Linux-Networking Digest #303, Volume #12 Fri, 20 Aug 99 15:13:59 EDT
Contents:
cbq route classifier problem (guenther)
Re: Ping Timeout? (Bill Pitz)
Re: Routing a 2nd real IP addr on a home net (Greg de Freitas)
Re: Delaying eth0 initiailization.... (Robert Lynch)
Re: Network Card on Deskpro 4000 ("Lee Sharp")
Re: Domain name? ("Michael Faurot")
Re: Why not C++ (Edwin Young)
Re: Help: Caching webpages... (Keith T. Garner)
Can Linux translate IPX<>IP? ("Jon Richardson")
Re: Can't get 2 NIC's to work (Frederic Pont)
Re: FTP and chroot (Jay Beattie)
problems with 2 networks (Tiberio, David)
Re: Restrict FTP users from deleting ("Pat Crean")
Re: Forwarding with ipchains ("Cedric Blancher")
Netgear FA310 and Micronics W6Li (John Meissen)
Re: Telnet thru' a firewall, any backdoor ??? (C.T.Nadovich)
Re: @Home with Muliple IP addresses and Linux (Stuart R. Fuller)
Re: Telnet thru' a firewall, any backdoor ??? (Mark Hahn)
Re: IPCHAINS forwarding and/or 2nd nic slowing down Internet access (Eric deRiel)
Re: NE2000 ISA help ("Bradley Dunn")
Re: looking for source for net2phone ip module (Hernan Coronel)
----------------------------------------------------------------------------
From: guenther <[EMAIL PROTECTED]>
Subject: cbq route classifier problem
Date: Fri, 20 Aug 1999 17:14:57 +0200
i have a problem configuring the routing based classifier. the ip command
always fails. i've also tried with the example in "Quality of Service
Support in Linux" by Saravanan Radhakrishnan but it won't work.
after configuring everything with tc i'll get the following error msg when
i add a route:
verleihnix# ip route add 194.152.160.71 via 194.152.160.71 flow 1:2
Error: either "to" is duplicate, or "flow" is a garbage.
i'll tried other examples too, but it's always the same error :-(
any suggestions ?
cu
-guenther
--
guenther starnberger, at-net
http://www.gst.priv.at/
------------------------------
From: Bill Pitz <[EMAIL PROTECTED]>
Subject: Re: Ping Timeout?
Date: Fri, 20 Aug 1999 15:40:18 GMT
Eran Dvey-Aharon <[EMAIL PROTECTED]> gave us the interesting posting of:
> The ping has timeout after 10 seconds. I use it just this way, and when
> have no connection simply wait ....
> You can modify the ping source to reduce that amount of time if it is
> critical to you.
I have been running, as a test, "ping -c 1 host" and it will sit and ping for
ever if the host is not alive. The packets do time out after 10 seconds if
I don't specify the -c, but it defeats the purpose of what I'm trying to do...
-Bill
------------------------------
From: Greg de Freitas <[EMAIL PROTECTED]>
Subject: Re: Routing a 2nd real IP addr on a home net
Date: Thu, 19 Aug 1999 14:46:59 +0100
Reply-To: [EMAIL PROTECTED]
Sounds like what you need is a floating IP addy.
--
Ciao 4 now, Greg.
# Email : mailto:[EMAIL PROTECTED] #
# Email : mailto:[EMAIL PROTECTED] #
# To Live, To Love, To Learn, To Leave A Legacy. #
------------------------------
Date: Fri, 20 Aug 1999 15:44:00 +0000
From: Robert Lynch <[EMAIL PROTECTED]>
Subject: Re: Delaying eth0 initiailization....
Speedy Fast wrote:
>
> On Fri, 20 Aug 1999 04:57:14 +0000, Robert Lynch <[EMAIL PROTECTED]>
> wrote:
>
> >Speedy Fast wrote:
> >>
> >> I am running a Realtek-8139 NIC in my Linux machine. At bootup it
> >> cannot find the NICs. It says somthing like "delaying eth0
> >> initalization". So once it's booted up I do a 'insmod rtl8139'
> >> followed by some other ifconfig stuff and what not.
