1

EDIT#2: I think I found the cause of the issue: in /proc/net/vlan/config there is an old entry for vlan620 which links to bond0 (I was trying to do bonding originally) plus an old vlan620 file again tied to bond0. How can these entries be removed since contents of /proc cannot be edited like normal files?

I am trying to use eth0 as 802.1q trunk interface - my configs are as follows:

ifcfg-eth0
# Intel Corporation 82571EB Gigabit Ethernet Controller
DEVICE=eth0
#SLAVE=yes
#MASTER=bond0
HWADDR=00:1b:21:ac:ae:d2
ONBOOT=yes
BOOTPROTO=none
TYPE=Ethernet

ifcfg-vlan620
#VLAN620 Config
VLAN=yes
VLAN_NAME_TYPE=VLAN_PLUS_VID_NO_PAD
DEVICE=vlan620
PHYSDEV=eth0
BOOTPROTO=static
ONBOOT=yes
IPADDR=172.16.37.7
NETMASK=255.255.255.224
GATEWAY=172.16.37.1

ifcfg-vlan621
#VLAN621 Configuration
VLAN=yes
VLAN_NAME_TYPE=VLAN_PLUS_VID_NO_PAD
DEVICE=vlan621
PHYSDEV=eth0
BOOTPROTO=static
ONBOOT=yes
IPADDR=172.16.37.39
NETMASK=255.255.255.224
GATEWAY=172.16.37.33

I then execute 'ifup eth0' then 'ifup vlan620'. The following error occurs when I try to bring up vlan620:

Device vlan620 does not seem to be present, delaying initialization.

However vlan621 can be brought up fine and can be used to access network.

When both interfaces are in down state (ifdown vlan620 and ifdown vlan621) I checked the system for files named vlan620 and vlan621.

find / -name vlan620
/proc/net/vlan/vlan620

A file exists under /proc/net even though interface is down. File contents

eth1  VID: 620   REORDER_HDR: 1  dev->priv_flags: 81
         total frames received         1379
          total bytes received        69170
      Broadcast/Multicast Rcvd          610

      total frames transmitted           83
       total bytes transmitted        17564
            total headroom inc            0
           total encap on xmit            0
Device: bond0
INGRESS priority mappings: 0:0  1:0  2:0  3:0  4:0  5:0  6:0 7:0
EGRESSS priority Mappings:

Doing the same lookup for vlan621 returns nothing and the interface works fine. The OS is RHEL5.5 32bit. Can anyone suggest a solution or a cause? Rebooting the system does not clean up the issue.

Note: the bond0 entries are commented out where relevant - I am trying to introduce bonding via eth0 and eth1 in active-backup mode but right now vlans dont work so I am trying to fix one problem at a time.

2
  • What distro are you using? What version? Can you create the VLANs by hand with vconfig?
    – David
    Feb 16, 2012 at 22:16
  • It's Red Hat 5.5 32bit. I actually have an idea what is causing the issue - see my edit #2 to the main question.
    – Lukasz
    Feb 16, 2012 at 22:52

1 Answer 1

1

Try using vconfig to remove the interfaces:

vconfig rem vlan620

Linux Journal has good article discussing VLAN support in Linux that may help.

1
  • David thank you that wasnt exactly it but 75% there - there was an entry in there for vlan620 but with interface eth0, so the command I had to use was vconfig rem eth0 instead - not sure how that happened but the problem is solved. Cheers
    – Lukasz
    Feb 17, 2012 at 19:31

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .