HOTSPOT
Your network contains an Active Directory domain named contoso.com. The domain
contains two member servers named Server1 and Server2. All servers run Windows Server
2012 R2.
Server1 and Server2 have the Network Load Balancing (NLB) feature installed. The servers
are configured as nodes in an NLB cluster named Cluster1. Both servers connect to the
same switch.
Cluster1 hosts a secure web Application named WebApp1. WebApp1 saves user state
information in a central database.
You need to ensure that the connections to WebApp1 are distributed evenly between the
nodes. The solution must minimize port flooding.
What should you configure? To answer, configure the appropriate affinity and the
appropriate mode for Cluster1 in the answer area.
Answer: 
Explanation:
http://technet.microsoft.com/en-us/library/cc782694(v=ws.10).aspx
How do You minimize port flooding?
You minimize port flooding by using unicast. However, you can use multicast if IGMP is used. Even though IGMP is not specified in the question, Microsoft is looking for the Multicast answer, and we are just supposed to assume IGMP is being used.
Rick, you’re a bit confused it seems: Unicast *maximizes* port flooding! 🙂 The best way to deal with this is, like you say, by using IGMP Multicast. If the switches don’t support it, however, the next best approach is to use Multicast with *static* ARP entries.
The exhibit of this question is false, check question 165 on this link :
http://www.testdumps.com/latest-microsoft-70-412-real-exam-download-161-170.html
Thanx MC
Affinity = None (Not in Above Exhibit)
Multicast = Evenly distribute Connections among Host.
Sometimes it:
Affinity = None or Single
Mode = Multicast or Multicast
these answers are right?
https://technet.microsoft.com/en-us/library/cc771709.aspx
Affinity: None – Any Node can respond to Request.
Mode: MultiCast – Packets sent to multiple Nodes in NLB Cluster
Multicast: Packets are sent to multiple computers simultaneously in a single transmission
from the source. In other words, when a host sends packets using multicasting, a single
set of packets is sent to all computers at once.
Unicast: each packet is sent to a single network destination identified
by a unique address. In other words, a host sends packets to a single computer.
There is NO ‘Client’ Setting for Affinity.
If you choose the multiple host filtering mode, you can then configure the affinity. When you
configure affinity, you can choose one of the three options:
• None : Any cluster node responds to any client request, even if the client is reconnecting
after an interruption. This option is suitable for stateless application, where the server
that is servicing the request does not have to remember the previous events to complete
the request. As a result, the client can jump from one server to another within the cluster
without problem.
• Single : A single cluster node handles all requests from a single client. This option is useful
for stateful applications where the status of a process or transaction is maintained
through the entire connection including when using SSL and e-commerce shopping cart
applications.
• Class C : A single node responds to all requests from a class C network (a network with a
Subnet of 255.255.255.0), often found when used with Multiple Proxy Servers. This type
of server is often used with cookie-based affinity or when a common database or session
state server is used.
This should be Affinity=None and Mode=Multicast. As I understand it, in NLB the unicast mode spoofs the same MAC address across all the nodes in the cluster, so all traffic to the individual nodes has to be sent across every switch port because it doesn’t know which port to contact the individual node on. In Multicast however, the original MAC address is kept and the clusters shared MAC is assigned as an additional address on each node. This means internal traffic can still be sent to individual nodes, while cluster traffic is still sent to the shared MAC & reaches each node. This reduces the amount of port flooding, and can be further improved by IGMP multicast if the switches support it
IMO The answer is None and Multicast
Not Single Single because the Web App saves user state information in a central (Back End) DB … See explanation below
Enabling either Single or Class C affinity ensures that only one cluster host handles all connections that are part of the same client session. This is important if the server application running on the cluster host maintains session state (such as “server cookies”) between connections.
At the same time, it is important to realize that this does not preserve session state with back-end databases in which many different transactions are occurring involving many different computers. Once the connection is ended, session state also ends.
Disabling affinity allows for improved load balancing because it allows multiple connections from the same client to be handled concurrently by different cluster hosts. To maximize scaled performance, disable client affinity (using the None option) when it is not needed
Multicast for the same reason Dave commented re the flooding which this article explains well
http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1006580
Question says ” Cluster1 hosts a secure web Application named WebApp1. WebApp1 saves user state
information in a central database.” ( secure web applications means SSL is in picture )
So Isnt it supposed to be Singe Affinity ?
Just pondering on James L’s explanation. Makes sense about the central database exception case !