> >>
> >> How can I get Linux to find the NICs at boot time?
> >
> >The couple of times I had this with a NE2000, I just edited
> >/etc/conf.modules so I had:
> >
> >alias eth0 ne
> >
> >(whereas it was previously set to "alias eth0 off", I believe). You
> >would do something like:
> >
> >alias eth0 rtl8139
>
> I have in my conf.modules
> options eth0 io=6100 irq=11
> alias eth0 rtl8139
> options eth1 io=6200 irq=10
> alias eth1 rtl8139
>
> And still no luck :(
Sorry; the problem with fixing stuff like this is you forget how you did
it. :)
Another place to look (on a RH system anyhow) is in:
/etc/sysconfig/network-scripts/ifcfg-ethx
the ONBOOT line:
DEVICE="eth0"
...
ONBOOT="yes" <-- this
might be the one that controls it.
Bob L.
--
Robert Lynch-Berkeley CA [EMAIL PROTECTED]
http://www.best.com/~rmlynch/
------------------------------
From: "Lee Sharp" <[EMAIL PROTECTED]>
Subject: Re: Network Card on Deskpro 4000
Date: Fri, 20 Aug 1999 10:11:36 -0500
[EMAIL PROTECTED] wrote in message <7pj1ag$abb$[EMAIL PROTECTED]>...
|I would like to install the network card on a Compaq Deskpro 4000.
|This is a Compaq netelligent 10/100 TX Embedded UTP Controler. My linux
|is RedHat 5.1.
|Where can i find the driver and how can i install the driver ?
|(Compilation, Configuration...)
It is a "tlan" card. It does suck. You will have to insert the module
manually... "insmod tlan" I never made it come up in a cold boot, but I
didn't work to hard at it. After all, how often do you need to boot Linux?
Just leave it up. :-)
The TI Thunder Lan card is also a CPU hog. Some real CPU load on this
card. It may be easier to find an older Dec Tulip card, and just use it. A
great card, with great support. But the Compaq one will work... A search
on deja with the keyword "tlan" will tell your more than you needed to know.
:-)
Lee
--
SCSI is *NOT* magic. There are *fundamental technical reasons* why it is
necessary to sacrifice a young goat to your SCSI chain now and then. * Black
holes are where God divided by zero. - I am speaking as an individual, not
as a representative of any company, organization or other entity. I am
solely responsible for my words.
------------------------------
From: "Michael Faurot" <[EMAIL PROTECTED]>
Subject: Re: Domain name?
Date: 20 Aug 1999 15:31:24 GMT
Yo <[EMAIL PROTECTED]> wrote:
: Linux box with 1 ip from Cable provider (came with a domain name like
: myhost.cableprovider.com) but I have also register for another domain
: name (mydomain.com) and I got a friend to host it for me which is up
: and running now That's mean myhost.cableprovider.com and mydomain.com
: can both access to my linux box by when I join IRC or some other thing
: It will display my domain name as myhost.cableprovider.com Is there a
: way I can change it my linux box to display as mydomain.com instead of
: myhost.cableprovider.com
Only with the cooperation of your ISP. When the IRC server gets a
connection from your machine, it does a reverse lookup to see what the
name is associated with the IP. Your ISP is authoritative where the
DNS PTR records are concerned for your IP.
: Please email me back
If someone is going to take the time to answer your question, you can
take the time to look for the answer in the news group.
--
==============================================================================
Michael | mfaurot | Rainy days and automatic weapons always get me down.
Faurot | atww.net |
------------------------------
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.development.system
Subject: Re: Why not C++
From: Edwin Young <[EMAIL PROTECTED]>
Date: 20 Aug 1999 15:32:43 +0100
David Schwartz <[EMAIL PROTECTED]> writes:
> You really do have to do one of two things:
>
> 1) Ignore the error cases and hope for the best, or
>
> 2) Consider every possible error case at every step and code the right
> behavior.
>
> Now, you can do either in C or C++. What is the third option?
