: PPP Problem: What does my ISP want? (Clifford Kite)
Local IP addresses (David Hinz)
Re: ip-routing / maquerading / dual internet gateway (Ben De Rydt)
Re: Firewall for linux (ByteMe)
Re: PPP Help Please... (Roger Plant)
Can't communicate through 2nd NIC ("Steve Snyder")
Re: Login BellSouth via Linux ? ([EMAIL PROTECTED])
Re: How to avoid down router? (ByteMe)
----------------------------------------------------------------------------
From: Christian Motschke <[EMAIL PROTECTED]>
Crossposted-To: comp.protocols.smb
Subject: Linux as router for 2 NT subnets ?
Date: 21 May 1999 15:28:21 +0200
Hello,
I use a linux server acting as a router between 2 subnets 192.168.1.0
and 192.168.42.0 . The Server is configured as router and all TCP/IP
protocols works well between two hosts in different subnets. The
server acts also as a domain-controller for the 192.168.42.0
domain. The NT-Domain of the 2 subnets are called differently. The
problem is, that clients in one subnet can't see the services of the
clients in the other subnet. Only the linux router will be seen. It is
not a problem of routing, because the communication that uses TCP/IP
between 2 clients in different subnets (like ftp, ping etc) works.
Someone suggested that I should install an IPX router on linux as
well. Is this really the only solution? (I allready tried it, with no
success)
TIA Christian
--
====================================================================
Christian Motschke [EMAIL PROTECTED] [EMAIL PROTECTED]
====================================================================
------------------------------
From: TS Stahl <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: Re: Modem and Sound Card Problem - RedHat 6 on Presario 5630
Date: Fri, 21 May 1999 10:59:23 -0500
The RH manual has a very good discussion of setting up ppp. Be warned, if you
have a 'winmodem' in the compaq, then you are screwed.
Happy computing.
Brian Stevenson wrote:
> Greetings,
>
> I have installed RedHat 6 over a variety of computers, and I was lucky
> enough to encounter no serious installation issues. However, when I tried to
> install RedHat 6 on my girlfriend's computer (a Compaq Presario 5630), I
> couldn't get the sndconf program to recognize the PCI sound card on her PC
> (all it said was that no PCI sound card was detected), even though that
> sound card is an ESS brand; Linux sees that it occupies IRQ 5, DMA 1, I/O
> Range $2000-$20FF. Needless to say, since we cannot get any sound playing,
> we are both very disappointed :-(
>
> A few days later, my girlfriend decided to bring her computer home, not
> hooking it to the LAN at her college dorm anymore. Now she can't get the
> modem to dial up to the Internet. It looks like the modem is recognized,
> and we have set up everything necessary (loginID, password, dial-up phone
> number), but the modem cannot be activated. Now the supposedly powerful box
> is just sitting there, looks pretty nice with the KDE interface and stuff,
> but cannot play any sound or connect to the Internet.
>
> I would be very grateful if somebody knowledgeable can give me some tips how
> to get the sound card recognized, playing sound, and the Linux box can
> dial-up to the ISP. Thank you very much.
>
> Brian Stevenson
--
Scott Stahl
MIS Asst.
Illinois Housing Development Authority
------------------------------
From: Clifford Kite <[EMAIL PROTECTED]>
Subject: Re: PPP Problem: What does my ISP want?
Date: Fri, 21 May 1999 08:33:40 -0500
Rand Simberg wrote:
>
> Here's the log:
> May 19 17:34:23 localhost pppd[1669]: pppd 2.3.7 started by root, uid
> 0
> May 19 17:34:24 localhost chat[1670]: abort on (BUSY)
> May 19 17:34:24 localhost chat[1670]: abort on (NO CARRIER)
> May 19 17:34:24 localhost chat[1670]: send (ATDT13108960011^M)
> May 19 17:34:24 localhost chat[1670]: expect (CONNECT)
> May 19 17:34:53 localhost chat[1670]: ATDT13108960011^M
> May 19 17:34:53 localhost chat[1670]: CONNECT
> May 19 17:34:53 localhost chat[1670]: -- got it
> May 19 17:34:53 localhost chat[1670]: send (^M)
> May 19 17:34:53 localhost chat[1670]: expect (login:)
> May 19 17:35:03 localhost chat[1670]: 115200^M
> May 19 17:35:03 localhost chat[1670]: netcom login:
> May 19 17:35:03 localhost chat[1670]: -- got it
> May 19 17:35:03 localhost chat[1670]: send (us,ppp,simberg^M)
> May 19 17:35:03 localhost chat[1670]: expect (Password:)
> May 19 17:35:13 localhost chat[1670]: ^M
You may have the wrong IRQ configured as the modem's IRQ for the device
node /dev/ttyS2 . The CONNECT '' expect/send sends a carriage return
when the CONNECT string from the modem is found. The ^M in the last
line above may be this carriage return and if this is so then the IRQ
(configured by setserial in a boot-up file) is not the the IRQ that the
modem actually uses. The time difference between the time the carriage
return is sent and the time ^M appears is 20 seconds which is
suspicious.
