This is a multi-part message in MIME format.
------=_NextPart_000_00F5_01C22E4D.046E6B60
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Hi folks,
After much research on the Internet I'm at a loss to get an explanation
as to why my ifconfig shows an output such as this :
eth0 Link encap:Ethernet HWaddr 00:10:DC:45:EB:E4
inet addr:209.239.20.9 Bcast:209.239.20.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:94828 errors:1 dropped:0 overruns:0 frame:0
TX packets:112200 errors:6091 dropped:0 overruns:0 carrier:12182
collisions:992 txqueuelen:100
RX bytes:13857807 (13.2 Mb) TX bytes:95681876 (91.2 Mb)
Interrupt:11 Base address:0x8000
(and this is just from being up for only 26 minutes!).
Why am I getting so many transmit errors ? And is there any reason why the
carrier errors are exactly double the transmit errors ? The card doesn't
hang
permanently - just for a second or two every once in a while - and that
makes
a difference in the overall speed for the server.
I have another server exactly like this one - but don't want to put it in
production
if it will generate the same errors. I cannot seem to generate the same
amount
of errors on my local network when testing huge downloads on this second
server.
There are a few - 1 or 2 - of the transmit errors - certainly not as many as
I see on
our production machine!
I'm running slackware 8.1 (kernel 2.4.18) and my card info is as follows :
==============================================================
root@server1:/usr/src/rtl# lspci
00:00.0 Host bridge: Silicon Integrated Systems [SiS]: Unknown device 0645
(rev 01)
00:01.0 PCI bridge: Silicon Integrated Systems [SiS] 5591/5592 AGP
00:02.0 ISA bridge: Silicon Integrated Systems [SiS] 85C503/5513
00:02.2 USB Controller: Silicon Integrated Systems [SiS] 7001 (rev 07)
00:02.3 USB Controller: Silicon Integrated Systems [SiS] 7001 (rev 07)
00:02.5 IDE interface: Silicon Integrated Systems [SiS] 5513 [IDE] (rev d0)
00:02.7 Multimedia audio controller: Silicon Integrated Systems [SiS]
SiS7012 PCI Audio Accelerator (rev a0)
00:07.0 Communication controller: Conexant: Unknown device 2f10 (rev 01)
00:0f.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C
(rev 10)
01:00.0 VGA compatible controller: nVidia Corporation NV11 [GeForce2 MX DDR]
(rev b2)
---
root@server1:/usr/src/rtl# ./rtl8139diag -ee
rtl8139-diag.c:v2.06 3/22/2002 Donald Becker (becker@scyld.com)
http://www.scyld.com/diag/index.html
Index #1: Found a RealTek RTL8139 adapter at 0xec00.
Decoded EEPROM contents:
PCI IDs -- Vendor 0x10ec, Device 0x8139.
PCI Subsystem IDs -- Vendor 0x10ec, Device 0x8139.
PCI timer settings -- minimum grant 32, maximum latency 64.
General purpose pins -- direction 0xe5 value 0x12.
Station Address 00:10:DC:45:EB:E4.
Configuration register 0/1 -- 0x8d / 0xc2.
EEPROM active region checksum is 0ac3.
EEPROM contents (64 words):
0x00: 8129 10ec 8139 10ec 8139 4020 e512 1000
0x08: 45dc e4eb 8d10 f7c2 8801 03b9 60f4 071a
0x10: dfa3 9836 dfa3 9836 03b9 60f4 1a1a 1a1a
0x18: 0000 f967 0000 0000 0000 0000 0000 2000
0x20: 0000 0000 0000 0000 0000 0000 0000 0000
0x28: 0000 0000 0000 0000 0000 0000 0000 0000
0x30: 0000 0000 0000 0000 0000 0000 0000 0000
0x38: 0000 0000 0000 0000 0000 0000 0000 0000
--
root@server1:/usr/src/rtl# ./rtl8139diag -m
rtl8139-diag.c:v2.06 3/22/2002 Donald Becker (becker@scyld.com)
http://www.scyld.com/diag/index.html
Index #1: Found a RealTek RTL8139 adapter at 0xec00.
The RTL8139 does not use a MII transceiver.
It does have internal MII-compatible registers:
Basic mode control register 0x1000.
Basic mode status register 0x782d.
Autonegotiation Advertisement 0x01e1.
Link Partner Ability register 0x0000.
Autonegotiation expansion 0x0000.
Disconnects 0x0000.
False carrier sense counter 0x0000.
NWay test register 0x0005.
Receive frame error count 0x0000.
--
(output from cat /proc/pci)
Bus 0, device 15, function 0:
Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139 (rev 16).
IRQ 11.
Master Capable. Latency=32. Min Gnt=32.Max Lat=64.
I/O at 0xec00 [0xecff].
Non-prefetchable 32 bit memory at 0xce012000 [0xce0120ff].
======================================================
I hope someone out there can help me or at least explain why this is
happening.
Thank you.
