cancel
Showing results for 
Search instead for 
Did you mean: 

Relationship between eager operators and node names when loading csv

asharpe
Node Clone

Hi all, does anyone know why the name of a node I create would be the difference between whether my query has an eager operator or not? For more context, I am attempting to create a graph from a csv using the LOAD CSV command and some MERGE statements. My first attempt worked and produced the graph I wanted just fine. However, when I suddenly changed the name of a node in a single MERGE statement, I got the following error which indeed caused problems with:

"The execution plan for this query contains the Eager operator, which forces all dependent data to be materialized in main memory before proceeding.
Using LOAD CSV with a large data set in a query where the execution plan contains the Eager operator could potentially consume a lot of memory and is likely to not perform well. See the Neo4j Manual entry on the Eager operator for more information and hints on how problems could be avoided."

I thought this error would be caused by maybe me adding a new MERGE statement that pulls aggregated data or something like that, so if any can share some ideas as to why this is happening that'd be great

1 ACCEPTED SOLUTION

asharpe
Node Clone

Ah, so of course I find the bug only after I make a post about the issue LOL (classic amirite?).

So the issue was that when I changed the name from "oldName" to "newName", I forgot to rename all the subsequent "oldName" nodes. As such, the erred query created nodes with both labels "oldName" and "newName". So I just renamed everything I was supposed to and now it works.

I hope this helps someone who encounters the same issue 🙂

View solution in original post

1 REPLY 1

asharpe
Node Clone

Ah, so of course I find the bug only after I make a post about the issue LOL (classic amirite?).

So the issue was that when I changed the name from "oldName" to "newName", I forgot to rename all the subsequent "oldName" nodes. As such, the erred query created nodes with both labels "oldName" and "newName". So I just renamed everything I was supposed to and now it works.

I hope this helps someone who encounters the same issue 🙂

Nodes 2022
Nodes
NODES 2022, Neo4j Online Education Summit

On November 16 and 17 for 24 hours across all timezones, you’ll learn about best practices for beginners and experts alike.