Copy a DATABASE on a causal cluster

[b_buchhold]

Assume I have a 4.0 cluster with several "graphs" (cypher: "DATABASE"s). I want to create a copy of one for them for experiments that will make modifications to the graph.

Initial Google searches have turned up neo4-admin copy --from-database=... to-database=... as exactly the thing I need (plus STOP DATABASE before and CREATE/START afterwards).

However, I could not find information on how to do this in a causal cluster. Does it have to be done on every machine or will the cluster sync the data if I do it on one machine? Are there other necessary steps?

[jggomez]

Hi,

If it is offline backup you must stop the instance but if it is online backup, it is not requiered.

How to do this in a causal cluster
https://neo4j.com/docs/operations-manual/current/backup/restoring/#backup-restoring-cluster

Thanks

[danny_oberoi]

Section 15.9 of Neo4j 4.0 documentation clearly explains that the
"The existing database must be stopped before copying from it, and the destination database must not yet exist"
https://neo4j.com/docs/operations-manual/current/tools/copy/

example :
bin/neo4j-admin copy --from-database=neo4j --to-database=persons --delete-nodes-with-labels="Movie"

[b_buchhold · Accepted solution]

Does it have to be done on every machine or will the cluster sync the data if I do it on one machine?

To answer this myself, in case anyone comes across this via Google.
From what I have experienced, neo4j-admin load has to be done on every machine. It is okay to neo4j-admin dump the code on any machine and write to a shared file system. though.