Hi David, Thank you very much for your response. I can see now what caused the deadlock.
However, I'm not really sure how I can solve this problem efficiently - as you mentioned, your proposed solution will effectively serialize my transactions. Unfortunately, it'd be very difficult for me to split each transaction into two in the application I'm currently working on (the code I posted in the first message was just a sample code I wrote to explain the problem more clearly) since there are many transactions which are "higher up the chain" and are not really nearby the node factory, which is where I want to use your lock-grabbing technique. I thought about creating a nested transaction just for the update for the factory node's "idseq" property, after which I'd commit (and thereby release the RW lock I grabbed when removing the non-existing property), but I now realize it can't work because nested transactions aren't "pure nested" and so the lock would remain until the parent transaction commits... Is there any way I could possibly release the lock I have for the node without committing the entire transaction? And since I presume such a way doesn't exist, do you happen to know of any other possible solutions for my problem? Thanks again, Ran. -- View this message in context: http://neo4j-community-discussions.438527.n3.nabble.com/Node-Id-generation-deadlock-tp3473118p3474747.html Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. _______________________________________________ Neo4j mailing list [email protected] https://lists.neo4j.org/mailman/listinfo/user