Kevin
------=_NextPart_000_00F5_01C22E4D.046E6B60
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Hi=20
folks,
After =
much research=20
on the Internet I'm at a loss to get an explanation
as to =
why my=20
ifconfig shows an output such as this :
eth0 Link encap:Ethernet =
HWaddr=20
00:10:DC:45:EB:E4 =20
inet=20
addr:209.239.20.9 Bcast:209.239.20.255 =20
Mask:255.255.255.0
&nb=
sp; UP=20
BROADCAST RUNNING MULTICAST MTU:1500 =20
Metric:1
RX=20
packets:94828 errors:1 dropped:0 overruns:0=20
frame:0
TX=20
packets:112200 errors:6091 dropped:0 overruns:0=20
carrier:12182
=20
collisions:992 txqueuelen:100=20
RX =
bytes:13857807=20
(13.2 Mb) TX bytes:95681876 (91.2=20
Mb)
=
Interrupt:11 Base=20
address:0x8000
(and =
this is just=20
from being up for only 26 minutes!).
Why am =
I getting so=20
many transmit errors ? And is there any reason why =
the
carrier errors are=20
exactly double the transmit errors ? The card doesn't hang =
permanently - just=20
for a second or two every once in a while - and that =
makes
a =
difference in the=20
overall speed for the server.
I have =
another=20
server exactly like this one - but don't want to put it in=20
production
if it =
will generate=20
the same errors. I cannot seem to generate the same=20
amount
of =
errors on my=20
local network when testing huge downloads on this second=20
server.
There =
are a few - 1=20
or 2 - of the transmit errors - certainly not as many =
as I see on=20
our =
production=20
machine!
I'm =
running=20
slackware 8.1 (kernel 2.4.18) and my card info is as follows=20
:
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
root@server1:/usr/src/rtl# =
lspci
00:00.0 Host bridge: Silicon Integrated Systems [SiS]: Unknown =
device=20
0645 (rev 01)
00:01.0 PCI bridge: Silicon Integrated Systems [SiS] =
5591/5592=20
AGP
00:02.0 ISA bridge: Silicon Integrated Systems [SiS]=20
85C503/5513
00:02.2 USB Controller: Silicon Integrated Systems [SiS] =
7001=20
(rev 07)
00:02.3 USB Controller: Silicon Integrated Systems [SiS] =
7001 (rev=20
07)
00:02.5 IDE interface: Silicon Integrated Systems [SiS] 5513 =
[IDE] (rev=20
d0)
00:02.7 Multimedia audio controller: Silicon Integrated Systems =
[SiS]=20
SiS7012 PCI Audio Accelerator (rev a0)
00:07.0 Communication =
controller:=20
Conexant: Unknown device 2f10 (rev 01)
00:0f.0 Ethernet controller: =
Realtek=20
Semiconductor Co., Ltd. RTL-8139/8139C (rev 10)
01:00.0 VGA =
compatible=20
controller: nVidia Corporation NV11 [GeForce2 MX DDR] (rev=20
b2)
---
root@server1:/usr/src/rtl# =
./rtl8139diag -ee
rtl8139-diag.c:v2.06 3/22/2002 Donald Becker (becker@scyld.com)
http://www.scyld.com/diag/i=
ndex.html
Index=20
#1: Found a RealTek RTL8139 adapter at 0xec00.
Decoded EEPROM=20
contents:
PCI IDs -- Vendor 0x10ec, Device=20
0x8139.
PCI Subsystem IDs -- Vendor 0x10ec, Device=20
0x8139.
PCI timer settings -- minimum grant 32, maximum =
latency=20
64.
General purpose pins -- direction 0xe5 value=20
0x12.
Station Address 00:10:DC:45:EB:E4.
=
Configuration=20
register 0/1 -- 0x8d / 0xc2.
EEPROM active region checksum is=20
0ac3.
EEPROM contents (64 words):
0x00: 8129 10ec 8139 10ec =
8139=20
4020 e512 1000
0x08: 45dc e4eb 8d10 f7c2 8801 03b9 60f4=20
071a
0x10: dfa3 9836 dfa3 9836 03b9 60f4 1a1a =
1a1a
0x18: 0000=20
f967 0000 0000 0000 0000 0000 2000
0x20: 0000 0000 0000 0000 =
0000 0000=20
0000 0000
0x28: 0000 0000 0000 0000 0000 0000 0000 =
0000
0x30: =20
0000 0000 0000 0000 0000 0000 0000 0000
0x38: 0000 0000 0000 =
0000 0000=20
0000 0000 0000
--
root@server1:/usr/src/rtl# =
./rtl8139diag -m
rtl8139-diag.c:v2.06 3/22/2002 Donald Becker (becker@scyld.com)
http://www.scyld.com/diag/i=
ndex.html
Index=20
#1: Found a RealTek RTL8139 adapter at 0xec00.
The RTL8139 does =
not use=20
a MII transceiver.
It does have internal MII-compatible=20
registers:
Basic mode control register =20
0x1000.
Basic mode status register =20
0x782d.