I guess the idea is to handle errors a little higher up the tree rather than
having to explicitly handle the errors at every level.
eg a standard (if rather badly written) C fubction could be
int foo()
{
fd = open("blah",O_RDONLY);
if(fd == -1)
{
return -1;
}
nbytes = read(buf,fd,1000);
if(nbytes == -1)
{
close(fd);
return(-1);
}
err = close(fd);
if(err == -1)
{
return(-1);
}
return 0;
}
the exception-based (C++) one would be
try{
foo();
}
except(err)
{
...
}
void foo()
{
file fd("blah",O_RDONLY); // throws an exception on failure
fd.read(buf,1000); // ditto
// file's destructor closes it on exit
}
This seems to me to have 2 advantages.
1) functions can return what you want instead of returning
an error code and modifying a pointer to the output.
2) you can handle errors higher up the tree(what you really care
about is that foo() failed, not exactly which function in foo()
went wrong).
Disadvantages:
1) If you want to handle errors at a very low level you end
up with lots of try{} blocks which are worse than if statements.
2) Exception handling is a disaster with objects which don't
have destructors (pointers to malloced memory, FILE *'s etc).
--
Edwin
------------------------------
From: [EMAIL PROTECTED] (Keith T. Garner)
Crossposted-To:
tw.bbs.comp.linux,alt.linux,comp.os.linux,comp.os.linux.misc,comp.os.linux.questions,info.ncsa-telnet,hk.comp.os.linux,hk.comp.os.unix,hk.comp.pc
Subject: Re: Help: Caching webpages...
Date: 18 Aug 1999 23:23:20 GMT
At one point, Jimmy Lio said something like:
> Right now my machines are connecting to the Internet thru a Linux box
> with IP Masquerading turned on. What should I do to make the Linux box
> cache web-pages?
Check out squid at http://squid.nlanr.net/Squid/
Keith
--
Keith T. Garner [EMAIL PROTECTED]
RIMS Data Center, Technical Architect [EMAIL PROTECTED]
"Unix doesn't have a monopoly on good ideas,
it just owns most of them." -- Alan Cox
------------------------------
From: "Jon Richardson" <[EMAIL PROTECTED]>
Subject: Can Linux translate IPX<>IP?
Date: Thu, 19 Aug 1999 00:00:31 +0100
Hi,
We're currently using a legacy database system which only uses IPX to talk
on the network. I've seen a couple of tools which run on a Windows machine
which 'translate' IP into IPX to allow users to telnet in.
I was wondering though if Linux can handle this natively? It doesn't seem
particularly hard to do but I can't find that kind of thing documented
anywhere... does anyone know how to do it?
Best Regards,
Jon Richardson
Systems Manager
Adrian Tesson Associates
------------------------------
From: Frederic Pont <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,comp.os.linux.setup
Subject: Re: Can't get 2 NIC's to work
Date: Thu, 19 Aug 1999 17:12:24 GMT
Ryan,
I've just installed 3 NICs on Linux and it's now working perfectly. The
3 NICs are :
1. eth0: 3Com 3c905B Cyclone 100baseTx (PCI)
2. eth1: 3Com 3c905B Cyclone 100baseTx (PCI)
3. eth2: NE2000 (ISA)
Some remaks about how to get it working :
1. I had to **compile the drivers into the kernel**. It worked with the
drivers as modules, but impossible to get them running directly at boot
time. I had to run 'modprobe 3c59x', 'modprobe ne', and
'/etc/rc.d/init.d/network restart' after boot.
NO PROBLEM with everything in the kernel.
2. **You cannot choose IRQ and IO for the 2 PCI NICs**. You'll have to
hope that they don't get the same IRQ at boot time. (control in
/proc/interrupts). If they get the same IRQ, you can try to put one of
the NICs on another PCI slot.
