Hi Boxuan, Issue seems to occurs when edge properties are retrieved: cache has expected size with g.V().outE().id() an not when I do g.V().outE().valueMap(); I am able to reproduce with following

Hi Vinayak, JanusGraph itself has spark as a dependency and the spark versions of the spark jars in the janusgraph/lib folder and of your spark cluster must match. Best wishes, Marc

---------- Forwarded message --------- From: Vinayak Bali <vinayakbali16@...> Date: Wed, 13 Jan 2021, 7:42 pm Subject: Hadoop and Spark not working with Janusgraph To:

On behalf of the JanusGraph Technical Steering Committee, we are excited to join LF AI & Data Foundation! We are now moving mailing lists to a new home under LF AI & Data. Mailing lists hosted on

Hi, Unfortunately, the janusgraph ref docs do not have full documentation for using scylladb, but in order to use scylladb you have to make sure the janusgraph configs are similar to the

Raised it in https://github.com/JanusGraph/janusgraph/issues/2351

When I replaced Cassandra with ScyllaDB it stopped working. Could it be that ScyllaDB is not compatible with either JanusGraph or Elasticsearch?

So it looks like the docker-compose-cql-es.yml works without a problem.

Hi, Can you also try with the tested docker-compose-cql-es.yml from: https://github.com/JanusGraph/janusgraph-docker#example-cassandra-es-with-docker-compose It has subtle differences with the yml

Hi, - yes I can estabilish ws connection to my server with a general ws test client. - the g in javascript prints this value: GraphTraversalSource { graph: Graph {}, traversalStrategies:

Hi Nicolas, Looks interesting. Your configs look fine and I couldn’t reproduce your problem. Could you provide some sample code to reproduce it? Best regards, Boxuan

Hi Timon, If I am understanding correctly, your code logic is like this: start_transaction(); read_vertex(type_v); read_vertex(version_v); // type_v ——hasVersion—> version_v if

Oops, that was a bit too quick, you did mention ConsistencyModifier.LOCK. The linked section reads: " When updating an element that is guarded by a uniqueness constraint,JanusGraph uses the following

Hi Timon You did not state whether you studied https://docs.janusgraph.org/advanced-topics/eventual-consistency/#data-consistency and whether that would fullfil your needs. Best wishes, Marc Op

OK, I see that you run gremlin-npm locally on your server. Are you sure that localhost is exposed in the gremlin server? What happens if you connect from your local machine towards the ip address of

Hi I am lost in the stream of messages. I gather that: you can establish a ws connection to your server with a general ws test client you can run a "const g = traversal().withRemote(remote);" in

Hi all, I have a question about locking data through Janusgraph. Here’s the problem: * We have a vertex with the label “type” with several properties * Types have one edge “hasVersion” to

Can you help me? It's really frustrating and I cannot find a solution. I can even give you ssh access so you can try to find a solution. Anything. I'm desperate.

It wasn't as stated in later posts..

Hi Jan can you elaborate on the method ? as far I've seen the new check is checking the name of the repo, and would expect is to start with string `cassandra` otherwise it would fail and not run. in