vlan doesn't work properly on 12.04 server

it s seems that the problem is from dhclient :

May 30 19:16:22 server32 dhclient: No DHCPOFFERS received.

Please attach your /etc/network/interfaces so we can reproduce this.

Thanks
chuck

I have attached the /etc/network/intefaces

with the some config,

works in Ubuntu 11.10

if you intend your vlan to actually come online, you're missing an "auto eth0.101" line in there.

I do that :
ifup eth0.101

"auto eth0.101" I tried to put in but its still the same problem . it s working just at the boot .
if I do : sudo /etc/init.d/networking restart
doesnt work anymore eth0.101 and I have to reboot all machine

Hi Claudio

I've reproduced your configuration (albeit with auto set for eth0.101 as well) and I'm not able to reproduce your issue.

eth0.101 comes up on boot and is restarted correctly using sudo /etc/init.d/networking restart:

sudo /etc/init.d/networking restart
 * Running /etc/init.d/networking restart is deprecated because it may not enable again some interfaces
 * Reconfiguring network interfaces... Removed VLAN -:eth0.101:-
ssh stop/waiting
ssh start/running, process 3232
Set name-type for VLAN subsystem. Should be visible in /proc/net/vlan/config
Added VLAN with VID == 101 to IF -:eth0:-
ssh stop/waiting
ssh start/running, process 3341
                                                                                                               [ OK ]

The messages from your syslog would indicate that the server is not able to retrieve a IP address from the DHCP server for eth0.101, interestingly it does do a release:

DHCPRELEASE on eth0.101 to 192.168.101.1 port 67

but is then not offered a new address.

Please could you check the log files on your DHCP server for more information as to why it does not response to the subsequent DHCPDISCOVER requests.

Marking 'Incomplete' pending your response. Please set back to 'New' once you have provided the requested information.

Thanks

I get the same , I have already upload my syslog on my first comment !

@James Page (james-page)
you got the some problem of mine .

Moving back to Incomplete.

Claudio, you misread James' comment, he said "and I'm not able to reproduce your issue".

James commet 'but is then not offered a new address.'

we talk about eth0.101

so we got the same issue !

Please stop messing with bug statuses.

I just spent a few minutes reproducing your setup and it works exactly as expected:
 - Changed my /etc/network/interfaces
 - Installed the vlan package
 - Rebooted
   - Got eth0 with a DHCP IP and no eth0.101 (as expected)
 - Ran "ifup eth0.101"
   - Got eth0.101 with DHCP IP
 - Ran /etc/init.d/neworking restart
   - Got eth0 with a DHCP IP and no eth0.101 (as expected)

So there's nothing wrong here, marking bug invalid.

reporduce that :
install vlan

Interface :
"auto eth0
iface eth0 inet static
        address 192.168.2.14
        netmask 255.255.255.0
        gateway 192.168.2.1
        dns-nameservers 192.168.2.1

auto eth0.43
iface eth0.43 inet dhcp"

then reboot you get the ip on vlan43 - perfect until here
then :
sudo /etc/init.d/networking restart

 * Running /etc/init.d/networking restart is deprecated because it may not enable again some interfaces
 * Reconfiguring network interfaces... Rather than invoking init scripts through /etc/init.d, use the service(8)
utility, e.g. service smbd reload

Since the script you are attempting to invoke has been converted to an
Upstart job, you may also use the reload(8) utility, e.g. reload smbd
Removed VLAN -:eth0.43:-
ssh stop/waiting
ssh start/running, process 2772
Set name-type for VLAN subsystem. Should be visible in /proc/net/vlan/config
Added VLAN with VID == 43 to IF -:eth0:-
Failed to bring up eth0.43.
                                                                         [ OK ]

ifconfig:
eth0 Link encap:Ethernet HWaddr *********
          inet addr:192.168.2.14 Bcast:192.168.2.255 Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:fe11:bfe4/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:1690 errors:0 dropped:123 overruns:0 frame:0
          TX packets:1460 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:157343 (157.3 KB) TX bytes:567141 (567.1 KB)

eth0.43 Link encap:Ethernet HWaddr *******
          inet6 addr: fe80::20c:29ff:fe11:bfe4/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
          RX packets:106 errors:0 dropped:0 overruns:0 frame:0
          TX packets:69 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0
          RX bytes:4896 (4.8 KB) TX bytes:37886 (37.8 KB)