3. You want to **fix IRQ and IO port for the ISA NE200 NIC**. As you
have everything in the kernel, you'll have to modify /etc/lilo.conf :
==============
boot=/dev/sda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
image=/boot/linux-2.2.9
label=linux-2.2.9
root=/dev/sda2
read-only
append="ether=0,0,eth0 ether=0,0,eth1 ether=10,0x280,eth2"
==============
the last line is what you want. For eth0 and eth1, you cannot set any
value for IRQ and IO. You can chose you IRQ for the ISA NIC. I'm not
sure about that, but the IO should be 0x280, 0x300, 0x320... for ISA.
You'll have to figure it out. In my BIOS I can book IRQs for ISA
devices, to make sure they're not going to be used by PCI devices a boot
time. IRQ 10 is booked for my NE2000 NIC.
4. here is what I get at boot time regarding my 3 NICs ('dmesg'):
==============
3c59x.c:v0.99H 11/17/98 Donald Becker
http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
eth0: 3Com 3c905B Cyclone 100baseTx at 0x1400, 00:50:04:14:82:f9, IRQ
15
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 3c905B Cyclone 100baseTx at 0x1480, 00:10:4b:69:7f:c8, IRQ 9
8K byte-wide RAM 5:3 Rx:Tx split, autoselect/Autonegotiate interface.
MII transceiver found at address 24, status 7809.
Enabling bus-master transmits and whole-frame receives.
ne.c:v1.10 9/23/94 Donald Becker ([EMAIL PROTECTED])
NE*000 ethercard probe at 0x280: 00 00 b4 38 e2 6e
eth2: NE2000 found at 0x280, using IRQ 10.
==============
5. I'm running RedHat5.2 with kernel 2.2.9
6. Hope it helps :-)
Fred
In article <[EMAIL PROTECTED]>,
<[EMAIL PROTECTED]> wrote:
> I have 2 identical 3com 905b NIC's. When I use just
> one to connect my machine to the outside world
> everthing is fine. As soon as I try to put both cards
> in nothing works. On bootup I get a delaying
> initialization message for both eth0 & eth1. Also adm
> takes forever to start when this happens.
>
> thx,
> Ryan
>
> ------------------ Posted via CNET Linux Help ------------------
> http://www.searchlinux.com
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Jay Beattie <[EMAIL PROTECTED]>
Subject: Re: FTP and chroot
Date: Fri, 20 Aug 1999 11:31:54 -0500
edit your /etc/ftpaccess files and add the following lines.
Also read man ftpaccess
restricted-uid USERNAME
guest-root /home/USERNAME
Greg Breheny wrote:
> The 'ftp' user on my RH52 system is setup so 'chroot' is issued to stop
> access to higher level directories.
>
> How can I set this up for another user, say 'jack'.
>
> Thanks
>
> Greg
------------------------------
From: Tiberio, David <[EMAIL PROTECTED]>
Subject: problems with 2 networks
Date: Fri, 20 Aug 1999 11:01:21 -0700
does this look correct? the TX and RX don't look right to
me!
I have 2 dsl lines and 2 nic cards for this machine, one
nic card on each dsl line
Kernel IP routing table
Destination Gateway Genmask Flags MSS
Window irtt Iface
216.46.85.16 * 255.255.255.240 U 1500
0 0 eth0
216.32.200.0 * 255.255.255.0 U 1500
0 0 eth1
loopback * 255.0.0.0 U 3584
0 0 lo
default 216.32.200.1 0.0.0.0 UG 1500
0 0 eth1
default 216.46.85.17 0.0.0.0 UG 1500
0 0 eth0
lo Link encap:Local Loopback
inet addr:127.0.0.1 Bcast:127.255.255.255
Mask:255.0.0.0
UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric:1
RX packets:65381538 errors:0 dropped:0 overruns:0
frame:0
TX packets:65381538 errors:0 dropped:0 overruns:0
carrier:0
Collisions:0
eth0 Link encap:Ethernet HWaddr 00:A0:CC:51:BE:DE
inet addr:216.46.85.20 Bcast:216.46.85.31
Mask:255.255.255.240
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:2111377 errors:0 dropped:0 overruns:0
frame:0
TX packets:74813 errors:0 dropped:0 overruns:0
carrier:0
Collisions:114
Interrupt:11 Base address:0xdc00
eth1 Link encap:Ethernet HWaddr 00:40:05:42:CF:49
inet addr:216.32.200.14 Bcast:216.32.200.255
Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:1711303 errors:0 dropped:0 overruns:0
frame:0
TX packets:4178344 errors:0 dropped:0 overruns:0
carrier:0
Collisions:88320
Interrupt:10 Base address:0xd880
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
------------------------------
From: "Pat Crean" <[EMAIL PROTECTED]>
Subject: Re: Restrict FTP users from deleting
Date: Fri, 20 Aug 1999 10:45:12 -0400
/etc/ftpaccess
Neel L. <[EMAIL PROTECTED]> wrote in message
news:mm3v3.11869$[EMAIL PROTECTED]...