Autonegotiation Advertisement =
0x01e1.
=20
Link Partner Ability register 0x0000.
Autonegotiation=20
expansion 0x0000.
=20
Disconnects &n=
bsp; =20
0x0000.
False carrier sense counter =20
0x0000.
NWay test=20
register =
;=20
0x0005.
Receive frame error =
count =20
0x0000.
--
(output from cat=20
/proc/pci)
=
Bus 0,=20
device 15, function 0:
Ethernet =
controller:=20
Realtek Semiconductor Co., Ltd. RTL-8139 (rev=20
16).
IRQ =
11.
=20
Master Capable. Latency=3D32. Min Gnt=3D32.Max=20
Lat=3D64.
I/O at 0xec00=20
[0xecff].
Non-prefetchable 32 bit =
memory at=20
0xce012000 [0xce0120ff].
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D
I hope =
someone out=20
there can help me or at least explain why this is =
happening.
Thank=20
you.
Kevin
------=_NextPart_000_00F5_01C22E4D.046E6B60--
From becker@scyld.com Fri Jul 19 08:31:01 2002
From: becker@scyld.com (Donald Becker)
Date: Fri Jul 19 07:31:01 2002
Subject: [realtek] Huge transmit/carrier errors with 8139C on linux
In-Reply-To:
Message-ID:
On Thu, 18 Jul 2002, Kevin Cassidy wrote:
> After much research on the Internet I'm at a loss to get an explanation
> as to why my ifconfig shows an output such as this :
...
> RX packets:94828 errors:1 dropped:0 overruns:0 frame:0
> TX packets:112200 errors:6091 dropped:0 overruns:0 carrier:12182
> collisions:992 txqueuelen:100
This looks like your link partner is forced to full duplex.
If so, there is a permanent fix. Slap the switch administrator just
behind the ear, barely catching the ear lobe, and say "what". Pause a
bit. Say "were". Another pause, regaining your composure. Say "you
thinking". Tell them that if their equipment doesn't work properly with
autonegotation, they should stop buying that brand. And the existing
equipment should be set to half duplex mode so that it will interoperate
properly with default configurations.
> Why am I getting so many transmit errors ?
Those are likely out-of-window collisions, as the link partner is
sending packets without doing CSMA/CD. Full duplex is only valid when
it's autonegotiated.
> root@server1:/usr/src/rtl# ./rtl8139diag -m
> rtl8139-diag.c:v2.06 3/22/2002 Donald Becker (becker@scyld.com)
> http://www.scyld.com/diag/index.html
> Index #1: Found a RealTek RTL8139 adapter at 0xec00.
> The RTL8139 does not use a MII transceiver.
> It does have internal MII-compatible registers:
> Basic mode control register 0x1000.
> Basic mode status register 0x782d.
> Autonegotiation Advertisement 0x01e1.
> Link Partner Ability register 0x0000.
No link partner information.
--
Donald Becker becker@scyld.com
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Second Generation Beowulf Clusters
Annapolis MD 21403 410-990-9993
From wgm@telus.net Sat Jul 27 06:21:00 2002
From: wgm@telus.net (Wm. G. McGrath)
Date: Sat Jul 27 05:21:00 2002
Subject: [realtek] Conflicted initialization of multiple 3Com 509s
Message-ID: <20020727021024.1f95c91c.wgm@telus.net>
Hi all,
I am setting up a small firewall with 3 3Com 509s in an old 486.
The 2.4.18 slackware kernel will boot and recognize one of the nics
and assign it to eth0 and 0x220. This card works and I can ssh into
the box from another on the same network.
Dmesg finds 3 ISA PnP cards but provides no information on
them.
snip
isapnp: Scanning for PnP cards...
isapnp: Card '3Com 3C509B EtherLink III'
isapnp: Card '3Com 3C509B EtherLink III'
isapnp: Card '3Com 3C509B EtherLink III'
isapnp: 3 Plug & Play cards detected total
Linux NET4.0 for Linux 2.4
snip
I have used el3 and 3c5x9 to reset the IOports and IRQs of the three
cards to 300,310 and 320, 5,7 and 9. And I've inserted an append
statement in lilo to assign the three nics to eth0 eth1 and eth2.
This produces the following initialization:
snip
eth0: 3c5x9 at 0x220, 10baseT port, address 00 60 97 e1 5e 85, IRQ
5. 3c509.c:1.18a 17Nov2001becker@scyld.com
http://www.scyld.com/network/3c509.html
eth1: 3c5x9 at 0x230, 10baseT port, address 00 60 97 e1 60 62, IRQ
10. 3c509.c:1.18a 17Nov2001becker@scyld.com
http://www.scyld.com/network/3c509.html
eth2: 3c5x9 at 0x240, 10baseT port, address 00 60 97 55 4a a1, IRQ
11. 3c509.c:1.18a 17Nov2001becker@scyld.com
http://www.scyld.com/network/3c509.html
snip
/proc/ioports and /proc/interrupts show only the one card, and the
three ports 220,230 and 240. There is no mention of ports
300,310,320, irqs 7,9 or eth1 or eth2.
