Hi all, I'm not a server engineer, so forgive me if my questions seem somewhat simple to you. I'm a network engineer and are planning a platform migration from Cisco Cat6500 to Cisco Nexus7K. Currently, we're using HSRPv1 for gateway redundancy, but when migrating to Nexus7K we expect to use HSRPv2 instead. Since HSRPv2 uses a different virtual Mac-address, I need to find out how to migrate from v1 to v2 without causing to much downtime for the serveres. When v2 is enabled and the master is elected, a gratuitous arp will be issued, but how will the servers react to this. Usually this is merely done to flush the CAM tables of the switches, but I'm trying to determine, how long the gateway-mac is held in the arp-table of the server, basically looking for the default arp-timeout. It seems that Windows 2003 and 2008 handle arp-caching in different ways, at least from the information I've been able to find. While 2008 seems to have a significantly lower arp-timeout, Windows 2003's is higher, but I'm unsure as to how long it actually caches an arp-reply. Also, does anyone have any recommendations pertaining to a procedure like this? We're really trying to avoid having to suspend the entire serveroperation, as we're looking at 2000+ servers on different subnets. Thanks.

Hi, Thank you for your post. ARP caching behavior has been changed in Windows 2008. The TCP/IP stack implementations in Windows 2008 comply with RFC4861 (Neighbor Discovery protocol for IP version 6 [Ipv6]) for both the IPv4 and the IPv6 Neighbor Discovery process. The ARP cache in Windows 2003: To minimize the number of broadcasts, ARP maintains a cache of IP address-to-media access control address mappings for future use. The ARP cache can contain both dynamic and static entries. Dynamic entries are added and removed automatically over time. Static entries remain in the cache until the computer is restarted. Each dynamic ARP cache entry has a potential lifetime of 10 minutes. New entries added to the cache are timestamped. If an entry is not reused within 2 minutes of being added, it expires and is removed from the ARP cache. If an entry is used, it receives two more minutes of lifetime. If an entry keeps getting used, it receives an additional two minutes of lifetime up to a maximum lifetime of 10 minutes. The Neighbor Cache in Windows 2008: In Windows Server 2008, ARP Cache is now known as Neighbor Cache. In the new Windows 2008 TCP/IP stack implementation, hosts create the neighbor cache entries when there is no matching entry in the neighbor cache. ARP cache entry for IPv4 is an example of a neighbor cache entry. After the entry is successfully created in the neighbor cache, the entry may change to the "Reachable" state if the entry meets certain conditions. If the entry is in the "Reachable" state, Windows 2008 TCP/IP hosts do not send ARP requests to the network. Therefore, Windows 2008 TCP/IP hosts use the information in the cache. If an entry is not used, and it stays in the "Reachable" state for longer than its "Reachable Time" value, the entry changes to the "Stale" state. If an entry is in the "Stale" state, the Windows 2008 TCP/IP host must send an ARP request to reach that destination. The "Reachable Time" value is calculated as follows: Reachable Time = BaseReachable Time × (A random value between MIN_RANDOM_FACTOR and MAX_RANDOM_FACTOR) RFC provides the following calculated results. BaseReachable Time 30,000 milliseconds (ms) MIN_RANDOM_FACTOR 0.5 MAX_RANDOM_FACTOR 1.5 Therefore, the "Reachable Time" value is somewhere between 15 seconds (30 × 0.5 seconds) and 45 seconds (30 × 1.5 seconds). If an entry is not used for a time between 15 to 45 seconds, it changes to the "Stale" state. Then, the host must send an ARP Request for IPV4 to the network when any IP datagram is sent to that destination. Hope these information would helpful. Best Regards, James

Hi James, Thanks for replying, this has definitely shed some light on what I was looking for. I'll try and do some testing and see how it goes from there. /Ulrich