Howto bind multicast listener to a specific network interface (NIC)?

All of our cluster machines have 2 network interfaces. One of the requirements is that all cluster-related traffic has to use a specific interface, leaving the other open for all non-clustering network traffic.
Thus, I've to bind the Coherence multicast listener to a specific interface.
Unfortunately I did not find a setting for this in the Coherence configuration XML file.
Is it possible to bind the multicast listener to a specific NIC, if yes, how can it be done?
Thanks a lot!
Best regards,
Wolf

Hi Wolf,
it can be specified as part of the unicast-listener configuration within the operational configuration:
http://coherence.oracle.com/display/COH34UG/unicast-listener
You need to specify the address and the port elements.
Alternatively, you can specify values for this as a Java property:
-Dtangosol.coherence.localhost=nicaddress -Dtangosol.coherence.localport=unicastportwhere nicaddress is the ip address of the network card going to the network you would like to direct Coherence traffic to. Unicast port is the udp port on which Coherence will listen. By default it is 8088 (or automatically incremente in case the port is already bound).
If your kernel is configured to use both ipv6 and ipv4 and you specified the ipv4 address of the network card then you might also need to specify:
-Djava.net.preferIPv4Stack=trueThe multicast listener and publisher will send and receive on the interface used by the unicast listener.
Best regards,
Robert

Similar Messages

Using a specific network interface

Hello, I would like to know how I can force Mac OS X to use a specific network interface while connecting to my mac mini (from the iMac):
I'm connected to the internet (I mean the local network) using the airport, and I connected my iMac with the mac mini using an ethernet cable (because I want to transfer files faster, and I don't want to reboot my mini in target disk mode), but when I connect to the mini (using the "share" feature) it seems to use the airport connection. Of course I could disable the airport interface to force it to use the ethernet one (and it works) but I don't want to...
So, how can I force a specific interface? I don't mind using a command line command to mount it initially btw.
Thanks

Hello,
Try using a different IP address range to put each interface in a different subnet. For example use 10.0.0.x and 255.255.255.0 for the Ethernet connection and 192.18.2.x and 255.255.255.0 for the Airport.
mrtotes

Setting applications to use specific network interfaces

How do I configure my computer such that specific applications will default to a different network interface than all the rest?
I want to set it up so that Adium always tries to use AirPort first and everything else uses ethernet if it's available.
Powerbook G4 1.5 Ghz Mac OS X (10.4.8) 1GB RAM

You can't (normally) get there from here.
What you're trying to do is affect the routing (route certain traffic through a specific interface). The issue is that routing works at the IP level, not the application level. This means that you can tell the system to route a.b.c.d (e.g. the address of the IM server you're connecting to) through a speciic router at w.x.y.z (e.g. the AirPort).
In order to do this, though, you need to have different IP address/subnet masks on each interface.
For example, assuming your wired ethernet is using 192.168.1.2/255.255.255.0 and your wired ethernet router is at 192.168.1.1, and your AirPort interface is using 10.1.1.2 with the AirPort router at 10.1.1.1 you can run the command:
<pre class=command>sudo route add a.b.c.d 10.1.1.1</pre>
and all traffic for a.b.c.d will go over the AirPort link.
Note that it doesn't matter what protocol you use - any traffic destined for a.b.c.d, whether it's AIM, HTTP, FTP or anything else - or what application you use - Adium, Safari, FireFox, etc. - they'll all use the AirPort link.
Note that this technique will NOT work if you have both interfaces on the machine configured in the same subnet, but this is not a supported configuration, so you shouldn't be doing that anyway.

Assigning certain applications to specific network interfaces

I connect to a wired and wireless network. Each have different uses.
I'd like to assign the web browsing and some other tasks to the wireless and the filesharing and NAS to the wired network.
Is there a way to assign this? or is it one or the other?

Just so we understand your setup:
You have two networks that you're connecting to, at
the same time?
Two routers/BaseStations?
Your airport card gets an IP like 10.0.1.x
Your ethernet gets an IP like...192.168.0.x
Something like that.

How to tell Transmission to use a specific network interface?

I don't see any option for it in the Preferences. Does anybody know?
Thank you.

Take a look in .config/transmission-daemon/settings.json. There are bind-address options to play with.