So what's up? Why are the cards being initialized like this? What
can I do to regain control and use the irqs, ports and devices I've
assigned and in general, track down the problem? I'd appreciate any
and all comments, suggestions, and thoughts on the matter.
bill
From DPDannemiller@mcleodusa.net Sun Jul 28 00:14:01 2002
From: DPDannemiller@mcleodusa.net (David Dannemiller)
Date: Sat Jul 27 23:14:01 2002
Subject: [realtek] Ethernet card config - Bering
In-Reply-To:
Message-ID:
I originally posted this to the leaf-user list, where it was suggested that
I post here.
My problem seems to be similar to this post:
http://www.scyld.com/pipermail/realtek/2002-January/001194.html
Here is additional info asked for by Donald Becker in that thread.
The machine is a Compaq Presario 7222. I don't know about bus master limits
for the motherboard.
The BIOS chip lists the following:
PL5600G
BIOS R1.03
This compaq uses an additional setup program:
SETUP for Presario
1.51/C Rev A
Softpaq 2141
Any help is appreciated.
DPD
----------
From: David Dannemiller
Date: Sat, 27 Jul 2002 00:04:27 -0500
To:
Subject: [leaf-user] Ethernet card config - Bering
I'm having a problem I think is related to ethernet card configuration.
My LEAF distribution is Bering_1.0-rc3 configured per chapter 4 of the users
guide for PPPoE.
Output of "uname -a" is:
Linux firewall 2.4.18 #4 Sun Jun 9 09:46:15 CEST 2002 i586 unknown
My 2 ethernet cards are both D-Link DFE-530TX+, which are PCI cards.
Relevant lines in /etc/modules are:
# /etc/modules: kernel modules to load at boot time.
#
# This file should contain the names of kernel modules that are
# to be loaded at boot time, one per line. Comments begin with
# a `#', and everything on the line after them are ignored.
#
# All modules should reside in /lib/modules, and you are required
# to list what you need *in the correct order*. Modprobe is not
# used and there is no dependacy checking.
# How they are loaded: insmod /lib/module/"$module".o $args
######################################################################
# More modules available from:
# http://leaf.sourceforge.net/devel/jnilo/bering/latest/modules/
######################################################################
# My ethernet cards
# needed by PCI modules
pci-scan
# D-Link card
rtl8139
# Linksys card
#tulip
# Modules needed for PPP/PPPOE connection
slhc
n_hdlc
ppp_generic
ppp_synctty
pppox
pppoe
# Masquerading 'helper' modules
# Other modules available in bering/modules/net/ipv4/netfilter
ip_conntrack_ftp
ip_conntrack_irc
ip_nat_ftp
ip_nat_irc
Shortly after boot, the following message is displayed:
eth0: Transmit timeout, status 0c 0005 media 00.
The ppp interface is not established. Output of "ip addr show" is:
1: lo: mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: dummy0: mtu 1500 qdisc noop
link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
3: eth0: mtu 1500 qdisc pfifo_fast qlen 100
link/ether 00:50:ba:cc:67:58 brd ff:ff:ff:ff:ff:ff
4: eth1: mtu 1500 qdisc pfifo_fast qlen 100
link/ether 00:50:ba:42:c7:b0 brd ff:ff:ff:ff:ff:ff
inet 192.168.1.254/24 brd 192.168.1.255 scope global eth1
Output of "ip route show" is:
192.168.1.0/24 dev eth1 proto kernel scope link src 192.168.1.254
Output of "lsmod" is:
Module Pages Used by
ip_nat_irc 2384 0 (unused)
ip_nat_ftp 2960 0 (unused)
ip_conntrack_irc 3056 1
ip_conntrack_ftp 3824 1
pppoe 6636 1
pppox 912 1 [pppoe]
ppp_synctty 4376 0 (unused)
ppp_generic 14920 1 [pppoe pppox ppp_synctty]
n_hdlc 5760 0 (unused)
slhc 4264 0 [ppp_generic]
rtl8139 11720 2
pci-scan 3048 1 [rtl8139]
Pings between the firewall machine and another machine on the LAN are
successful in both directions. Ping from the firewall machine to the
Internet fails - returns nothing.
The D-Link ethernet cards are PnP. The BIOS defaulted both cards to the
same IRQ. I changed 1 of the cards to a different IRQ, but the behavior
described above did not change.