******* is my MAC

Instead of calling "/etc/init.d/networking restart" which has was clearly stated in the error message, is deprecated, call:
 - ifdown -a -v
 - ifup -a -v

That should show what the problem is.

ifdown -a -v:
ifconfig: empty
/proc/net/vlan/config
VLAN Dev name | VLAN ID
Name-Type: VLAN_NAME_TYPE_RAW_PLUS_VID_NO_PAD

ifup -a -v

Configuring interface lo=lo (inet)
run-parts --verbose /etc/network/if-pre-up.d
run-parts: executing /etc/network/if-pre-up.d/vlan
run-parts: executing /etc/network/if-pre-up.d/wireless-tools
run-parts: executing /etc/network/if-pre-up.d/wpasupplicant
ip link set dev lo up
run-parts --verbose /etc/network/if-up.d
run-parts: executing /etc/network/if-up.d/000resolvconf
run-parts: executing /etc/network/if-up.d/avahi-daemon
run-parts: executing /etc/network/if-up.d/ip
run-parts: executing /etc/network/if-up.d/ntpdate
run-parts: executing /etc/network/if-up.d/openssh-server
run-parts: executing /etc/network/if-up.d/samba
run-parts: executing /etc/network/if-up.d/upstart
run-parts: executing /etc/network/if-up.d/wpasupplicant
Configuring interface eth0=eth0 (inet)
run-parts --verbose /etc/network/if-pre-up.d
run-parts: executing /etc/network/if-pre-up.d/vlan
run-parts: executing /etc/network/if-pre-up.d/wireless-tools
run-parts: executing /etc/network/if-pre-up.d/wpasupplicant
ip addr add 192.168.2.14/255.255.255.0 broadcast + dev eth0 label eth0
ip link set dev eth0 up
 ip route add default via 192.168.2.1 metric 100 dev eth0
run-parts --verbose /etc/network/if-up.d
run-parts: executing /etc/network/if-up.d/000resolvconf
run-parts: executing /etc/network/if-up.d/avahi-daemon
run-parts: executing /etc/network/if-up.d/ip
run-parts: executing /etc/network/if-up.d/ntpdate
run-parts: executing /etc/network/if-up.d/openssh-server
ssh stop/waiting
ssh start/running, process 13101
run-parts: executing /etc/network/if-up.d/samba
run-parts: executing /etc/network/if-up.d/upstart
run-parts: executing /etc/network/if-up.d/wpasupplicant
Configuring interface eth0.40=eth0.40 (inet)
run-parts --verbose /etc/network/if-pre-up.d
run-parts: executing /etc/network/if-pre-up.d/vlan
Set name-type for VLAN subsystem. Should be visible in /proc/net/vlan/config
Added VLAN with VID == 40 to IF -:eth0:-
run-parts: executing /etc/network/if-pre-up.d/wireless-tools
run-parts: executing /etc/network/if-pre-up.d/wpasupplicant

dhclient3 -e IF_METRIC=100 -pf /var/run/dhclient.eth0.40.pid -lf /var/lib/dhcp/dhclient.eth0.40.leases -1 eth0.40
Failed to bring up eth0.40.

We are having the same problem.

After restarting a VLAN interface in 12.04 (ifdown/ifup) we cannot communicate with other hosts in the the same VLAN.
The ifup command does not report any error. It says:
"Added VLAN with VID == XX to IF -:ethX:-" but no communication with hosts in the VLAN.
The only way to fix it is to reboot.

I found our problem. Seems to be related to the intel e1000 driver.

Running the following fixed it for us:
 ethtool -K ethX rxvlan off

Found the info here:
https://bugzilla.redhat.com/show_bug.cgi?id=754589

Claudio can you confirm that the following:
dhclient3 -e IF_METRIC=100 -pf /var/run/dhclient.eth0.40.pid -lf /var/lib/dhcp/dhclient.eth0.40.leases -1 eth0.40
echo $?

Gives you a > 0 value?

Then, you can perhaps try:
dhclient3 -e IF_METRIC=100 -pf /var/run/dhclient.eth0.40.pid -lf /var/lib/dhcp/dhclient.eth0.40.leases -1 eth0.40 -d

Which should give you more details on why it's failing.

If Pär Näsberg's workaround works for you too, then it's a kernel bug and I'll close that bug or re-assign to kernel.

No reply in over a year, seems like this was a kernel bug related to the e1000 driver, not a userspace bug.