Dual network interface when joining multicast group

Hello,
I've have a machine with two networks interfaces, eth0 and eth1. The route table is configured to send all the multicast traffic through eth1.
This is
route add -net 224.0.0.0 netmask 240.0.0.0 dev eth1
I have the following code that joins to a multicast address to listen incoming messages in my Java application
InetAddress group = InetAddress.getByName(multiCastAddress);
socket = new MulticastSocket(Integer.parseInt(multiCastPort));
socket.joinGroup(group);
The result is that the application can not see the multicast traffic since the joinGroup operation is always done on eth0, guess because JVM selects the first one available.
If I change the route entry to use eth0 for all the multicast traffic, the same code works perfectly, and from my Java application I can see the multicast traffic.
This is
route add -net 224.0.0.0 netmask 240.0.0.0 dev eth0
Is there any way to force to join a specific network interface?
Thanks

The answer is staring you in the face in the Javadoc:
http://java.sun.com/j2se/1.5.0/docs/api/java/net/MulticastSocket.html#joinGroup(java.net.SocketAddress,%20java.net.NetworkInterface)
That bug report only concerns setNetworkInterface(), which the OP isn't using, on Linux and IPv6, and it's dated 2002.

Unable to receive multicast on a multi network interface host

Can anyone please help?
I am not able to receive multicast on a multi network interface host.
This is the code:
public class LegacyMulticast {
public static void main(String[] args) throws IOException {
InetSocketAddress group = new InetSocketAddress("224.10.20.30", 12345);
NetworkInterface iface = NetworkInterface.getByName("eth3");
MulticastSocket socket = new MulticastSocket();
socket.setSoTimeout(5000);
socket.joinGroup(group , iface);
byte[] b = new byte[4096];
DatagramPacket datagram = new DatagramPacket(b, b.length);
try {
socket.receive(datagram);
System.out.println("Success");
catch (SocketTimeoutException ste) {
System.out.println("Failure");
socket.leaveGroup(group, iface);
The funny thing is that when I start this code, Wireshark detects IGMP join on the correct network interface caused by joinGroup call, followed by constant UDP traffic on correct port which I aim to receive and finally I see correct IGMP leave caused by leaveGroup call and the UDP traffic stops.
I just don't receive anything on socket.receive call.
I am trying to figure this one out for days now. Pls help someone :)
P.S. I suspected security, but System.getSecurityManager() call returns null.
P.P.S. I also tried implementing this with NIO, but with same result.

public class LegacyMulticast {
public static void main(String[] args) throws IOException {
InetSocketAddress group = new InetSocketAddress("224.10.20.30", 12345);
NetworkInterface iface = NetworkInterface.getByName("eth3");
MulticastSocket socket = new MulticastSocket(12345);
socket.setSoTimeout(5000);
socket.joinGroup(group , iface);
byte[] b = new byte[4096];
DatagramPacket datagram = new DatagramPacket(b, b.length);
try {
socket.receive(datagram);
System.out.println("Success");
catch (SocketTimeoutException ste) {
System.out.println("Failure");
socket.leaveGroup(group, iface);
This is still not working.
The multicast is coming from the network (not localhost), OS is Windows 7.
When I start a multicast stream localy via VLC application on this same group and port, I receive the traffic in this Java application OK.
Once again, Wireshark detects IGMP join and leave and multicast traffic flow on eth3 network interface initiated by the Java application, but it just doesn't see the UDP datagrams.

Mount filesystems (NFS & SSHFS) on specific network comming up

Hi,
I'd like to mount some NFS and SSHFS shares when a specific network/interface has a connection. How can I do this?
Thanks for your help in advance.

The problem was how the exports were configured from the Synology box. I tried to specify access for my whole subnet using 192.168.1.* as the host specification. A more thorough reading of the /etc/exports man page statement on wildcards notes the following: "Wildcards should not be used with IP addresses; however, it is possible for them to work accidentally if reverse DNS lookups fail." In my case, my servers had DNS (and therefore presumable revers DNS) entries, and therefore mounts failed. My clients (largely laptops) get a DHCP address from my router, and don't have DNS entries in my domain. Therefore, the mounts "worked accidentally" from my clients, and not from my servers.
I'm now adding "works accidentally" to my lexicon of strange & wonderous software development feature terms, along with "bug-like feature". Now that I've calmed down enough from the lost week of spare time, I should be able to submit an RFE to Synology without using any profanity. Had I entered the offending wildcards by hand-editing the exports, I'd hold myself entirely responsible. Since I did this with the Synology UI, it seems reasonable to expect that I'd at least see a warning that my host spec was potentially not going to achieve the intended results.

Network Interface selection

Hello,
I'd like to use specific network interfaces for different applications. For example using always wi-fi for skype, only one browser on a specified VPN and so on...
I thought to do it using ipfw, but it seems it is not possible to set rules on the base of process ID.
Is there any tool allowing to do such a thing? Any other suggestion would be well accepted!
Thanks in advance

I am not aware of any tool that can accomplish exactly what you ask. The command line tool "route" can setup protocol-based routes. You could have skype go to one interface and http go through another. But I don't know of any way to make it application-specific.