> Is there anyway I can manipulate the CHMOD so that users can
upload/download
> but NOT delete files? I've been stuck on this SO long, I cannot figure it
> out! Thanks!!
>
> Neel
>
>
------------------------------
From: "Cedric Blancher" <[EMAIL PROTECTED]>
Subject: Re: Forwarding with ipchains
Date: Fri, 20 Aug 1999 19:30:38 +0200
Calum Lawler <[EMAIL PROTECTED]> a �crit dans le message :
[EMAIL PROTECTED]
> I am trying to forward any website requests to another machine.
> 192.168.1.10 (RH6) should forward any packets arriving in port 80 to
> 192.168.1.41 (QNX) where the web server is... My ipchains rule was
> implemented as follows:
> ipchains -A forward -b -p TCP -s 0/0 www -d 192.168.1.41 www
>
> Nothing gets forwarded.
Yes, sure !
You rule means nothing as there is no target. Moreover, even if an
ACCEPT target was present, it would mean :
accept, both direction, paquets from anywhere:80 to 192.168.1.41:80
If yhou want to redirect, you will have to use ipmasqadm package and do
this :
ipmasqadm portfw -a -p TCP -L $gateway 80 -R 192.168.1.41 80
That's all.
------------------------------
From: [EMAIL PROTECTED] (John Meissen)
Crossposted-To: comp.os.linux.hardware
Subject: Netgear FA310 and Micronics W6Li
Date: 20 Aug 1999 18:10:50 GMT
I recently acquired a dual PPro system with a Micronics
W6Li motherboard. I installed Slackware 4.0, rebuilt the kernel
for SMP support, and it works very well, thank you.
My other system has two Netgear FA310tx NICs with the Digital
chip, and has no problems. I went shopping for a decent card
for the new system, and picked up one of the new FA310tx's
with the clone chip (in spite of the problems I'd had trying
to use them before). I rebuilt the kernel with tulip support.
When I first booted I got an error message from the BIOS (long
before Linux booted, before the SCSI BIOS initialized), to the
effect of (and this is from memory)
Resource Conflict - NIC Interface detected in PCI slot != 0
and the card wouldn't work. The driver finds it, but it just
doesn't work. I moved the card to slot 0 and everything works
fine.
Now, I could say OK, I have a workaround, but I need =2= NICS
in this machine.
So, is this some new restriction with this NIC, or is there
some bizarre restriction in the BIOS for this motherboard?
john-
------------------------------
Crossposted-To:
comp.sys.sun.admin,comp.unix.admin,comp.os.linux.setup,comp.unix.misc,comp.unix.questions,comp.os.ms-windows.nt.admin.networking,comp.os.ms-windows.apps.winsock
From: [EMAIL PROTECTED] (C.T.Nadovich)
Subject: Re: Telnet thru' a firewall, any backdoor ???
Date: Fri, 20 Aug 1999 17:32:25 GMT
[EMAIL PROTECTED] () writes:
>Dear TCP/IP Gurus,
>I have several machines on a subnet "behind" a corporate firewall
>which disables any telnet access to the outside world.
>The firewall allows HTTP and POP3 but denies FTP and Telnet.
>Is there any __backdoor__ which allows our machines telnet'ing
>to the outside world ?
