10:48:41 WARN org.janusgraph.graphdb.database.idassigner.placement.PartitionIDRange - Individual key range is too small for partition block - result would be empty; hence ignored: KeyRange(left:

thanks a lot 在 2018年8月9日星期四 UTC+8下午10:57:55，Alexandr Porunov写道：

Great to see this release! TinkerPop 3.3 support also implies support for Kerberos authentication towards gremlin server and audit logging of gremlin queries, both of which should be interesting for

I was able to reproduce your issue with the last script. This looking like a bug in the Lucene support. I wasn't able to reproduce the behavior in Elasticsearch. Could you please open an issue?

As an update on this (although it may be a different problem or even a red herring) I managed to create a failing add() traversal that threw an exception message: "Could not commit transaction due to

What are you exactly looking for? If you want to use custom id assignment, then you should set ALLOW_SETTING_VERTEX_ID to true in

** updated script

Hi Thanks I have also tried on the latest version i.e. 0.3.0 it works fine for string. However When I try the same with other data types the issue still persists in janus graph. like when i tried the

do you have a good example with more details? or share me some code ? 在 2018年8月9日星期四 UTC+8上午1:09:12，Alexandr Porunov写道：

Bit of a strange one this... I have a simple graph with a cassandra/elasticsearch back-end that after a few days of light (dev) usage suddenly starts acting strange when adding vertices. I hope the

Hi I am totally new for janus graph so wanted to start a very new project from scratch but unfortunately not getting proper guidance to start. I read the document page itself and did some try with

You can increase your id blocks with the next parameter: ids. block-size=1000000 No, it isn't a bottleneck. Even so, you can create your own strategy of id assignment. For example, you can implement

On Wednesday, August 8, 2018 at 10:45:47 AM UTC+3, jcbms wrote: use mapreduce on yarn for data loading ,I have almost 14 containers ,each container has one janusgraph instance for data loading ,I

I was using 0.3.0.

Rather than mvnrepository, I tend to use search.maven.org. https://search.maven.org/#artifactdetails%7Corg.janusgraph%7Cjanusgraph-es%7C0.3.0%7Cjar I'm not sure why mvnrepository added that test

The data is still stored in the storage backend (Cassandra, HBase). You could run a reindex operation to get ES populated with the data after addressing some of the possible data overload situations

Thanks a lot Daniel. Works like a charm. - Mitesh

use mapreduce on yarn for data loading ,I have almost 14 containers ,each container has one janusgraph instance for data loading ,I meet this problem ,is this mean Id assing is broken or is the

Hi Actually I am not getting the vertex back on the line 28 as in your response after last commit. I mean I am inly able to fetch the vertex using _label property not by externalId, name.I have not

Tks, i see it, i mean that only have a test scope dependency for janusgraph-es on mvnrepository, it's wrong, it need to be rewritten for remove the scope for test, As you say, it's only a test scope