You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@httpd.apache.org by "Baca-Overly, Connie" <co...@nrel.gov> on 2003/06/13 22:06:11 UTC

RE: [users@httpd] LAN team says I have 2 IP's w/1 MAC address, bu t I have 2 cables with 2 separate IP's. What gives?

For Solaris systems we have our network engineers lock down the port to 100
MB Full Duplex.  Additionally I force disable autonegotiate and force 100MB
full duplex.  View the /etc/system file to determine interface settings;
they should be:

set hme:adv_autoneg_cap = 0
set hme:adv_100fdx_cap = 1
set hme:adv_100hdx_cap = 0
set hme:adv_10fdx_cap = 0
set hme:adv_10hdx_cap = 0

To determine what your system has the interface set at use the ndd command:
example ndd -get /dev/adv_100fdx_cap
a return of 1=ON and 0=OFF

To set use the ndd -set /dev/hme/adv_100fdx_cap 1
for each of the files in the /etc/system file

Good Luck
Best regards,

Connie Overly
-----Original Message-----
From: Hulslander, Ryan [mailto:Ryan.Hulslander@ps.net]
Sent: Friday, June 13, 2003 12:45 PM
To: users@httpd.apache.org
Subject: [users@httpd] LAN team says I have 2 IP's w/1 MAC address, but
I have 2 cables with 2 separate IP's. What gives?


I'm really, REALLY hoping someone can shed some light on this. 

The hardware is a Soyo K-333 Dragon MB running SuSE 8.0 w/Apache2 as well as
Samba. I have 2 NIC's, the onboard NIC and a D-Link 530TX+ card. I have two
IP addresses, 155.16.78.54 & .55  on eth0 I get lots of collisions, errors,
etc., and barely any traffic on eth1. I have one "site.net" address and one
site.com" address, both configured as virtual hosts in Apache. 

The problem I have is that both NIC's seem to be set at 10MB half-duplex,
and mii-tool won't get them to negotiate to 100 full-duplex, nor will it
force it. The LAN team says it sees both cables coming in to the switch, and
that both IP's are being seen as the same MAC address?! (I pasted their
exact comments to me below)

The performance isn't really noticeable for Apache until I tried to turn on
Samba, and then it just about craters it.

Here's the dump of the ifconfig:

eth0      Link encap:Ethernet  HWaddr 00:05:5D:D0:29:80  
          inet addr:155.16.78.54  Bcast:155.16.78.255  Mask:255.255.255.0
          inet6 addr: fe80::205:5dff:fed0:2980/10 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:23619268 errors:1053 dropped:0 overruns:0 frame:0
          TX packets:3752750 errors:380508 dropped:0 overruns:0
carrier:761016
          collisions:105122 txqueuelen:100 
          RX bytes:1116970680 (1065.2 Mb)  TX bytes:2828485063 (2697.4 Mb)
          Interrupt:10 Base address:0x2000 

eth1      Link encap:Ethernet  HWaddr 00:50:2C:04:DA:D5  
          inet addr:155.16.78.53  Bcast:155.16.78.255  Mask:255.255.255.0
          inet6 addr: fe80::250:2cff:fe04:dad5/10 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:20789796 errors:0 dropped:0 overruns:0 frame:0
          TX packets:504 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100 
          RX bytes:411341177 (392.2 Mb)  TX bytes:30288 (29.5 Kb)
          Interrupt:11 Base address:0x4000 

The modules.conf file load "8139too" driver for both interfaces, and the
mii-tool reports on both:

forum:/etc # mii-tool eth0
eth0: 10 Mbit, half duplex, no link
forum:/etc # mii-tool eth1
eth1: 10 Mbit, half duplex, no link

No amount of trying to force the eth0 or eth1 interfaces to 100MB work.

The comments I got from the LAN guy was:

-------------

After clearing the IP ARP tables on both routers there is something really
strange going on, see information below.  The r4-pscptc4 router is the
gateway router 155.16.78.1.

r3-pscptc3#sh ip arp 155.16.78.54
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  155.16.78.54            0   0005.5dd0.2980  ARPA   Vlan78
r3-pscptc3#sh ip arp 155.16.78.53
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  155.16.78.53            0   0005.5dd0.2980  ARPA   Vlan78

r4-pscptc4#sh ip arp 155.16.78.53
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  155.16.78.53            1   0050.2c04.dad5  ARPA   Vlan78
r4-pscptc4#sh ip arp 155.16.78.54
Protocol  Address          Age (min)  Hardware Addr   Type   Interface
Internet  155.16.78.54            1   0005.5dd0.2980  ARPA   Vlan78

The bottom line of the above information is even after clearing both routers
ARP tables (twice) the r3-pscptc3 router is still seeing both IP addresses
as the same MAC address.  Routers don't just show this type of behavior for
any reason.  The MAC address 00-50-2c-04-da-d5 for 155.16.78.53 is not even
known to the CAM table on either switch (s3 or s4).  As a suggestion I would
check with the OS vendor to see if there are any known issues.

------------

I'm clueless. I've pulled up some docs regarding similar situations and it
talks about "iptables" or some such. Is there some kind of mapping someplace
to make sure the right NIC responds to the right IP's, etc., so my website
and Samba can start working right?

Thank You, Thank You, Thank You!

Ryan Hulslander
"Completely Baffled by this in Dallas"

---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org

---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org