0

I am using total 4 nodes 2 in 1a and 2 is in 1b availability zone in singapore region. Right now I am using only 1 seed. Suppose it goes down then how can i recover my cluster.

OR, Suppose I will be using 2 seed nodes and both nodes will go down then how can I recover my my seed nodes.

I am using I class family instances with ssd.

saranjeet singh
  • 868
  • 6
  • 17
  • I got the answer. The seed node designation has no purpose other than bootstrapping the gossip process for new nodes joining the cluster. Seed nodes are not a single point of failure, nor do they have any other special purpose in cluster operations beyond the bootstrapping of nodes. :) – saranjeet singh Apr 01 '14 at 18:37

1 Answers1

0

If you have a single seed node and it goes down, you will have problems to add a replacement node; you'll have to specify another node as the new seed.

You should have at least one seed node in each DC, and ideally two in each DC. Then, when you add nodes there will be a DC-local seed node available to bootstrap the new node into the cluster.

djatnieks
  • 734
  • 3
  • 11