Installing only 1 ethernet card at a time allowed the PPP interface to be
established. Output of "ip addr show" is:
1: lo: mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: dummy0: mtu 1500 qdisc noop
link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
3: eth0: mtu 1500 qdisc pfifo_fast qlen 100
link/ether 00:50:ba:cc:67:58 brd ff:ff:ff:ff:ff:ff
4: ppp0: mtu 1492 qdisc pfifo_fast qlen 3
link/ppp
inet 199.174.15.61 peer 199.174.15.1/32 scope global ppp0
Output of "ip route show" is:
199.174.15.1 dev ppp0 proto kernel scope link src 199.174.15.61
default via 199.174.15.1 dev ppp0
Output of "lsmod" is:
Module Pages Used by
ip_nat_irc 2384 0 (unused)
ip_nat_ftp 2960 0 (unused)
ip_conntrack_irc 3056 1
ip_conntrack_ftp 3824 1
pppoe 6636 1
pppox 912 1 [pppoe]
ppp_synctty 4376 0 (unused)
ppp_generic 14920 3 [pppoe pppox ppp_synctty]
n_hdlc 5760 0 (unused)
slhc 4264 0 [ppp_generic]
rtl8139 11720 1
pci-scan 3048 1 [rtl8139]
I tried one of the D-Link boards with a Linksys board. In that
configuration I could force which board was eth0 and eth1 by the order in
which modules were listed in /etc/modules. I could get either the LAN or
the WAN to work (meaning successful pings), but not both at the same time.
Any assistance is appreciated.
DPD
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
------------------------------------------------------------------------
leaf-user mailing list: leaf-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/leaf-user
SR FAQ: http://leaf-project.org/pub/doc/docmanager/docid_1891.html
From wgm@telus.net Sun Jul 28 03:54:01 2002
From: wgm@telus.net (Wm. G. McGrath)
Date: Sun Jul 28 02:54:01 2002
Subject: [realtek] Disabling PnP on 3Com 509Bs
Message-ID: <20020727154140.33ca0bda.wgm@telus.net>
Hi all,
Further to my above question, additional reading suggests one cause
might be ISA PnP. So, I'd like to give that a try. How can I disable
ISA PnP on a 3Com 509B? If I do this presumably the linux probe can
then do it's thing at boot. The motherboard is a simple AMI bios 486
- no slot control, no address specification, no IRQ selection, no
PnP OS feature - so I can't do it in hardware. I don't have dos on
this box either, so using a dos utility isn't an option.
All advice appreciated,
bill
From wgm@telus.net Sun Jul 28 21:05:00 2002
From: wgm@telus.net (Wm. G. McGrath)
Date: Sun Jul 28 20:05:00 2002
Subject: [realtek] Restarting 3Com driver
Message-ID: <20020728165357.6f50a704.wgm@telus.net>
Hi all,
The saga continues I guess.
I'm not sure if there isn't an easier way to turn PnP off, but I've
found isapnptools. I've used pnpdump to create a config file
/etc/isapnp.conf that gets read during boot. This file is pretty
much complete by default and all I had to do was tweak it in a few
places to adjust the addresses and irqs. It may not disable PnP, but
at least it allows me to reconfigure the nics.
But as you would expect from Murphy, rebooting using isapnp.conf
caused a fatal error because the cards are already activated by PnP.
So I inserted an (ACT N) command into the file to turn off the
devices before resetting the address on the card and checking them.
This works and rebooting displays messages to stout during boot that
show that the nics have been changed. Fingers crossed.
But Murphy gets his revenge. With the driver turned off, networking
doesn't work, whereas it did before. So how and where can I restart
the 3Com driver. I think it is called 3c509, and I've built it into
the kernel.
bill
From becker@scyld.com Mon Jul 29 13:28:01 2002
From: becker@scyld.com (Donald Becker)
Date: Mon Jul 29 12:28:01 2002
Subject: [realtek] Ethernet card config - Bering
In-Reply-To:
Message-ID:
On Sat, 27 Jul 2002, David Dannemiller wrote:
> I originally posted this to the leaf-user list, where it was suggested that
> I post here.
> Linux firewall 2.4.18 #4 Sun Jun 9 09:46:15 CEST 2002 i586 unknown
>
> My 2 ethernet cards are both D-Link DFE-530TX+, which are PCI cards.
> Relevant lines in /etc/modules are:
...
> # needed by PCI modules
> pci-scan
You shouldn't need to list this explicitly. The PCI scanning and power
management support is automatically loaded before the rtl8139.o module.
> # D-Link card
> rtl8139
..
> Shortly after boot, the following message is displayed:
>
> eth0: Transmit timeout, status 0c 0005 media 00.
There should have been other messages before this.
What is the detection message?
What is the packet count when this happens? ('cat /proc/net/dev/')
What does 'mii-diag' report?
http://www.scyld.com/diag/index.html
ftp://ftp.scyld.com/pub/diag/
> The D-Link ethernet cards are PnP.
Note that all PCI cards are "PnP".
PCI "PnP" is unrelated to the actions taken with ISA Plug-n-Pray.
> The BIOS defaulted both cards to the same IRQ.
This is not a problem with PCI cards. In ancient times there were a few
broken SCSI drivers, but those have long since been fixed.
> I tried one of the D-Link boards with a Linksys board. In that
> configuration I could force which board was eth0 and eth1 by the order in
> which modules were listed in /etc/modules. I could get either the LAN or
> the WAN to work (meaning successful pings), but not both at the same time.
This points to a routing or firewall problem.
