Inconsistent query result due to communication failure between nodes

Hi All,

We are using transport client to connect to ES. While querying on ES, we
are getting inconsistent results:

We are using cluster of two nodes.

For example:
When we are firing query first time, the number of hits is 50 while
executing the same query second time returns 100 hits. This is happening
because the transport client execute the request in round robin. So, first
query goes to NodeA while other one goes to NodeB.

The results are coming inconsistent because nodes are not able to
communicate with each other.

I would like to know the reasons of connection lost / communication failure
between the nodes.

Thanks,
Ankit Jain
iLabs

--
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.

What I've seen in the past about this is:
Network failure between both nodes
Out of memory exception

So shards are not synchronized anymore.

Basically it ends up to raise split brain issue. It might happen when using only 2 nodes.

If you know that NodeA have the right number of docs, stop NodeB, remove in /data dir the bad index. Restart NodeB.
It will recover docs from NodeA.

My 2 cents

--
David Pilato | Technical Advocate | Elasticsearch.com
@dadoonet | @elasticsearchfr | @scrutmydocs

Le 26 juin 2013 à 12:04, Ankit Jain ankitjaincs06@gmail.com a écrit :

Hi All,

We are using transport client to connect to ES. While querying on ES, we are getting inconsistent results:

We are using cluster of two nodes.

For example:
When we are firing query first time, the number of hits is 50 while executing the same query second time returns 100 hits. This is happening because the transport client execute the request in round robin. So, first query goes to NodeA while other one goes to NodeB.

The results are coming inconsistent because nodes are not able to communicate with each other.

I would like to know the reasons of connection lost / communication failure between the nodes.

Thanks,
Ankit Jain
iLabs

--
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.

Hi David,

Thanks for the reply.

~Ankit

On Wednesday, 26 June 2013 10:13:38 UTC, David Pilato wrote:

What I've seen in the past about this is:
Network failure between both nodes
Out of memory exception

So shards are not synchronized anymore.

Basically it ends up to raise split brain issue. It might happen when
using only 2 nodes.

If you know that NodeA have the right number of docs, stop NodeB, remove
in /data dir the bad index. Restart NodeB.
It will recover docs from NodeA.

My 2 cents

--
David Pilato | Technical Advocate | Elasticsearch.com
@dadoonet https://twitter.com/dadoonet | @elasticsearchfrhttps://twitter.com/elasticsearchfr
| @scrutmydocs https://twitter.com/scrutmydocs

Le 26 juin 2013 à 12:04, Ankit Jain <ankitj...@gmail.com <javascript:>> a
écrit :

Hi All,

We are using transport client to connect to ES. While querying on ES, we
are getting inconsistent results:

We are using cluster of two nodes.

For example:
When we are firing query first time, the number of hits is 50 while
executing the same query second time returns 100 hits. This is happening
because the transport client execute the request in round robin. So, first
query goes to NodeA while other one goes to NodeB.

The results are coming inconsistent because nodes are not able to
communicate with each other.

I would like to know the reasons of connection lost / communication
failure between the nodes.

Thanks,
Ankit Jain
iLabs

--
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 elasticsearc...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.

--
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.