On 2'nd Sept I had upgraded my Neo4j version from 3.3.9(CE) to 3.5.21(CE).
From there onwards my store size has been increasing in increments of 4GB every day. I am pretty sure that I am not adding that much data daily.
For further investigati...
I have a java based microservice which uses Neo4j as the DB. Every request to the service either updates or create a new Node labelled as Device which has ~25 properties.
The DB has close to 4M Device Nodes.
I am using SDN(Spring-Data-Neo4j) s...
Unfortunately, query logging is an enterprise-only feature and I am using a community edition.
Regarding the debug logs, below is the log snippet from the latest occurrence of the issue.
2020-09-02 12:03:51.349+0000 INFO [o.n.k.i.t.l.c.Check...
UPDATE: Have upgrade the neo4j DB version from 3.3.2 to 3.5.21 and the migration is successful but the query still takes time.
The migration rules out any possibility of a query spiking due to a bug in any of the older versions.
My initial question was for the time taken in save method of SDN for Device node. On your follow up question of queries which are being run when Device node is saved to which I mentioned a query(i.e. the cypher for saving User Node) which...
The merge query which I mentioned in my previous reply is itself an individual transaction and not part of any other transaction.
I moved to writing cypher from .save() function to mitigate the huge amount of time taken by .save() func...