Hi all. I have currently an OpenVZ server: uname -a Linux vader8.superhost.pl 2.6.32-042stab055.16 #1 SMP Fri Jun 8 19:22:28 MSD 2012 x86_64 x86_64 x86_64 GNU/Linux cat /etc/redhat-release CentOS release 6.2 (Final) lspci | grep -i eth 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 03) cd /etc/sysconfig/network-scripts cat ifcfg-eth0 DEVICE="eth0" BOOTPROTO=none NM_CONTROLLED="no" ONBOOT="yes" TYPE="Ethernet" HWADDR=00:25:22:0D:C2:2A DOMAIN=domain.eu IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="System eth0" UUID=5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03 BRIDGE=vmbr0 cat ifcfg-eth0.20 DEVICE=eth0.20 BOOTPROTO=none ONBOOT=yes VLAN=yes BRIDGE=vmbr20 cat ifcfg-vmbr20 DEVICE=vmbr20 BOOTPROTO=none DHCPCLASSIPADDR=10.20.0.108 NETMASK=255.255.0.0 ONBOOT=yes TYPE=Bridge cat ifcfg-vmbr0 DEVICE=vmbr0 BOOTPROTO=none DHCPCLASSIPADDR=195.218.152.219 NETMASK=255.255.255.0 ONBOOT=yes TYPE=Bridge cat ifcfg-vmbr0:1 DEVICE=vmbr0:1 BOOTPROTO=none IPADDR=192.168.2.203 NETMASK=255.255.255.0 ONBOOT=yes cat ifcfg-venet0 DEVICE=venet0 ONBOOT=yes STARTMODE=onboot route -n Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 0 vmbr0 195.218.152.0 0.0.0.0 255.255.255.0 U 0 0 0 vmbr0 10.20.0.0 0.0.0.0 255.255.0.0 U 0 0 0 vmbr20 169.254.0.0 0.0.0.0 255.255.0.0 U 1003 0 0 vmbr0 169.254.0.0 0.0.0.0 255.255.0.0 U 1006 0 0 vmbr20 0.0.0.0 193.218.152.1 0.0.0.0 UG 0 0 0 vmbr0 iptables -L Chain INPUT (policy ACCEPT) target prot opt source destination Chain FORWARD (policy ACCEPT) target prot opt source destination Chain OUTPUT (policy ACCEPT) target prot opt source destination So I have a setup in which: - eth0 is my physical interface - it is bridged with vmbr0 with ip 195.218.152.219 and this address is accessible across the network - eth0.20 is a logical interface in VLAN 20 on eth0 - it is bridged with vmbr20 with ip 10.20.0.108 and this address IS NOT accessible across the network - vmbr0 bridge has an alias vmbr0:1 with ip 192.168.2.203 and this address is accessible across the network When I set up the network for testing: - eth0 as a physical interface 195.218.152.219 - eth0.20 as a VLAN interface 10.20.0.108 both addresses are accessible across the network (VLANs on the switch are set up correctly, VLAN 1 untagged, VLAN 20 tagged). Do you see an error in my configuration? Why is 10.20.0.108 not available? Best regards, Rafal.
m.roth at 5-cent.us
2012-Jun-25 14:18 UTC
[CentOS] CentOS 6 bridging problem. - semi-unrelated
Rafa? Radecki wrote:> Hi all. > > I have currently an OpenVZ server: > > uname -a > Linux vader8.superhost.pl 2.6.32-042stab055.16 #1 SMP Fri Jun 8 > 19:22:28 MSD 2012 x86_64 x86_64 x86_64 GNU/Linux > > cat /etc/redhat-release > CentOS release 6.2 (Final)<snip> I don't even remember that kernel for 6.2. The last five or six are all 2.6.32-220.x; before that was a 32-118, and I *think* it started with -76 (correct me if I'm wrong, folks.) mark
On 06/25/2012 05:22 AM, Rafa? Radecki wrote:> Do you see an error in my configuration? Why is 10.20.0.108 not available?Not immediately, but check the output of the 'ip' tools. ifconfig and route are deprecated: ip route show ip addr show Finally, see if there's any incoming traffic on the tagged interface: tcpdump -n -i eth0.20 ... while you ping the assigned address.
On 06/25/2012 05:22 AM, Rafa? Radecki wrote:> Do you see an error in my configuration? Why is 10.20.0.108 not available?I don't have a lot of managed switches around to do extensive testing. The closest test I can do is on 5.8 with VLAN 2 and different IPs. Based on that, everything seems like it should work. Since you're able to set up the addresses on the interface without a bridge, the likelihood of a driver problem seems fairly low. Did you ever send the output of "brctl show"? You should definitely be able to run tcpdump on eth0.20 and see any traffic on that interface. You said that you saw none when you tried to ping the Linux host from the network. What about the reverse? Do you see data go out eth0.20 when you try to ping an address in the attached subnet from the Linux host?