0

I am starting a sub-interface off of my existing eth0, and its supposed to be an instant thing, so I'm trying to avoid using a config file. When I issue:

ifconfig eth0:1 192.168.0.2 up

The interface comes up, and works, but my DNS resolution is lost, and I can't ping anything even say www.google.com, yet I could before the interface was brought up. The configuration for the initial interface is: 

[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"

Am I not able to resolve DNS names after I start the sub-interface because they're on the same network?

This is the output of ip addr show

$ ip addr list

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 00:50:56:97:0c:06 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.1/24 brd 10.162.111.255 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
  • Look at the routing table, befote and after. My guess is that your neglegting to include a mask is trashing your routes. – Zoredache Sep 12 '13 at 15:24
  • I might have just solved this issue as well as my other one, I'm testing some more, I will post my answer in a bit. – FilBot3 Sep 12 '13 at 16:07
  • 1
    BTW, you really should avoid using `ifconfig` and start learning how to use `ip` instead. The equivalent command would be `ip addr add 192.168.0.2/24 dev eth0`. – Zoredache Sep 12 '13 at 16:30
  • What's the added benefit of using that? – FilBot3 Sep 13 '13 at 13:41
  • I used the `ip addr add 192.168.02/24 dev eth0` command, and now the ip address is not showing, nor can I use `ifconfig eth0:1 192.168.0.2 netmask 255.255.255.0 up` command. Tells me: `SIOCGIFADDR: Cannot assign requested address SIOCSIFBROADCAST: Cannot assign requested address SIOCSIFFLAGS: Cannot assign requested address` and then when i use the command you suggested with `up` it tells me its garbage. How do I remove the file now since `RTNETLINK answers: File exists` – FilBot3 Sep 13 '13 at 16:18
  • Use `ip addr` to list addresses for an interface. As for why. See http://serverfault.com/questions/458628/should-i-quit-using-ifconfig – Zoredache Sep 13 '13 at 17:03
  • Even when I did the `ip addr list` the interface `eth0:1` didn't show up. – FilBot3 Sep 13 '13 at 18:24
  • No it doesn't show up. **ifconfig lies to you** because it is using a deprecated kernel API. Sub-interfaces are not really interfaces. ip addr correctly reports how the kernel handles additional addresses assigned to an interface. eth0:n is not a separate interface. You need to accept that it isn't another interface, or you will see behaviors you don't understand, because you have invalid expectations. Using `ip` instead of `ifconfig` helps understand what is really going on. – Zoredache Sep 13 '13 at 18:29
  • Ok. Sorry for being a pain. I'm reading up on `ip` right now. – FilBot3 Sep 13 '13 at 18:31
  • Just posted the output of `ip addr show` – FilBot3 Sep 13 '13 at 19:00
  • What the heck? That is weird. – Zoredache Sep 13 '13 at 19:55

1 Answers1

1

Turns out, in relation to this post, the answer I got that helped fix that issue, also fixed this issue. Instead of just issuing:

ifconfig eth0:1 192.168.0.2 up

I issued the command with a netmask at the end:

ifconfig eth0:1 192.168.0.2 netmask 255.255.255.0 up

and it worked. I can ping any DNS name, traceroute and all that jazz.

Community
  • 1
FilBot3
  • 234
  • 4
  • 18
  • However, now that I have the ability to ping FQDN names, I cannot access my website provided by the IP I am bringing up. – FilBot3 Sep 16 '13 at 16:05
  • Turns out that there were multiple machines on my network with that same IP, just no FQDN with them. Someone who is running my network is not properly maintaining it. – FilBot3 Sep 20 '13 at 13:08