Server Redundancies

Answered Question
Dec 11th, 2006
User Badges:

Hi,


I wish to have server redundencies in my network, if one server fails/powered off the other server should seamlessly take over and so on.


Can anyone please tell me a solution/design where this is possible.


Which cisco router / switch would i need.



Regards,

Amol M.

Correct Answer by kamal-learn about 10 years 8 months ago

hi

hi

the FIRST solution that is not related to any network vendor is clustring you buy two servers with the same hardware the same software the same config use the clustering software to make them behaving as a cluster in this scenario if one fails the other takes over.


the SECOND which is cisco solution called SLB (do a lookup for the SLB word),it stands for server load balancing ,that provides load balancing between a series of servers those servers are represented to the network as a virtual server with one ip address so if one fails the switch detects that and forwards

the request to the alive servers in a round robin fashion with respect to the connections already opened(in a case of a file transfer for instance) this feature is supported on 6500 and 4980G at the time i studied for my BCMSN 2005.


the THIRD solution is using NAT or what we call TCP LOAD DISTRIBUTION ,you use here (nat pool in a rotary fashion and ip nat destination...) do a lookup on that subject



HTH

please do rate if it does clarify


  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4.5 (2 ratings)
Loading.
Correct Answer
kamal-learn Tue, 12/12/2006 - 00:05
User Badges:
  • Bronze, 100 points or more

hi

hi

the FIRST solution that is not related to any network vendor is clustring you buy two servers with the same hardware the same software the same config use the clustering software to make them behaving as a cluster in this scenario if one fails the other takes over.


the SECOND which is cisco solution called SLB (do a lookup for the SLB word),it stands for server load balancing ,that provides load balancing between a series of servers those servers are represented to the network as a virtual server with one ip address so if one fails the switch detects that and forwards

the request to the alive servers in a round robin fashion with respect to the connections already opened(in a case of a file transfer for instance) this feature is supported on 6500 and 4980G at the time i studied for my BCMSN 2005.


the THIRD solution is using NAT or what we call TCP LOAD DISTRIBUTION ,you use here (nat pool in a rotary fashion and ip nat destination...) do a lookup on that subject



HTH

please do rate if it does clarify


Actions

This Discussion