Note that the Linksys card requires a bus-master slot, so if it works
there is not a PCI slot problem.
--
Donald Becker becker@scyld.com
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Second Generation Beowulf Clusters
Annapolis MD 21403 410-990-9993
From luc.barbier@ccs.be Wed Jul 31 10:25:01 2002
From: luc.barbier@ccs.be (Luc Barbier)
Date: Wed Jul 31 09:25:01 2002
Subject: [realtek] rtl8190's configuration from modules.conf
Message-ID:
Hello,
I'm using the Suse Linux 7.3 and I try to force the half-duplex mode
of the rtl8139 network card.
I think it is possible to do it from the file /etc/modules.conf.
Any help ?
Thank you.
Luc.
Luc.barbier@ccs.be.
From becker@scyld.com Wed Jul 31 11:18:01 2002
From: becker@scyld.com (Donald Becker)
Date: Wed Jul 31 10:18:01 2002
Subject: [realtek] rtl8190's configuration from modules.conf
In-Reply-To:
Message-ID:
On Wed, 31 Jul 2002, Luc Barbier wrote:
> I'm using the Suse Linux 7.3 and I try to force the half-duplex mode
> of the rtl8139 network card.
Half duplex mode is standard Ethernet, and thus the default.
Why do you think that you need to force half duplex? You can limit the
advertised capabilities, thus avoiding negotiating to a full duplex
mode. But there is rarely a reason to do this.
--
Donald Becker becker@scyld.com
Scyld Computing Corporation http://www.scyld.com
410 Severn Ave. Suite 210 Second Generation Beowulf Clusters
Annapolis MD 21403 410-990-9993
From DPDannemiller@earthlink.net Wed Jul 31 23:44:02 2002
From: DPDannemiller@earthlink.net (David Dannemiller)
Date: Wed Jul 31 22:44:02 2002
Subject: [realtek] Ethernet card config - Bering
In-Reply-To:
Message-ID:
> From: Donald Becker
> Date: Mon, 29 Jul 2002 12:27:50 -0400 (EDT)
> ...
>> The BIOS defaulted both cards to the same IRQ.
>
> This is not a problem with PCI cards. In ancient times there were a few
> broken SCSI drivers, but those have long since been fixed.
OK. I let the BIOS set both cards to IRQ 11.
>> # needed by PCI modules
>> pci-scan
>
> You shouldn't need to list this explicitly. The PCI scanning and power
> management support is automatically loaded before the rtl8139.o module.
When I take this line out, the output of "ip addr" is:
1: lo: mtu 16436 qdisc noqueue
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: dummy0: mtu 1500 qdisc noop
link/ether 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff
So I put it bsck in.
>> Shortly after boot, the following message is displayed:
>>
>> eth0: Transmit timeout, status 0c 0005 media 00.
>
> There should have been other messages before this.
> What is the detection message?
I'm not sure what you mean by detection message (newbie). Here are the
contents of syslog:
Jul 31 19:54:19 firewall syslogd 1.3-3#31.slink1: restart.
Jul 31 19:54:20 firewall kernel: klogd 1.3-3#31.slink1, log source =
/proc/kmsg started.
Jul 31 19:54:20 firewall kernel: Cannot find map file.
Jul 31 19:54:20 firewall kernel: Loaded 48 symbols from 12 modules.
Jul 31 19:54:20 firewall kernel: Linux version 2.4.18 (root@debian) (gcc
version 2.95.2 20000220 (Debian GNU/Linux)) #4 Sun Jun 9 09:46:15 CEST 2002
Jul 31 19:54:20 firewall kernel: BIOS-provided physical RAM map:
Jul 31 19:54:20 firewall kernel: BIOS-e801: 0000000000000000 -
000000000009f000 (usable)
Jul 31 19:54:20 firewall kernel: BIOS-e801: 0000000000100000 -
0000000008000000 (usable)
Jul 31 19:54:20 firewall kernel: On node 0 totalpages: 32768
Jul 31 19:54:20 firewall kernel: zone(0): 4096 pages.
Jul 31 19:54:20 firewall kernel: zone(1): 28672 pages.
Jul 31 19:54:20 firewall kernel: zone(2): 0 pages.
Jul 31 19:54:20 firewall kernel: Kernel command line: BOOT_IMAGE=linux
initrd=initrd.lrp init=/linuxrc root=/dev/ram0 boot=/dev/fd0u1680:msdos
PKGPATH=/dev/fd0u1680
LRP=root,etc,local,modules,ppp,pppoe,shorwall,dnscache,weblet,dhcpd
Jul 31 19:54:20 firewall kernel: Initializing CPU#0
Jul 31 19:54:20 firewall kernel: Detected 59.830 MHz processor.