You can determine with certainty whether the IRQ is the problem by using
'' ATZ as the first chat expect/send and OK ATDT13108960011
as
the second. Note the time difference between the time when the ATZ is
sent and the time the OK appears. A difference of more than 1 or 2
seconds
means the IRQ of the device file is misconfigured (19 seconds is common
for a misconfigured IRQ).
It could be something else though. The carriage return sent by using ''
as a send in an expect/send sometimes confuses an ISP. It's better to
use
'\d\c' which adds a small delay and doesn't send a carriage return.
---
Clifford Kite Not a guru.
(tm)
[EMAIL PROTECTED]
------------------------------
From: David Hinz <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.sys.next.sysadmin
Subject: Local IP addresses
Date: Fri, 21 May 1999 20:27:19 GMT
In the IP addressing scheme isn't there a set of IP addresses that can
be assigned to a local system so that they aren't seen by the outside
world. I'm trying to network my home systems together and I need to
assign IP addresses to each system. But I also want to be able to get
to the Internet via PPP. But I don't my local IP address visible beyond
my local network.
It seems like they were 192.???.???.??? addresses.
Thanks,
david hinz
please cc any responses to [EMAIL PROTECTED]
------------------------------
From: Ben De Rydt <[EMAIL PROTECTED]>
Crossposted-To: be.comp.os.linux,nl.comp.os.linux
Subject: Re: ip-routing / maquerading / dual internet gateway
Date: Fri, 21 May 1999 22:28:44 +0200
Lieven Van Acker wrote:
> CENTRAL LINUX GATEWAY
>
> | | | | |
> | | | | ---> eth0: Internet link 1: leased line, public fixed IP
> addr (ISP1)
> | | | ------> eth1: Internet link 2: cable-modem, dhcp configured
> (ISP2)
> | | ---------> eth2: DMZ
> | ------------> eth3: Internal subnet 1, private address range,
> masqueraded
> ---------------> eth4: Internal subnet 1, private address range,
> masqueraded
>
> What we want to reach is explained simply:
>
> All incoming traffic ( to the gateway and DMZ ) has to come in via
> internet link 1 (Leased Line).
> All outgoing traffic except SMTP traffic has to go via internet link 2
> (Cable-Modem).
> All outgoing SMTP traffic has to be routed via the Leased Line.
>
> Outgoing traffic from our private subnets is masqueraded before is
> leaves via eth1 and is blocked to go via eth0.
In het Nederlands want dat ligt me beter:
Ik begrijp de vraag niet helemaal, maar 'k zal toch proberen.
De eenvoudigste oplossing lijkt mij twee routers te gebruiken, waarbij
��n router (router 1) deel uitmaakt van het netwerk dat de tweede bedient
(router 2).
router 1:
- eth0: kabelmodem
- eth1: deel van het netwerk van router 2 bv.
- eth2: intern subnet 1
- eth3: intern subnet 2
router 2:
- eth0: leased line
- eth1: intern netwerk: bevat mailserver, ... en router 1.
Gebruik static routing:
router 1:
- route naar netwerk met router 2 via eth1
- default route naar Internet Provider kabelmodem (masquerading) via eth0
- routes naar de interne subnetten via eth2 en eth3
router 2:
- default route naar Internet Provider leased-line via eth0 (hier ook
masquerading?)
- route naar alle subnetten via router 1 -> eth1
op de subnetten:
- gateway instellen op router 1
- mailserver instellen op de mailserver (router 1 zal dit zelf routen)
op het netwerk van router 2:
- gateway instellen op router 2
Zo zal alle verkeer van de subnetten langs de kabelmodem gaan, en alle
verkeer van de mailserver langs de leased line. Tevens heb je een klein
netwerkje met de mailserver e.d. dat je goed kan afschermen van de
interne subnetten.