>Please drop a copy of your hint to my email
> [EMAIL PROTECTED]
It depends if it is a HTTP proxy or a port filter, but you can do it if
you have an accomplice machine with a shell account out on the
net.
If your firewall is a port filter, you simply need an accomplice machine
out on the net that serves telnet on port 80. That's easy to arrange.
If it's a proxy you can often use a POST to get a clear TCP channel. In
this case you need something a little more elaborate at your remote
accomplice, but it's not a biggie.
--
Secure webmail-news-voicemail-fax with JTAN ProShell
http://www.jtan.com/proshell - Unix SSHells for Pros
------------------------------
From: [EMAIL PROTECTED] (Stuart R. Fuller)
Subject: Re: @Home with Muliple IP addresses and Linux
Reply-To: [EMAIL PROTECTED]
Date: Fri, 20 Aug 1999 18:10:01 GMT
netmail.home.com ([EMAIL PROTECTED]) wrote:
: Hi all,
:
: I've been an Comcast@home subscriber for over half a year, with a RH
: Linux box acting as a gateway for the rest of my network, no problems.
: Now, I am considering getting another IP address. Does anyone have
: exerience with this, and does anyone know if I would have to have 3
: NICs, or 4 in the gateway computer, etc, any other details, advantages,
: disadvantages to doing this?
What is your goal in seeking another IP address?
Stu
------------------------------
From: Mark Hahn <[EMAIL PROTECTED]>
Crossposted-To:
comp.sys.sun.admin,comp.unix.admin,comp.os.linux.setup,comp.unix.misc,comp.unix.questions,comp.os.ms-windows.nt.admin.networking,comp.os.ms-windows.apps.winsock
Subject: Re: Telnet thru' a firewall, any backdoor ???
Date: 20 Aug 1999 17:05:19 GMT
> The firewall allows HTTP and POP3 but denies FTP and Telnet.
> Is there any __backdoor__ which allows our machines telnet'ing
> to the outside world ?
it's not a backdoor, but yes, of course: anyone inside can simply
"telnet someweirdmachine http", and if someweirdmachine happens
to be running telnetd on port 80 (or IRC, or a mud, etc), then
your inside machines are indeed telnet'ing to the outside.
in other words, you're filtering on port numbers, which doesn't
prevent someone from using any other protocol on the same port.
assuming that the destination port is being used for its assigned purpose
is pretty much "security through ignorance" (non-security, that is.)
------------------------------
Crossposted-To: comp.os.linux.security
Subject: Re: IPCHAINS forwarding and/or 2nd nic slowing down Internet access
From: Eric deRiel <[EMAIL PROTECTED]>
Date: Fri, 20 Aug 1999 18:12:14 GMT
Eric deRiel <[EMAIL PROTECTED]> writes:
> Ricky Ng-Adam <[EMAIL PROTECTED]> writes:
>
>[...]
>> Everything is OK until I load a web page
>> from 192.168.1.2. Then, the ping time goes higher (and the network is
>> much
>> slower!)...
>
> I have a similar problem, though not exactly identical: traffic
> through eth0 (connected to the Internet) slows down considerably over
> time until the device is stopped and then restarted. In my case, the
> types of data transfers involved range from http to X to SMB -- never
> any one thing, but the slowdown will eventually occur even if no data
> are transferred.