Why not use application specific multicast listener in appserver cluster

Hello Friends,
I am using oracle9iAs. we have basically two enterprise application. and often they communicate to each other, currently we have message bean sending a message with is nothing more than 2 integer values. The second application uses these two integers to update specific data cache.
First of All I am not aware or have come across any documentation on JMS clustering in Oralce9iAS. they still are very much behind Weblogic interms of feature implementation, If I wrong correct me here.
Now coming to my requirement, we are clustering application servers now and I need to be able to send messages to all
the servers in the cluster, lets say message generated by server 1 and application A to Application B on all other servers in cluster including the host.
well My dilemma is I want to use JMS but How to subscribe to different topics on cluster if topic is not only cluster able,.. if topics themselves running in clustered servers are not clustered.
so the easiest way for me to do this is to use a Multicast Listener on Application B on all the servers in the cluster and send message to it. well I know I will using socket here,... but I do also have control over transaction scope, this message is only sent upon successful completion of transaction on Application A.
now I ask my question. In this situation do things is there any risk that I am taking by using MulticastListner in my app?
your thoughts and ideas are appreciated
Thanks
Jay

Scrap JMS. Use plain java.net or JINI.
You could create a Peer-to-Peer Chain:
Server A -> Server B -> Server C -> Server A
On each Server have a server listening for messages (from the Server to the left in the chain) and a client that sends messages (to the Server to the right in the chain) and an application that links the server and client on each Server.
So,
Server A's "server" listens for messages from the "client" on Server C, and so on.
This is vary efficent. A simple send-receive chain that is clusterd within a small (or large) group of Servers. You could also try Bi-Directional Parallel Communications BDPC. That would basically have Server A send messages to both Servers - one to the left and one to the right, then at the same time, listen for messages from the Server on the left and the Server on the right. This is better suited for large-scale clusters where a Uni-directional communications model would take too long to propogate messages to every member of the group.
But, that's just me.

ICMPv6 'Multicast Listener Report' messages flooding the local network

Hi all,
We have a small number of 93z desktops scattered around our school. There seems to be an issue with the ethernet adapters running the Intel I217 chipset. When these machines are in sleep one or more start sending IPV6 "Multicast Listener Report Message v2" packets to the network (observed by Wireshark). Then they begin to snowball and shortly thereafter they are flooding the network, bringing it to its knees.
We had the same issue on some Dell workstations, however, Dell released a firmware update to address this issue with this chipset.
Just wondering if this is an issue Lenovo is aware of and/or has addressed. We've worked around it on Lenovo's machines by disabling all power options on the ethernet adapters but we shouldn't have to do that. If Dell can provide a firmware fix why can't Lenovo (assuming that they haven't).
I know I'm not the only one - https://www.google.com.au/?complete=0#complete=0&q=intel+I217+flood&safe=active

I can't tell you which drive to use because Lenovo are terrible with drivers, however, the best way to workaround the issue is to disable the power management settings on the adapter...
We had the same problem on some Dell T1700 workstations, Dell moved very quickly to release an updated BIOS and driver that resolved it. Intel, I believe, have also corrected this in their native drivers but I'm unsure (and unwilling to test) if Lenovo have, I don't believe they had as of a month ago as we reimaged some machines and updated them with latest drivers through the update tool and then we started seeing multicast traffic again.
Lenovo support is rubbish.

Use VPN connection as a listen network interface in Web Application proxy

I have a test environment: domain in hyper-v with Sharepoint and Office Web Apps servers (all under Windows 2012 - Windows 2012 R2).
Because my home ISP does not permit some inbound ports (80,443) in a gate machine (under Windows 2012 R2) I create a vpn connection (by "setup a new connection or network") to my outside vpn server. On this vpn server the ports forwarding is configured
and work fine (f.e. default IIS site is visible).
I try to public my Sharepoint 2013 Foundation in Internet over this vpn connection and faced with the problem - WAPx (Web application proxy) does not bind to this vpn connection, only to traditional network interfaces.
The question is how to make listening WAPx the VPN interface?

Hi,
Thank you for posting in Windows Server Forum.
Please check beneath thread and article might helpful in your case.
Configure a reverse proxy device for SharePoint Server 2013 hybrid
http://technet.microsoft.com/en-us/library/dn607304(v=office.15).aspx
Forcing VPN users through a proxy
http://social.technet.microsoft.com/Forums/en-US/5a6a502d-4583-4c51-8486-3af982ba92da/forcing-vpn-users-through-a-proxy?forum=winserverNIS
What’s New in 2012 R2: People-centric IT in Action - End-to-end Scenarios Across Products
http://blogs.technet.com/b/in_the_cloud/archive/2013/07/17/people-centric-it-in-action-end-to-end-scenarios-across-products.aspx
Hope it helps!
Thanks,
Dharmesh

Associating NAT with a specific network service definition... how?