Jul 31 19:54:20 firewall kernel: Console: colour VGA+ 80x25
Jul 31 19:54:20 firewall kernel: Calibrating delay loop... 119.19 BogoMIPS
Jul 31 19:54:20 firewall kernel: Memory: 126884k/131072k available (864k
kernel code, 3800k reserved, 214k data, 60k init, 0k highmem)
Jul 31 19:54:20 firewall kernel: Dentry-cache hash table entries: 16384
(order: 5, 131072 bytes)
Jul 31 19:54:20 firewall kernel: Inode-cache hash table entries: 8192
(order: 4, 65536 bytes)
Jul 31 19:54:20 firewall kernel: Mount-cache hash table entries: 2048
(order: 2, 16384 bytes)
Jul 31 19:54:20 firewall kernel: Buffer-cache hash table entries: 8192
(order: 3, 32768 bytes)
Jul 31 19:54:20 firewall kernel: Page-cache hash table entries: 32768
(order: 5, 131072 bytes)
Jul 31 19:54:20 firewall kernel: CPU: Before vendor init, caps: 000001bf
00000000 00000000, vendor = 0
Jul 31 19:54:20 firewall kernel: Intel Pentium with F0 0F bug - workaround
enabled.
Jul 31 19:54:20 firewall kernel: CPU: After vendor init, caps: 000001bf
00000000 00000000 00000000
Jul 31 19:54:20 firewall kernel: CPU: After generic, caps: 000001bf
00000000 00000000 00000000
Jul 31 19:54:20 firewall kernel: CPU: Common caps: 000001bf
00000000 00000000 00000000
Jul 31 19:54:20 firewall kernel: CPU: Intel Pentium 75 - 200 stepping 05
Jul 31 19:54:20 firewall kernel: Checking 'hlt' instruction... OK.
Jul 31 19:54:20 firewall kernel: POSIX conformance testing by UNIFIX
Jul 31 19:54:20 firewall kernel: PCI: BIOS32 entry (0xc00fbfe0) in high
memory, cannot use.
Jul 31 19:54:20 firewall kernel: PCI: Using configuration type 1
Jul 31 19:54:20 firewall kernel: PCI: Probing PCI hardware
Jul 31 19:54:20 firewall kernel: Linux NET4.0 for Linux 2.4
Jul 31 19:54:20 firewall kernel: Based upon Swansea University Computer
Society NET3.039
Jul 31 19:54:20 firewall kernel: Initializing RT netlink socket
Jul 31 19:54:20 firewall kernel: Starting kswapd
Jul 31 19:54:20 firewall kernel: pty: 256 Unix98 ptys configured
Jul 31 19:54:20 firewall kernel: Serial driver version 5.05c (2001-07-08)
with MANY_PORTS SHARE_IRQ DETECT_IRQ SERIAL_PCI enabled
Jul 31 19:54:20 firewall kernel: ttyS00 at 0x03f8 (irq = 4) is a 16550A
Jul 31 19:54:20 firewall kernel: Software Watchdog Timer: 0.05, timer
margin: 60 sec
Jul 31 19:54:20 firewall kernel: block: 128 slots per queue, batch=32
Jul 31 19:54:20 firewall kernel: RAMDISK driver initialized: 16 RAM disks of
4096K size 1024 blocksize
Jul 31 19:54:20 firewall kernel: Floppy drive(s): fd0 is 1.44M
Jul 31 19:54:20 firewall kernel: FDC 0 is a post-1991 82077
Jul 31 19:54:20 firewall kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Jul 31 19:54:20 firewall kernel: IP Protocols: ICMP, UDP, TCP, IGMP
Jul 31 19:54:20 firewall kernel: IP: routing cache hash table of 1024
buckets, 8Kbytes
Jul 31 19:54:20 firewall kernel: TCP: Hash tables configured (established
8192 bind 8192)
Jul 31 19:54:20 firewall kernel: Linux IP multicast router 0.06 plus PIM-SM
Jul 31 19:54:20 firewall kernel: ip_conntrack version 2.0 (1024 buckets,
8192 max) - 312 bytes per conntrack
Jul 31 19:54:20 firewall kernel: ip_tables: (C) 2000-2002 Netfilter core
team
Jul 31 19:54:20 firewall kernel: NET4: Unix domain sockets 1.0/SMP for Linux
NET4.0.
Jul 31 19:54:20 firewall kernel: RAMDISK: Compressed image found at block 0
Jul 31 19:54:20 firewall kernel: Freeing initrd memory: 400k freed
Jul 31 19:54:20 firewall kernel: VFS: Mounted root (minix filesystem).
Jul 31 19:54:20 firewall kernel: Freeing unused kernel memory: 60k freed
Jul 31 19:54:20 firewall kernel: VFS: Disk change detected on device
fd(2,44)
Jul 31 19:54:20 firewall kernel: rtl8139.c:v1.17 1/28/2002 Donald Becker,
becker@scyld.com.
Jul 31 19:54:20 firewall kernel: http://www.scyld.com/network/rtl8139.html
Jul 31 19:54:20 firewall kernel: eth0: D-Link DFE-538TX (RealTek RTL8139) at
0x1000, IRQ 11, 00:50:ba:cc:67:58.