Groeten,
Ben
=====================================================================
mailto:[EMAIL PROTECTED] ---------- http://users.pandora.be/bdr/
------------------------------
From: ByteMe <[EMAIL PROTECTED]>
Subject: Re: Firewall for linux
Date: Fri, 21 May 1999 12:29:10 -0700
Alamin wrote:
> I am new to linux and have been looking for a firewall for a RedHat linux
> box that I want to install. I basicly have 2 networks, one that has a
> internet connection(perment) and the internal network. Would like to join
> these two so that I can transfer data to and from the internet connection
> box. The internet connection bix is running window nt and IIS 4.0. The
> website on this box accesses SQL 6.5 and 7.0 databases. I guess I am looking
> for something like MS Proxy or a firewall package.
>
> Thanks
>
> Steve Berra
ipfwadm generaly does come with linux distributions , weather you installed
it or not , who knows ( do a rpm -q ipfwadm ) to see weather it is
installed .
there is more that you have to do than just install ipfwadm. If you are running
a new kernel
( 2.0.XX ) then you do not have to recompile it ( if forget the exact kernel
ver. that had ip_forwarding pre compiled into the kernel )
The following assumes that you have RH 5.2 or lower ( they made some changes in
Ver. 6.0 )
specificaly they use ipchains instead of ipfwadm but most of this should apply.
( I have'nt played with 6.0 yet , I'm leary of it )
required :========================================================
( SETTING UP THE NET CARDS )
2 nic's installed
don't use cheap cards as linux has a tendancy blow them out
!!
( i.e use 3com , SMC, NE2000 or something high qual. )
( pci cards are also much harder to setup , it depends on
needs)
you need to know the IRQ & IO that the cards are using
( do not autoprobe the cards )
edit the file ( /etc/conf.modules )
/* (there should be an entry already in there something
similar to: ) */
/* alias eth0 ( xx ) where xx is
the lib module for your card */
/* options ( xx ) io=0x300 irq=5 ne = ne2000 card,
wd = western digital */
( if you are using two different cards )
alias eth0 ne
alias eth1 wd
options ne io=0x300
options wd io=0x280
(else if both cards the same)
alias eth0 ne
alias eth1 ne
options ne io=0x300,0x280
look at the HOWTO's for seting up ethernet cards for more
info
===================================================================
edit the file ( /etc/lilo.conf )
( just above the line that starts with /*
image=/boot/vmlinuz */ or similar )
add the line : append="ether=5,0x300,eth0
ether=3,0x280,eth1"
/*
boot=/dev/hda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
append="ether=5,0x300,eth0 ether=3,0x280,eth1"
image=/boot/vmlinuz-2.0.34-0.6
(blah blah blah and so fourth)
=================================================
create the file ( /etc/sysconfig/network-scripts/( ifcfg-eth1) )
DEVICE=eth1
IPADDR=xxx.xxx.xxx.xxx
NETMASK=255.255.255.0
NETWORK=xxx.xxx.xxx.0
BROADCAST=xxx.xxx.xxx.255
ONBOOT=yes
( you should aready have a file
/etc/sysconfig/network-scripts/ifcfg-eth0 )
( follow the syntax in that file if different than above )
===================================================================
edit the file /etc/sysconfig/network
NETWORKING=yes
FORWARD_IPV4=true ( this line is probably set to false )
HOSTNAME=( somewhere.athome.com )
DOMAINNAME= ( athome.com )
GATEWAY= ( you need to find this info out from your ISP )
GATEWAYDEV=eth0
( this sets up your overall networking options for the
system )
===============================================================
edit the file /etc/hosts
add :
(your.static.ip) ( your hostname )
/*
127.0.0.1 localhost
localhost.localdomain
( static ip for eth0) ( hostname) (
hostname.domain.com )
( static ip for eth1) ( hostname )
(hostname.domain.com )
*/
(this is required at first because some deamons i.e
sendmail have a difficult time)
( figuring out which card to listen to the next time you
boot, and may hang )
( later you can delete these enrtires , but that gets
into security issues as to why)
( you should not have any trusted name entries , something
you will have to)
(pay serious attention too later after you get it all
setup properly )
==================================================================
you can rebbot your machine at this point too make sure that your NIC's init
correctly,
if they don't, no point in going on until this issue is resolved !!!
try to ping both cards to make sure they have initialized after booting.
if they have at this point you should be able to ping out to the internet (
try 208.1363108.1 )
or any static ip out there.
