We're having an issue with UDP broadcast over a VLAN that has three IP's configured on it. This is on a Cat 6509.
ip address 10.10.249.1 255.255.255.0 secondary
ip address 10.10.250.1 255.255.255.0 secondary
ip address 192.168.101.1 255.255.255.0
no ip redirects
ip pim sparse-mode
The device broadcasting is 192.168.101.34.
We've added the global config:
ip forward-protocol udp 4444
And we've tried interface commands like:
ip helper-address 192.168.101.34
ip directed broadcast
Nothing seems to help, though. Are there any inherit limitations due to the multinet? Devices on the 192.168.101.0/24 subnet receive the broadcast fine. Devices on the other two subnets do not. Basic routing between all subnets is fine.
You shouldn't need to configure anything on the L3 vlan interface because all clients are in the same vlan and a vlan is a broadcast domain so the switch should flood out the broadcast to all ports in that vlan.
Is the broadcast address 255.255.255.255 or is it a directed broadcast ie. 192.168.101.255. Either way every device should receive the frame at L2 because the switch should flood the broadcast to all switch ports in that vlan, other than the one it was received on.
Thanks for the reply. This seems to me fundamentally correct and exactly what I would think. The broadcast address for that L3 vlan interface is 255.255.255.255.
If I connect a workstation to a port in VLAN 11 and give it the IP address of 192.168.101.128 my network stream in VLC at udp://@:4444 works fine. If I change the IP address on that workstation to anything in the 10.10.249.x or 10.10.250.x, the stream immediately stops. What's weird is it doesn't give me any errors that you might typically see in VLC if say a firewall is preventing access to the stream.
I guess I'm looking at this incorrectly and the issue isn't with broadcast forwarding at all.
I know this is a longshot, but have you tried to add a secondary address to the source vlc workstation? In other words, if the ip address is 192.168.1.2, add 10.10.249.2 and 10.10.250.2 to the workstation as secondary addresses.
*** Please rate all useful posts ***
I'm going under the presumption now that the Barix is doing a directed broadcast to the 192.168.1.0/24 subnet. So, I'm guessing the switch is forwarding that as a directed-broadcast even though the L3 interface broadcast address is 255.255.255.255 and the switch has not been configured to forward directed broadcasts??? Doesn't make sense to me. I'm tempted to attempt to configure the Barix broadcast to 255.255.255.255. The limited documentation that I've read suggests it is broadcasting to 192.168.0.255.
That was my thinking as well. A directed broadcast would still get to all the devices at L2 but would be dropped by non 192.168.101.x clients when they realised it wasn't addressed to them.
With a 255.255.255.255 address all hosts on that vlan would then have to process the packet.
Thanks for the reply. This actually might be a good, temporary work-around as we migrate to the new IP range; however, with Win 7 it seems it is no longer possible, or at least easy, to add a secondary IP address to an interface with DHCP enabled. If you do it with static, and go back to DHCP it wipes out all secondary IP's. Maybe there's a way to do it in the registry, but I've not researched that yet.
Thank you though,
The reason I suggested it is that we have multiple subnets on one of our vlans like you, and dhcp will not work for anything other than the primary subnet unless we have a secondary address on the dhcp server to cover all of the secondary addresses as well. It may be your permanent solution, and not just a temporary workaround if that's the only way to get it to work. You may want to statically assign this host (I'm talking about the vlc host only) an address and then put the secondary address on it to test.
*** Please rate all useful posts ***
If you're running 2008R2 DHCP, you can fix the DHCP issue that you're having by configuring a DHCP super scope on the DHCP server. We ran into that problem too and managed to get DHCP working on all three subnets. I know it works on 2008R2 but not sure on other versions.
We're mandated to get off the 192.168.101.0 subnet; so, if we can't get it working this way we may need to do a complete cutover including re-IP-ing all devices involved in this dilemma. It's more complicated than I'm stating as this is basically a troubleshooting step to see why it broke. There's an application server involved, an audio streaming server, and some Barix devices. We're attempting to get all the workstations cutover first when we ran into this issue.
Not sure if this is applicable to this post but we have used in the past dhcp smart-relay on interfaces with primary and secondary addresses this way dhcp services all and not just the primary ip address range.
Please don't forget to rate any posts that have been helpful.