>
> A sample session (w.x.y.z is the gateway provided by the ISP):
>
> [fruitbat@kiwi network-scripts]$ ping w.x.y.z
> PING w.x.y.z (w.x.y.z): 56 data bytes
> 64 bytes from w.x.y.z: icmp_seq=0 ttl=63 time=44.0 ms
> 64 bytes from w.x.y.z: icmp_seq=1 ttl=63 time=27.7 ms
> 64 bytes from w.x.y.z: icmp_seq=2 ttl=63 time=38.3 ms
>
> --- w.x.y.z ping statistics ---
> 3 packets transmitted, 3 packets received, 0% packet loss
> round-trip min/avg/max = 27.7/36.6/44.0 ms
> [fruitbat@kiwi network-scripts]$ ping w.x.y.z
> PING w.x.y.z (w.x.y.z): 56 data bytes
> 64 bytes from w.x.y.z: icmp_seq=0 ttl=63 time=49.8 ms
> 64 bytes from w.x.y.z: icmp_seq=1 ttl=63 time=557.5 ms
> 64 bytes from w.x.y.z: icmp_seq=2 ttl=63 time=97.5 ms
> 64 bytes from w.x.y.z: icmp_seq=3 ttl=63 time=117.5 ms
> 64 bytes from w.x.y.z: icmp_seq=4 ttl=63 time=77.6 ms
> 64 bytes from w.x.y.z: icmp_seq=5 ttl=63 time=97.5 ms
> 64 bytes from w.x.y.z: icmp_seq=6 ttl=63 time=97.5 ms
> 64 bytes from w.x.y.z: icmp_seq=7 ttl=63 time=77.6 ms
> 64 bytes from w.x.y.z: icmp_seq=8 ttl=63 time=97.4 ms
>
> --- w.x.y.z ping statistics ---
> 9 packets transmitted, 9 packets received, 0% packet loss
> round-trip min/avg/max = 49.8/141.1/557.5 ms
> [fruitbat@kiwi network-scripts]$ ping w.x.y.z
> PING w.x.y.z (w.x.y.z): 56 data bytes
> 64 bytes from w.x.y.z: icmp_seq=0 ttl=63 time=1127.2 ms
> 64 bytes from w.x.y.z: icmp_seq=1 ttl=63 time=147.2 ms
> 64 bytes from w.x.y.z: icmp_seq=2 ttl=63 time=187.1 ms
> 64 bytes from w.x.y.z: icmp_seq=3 ttl=63 time=1130.0 ms
> 64 bytes from w.x.y.z: icmp_seq=4 ttl=63 time=134.6 ms
> 64 bytes from w.x.y.z: icmp_seq=5 ttl=63 time=266.9 ms
> 64 bytes from w.x.y.z: icmp_seq=6 ttl=63 time=97.4 ms
>
> --- w.x.y.z ping statistics ---
> 7 packets transmitted, 7 packets received, 0% packet loss
> round-trip min/avg/max = 97.4/441.4/1130.0 ms
> [fruitbat@kiwi network-scripts]$ ping w.x.y.z
> PING w.x.y.z (w.x.y.z): 56 data bytes
> 64 bytes from w.x.y.z: icmp_seq=0 ttl=63 time=1031.3 ms
> 64 bytes from w.x.y.z: icmp_seq=1 ttl=63 time=1121.3 ms
> 64 bytes from w.x.y.z: icmp_seq=2 ttl=63 time=1043.9 ms
> 64 bytes from w.x.y.z: icmp_seq=3 ttl=63 time=1160.7 ms
>
> --- w.x.y.z ping statistics ---
> 6 packets transmitted, 4 packets received, 33% packet loss
> round-trip min/avg/max = 1031.3/1089.3/1160.7 ms
> [fruitbat@kiwi network-scripts]$ sudo ./ifdown eth0; sudo ./ifup ifcfg-eth0
> Password:
> [fruitbat@kiwi network-scripts]$ ping w.x.y.z
> PING w.x.y.z (w.x.y.z): 56 data bytes
> 64 bytes from w.x.y.z: icmp_seq=0 ttl=63 time=21.5 ms
> 64 bytes from w.x.y.z: icmp_seq=1 ttl=63 time=40.2 ms
> 64 bytes from w.x.y.z: icmp_seq=2 ttl=63 time=20.3 ms
> 64 bytes from w.x.y.z: icmp_seq=3 ttl=63 time=21.2 ms
> 64 bytes from w.x.y.z: icmp_seq=4 ttl=63 time=20.0 ms
>
> --- w.x.y.z ping statistics ---
> 5 packets transmitted, 5 packets received, 0% packet loss
> round-trip min/avg/max = 20.0/24.6/40.2 ms
> [fruitbat@kiwi network-scripts]$
>
> These were taken over the course of about two hours. It looks like
> something very specific, but I haven't been able to figure it out.