I don't know if this is possible, but if it is I would value any help in configuring it on SLS 10.6.2
I have SLS 10.6.2 server running on a Mac Mini. It connects to multiple static IPs via a number of 'virtual' network services using its single ethernet connection. This achieved by simply creating new services in system preferences / network associated with the ethernet interface, and defining each to have its own fixed IP within our assigned static IP range. This allows the mini to listen for and respond to multiple public IPs. The various bits of software on the mini are configured to 'listen' to their own IP only. The approach works well for this purpose.
We are using SLS VPN to allow connections to the server from remote locations, and currently provide these with access to the public internet by setting up the VPN to assign one of the unused static IP address in our block to the incoming connection. It works fine unless there is more than one person wanting to use the VPN connection...
So I would like to set up a NAT configuration on the unused IP address and associate this with a private network range (e.g. 192.168.x.x), and then get the VPN to allocated incoming connections on of the range of IP's.
When I try to configure NAT using server preferences, I only have the option of attaching it to one of the defined network interfaces (e.g. ethernet, airport, firewire), but not to a specific service attached to the ethernet connection.
Is it possible to do this? If so how?
Alternatively, is it possible to configure an unused network connection (e.g. the FW connection) to point to the required fixed IP, even if it is not physically connected to anything? If I could do this, I could side-step the problem.
Thanks in advance for any help.
Message was edited by: Gavin Lawrie

Get yourself a server-grade firewall (VPN, DMZ, port-mapping, syslog, etc) and move on.

Multicast Listener Report Message purpose

Hi everybody.
An ipv6 host performs dad before starts using ipv6 address. For example our router is an ipv6 client performing DAD as shown below:
::     ff02::1:ff00:2222         ICMPv6 78 Neighbor Solicitation for fe80::200:ff:fe00:2222
fe80::200:ff:fe00:2222         ff02::1 ICMPv6 86 Neighbor Advertisement fe80::200:ff:fe00:2222 (ovr) is at 00:00:00:00:22:22
fe80::200:ff:fe00:2222          ff02::16 ICMPv6 90 Multicast Listener Report Message v
What is the purpose of " Multicast listener report message" above ?
The " multicast listener report message" in expanded form shown below:
No.     Time        Source                Destination           Protocol Length Info
     11 58.984000   fe80::200:ff:fe00:2222 ff02::16              ICMPv6   90     Multicast Listener Report Message v2
Frame 11: 90 bytes on wire (720 bits), 90 bytes captured (720 bits)
Ethernet II, Src: 00:00:00_00:22:22 (00:00:00:00:22:22), Dst: IPv6mcast_00:00:00:16 (33:33:00:00:00:16)
Internet Protocol Version 6, Src: fe80::200:ff:fe00:2222 (fe80::200:ff:fe00:2222), Dst: ff02::16 (ff02::16)
Internet Control Message Protocol v6
    Type: Multicast Listener Report Message v2 (143)
    Code: 0
    Checksum: 0x2bc5 [correct]
    Reserved: 0000
    Number of Multicast Address Records: 1
    Multicast Address Record Changed to exclude: ff02::1:ff00:2222
        Record Type: Changed to exclude (4)
        Aux Data Len: 0
        Number of Sources: 0
        Multicast Address: ff02::1:ff00:2222 (ff02::1:ff00:2222)
================================================================
Thanks and have a great weekend.

I think the MLD report is to ensure that the MAC address isn't subscribed to any stale multicast group memberships left over from the previous client network configuration.
--- Jim Leinweber, WI State Lab of Hygiene

Streaming Photos and listen to a specific playlist ?

In the new software version for ATV is it possibile to Stream Photos and listen to a specific playlist (stream) like in the older version ? If yes, can somebody give a help ??? (without sync libraries)

I want that my client and server program can
communicate with each other without configuring the
router.
I don't want to make a chatprogram but used this as
an example how communication is possible without
configuring the router.Nope.
>
My server and client program are both running on a
different computer at 2 different LAN's. The server
has 2 buttons and if one is pressed the client
receives a message. This works in a LAN, but not how
I want.
Okay. Although the statement gets a little odd there at the end.
If you want Computer A to send a message to Computer B and they are in different networks then the router for the network where computer B is must be configured to allow and forward through connections from Computer A to computer B.
And vice-versa.
The alternative is to do the following.
Have a third computer (let's call it C) that is a server and is available through a public IP.
Then when Computers A and B come online they open connections to C. When A wants to send a message to B... A sends to C and C sends to B.
This would not require any network reconfigurations.

Howto bind multicast listener to a specific network interface (NIC)?

Similar Messages

Maybe you are looking for