and then you should be able to ping your work station ( providing that it is
setup according to
to your private net settings at this point !!
and then try to ping from your work station to your eth1 static ip on your
linux box.
if (yes ) {
continue on();
}
setting up two network cards the first time can be very frustrating , the above
has worked on every machine i've done in the past ( which is quite a few ) read
the HOWTO's
====================================================================
edit the file /etc/rc.d/rc.local
go to the end of the file and add:
/sbin/insmod /lib/modules/2.0.34-0.6/ipv4/ip_masq_ftp.o
/sbin/insmod /lib/modules/2.0.34-0.6/ipv4/ipip.o
/sbin/insmod /lib/modules/2.0.34-0.6/ipv4/ip_alias.o
( there are more mods you can load for other services the
main ones here)
(are the ip_masq_ftp.o & ipip.o , this enables your
forwarding modules that)
( ipfwadm needs too accomplish masquerading and forwarding
)
(you can compile all of this directly into your kernel if
you want, it's up to you )
then add:
/sbin/ipfwadm -F -p deny
/sbin/ipfwadm -F -i accept -S (your.workstation.ipaddr)/32
-D 0.0.0.0/0 -m
( -S is the source of the packet)
(/32 says that it's a full ip addr. and not the network
addr.)
(-D is for the destination addr)
(0.0.0.0/0 is for any A,B,C network i.e nomask)
(-m is for masquerade )
( the syntax here may be different than others look under
Firewall-HOWTO for )
( a better description of all the possibilties , this is
just to get you up and running)
( DO NOT use this permanently for your firewall policies)
(obviosly your workstation has to broadcast on the same
network as your gateway)
(address which will be your address on eth1 not your ISP's
gateway addresss)
(set your workstions gateway pointer to your eth1 cards
address)
( in general your eth0 card should be hooked to the
internet & eth1 will be )
( on your private network at home or the office )
( ipfwadm uses the static routes decided when your machine
boots which gets setup)
(by ifconfig using the scripts found in
/etc/sysconfig/network-scripts dir)
(no need to setup up the routes manualy because we created
)
(the ifcfg-eth1 script already)
(ifconfig will read this script at each boot )
I believe I have covered everything , at least this will
give you something
to start working with
------------------------------
From: [EMAIL PROTECTED] (Roger Plant)
Subject: Re: PPP Help Please...
Date: Fri, 21 May 1999 13:43:30 GMT
On Thu, 20 May 1999 15:17:02 GMT, [EMAIL PROTECTED] (Greg Aeschliman)
wrote:
>I'm starting to go a bit crazy trying to get connected to my ISP here....
>
> May 20 18:25:45 localhost pppd[6343]: Connect: ppp0 <--> /dev/cua2
> May 20 18:26:16 localhost pppd[6343]: Receive serial link is not 8-bit
>clean:
>
> May 20 18:26:16 localhost pppd[6343]: Problem: all had bit 7 set to 0
>
Not being 8 bit clean is generally bad, I can't think of an obvious
reason this might occur.
One option may be to dial into the host using minicom, and make sure
it's actually sending ppp type stuff. (Some hosts want you to enter a
password, and account (by hand) before they will start ppp)
Generally ppp stuff seems to look like lots of }}}'s
You could try using ttyS2 rather than cua2, but I can't see that
fixing it.
While pppd is trying to connect try a stty -a < /dev/ttyS2 (or
/dev/cua2).
And make sure it is saying cs8, and all the other parameters look ok.
If it isn't something must have changed it, because pppd sets it up
for cs8.
That's all my guesses
Good luck
Regards
Roger
===========================================================
Roger Plant :-) Email: [EMAIL PROTECTED]
===========================================================
------------------------------
From: "Steve Snyder" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Can't communicate through 2nd NIC
Date: Fri, 21 May 1999 14:32:50 -0400 (EST)
Reply-To: "Steve Snyder" <[EMAIL PROTECTED]>
I am attempting to configure my server (RedHat v5.2 /w kernel v2.2.9) to
act as a gateway to @Home through a cable modem.
Apart from any name resolution issues that I have not yet addressed, I
can't even communicate through the 2nd NIC using a static IP address. Note
that I am not yet trying to communicate across my LAN, but to just
establish communications locally on the server with the Internet via the
cable modem.
Attempting to ping @Home's name server (IP = 24.4.162.33) fails. The ping
just hangs until killed. Attempts to telnet to the same IP address gets me
the message "Unable to connect to remote host: No route to host". Hmm...
