Neo4j Enterprise 4.4.7
Causal cluster with 3 nodes (1 leader + 2 followers)
Data: 10M+ Nodes and 10M+ Rels
It seems that our cluster has entered a corrupted state. We have a daily process in place that updates our database with new data.
Since last Saturday it started failing.
After investigating the issue it looks like the database is keeping a reference to a deleted node (??)
This seems to happen only when executed from the shell script batched query. If I execute the CREATE query manually it won't fail
Some other SO questions that I think are related to this one:
Neo4j 2.0 unique constraint error "node already exists" when it does not
