Hi!
I have set up a ElasticSearch cluster with NLB (Network load balancing). It
all works very well, but I have a question about when/if a node goes down.
ElasticSearch, should then choose another node as master, the most eligible
on? Right? But what happens for me is this.It is to seperate scenarious
that I can't wrap my head around.
-
Ok, I shutdown the master node. Then the virtual ip stops responding and
I get 404. It seems like NLB is still trying to direct all the traffic to
the master node which I shutdown. The host is still up, only the
elasticsearch application is shutdown. So how do I make NLB try the other
host? Which infact is up and running. -
I shutdown one node, not the master. The virtual ip setup with NLB still
responds and I can access the head plugin via it. So far so good. But, if I
try to access my site (asp.net mvc using NEST) using the cluster IP get
this.
No connection could be made because the target machine actively refused it
[VIRTUAL IP]
.Which is very strange, since the virtual ip is up and running and I can
access everything via the head plugin.
So the questions are, how do I tell NLB to try the other host when
elasticsearch master node is down, but the server is still up, and why does
my cluster become inaccessible when I shut down one of the other nodes in
it. Am I missing some configuration?
Thanks!
//Tommy
--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.