1

I created a sub-interface on a RHEL box, eth0:1, and copied the eth0 config and changed all the settings to reflect eth0:1 and the IP address. However, when I issue ifup eth0:1 I receive this error:

[root@server-1 ~]# ifup eth0:1
Error, some other host already uses address 192.168.0.2.

[root@server-1 ~]# ping -c 1 192.168.0.2
PING 192.168.0.2 (192.168.0.2) 56(84) bytes of data.

--- 192.168.0.2 ping statistics ---
1 packets transmitted, 0 received, 100% packet loss, time 10000ms

However, when I issue: ifconfig eth0:1 192.168.0.2 up the command works and no errors. I would use the later command, but when I use it, it interferes with my primary, eth0, interface, and messes up the routing of my DNS traffic because its on the same subnet as the sub-interface eth0:1. I will make another post about the DNS issue.

These are the configs of the interfaces

[root@server-1 network-scripts]# cat ifcfg-eth0
DEVICE="eth0"
BOOTPROTO="static"
HWADDR="00:50:56:AF:0C:06"
IPADDR="192.168.0.1"
IPV6INIT="yes"
NETMASK="255.255.255.0"
ONBOOT="yes"
TYPE="Ethernet"
DNS1="192.168.2.10"
DNS2="192.168.3.10"

[root@server-1 network-scripts]# cat ifcfg-eth0:1
DEVICE="eth0:1"
BOOTPROTO="static"
HWADDR="00:50:56:AF:0C:06"
IPADDR="192.168.0.2"
IPV6INIT="yes"
NETMASK="255.255.255.0"
ONBOOT="no"
TYPE="Ethernet"
DNS1="192.168.2.10"
DNS2="192.168.3.10"

ip addr show

[root@server-1 network-scripts]# ip addr show
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc noqueue state UNKNOWN
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:50:56:AF:0c:06 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.1/24 brd 192.168.0..255 scope global eth0
    inet 192.168.2.1/30 brd 192.168.2.3 scope global eth0:2
    inet 192.168.2.5/30 brd 192.168.2.7 scope global eth0:3
    inet6 fe80::250:56ff:fe97:c06/64 scope link
       valid_lft forever preferred_lft forever
FilBot3
  • 234
  • 4
  • 18
  • Can you post the contents of the `ifcfg-eth0:1` file? – John Sep 12 '13 at 14:43
  • Configs posted. – FilBot3 Sep 12 '13 at 15:01
  • Did you remove/change MAC addresses (HWADDR)? – dsmsk80 Sep 12 '13 at 15:07
  • No, I copied what is in the config files. When I created the configs, i just did a `cp ifcfg-eth0 ifcfg-eth0:1 then edited the name and IP address. – FilBot3 Sep 12 '13 at 15:08
  • Please send output of "ip addr show" from that host. You can have multiple IP addresses on single interface without any sub-interfaces. – dsmsk80 Sep 12 '13 at 15:09
  • Added `ip addr show` output. – FilBot3 Sep 12 '13 at 15:13
  • Simplify/remove the useless directives in the ifcfg-eth0:1, leave only DEVICE=eth0:1, IPADDR=192.168.0.2, IPV6INIT=yes, NETMASK=255.255.255.0, ONBOOT=yes, issue service network restart and test it again – dsmsk80 Sep 12 '13 at 15:17
  • I shouldn't leave the HWADDR in there as well? That was my experience with the other sub-interfaces I've had to create. – FilBot3 Sep 12 '13 at 15:19
  • It's not necessary, you create a sub-interface of some interface which already has HWADDR defined. – dsmsk80 Sep 12 '13 at 15:24
  • So far, even after trimming the `ifcfg-eth0:1` file down to just those few lines, I'm getting that network interface is in use error. However, when I specified the `netmask` along with the ifconfig command, everything seems to be working just fine. I'm still testing a bit. The command I used was `ifconfig eth0:1 192.168.0.2 netmask 255.255.255.0 up` and I can resolve DNS names and such unlike before. – FilBot3 Sep 12 '13 at 15:47

2 Answers2

3

This message is generated either by /etc/sysconfig/network-scripts/ifup-eth or /etc/sysconfig/network-scripts/ifup-aliases which process ifcfg-ethX configuration files to configure network interfaces.

When you check the scripts more carefully you will find out that this error appears after arping command which is trying to find out if the IP addresses specified in ifcfg-ethX files are unique in the network.

The scripts try to avoid of duplicate configurations in the network. Ifconfig doesn't care, it just add IP address to the interface. Check your configuration files twice as there may be some IP address twice.

Finally, it's worth to mention that you can put ARPCHECK=no directive to ifcfg-ethX file to disable this check.

dsmsk80
  • 5,757
  • 17
  • 22
  • Some how, someone had the same IP address as the server I was assigned, so the provisioning team double assigned an IP address. – FilBot3 Apr 10 '15 at 16:18
1

I worked this out after reading dsmsk80 answer and looking in the ifup-eth script. The key line is this:

    /sbin/arping -c 2 -w 3 -D -I <INTERFACE> <VLAN>

So for the OP example:

    /sbin/arping -c 2 -w 3 -D -I eth0:1 192.168.0.2

Something I used this on recently returned this:

    > /sbin/arping -c 2 -w 3 -D -I eth0.1508 192.168.8.1
    ARPING 192.168.8.1 from 0.0.0.0 eth0.1508
    Unicast reply from 192.168.8.1 [00:1C:C4:A1:D8:39]  0.605ms
    Sent 1 probes (1 broadcast(s))
    Received 1 response(s)

I could then take this MAC (00:1C:C4:A1:D8:39) and look it up on the dynamic addresses section of my switch. This in turn told me the VLAN and port numbers of the interface that already used that IP.

Wookie
  • 111
  • 2