Apoc.import.csv does not import all nodes from csv

Neo4j Graph Platform Procedures & APOC
1

Hello,
I'm currently running Neo4j version 4.0 and installed plugin APOC 4.0.

  • I'm trying to load bunch of csv files using this CALL string. All csv files are ready in neo4j/plugins folder.
    image

  • However when I run this query, I get this result:

    image
    image1444×341 20.5 KB

  • So I try to grep this id in file directory, which shows results:

    image
    image1347×171 11.8 KB

    As you can see, a node with id 17997 is existent and has some relationships.
    Header in processes.csv is id:ID|NodeName:STRING|Characteristics:STRING[]|CommandLine:STRING|Pid:INT|CreationTime:localtime|:LABEL and header in relationships.csv is :START_ID|:END_ID|:TYPE

  • I investigated more and ran only this query:

    image
    image1216×410 31.2 KB

    which loads only nodes from processes.csv. As you can see, it finished with loading 1306 nodes.

  • But actually in processes.csv are 1667 nodes, here is proof (number of lines in file minus one for header):
    image

So it's obvious, that the first problem is caused by some of the nodes not loaded properly.
My question is why are they not loaded and how come that procedure does not return any errors.

Thanks, Petr

2

All the .csv file must reside in imports folder and not plugins folder.

3

Im sorry, I meant neo4j/import folder. All csv files truly are in import folder.
Otherwise all those 1306 would't be imported.

4

Hi @muzos07,
Could you check if you have duplicate nodes in the file and difference is same?

5

I'm forced to close this thread since I'm not currently able to replicate this problem. I accidentally deleted original dataset. I will post here if I encounter this problem again