getTotalHits isn't the same on the same search

Elastic Stack Elasticsearch
1

On few sequential search requests getTotalHits returns different values.
Why?
Of course, there wasn't any indexing between requests.More over, there
wasn't any indexing since ES start :slight_smile:

--
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.

2

I've seen that in the past on a cluster with 2 nodes having shards with not the same number of docs due to a split brain issue.
Do you have one node or many?

Check number of docs per shard and look if there is a difference between primary and replicas.

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

Le 23 juil. 2013 à 12:19, Andrew Gaydenko andrew.gaydenko@gmail.com a écrit :

On few sequential search requests getTotalHits returns different values. Why?
Of course, there wasn't any indexing between requests.More over, there wasn't any indexing since ES start :slight_smile:

--
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.

3

On Tuesday, July 23, 2013 2:31:44 PM UTC+4, David Pilato wrote:

I've seen that in the past on a cluster with 2 nodes having shards with
not the same number of docs due to a split brain issue.
Do you have one node or many?

Check number of docs per shard and look if there is a difference between
primary and replicas.

I play with the only node with

index.number_of_shards: 1
index.number_of_replicas: 0

in config.

--
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.

4

That should not happen with the same exact query with no indexing / refreshing in the mean time.

What does your query look like?

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

Le 23 juil. 2013 à 12:44, Andrew Gaydenko andrew.gaydenko@gmail.com a écrit :

On Tuesday, July 23, 2013 2:31:44 PM UTC+4, David Pilato wrote:
I've seen that in the past on a cluster with 2 nodes having shards with not the same number of docs due to a split brain issue.
Do you have one node or many?

Check number of docs per shard and look if there is a difference between primary and replicas.

I play with the only node with

index.number_of_shards: 1
index.number_of_replicas: 0

in config.

--
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.

5

On Tuesday, July 23, 2013 2:52:59 PM UTC+4, David Pilato wrote:

That should not happen with the same exact query with no indexing /
refreshing in the mean time.

You are absolutely right! It is my fault - the reason is in my code.
Please, forgive me.

--
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.