0
votes

I've imported the data into the database using batch-import. Starting the neo4j server, I can se at the dashboard that all the nodes and relationships have been imported. When I try to execute the cypher query, however, it returns the message Invalid encoding: '12'

neo4j-sh (0)$ start a = node(1) return a
==> Invalid encoding '12'
neo4j-sh (0)$ 

In the past I have already successfully imported the data using batch-import and now, following the same steps I can't figure out what causes the error.

The file nodes.csv contains the data in the following format:

node_id msisdn:string:users
1   000000F8BE951D6DE6480F4AFDFB670C553E47C0
2   0000021449360C1A398ED9A18800B2B13AA098A4
3   00000DABDE4C555FC82F7D534835247B94873C2C
4   00001BE4128DB41729365A41D3AC1D019E5ED8A6
5   00002506A1BC28F5DAE937703106CE6B39B857A0
6   00002781A2ADA816CDB0D138146BD63323CCDAB2

and relationships.csv the following:

calling_party   called_party    connection_type:string:connection   consecutive_day:int:connection  called_network_id:int:connection    calls_count:int duration_sum:int
6   209339  CALLED  3   9800    1   532
6   667602  CALLED  25  9800    1   31
6   917611  CALLED  54  9811    2   17
6   1057687 CALLED  14  9800    1   29
6   1070735 CALLED  41  9800    1   285
6   1070735 CALLED  43  9800    1   18
6   2106202 CALLED  29  9802    1   26
6   2106202 CALLED  0   9802    1   10

The settings in batch.properties are configured as following:

use_memory_mapped_buffers=true
neostore.nodestore.db.mapped_memory=100M
neostore.relationshipstore.db.mapped_memory=500M
neostore.propertystore.db.mapped_memory=4G
neostore.propertystore.db.strings.mapped_memory=200M
neostore.propertystore.db.arrays.mapped_memory=0M
neostore.propertystore.db.index.keys.mapped_memory=15M
neostore.propertystore.db.index.mapped_memory=15M
batch_import.node_index.users=exact
batch_import.relationship_index.connection=exact

I am not sure whether the wrong format might be related to the fact that I prepared the csv files on Windows machine while the neo4j and batch-import is running on Linux machine.

Anyway I would be really thankful if anyone could help to solve the problem.

1

1 Answers

0
votes

Perhaps a \r line-ending? I fixed that yesterday, perhaps you want to re-import your data with the latest build?

Or clean the line ending before importing with dos2unix or tr -d '\r' <input >output