Maybe the answer is Affinity = None
I remained fairly puzzled about how multicast could produce less port flooding than unicast. So for any other doubters like me, straight from the horse’s mouth:
“Although the unicast method works in all routing situations, it has the following disadvantages:
A second network adapter is required to provide peer-to-peer communication between cluster hosts.
If the cluster is connected to a switch, incoming packets are sent to all the ports on the switch, which can cause switch flooding.”
https://technet.microsoft.com/en-us/library/cc782694%28v=ws.10%29.aspx
I think the key word here is “Secure” Web Application
If it were HTTP then “None” would be correct for Affinity
Since it is HTTPS the Affinity should be “Single”
https://technet.microsoft.com/en-us/library/cc759039%28v=ws.10%29.aspx
excelent simply brilliant…
Fantastic, thanks!
Nice link thanks, the state “stored in a central database” can send you towards None as any node should be able to serve a request in that case. But as you say, the mention of secure/HTTPS is critical.
I will confuse things a little, the Affinity could also be Class C/Network. No mention if this app is an Intranet or Internet site.
“It is not efficient for Internet clients to use Single affinity because, in Single affinity, NLB load balances each client by the client’s entire IP address, which can span a broad range. By using Class C affinity, NLB associates clients with only the same Class C portion of the IP address with particular members. Therefore, you essentially reduce the range of IP addresses by which NLB load balances clients.”
https://technet.microsoft.com/en-us/library/bb687542.aspx
I think Mc Lovin is right on the spot.
I agree with the single affinity, as it is a secure web app (https) which is not supported by none.
I thought that unicast would minimize port flooding, unless multicast with IGMP is enabled, but that is not an option.
So my opinion is single/unicast
I also tend to back the idea of “single affinity” because of ‘Secure web app” and saving of “user state” in a database which needs be done on per connection and the fact that “even load distribution” is needed on per ” client connection” to WebApp1 and not per html pages.
As for the mode ” Multicast” since that is the only way to avoid flooding.
I think the question is assuming that you are using a switch that has support for IGMP mode (I believe most switches support this these days anyway). Therefore I have changed my mind an now think it is single/multicast.
This question was in my exam today, as it is stated above but it stated that it has system-state data. I went for Single (So state data isn’t lost and multicast)
Helpful info. Fortunate me I discovered your website by accident, and I’m shocked why this coincidence didn’t came about earlier! I bookmarked it.|
Affinity should be set to None, having it like this allows every node to respond to incoming requests.
Affinity: NONE
Mode: Multicast
I agree with Joe that single/unicast is correct.
I watched the CBTNuggets70-412 Exam Videos and this is what Greg says about the Unicast, Multicast and IGMP Mulitcast:
“Unicast is in many ways probably the best of that three in terms of the amount of traffic that is generated on the subnet. Multicast is one of those protocols that the network team loves to hate, because Multicast has a tendency to flood the switches. This flooding can be minimized when your switches support IGMP multicast….”
From:
https://msdn.microsoft.com/en-us/library/bb742455.aspx
“when Network Load Balancing uses multicast mode, switches often flood all ports by default to deliver multicast traffic.”
Looks like Multicast is the right answer.
From msdn:
https://msdn.microsoft.com/en-us/library/bb742455.aspx
Network Load Balancing’s unicast mode induces switch flooding in order to simultaneously deliver incoming network traffic to all cluster hosts. Also, when Network Load Balancing uses multicast mode, switches often flood all ports by default to deliver multicast traffic. However, Network Load Balancing’s multicast mode gives the system administrator the opportunity to limit switch flooding by configuring a virtual LAN within the switch for the ports corresponding to the cluster hosts
From Technet:
https://social.technet.microsoft.com/Forums/windowsserver/en-US/962906a8-8193-4026-aed2-14df2d57de33/network-load-balancing-modes?forum=ws2016
In Unicast mode, NLB reassigns the station MAC (media access control) address of the network adapter for which it is enabled and all cluster hosts are assigned the same MAC address. Unicast mode induces switch flooding, where all switch ports are flooded with NLB traffic, even ports to which non-NLB servers are attached.
In multicast mode each cluster host is gonna keep original MAC address and it’s going to assing an additional multicast mac address to each newtork adapter so you will have one network adapter in server and each node in cluster will have same multicast mac address. This makes it easier for networks switches and routers that supports multicast to be able to understand how to communicate with the particular host. Multicast doesn’t really solve the network flooding problem.
IGMP multicast – this is the special form of multicast form which will prevent switches from being flooded with network traffic. When you deploy igmp traffic will be forwarded only through switch ports that participate in NLB cluster. What that means that you will need switches that supports that functionality.