Jul 31 19:54:20 firewall kernel: eth1: RealTek RTL8139C Fast Ethernet at
0x1800, IRQ 11, 00:50:ba:42:c7:b0.
Jul 31 19:54:20 firewall kernel: CSLIP: code copyright 1989 Regents of the
University of California
Jul 31 19:54:20 firewall kernel: HDLC line discipline: version $Revision:
3.3 $, maxframe=4096
Jul 31 19:54:20 firewall kernel: N_HDLC line discipline registered.
Jul 31 19:54:20 firewall kernel: PPP generic driver version 2.4.1
Jul 31 19:54:20 firewall kernel: grsec: time set by (hwclock:25905) UID(0)
EUID(0), parent (S50hwclock:4689) UID(0) EUID(0)
Jul 31 19:54:20 firewall dhcpd: Internet Software Consortium DHCP Server
2.0pl5
Jul 31 19:54:20 firewall dhcpd: Copyright 1995, 1996, 1997, 1998, 1999 The
Internet Software Consortium.
Jul 31 19:54:20 firewall dhcpd: All rights reserved.
Jul 31 19:54:20 firewall dhcpd:
Jul 31 19:54:20 firewall dhcpd: Please contribute if you find this software
useful.
Jul 31 19:54:20 firewall dhcpd: For info, please visit
http://www.isc.org/dhcp-contrib.html
Jul 31 19:54:20 firewall dhcpd:
Jul 31 19:54:20 firewall dhcpd: Listening on
LPF/eth1/00:50:ba:42:c7:b0/192.168.1.0
Jul 31 19:54:20 firewall dhcpd: Sending on
LPF/eth1/00:50:ba:42:c7:b0/192.168.1.0
Jul 31 19:54:20 firewall dhcpd: Sending on Socket/fallback/fallback-net
Jul 31 19:54:37 firewall root: Shorewall Started
Jul 31 19:54:38 firewall /usr/sbin/cron[28647]: (CRON) INFO (pidfile fd = 3)
Jul 31 19:54:38 firewall /usr/sbin/cron[20782]: (CRON) STARTUP (fork ok)
Jul 31 19:55:19 firewall kernel: eth0: Transmit timeout, status 0d 0004
media 00.
Jul 31 19:55:19 firewall kernel: eth0: Tx queue start entry 4 dirty entry
0, full.
Jul 31 19:55:19 firewall kernel: eth0: Tx descriptor 0 is 0008a03c. (queue
head)
Jul 31 19:55:19 firewall kernel: eth0: Tx descriptor 1 is 0008a03c.
Jul 31 19:55:19 firewall kernel: eth0: Tx descriptor 2 is 0008a03c.
Jul 31 19:55:19 firewall kernel: eth0: Tx descriptor 3 is 0008a03c.
Jul 31 19:55:19 firewall kernel: eth0: MII #32 registers are: 1100 782d 0000
0000 01e1 45e1 0001 0000.
Jul 31 20:00:01 firewall /USR/SBIN/CRON[16638]: (root) CMD
(/etc/multicron-p)
Jul 31 20:03:22 firewall kernel: VFS: Disk change detected on device fd(2,0)
Jul 31 20:03:57 firewall kernel: VFS: Disk change detected on device fd(2,0)
> What is the packet count when this happens? ('cat /proc/net/dev/')
The file /proc/net/dev is size zero.
> What does 'mii-diag' report?
> http://www.scyld.com/diag/index.html
> ftp://ftp.scyld.com/pub/diag/
I'll ask that someone on the LEAF list to compile mii-diag for me.
>> I tried one of the D-Link boards with a Linksys board. In that
>> configuration I could force which board was eth0 and eth1 by the order in
>> which modules were listed in /etc/modules. I could get either the LAN or
>> the WAN to work (meaning successful pings), but not both at the same time.
>
> This points to a routing or firewall problem.
>
> Note that the Linksys card requires a bus-master slot, so if it works
> there is not a PCI slot problem.
I wasn't very clear here. When I had the D-Link board and the Linksys
boards installed together, I could only get one working - the D-Link board.
By swapping the order the driver modules were listed in /etc/modules, I
could force the boards to swap assignments
D-Link on eth0 and Linksys on eth1
vs
Linksys on eth0 and D-Link on eth1
But always, the interface (eth0 or eth1) assigned to the D-Link board
worked, and the interface assigned to the Linksys board didn't. This made
me suspect a hardware problem with the Linksys board, but it passed the
diagnostic program that came packaged with the board, and it worked if
installed alone.
Two more data points:
1) I downloaded Freesco for pppoe and ran setup with the 2 D-Link board
configuration. Freesco only identifies 1 of the 2 boards (i.e. ifconfig has
an entry for eth0, but not for eth1). The ppp0 interface works (i.e. I can
successfully ping the gateway).
2) I installed 1 of the D-Link boards and an ISA board, and brought the
system up with Freesco. Both interfaces seem to work - yea.
I would like to get the ISA board back to its other application. Any help
debugging the 2 PCI configuration is appreciated.
DPD