Second interface is up and running but there is no "running" flag 1

[sunny1504]

I have two interface on my Sun-Fire-V440
The first interface is ce0 and the other is ce1
both the interface are plumbed.
If you notice that the first interface ce0 had the flag UP,BROADCAST,RUNNING,MULTICAST,IPv4 but the second interface ce1 has the flag UP,BROADCAST,MULTICAST,IPv4.
here in the second interface the flag RUNNING is missing.

please see the below mentioned output of ifconfig

#ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
ce0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 192.168.8.29 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51
ce1: flags=1000803<UP,BROADCAST,MULTICAST,IPv4> mtu 1500 index 20
inet 192.168.8.23 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:52


Because of this I cannot ping the second interface from other machine please can some one suggest.

thanks
Sunny

 

[cndcadams]

sunny1504;

Sorry but I need to start with the basic questions.

Did you set up a /etc/hostname.ce1 file?

what does your /etc/hosts file look like?

Thanks

CA

 

[tuka]

ce0 show a flag of 1000843
cd1 show a flag of 1000803 (no resources allocated)
As CNDCADAM stated in the earlier post check your setup.

Keyword(s):ifconfig, flag

Description Top


Document Body Top

This document explains the flags value in ifconfig -a output. For example:

# ifconfig -a

lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
hme0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 129.149.141.67 netmask ffffff00 broadcast 129.149.141.255


The flags value is a hexadecimal representation of the state of the interface. Each number in the value is distinct. In the above example, both flags= values are comprised of 7 different hex values, 1,0,0,0,8,4,9. To find out what the value means, follow the chart below: For example if the flags value is 1000849, that means the interface card is:

0x00000008 + 0x00000001 = is a loopback and the card is up
0x00000040 = resources are allocated
0x00000800 = card supports multicasting
=========
0x00000849


Previous to Solaris[TM] 8, the flags value only showed three digits, e.g., flags=843. Below is the chart excerpt from /usr/include/net/if.h:

0x00000001 /* interface is up */
0x00000002 /* broadcast address valid */
0x00000004 /* turn on debugging */
0x00000008 /* is a loopback net */
0x00000010 /* interface is point-to-point link */
0x00000020 /* avoid use of trailers */
0x00000040 /* resources allocated */
0x00000080 /* no address resolution protocol */
0x00000100 /* receive all packets */
0x00000200 /* receive all multicast packets */
0x00000400 /* protocol code on board */
0x00000800 /* supports multicast */
0x00001000 /* multicast using broadcast address */
0x00002000 /* non-unique address */
0x00004000 /* DHCP controls this interface */
0x00008000 /* do not advertise */
0x00010000 /* Do not transmit packets */
0x00020000 /* No address - just on-link subnet */
0x00040000 /* interface address deprecated */
0x00080000 /* address from stateless addrconf */
0x00100000 /* router on this interface */
0x00200000 /* No NUD on this interface */
0x00400000 /* Anycast address */
0x00800000 /* Do not exchange routing info */
0x01000000 /* IPv4 interface */
0x02000000 /* IPv6 interface */
0x04000000 /* Mobile IP controls this interface */
0x08000000 /* Don't failover on NIC failure */
0x10000000 /* NIC has failed */
0x20000000 /* Standby NIC to be used on failures */
0x40000000 /* Standby active or not ? */
0x80000000 /* NIC has been offlined */

 

[sunny1504]

Below is my configuration file.

bash-2.05# more /etc/hostname.ce1
sun-12
bash-2.05# more /etc/hostname.ce0
sun

bash-2.05# more /etc/inet/hosts
127.0.0.1 localhost
192.168.8.29 sun loghost
192.168.8.23 sun-12

 

[coffeysm]

Did you try ifconfig ce1 sun-12 netmask + broadcast + up? Just to see what happens?

 

[cndcadams]

sunny1504;

Your problem is that you did not put the entry for;
192.168.8.23 sun-12
in your /etc/hosts file.