> Any suggestions?
I'll offer a few suggestions myself, then, since everyone else seems
to be stumped.
Could my IP chain rules be causing a slowdown of this magnitude?
If so, what should I look for to find the source of the error?
Alternately, could this be the result of heavy ssh traffic using up
/dev/random? If so, what can I use intead? /dev/urandom? Or am I
completely mistaken about the use of random keys in RSA-type
schemes?
The machine has a very small amount of memory (8M), and while I've
always compiled the kernel down to minimum and disabled everything
I possibly could (also leading to good security), and never had
this sort of problem until very recently, I have noticed that
/proc/meminfo generally reports about 400k free at any time. I
never made much of a close study of the machine's memory usage
before, so I have nothing to compare to: is that abnormally small?
Could there be some kind of hidden kernel function at work,
something that was trying to operate on an ever-increasing pile of
historical net traffic data?
Could the device driver be misconfigured somehow? Are there
settings that would inadvertently cause this sort of behavior on
slow or low-memory systems?
Could it be some kind of hardware-level synchronization problem
with my cable modem?
Unfortunately I don't know where to begin to research any of these
possibilities, as I seem to have exhausted what's availble amongst the
HOWTOs, docs, and man pages. In the meantime, I continue to observe
ever-increasing latency, as ping times to the local gateway go from
under 20 ms to over 3500 and plummet back down to 20 again when I
restart the device.
I apologize for what is essentially a repost, but I have no other
recourse than Usenet, as the cable company has concluded that their
equipment is functioning normally.
eric
--
"We come and go alone, why do they need to know?"
------------------------------
From: "Bradley Dunn" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.hardware,redhat.hardware.arch.intel
Subject: Re: NE2000 ISA help
Date: Fri, 20 Aug 1999 11:08:07 -0400
I had the same problems until yesterday since others helped me, I will try
to help yall. I have a Linksys Ether16 which is ne2000 compatable it is also
ISA
Here is what I did, first I booted into DOS then I loaded the setup program
that came with my card. I took it out of plug and play mode, but first I
looked at the current
settings of the card. My card was being assigned IO address 0x240 and IRQ 3,
so I set it to use these settings every time.
I then booted into Linux and edited /etc/conf.modules and added the
following
options ne io=240 irq=3
alias eth0 ne
I then saved the file and did a modprobe /etc/conf.modules
and it found the card. Out of curiosity I rebooted to see if it would find
it.. and it did...
try it.. it works....
Brad
>
> My NE2000 card has 3 settings jumperless, autodetect and plug-n-play. I
have
> tried jumperless and autodetect and still no luck finding the card with
> Redhat 5.2. From the RedHat site FAQ I tried:
>
> insmod 8390
> insmod ne io=0x0280 irq=11
>
> Got back:
> ne.c: No NE*000 card found at i/o = 0x280
> /lib/modules/preferred/net/ne.0 : init_module: Device or resource busy
>
> These were the IO and IRQ reported under Win98 where it works fine in
> autodetect or PnP (gone in jumperless)
>
> I also have a SoundBlaster 32 card that isn't found - could these two
cards
> be conflicting?
>
> Thanks for any advice!
> Simon Anderson
>
>
------------------------------
From: Hernan Coronel <[EMAIL PROTECTED]>
Subject: Re: looking for source for net2phone ip module
Date: Fri, 20 Aug 1999 17:50:11 GMT
In article <[EMAIL PROTECTED]>,
"Darrel Davis" <[EMAIL PROTECTED]> wrote:
> Has anybody seen source for a module for ip_masq_net2phone?
> I'd really like to find it since we are using Linux as a firewall and
> I need to punch net2phone through it.
Hi! Are you sure this code exists? I did my research on the net and it
looks that it is not available yet. Someone posted something about
a "general handler" that may do the trick but I couldn't find anything
on that generic handler neither.
If anybody knows how to fix this please mail me. Thanks.
Bye, Hernan
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
** 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
******************************
