I can sure there is only one master node exists and 16 nodes work as one 
cluster.
I think I know what it happened. I found that when successive requests, 
elasticsearch will execute query by once primary shards once replica shards 
at default configuration.
But what I can not understand is why the primary shads and the replica 
shards given different result at the same time point?
This happened when I index some new documents but not refresh, if I refresh 
the cluster, then the primary shards and the replica shards will give the 
same result.

On Thursday, August 28, 2014 6:41:22 PM UTC+8, Greg Murnane wrote:
>
> This is a symptom that could happen with bad GC events, or with split 
> brain. Can you look at the GC logging output to see how long the stop the 
> world pauses you're seeing are? You can also run a query like " curl -XGET '
> http://localhost:9200/_cluster/state/master_node?local=true' " on each of 
> the nodes to make sure that they agree on which one is the master node.
>
> Look also at wait CPU and disk utilization when you run a query. Unless 
> you have a physical disk for each node on this system, it's likely that 
> there can be IO contention with 16 nodes querying the disks.
>
> If all that looks ok, if you are running replicas, then you can try 
> pulling out a replica and an original, and loading them into an isolated ES 
> node on another system, and query there. It's possible that some of the 
> replicas could be corrupted, and this would allow you to detect that.
>
> -----
>
> Out of curiousity, though, I wonder what the purpose of running so many 
> nodes on a single machine is. ES is very effective at using the entire CPU 
> with only one node, and replicating your heap size 16 times, adding IO 
> contention, and splitting the cache 16 ways all seem like they would hurt 
> performance immensely.
>
> The information transmitted in this email is intended only for the 
> person(s) or entity to which it is addressed and may contain confidential 
> and/or privileged material. Any review, retransmission, dissemination or 
> other use of, or taking of any action in reliance upon, this information by 
> persons or entities other than the intended recipient is prohibited. If you 
> received this email in error, please contact the sender and permanently 
> delete the email from any computer.

-- 
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+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/d86c4ac0-910f-4d00-8276-bec8aed220ad%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to