Too few shards causing transport errors

Hello,

I am running elastcisearch 1.4.3 and injecting into my indices using
netflix suro.

I am creating a template that sets the number of shards per node per index
to be exactly one (2 shards, 1 replica with 4 data nodes), therefore only 2
of the 4 nodes have primary shards.

For some reason this setup causes a NoNodeAvailableException when I am
indexing. As soon as I bring the shard count up to at least 4 shards with 1
replica the problem is solved.

Our goal was to have as few shards per node as possible in order to
maximize performance we can't seem to get it to work with only 1 shard per
node (per index)

Any thoughts on what might be causing this?

Some info on our setup:
We have 4 m1.xlarge instances in AWS (15GB Ram, 4core CPUs, 1.6TB HDD)
7.5 GB Memory dedicated to ES
mlockall: true
Index Refresh interval: 30s
Mulicast Discovery: Disabled

--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/5afdd9ca-cde5-464b-96a9-293e884568a9%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.