I confirmed this by setting up a bge port and only entering it into /etc/inet/hosts and got the same results you are having.

I have never really used /etc/inet/hosts and have not read up on what is it's purpose. Maybe one of the other Techs/Administrators can give us some info on that.

I always use /etc/hosts for setting up hosts and ips.

Thanks

CA

 

[cndcadams]

sunny1504;

Ignore my last post; /etc/hosts is a symbolic link /etc/inet/hosts.

Reason my port did not show as running is because it was not plugged into the hub in another room. I should have checked that before I posted.

Are you sure your cable is good?

Thanks

CA

 

[sunny1504]

bash-2.05# ifconfig ce1 netmask + broadcast + up
Setting netmask of ce1 to 255.255.255.0

bash-2.05# ifconfig ce1 sun-12 netmask + broadcast + up
ifconfig: sun-12: bad address

bash-2.05# ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
ce0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 192.168.8.29 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51
ce1: flags=1000803<UP,BROADCAST,MULTICAST,IPv4> mtu 1500 index 20
inet 192.168.8.23 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51

bash-2.05# netstat -rn

Routing Table: IPv4
Destination Gateway Flags Ref Use Interface
-------------------- -------------------- ----- ----- ------ ---------
10.X.X.254 192.168.8.252 UGHD 1 1
192.168.8.0 192.168.8.29 U 1 14686 ce0
192.168.8.0 192.168.8.23 U 1 0 ce1
224.0.0.0 192.168.8.29 U 1 0 ce0
default 192.168.8.253 UG 1 13440
127.0.0.1 127.0.0.1 UH 31095216904 lo0
bash-2.05#



bash-2.05# arp -a

Net to Media Table: IPv4
Device IP Address Mask Flags Phys Addr
------ -------------------- --------------- ----- ---------------
ce0 sun.x.x.com 255.255.255.255 SP 00:03:ba:56:0b:51
ce1 sun-12.x.x.com 255.255.255.255 SP 00:03:ba:56:0b:51

 

[sunny1504]

notice the below mentioned behaviour

bash-2.05# ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
ce0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 192.168.8.29 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51

*In the below mentioned command I have plumbed the ce1 from scratch**

bash-2.05# ifconfig ce1 plumb 192.168.8.23 netmask 255.255.255.0

**After the above step I can see the running flag on the interface ce1 as mentioned below**

bash-2.05# ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
ce0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 192.168.8.29 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51
ce1: flags=1000842<BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 21
inet 192.168.8.23 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51

**Here is the problem after the interface is set with the flag UP the flag RUNNING disappear **

bash-2.05# ifconfig ce1 up
bash-2.05# ifconfig -a
lo0: flags=1000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4> mtu 8232 index 1
inet 127.0.0.1 netmask ff000000
ce0: flags=1000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4> mtu 1500 index 2
inet 192.168.8.29 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51
ce1: flags=1000803<UP,BROADCAST,MULTICAST,IPv4> mtu 1500 index 21
inet 192.168.8.23 netmask ffffff00 broadcast 192.168.8.255
ether 0:3:ba:56:b:51
bash-2.05#


Can any one tell me is there something wrong with what I am doing.

 

[cndcadams]

sunny1504;

1) have you tried a reboot?

2) I followed your procedure and if the cable was plugged in it worked fine.

3) I shutdown the port, unplumbed it, removed the cable.
Then I plumbed it using your procedure and it showed as running. Then I marked it as up and running no longer show up. Then I plugged cable in and it marks it as running.

The only things I can think is your cable is bad, the hub/switch port is not functioning, the ce1 port is having an issue.

Does the link light come on when the cable is plugged in?


Thanks

CA

 

[sunny1504]

Cndcadams,

The link light comes' up and the system can't be rebooted as its a production server.