Device eth0 (IP = 192.168.0.12) is connected to my LAN's hub and has been
working all alone. Device eth1 (IP = 24.4.162.173, assigned by @Home) is
the NIC connected to the cable modem. The LEDs on the 3Com USR cable modem
indicate that it is communicating with my NIC and that it recognizes the
signal on the coax cable.
Here is where I'm at:
At boot time
==========
eth0: 3Com 3c905B Cyclone 100baseTx at 0xe400, 00:10:4b:9a:82:e5, IRQ 11
8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface.
MII transceiver found at address 24, status 786d.
MII transceiver found at address 0, status 786d.
Enabling bus-master transmits and whole-frame receives.
eth1: 3Com 3c900 Boomerang 10Mbps Combo at 0xe800, 00:60:97:c8:01:c8, IRQ 10
8K word-wide RAM 3:5 Rx:Tx split, autoselect/10baseT interface.
Enabling bus-master transmits and whole-frame receives.
#cat /etc/sysconfig/network
=======================
NETWORKING=yes
FORWARD_IPV4=yes
HOSTNAME="corona.snydernet.lan"
DOMAINNAME=snydernet.lan
GATEWAY=24.4.162.173
GATEWAYDEV=eth1
#cat /etc/sysconfig/network-scripts/ifcfg-eth0
=======================================
DEVICE="eth0"
IPADDR="192.168.0.12"
NETMASK="255.255.255.0"
NETWORK=192.168.0.0
BROADCAST=192.168.0.255
ONBOOT="yes"
BOOTPROTO="none"
#cat /etc/sysconfig/network-scripts/ifcfg-eth1
=======================================
DEVICE="eth1"
IPADDR="24.4.162.173"
NETMASK="255.255.255.0"
ONBOOT="yes"
BOOTPROTO="none"
#netstat -nr
==========
Destination Gateway Genmask Flags MSS Window irtt Iface
192.168.0.12 0.0.0.0 255.255.255.255 UH 0 0 0 eth0
24.4.162.173 0.0.0.0 255.255.255.255 UH 0 0 0 eth1
192.168.0.0 0.0.0.0 255.255.255.0 U 0 0 0 eth0
24.4.162.0 0.0.0.0 255.255.255.0 U 0 0 0 eth1
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
0.0.0.0 24.4.162.173 0.0.0.0 UG 0 0 0 eth1
Can someone advise me on what to configure just to get basic communications
going? If there's any info missing from above I would happily provide it.
Thank you.
***** Steve Snyder *****
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: bellsouth.net.support
Subject: Re: Login BellSouth via Linux ?
Date: Fri, 21 May 1999 16:33:35 -0400
Bellsouth uses Username:(use sername:) instead of login:
they also use PAP
Wild Cat wrote:
> Hi there !
>
> I try to use BellSouth with my LinuxBox and X-ISP. I believe that I have some
> troubles to know exactly want BellSouth requires for the connection. The link
> is made, but it stucks at <ogin:>. Apparently, the login is not sent or not
> received. I do not know if BellSouth requires a PAP-Secrets or CHAP-Secrets.
> In that case would they (BellSouth) mind giving their remotename ?
>
> Or if somebody has a easier way to connect to Internet via BellSouth, I will
> be glad to know it. :-)
>
> Thanks you all in advance.
>
> Michael
>
> (PS: This message has been posted in bellsouth.net.support and
> comp.os.linux.networking)
------------------------------
From: ByteMe <[EMAIL PROTECTED]>
Crossposted-To: [EMAIL PROTECTED]
Subject: Re: How to avoid down router?
Date: Fri, 21 May 1999 12:39:27 -0700
[EMAIL PROTECTED] wrote:
> In article <[EMAIL PROTECTED]>,
> Roger Rouse <[EMAIL PROTECTED]> wrote:
> > My ISP has nothing to do with Sprint.
>
> -- [Your traceroute output from a previous post] --
> 1 207.225.190.254 (207.225.190.254) 25.591 ms 18.83 ms 20.535 ms
> 2 premiumB6.phnx.uswest.net (209.180.138.6) 20.268 ms 19.28 ms 25.765
> ms
> 3 206.80.192.219 (206.80.192.219) 20.885 ms 22.12 ms 18.763 ms
> 4 sl-gw12-ana-1-1-1.sprintlink.net (144.228.207.205) 26.411 ms 28.484
> ms 28.641 ms
> [...]
