Boolean query does not return expected data in Elasticsearch - elasticsearch

Question

I have the following document in Elasticsearch as reported by Kibana:
{"deviceId":"C1976429369BFE063ED8B3409DB7C7E7D87196D9","appId":"DisneyDigitalBooks.PlanesAdventureAlbum","ostype":"iOS"}
Why the following query does not return success?
[root#myvm elasticsearch-1.0.0]# curl -XGET 'http://localhost:9200/unique_app_install/_search?pretty=1' -d '
{
"query" : {
"bool" : {
"must" : [ {
"term" : {
"deviceId" : "C1976429369BFE063ED8B3409DB7C7E7D87196D9"
}
}, {
"term" : {
"appId" : "DisneyDigitalBooks.PlanesAdventureAlbum"
}
}, {
"term" : {
"ostype" : "iOS"
}
} ]
}
}
}'
Here is the response from Elasticsearch:
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 5,
"successful" : 5,
"failed" : 0
},
"hits" : {
"total" : 0,
"max_score" : null,
"hits" : [ ]
}
}
As a side question, is this the fastest way to query the data in my case?
Thx in advance.
UPDATE:
Could it be related to the fact that I used the following mapping for this index?
curl -XPOST localhost:9200/unique_app_install -d '{
"settings" : {
"number_of_shards" : 5
},
"mappings" : {
"sdk_sync" : {
"properties" : {
"deviceId" : { "type" : "string" , "index": "not_analyzed"},
"appId" : { "type" : "string" , "index": "not_analyzed"},
"ostype" : { "type" : "string" , "index": "not_analyzed"}
}
}
}
}'

Answer 1

Check if the type of your document was right while inserting: sdk_sync.
I have used your items and for me it works. Using the following curl request give the right response for me:
curl -XPOST localhost:9200/unique_app_install/sdk_sync/1 -d '{
"settings" : {
"number_of_shards" : 5
},
"mappings" : {
"sdk_sync" : {
"properties" : {
"deviceId" : { "type" : "string" , "index": "not_analyzed"},
"appId" : { "type" : "string" , "index": "not_analyzed"},
"ostype" : { "type" : "string" , "index": "not_analyzed"}
}
}
}
}'
curl -XPOST localhost:9200/unique_app_install/sdk_sync/1 -d '{
"deviceId":"C1976429369BFE063ED8B3409DB7C7E7D87196D9",
"appId":"DisneyDigitalBooks.PlanesAdventureAlbum",
"ostype":"iOS"
}'
curl -XGET 'http://localhost:9200/unique_app_install/_search?pretty=1' -d '
{
"query" : {
"bool" : {
"must" : [ {
"term" : {
"deviceId" : "C1976429369BFE063ED8B3409DB7C7E7D87196D9"
}
}, {
"term" : {
"appId" : "DisneyDigitalBooks.PlanesAdventureAlbum"
}
}, {
"term" : {
"ostype" : "iOS"
}
} ]
}
}
}'

Answer 2

Unless you specify the field NOT to be analyzed, every fields are analyzed by default.
It means that deviceId "C1976429369BFE063ED8B3409DB7C7E7D87196D9" will be indexed as "c1976429369bfe063ed8b3409db7c7e7d87196d9" (lower case).
You have to use term query or term filter with string in LOWER CASE.
That is the reason why you should specify {"index": "not_analyzed"}
for the mapping.