Is it possible to execute the script /etc/rcS.d/S30network.sh instead of rebooting, without causing any downtime.

thanks

 

[sunny1504]

FYI: After configuring the second device I have still not rebooted the system.

Also Did you notice that both my interface have the same mac id, does that cause any kind of problem? .

I have another system with two interface again both the interface are working fine and they share the same mac id.

 

[cndcadams]

sunny1504;

found the below on the web;

The same ethernet hardware (MAC) address is used by default on Sun systems with multiple ethernet interfaces. This works fine for failover load balancing or failover groups. It presents a problem when the two interfaces are connected to the same subnet.
Directions - To enable uniqe MAC addresses on multiple Sun ethernet interfaces, the BootPROM variable local-mac-address? needs to be set to true. This can be done from the ok prompt or by issuing the following command from the Solaris shell as superuser:
eeprom"local-mac-address?=true"

I tried the below;

I have a v240 that has the local-mac-address set to true so my network ports are assigned there own mac addresses. I changed the local-mac-address to false and rebooted.
The devices come up with the same mac address and both on the same subnet and I had no problems pinging to either ip address.

The older systems always assigned the same mac to the ethernet ports. Just recently when I was working on my v240 did I realize each port had it's own mac address. I am not a network admin so I really can't say what problems may be caused by the same mac on multiple ethernet ports on the same subnet.

Is your other system also a V440?
Are the ethernet ports set up on the same subnet?
What is your local-mac-address setting? (run eeprom from OS to check this)

I will keep looking though.

Thanks

CA

 

[cndcadams]

sunny1504;

I found a couple things on the web. Your problem may be related to the same mac address for each port.

http://www.filibeto.org/sun/sunray-users/lib/sunsolve/16733.html

http://www.foundrynet.com/services/faqs/Sun.html

http://sunportal.sunmanagers.org/pipermail/summaries/2004-March/005126.html

I am only using a hub, no switch so I really can't mess around much more with this stuff.

Let me know what you figure out, very interested in the resolution.

Maybe some off the other techs/Administrators can give us a little more info.

Thanks

CA

 

[coffeysm]

I would not run that S30network script again once the system is up and running. You should be able to run the ifconfig command using the name listed in /etc/hosts (sun-12), I would check that file for a typo or something also. I do not think having the same MAC address will interfere, because Solaris uses the systems MAC address. But you can do eeprom set-local-mac-address=true, then if you know the MAC address set it manually with the ifconfig command or reboot. I would also clean out your ARP tables if you do it manually:

sh
for S in `arp -a | awk '{print $2}'`
do
arp -d $S
done

 

[coffeysm]

I also noticed in your first post your interfaces had different MAC addresses, in your second post they had the same MAC address. Did you do anything else other then what was suggested? I would clear out your ARP cache anyway, just in case.

 

[sunny1504]

Cndcadams,

You are right the other system is Sun-Fire-V440 and the local-mac-address was false. Now I have set the local-mac-address to true from os using eeprom for the system having both the interface on the same subnet.

You are absolutely right I do have similar setup, both the interface have the same mac id and also both the interface are pointing to different subnet.

Also as I told you the link light comes' up on the second interface.

Is there any other way out..

 

[cndcadams]

sunny1504;

coffeysm make a good point about the mac address being different in the first post compared to later posts (good eye coffeysm).

Also I don't think the local-mac-address setting will take affect unless system is rebooted. I tried changeing it then unplumb and plumb the device and it still keeps the same mac setup.

Ok so the link light comes up. (should be working)

Have you tried another cable and port on your switch/hub?

Thanks

CA

 

[sunny1504]

I changed the macid with the ether option.

 

[coffeysm]

I would unplumb it, clean your ARP cache out, then replumb it up. Try running only the plumb command first then bringing it up, not sure if it will help but a few times I had to run plumb by itself first. Let it use the MAC address that was built into the card. I would also check the duplex settings, link speed, etc...You could also try snooping the interface to see if any traffic passes through:

snoop -d ce1