swithcover between failed n/w to alive n/w

[Singar]

Hi,

1) What are the nowadays trends to switchover between two redundant links ? I know about VRRP and HSRP (cisco's).

2) how does the traffics destined towards backup network will come back to the primary network link when the primary n/w is up.

Thanks
singar

[U.SivaKumar_2 · Accepted solution]

Hi,

Dynamic routing protocols will very well help you to achieve redundancy. I think OSPF is best of it's kind for parallel link load balancing and failover.

you can achieve redundancy using static routes
with different preferences pointing to same destination in cisco router or a unix host to a certain extent.

The two types of failover with network devices.
stateful failover and stateless failover.
In stateful , after the failover to backup node
the backup node monitor the state of the primary node . When primary node comes online , backup node will suicide.

In stateless failover , backup node is allowed
to be active even if the primary comes back online.

During failover of primary node to backup node , the system's MAC addresses change and the system sends out gratuitous ARP requests to flush the old MAC addresses from the ARP caches on attached IP devices, and update the caches with the new MAC addresses.

I would like to quote cisco's technical papers
about cisco PIX failover

"The standby unit uses the failover IP address and the MAC addresses of the Secondary unit. If a switchover occurs, the units swap the IP address and MAC addresses they are using so as to replace each other's presence on the network. This action is invisible to the network. The IP to MAC address relationships remain exactly the same, so no ARP tables in the network need to time out or be changed. No other piece of network equipment needs to know about the redundancy or that a switchover occurred. Note that the system IP and the failover IP addresses must be on the same subnet, so there may not be a router between the two units."

Also about IP failover in switched networks.

"There are two issues to address in switched environments. First, the switch needs to learn that a particular MAC address has moved from one port to another. Each unit (unless the unit is failed) transmits a series of failover messages on each interface using its new MAC and IP addresses, allowing the switch to update its internal MAC tables. We strongly recommend that customers enable portfast on all switch ports that connect to PIX interfaces. In addition, channeling and trunking should be disabled on these ports. Thus, if the PIX's interface goes down during failover, the switch will not have to wait 30 seconds while the port is transitioned from a listening to learning to forwarding state.

This blocking of network traffic brings us to the second issue. If the "hello" packets being sent by failover don't get forwarded, each unit thinks something is wrong and begins testing its interfaces. As stated earlier, this results in one unit failing because the test results are "if I'm okay, then you must be failed." To get around this problem, any time a switchover takes place, the units enter a "waiting" state. In this state network traffic is free to flow through the active unit, but failover waits for two "hello" messages to be received before monitoring interfaces again. This allows the switch to enter a blocking state without disrupting failover. Once the second "hello" message is heard, failover resumes normal monitoring of its interfaces.

For PIX Software version 5.2 and later, when a device changes state from standby to active, or from active to standby, a gratuitous ARP is set to each network interface to rebroadcast the new IP and MAC addresses."

Hope you are satisfied with this level of technology.

regards,
U.SivaKumar