> --
>
> Sprint is one of your ISP's peering partners, as is evidenced from the
> fact that there is a direct link between one of your ISP's routers
> [namely, 206.80.192.219 - owned by USWest] and one of Sprint's routers
> [namely, 144.228.207.205]. This means that they have an agreement with
> Sprint to carry some of their traffic. They either pay for this
> outright, or trade for it in exchange for other services.
>
> > This downed router is screwing me over and I have no options!
>
> Yes, such is life.
>
> > It's extremely frustrating.
>
> Indeed. Be glad you're not one of the engineers who have to *fix* such
> problems.
>
> > I do belive the purpose of having routers is to
> > avoid downed connections right?
>
> No. The purpose of routers is to provide connectivity between different
> networks. They make up the foundation of the "Internet", which is a
> network-of-networks, connected by machines which used to be called
> "gateways" but which are now called "routers". One of the features of
> most routing protocols [which define software used by routers to
> determine where packets go] is the ability to change routes based on
> which connections are live. Avoiding downed connections is a _feature_
> of routers, not their purpose. Some routers have only one upstream
> connection [their "default route"], and so they do not possess this
> feature.
>
> > I cannot believe that there is only one way
> > for me to get from point A to point B.
>
> How many ways there are for you to get from point A to point B depends
> on how the intervening machines are hooked up. There is almost always
> only one way to get to your default gateway [unless you have more than
> one NIC, and are multi-homed]. Your default gateway is connected to at
> least one other upstream router, which is also connected to at least one
> more router, and so forth. If each router upstream from you is
> connected to only one other router, and point B is a host on the network
> connected to one of those routers, there will be only one path. Even if
> some of the routers have more than one upstream path, there may still
> only be one way to get from A to B. It all depends on what routes are
> being announced.
>
> > There must be other routes.
>
> Usually, but not always.
>
> > I should be able to access these routes especially since I'm
> > paying for it.
>
> You are not "paying for it". You are paying for connectivity to your
> ISP's routers, nothing more. Your ISP has seperate agreements with
> other providers, like Sprint, probably UUNet, and others. *They* pay
> these companies [or provide like service] to carry your traffic outside
> their network. In turn, each of these companies may have peering
> arrangements with yet more providers, and *they* foot the bill for that
> part of the journey. The internet is more like a network of toll-roads,
> except you only have to pay the first toll yourself -- all the others
> are paid by the owner of the previous toll-road. Quite a bit of bang
> for your buck, you should be happy. Setting up a direct connection
> between, say, San Francisco and Italy would cost you quite a bit more
> than your usual rate from your ISP.
>
> > Other people's machines should not be allowed to grab my
> > packets and throw them into the bit bucket!!!
>
> It's not the fault of the downed router, it's the fault of the *prior*
> router that your packets get lost. Routers do not "grab" packets, they
> are sent to each router by the one before it. Which router to send your
> packets to is decided by the current router, and so you should be angry
> about the "last hop" router, not the downed one. Routers fail, it's a
> fact of life. What causes problems is when the *other* routers still
> try to send traffic to the downed machine, because they've failed to
> update their own routing tables correctly.
>
> > That is unacceptable.
>
> Simply saying something is "unacceptable" doesn't make it any less true.
>
> > Who at Sprint is responsible for this router?
>
> I have no idea. I'm sure Sprint has a 24-hour staff whose sole purpose
> it is to monitor such things.
>
> > How do I contact them?
>
> That would most likely be pointless. Like I said, routers fail all the
> time. Almost every aspect of their proper functioning is completely
> automated, since keeping track of [possibly] several million different
> available routes, all of which are changing on a very frequent basis, is
> a task only a computer could accomplish in a reasonable amount of time.
> By the time you notice that your connection is down, I assure you some
> poor engineer somewhere has already been paged and is working on the
> problem. Sometimes there really *is* only one way to get to point B
> [because of a route bottleneck somewhere, for example if the destination
> machine is deep inside a private network], and there isn't anything that
> can be done except fix the downed router [in those cases it's okay to be
> pissed at the down machine :)]. If the box has completely failed [fried
> power supply or whatnot], it will need to be replaced, or the network
> rewired. Both of those things can take some time.
>
> My advice is to wait patiently, or get up from your computer and go for
> a walk. If you find that too difficult, and feel the urge to do
> *something*, start a log of which machines fail at what times and for
> how long. Or take up knitting. Or something.
>
> --
> -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.---
That was a very good response bill , I hope more people read it !!!!!!
------------------------------
** 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
******************************
