This could be a browser issue actually. I tried running your query directly
via a Cypher ExecutionEngine, and no matter how many times I tried, it
always gave me the 860 results back.
On Tue, Jun 24, 2014 at 10:20 AM, shai katzir wrote:
> It's getting a little weird.
> I run the next query :
>
Yes, you can still use legacy indexes for fulltext, numeric indexes and
relationship indexes, but please note that writing to those is still a
manual process (at least for numeric indexes).
Otherwise you can configure node_auto_index or relationship_auto_index to
be semi-automatic.
See: http://jex
You're welcome.
I created index in neo4j shell. It works. But the command schema --await
must be schema await -l Item -p itemId.
Another question is: as I understand, until now neo4j doesn't support
schema index for relationship index, numeric range...
Could i make legacy index in neo4j shell
Thank you Michael. I have read/watched most of the resources you linked to (all
very useful), so I presume I have a good enough understanding of the language.
I guess I wanted confirmation of my understanding and assumptions in practice.
Regarding the large vs. small write queries, I don’t have
Hi,
there is a number of resources for optimizing Cypher queries.
We had the webinar: http://watch.neo4j.org/video/84900121
and blog posts from Wes: http://wes.skeweredrook.com/cypher/ and Mark
http://www.markhneedham.com/blog/category/databases-2/neo4j/
and Luanne
http://thought-bytes.blogspot.d
Afaik you have to setup a load balancer of some kind (or you use smart
clients).
That can be something like dns-round-robin if you're only reading, but if
you also write to the cluster, you'll have to look into consistent routing
(writing to master).
See here: http://docs.neo4j.org/chunked/milest
Hi,
I have a HA cluster containing 2 Neo4j instances running on VMs that are
included in a Windows Failover Cluster. Both VMs run Windows Server 2012 R2.
My question is: do I still have to set up a load balancer for the HA
Cluster or is the load balancer running on cluster sufficient?
Thanks,
For mostly write-based queries, does the query engine perform better with
one medium/large Cypher query or several smaller queries in succession? I
am using the HTTP transactional endpoint and sending all statements in one
request, so latency is constant.
Also, is there a general resource/listi
If we do from our code then how to stop server properly.
And if
We unlock the database by code Please share it.
On Saturday, April 26, 2014 7:46:47 PM UTC+5:30, Michael Hunger wrote:
>
> I think you don't shut down the database correctly at restart so it is
> still running in the background
>
>
You could probably call prettify on them, which uses a simpler but similar
parser.
Otherwise right now there is no way since the parse methods were removed.
Perhaps it makes sense to raise a github issue for a "validate()" or
compile() method on execution-engine.
Which would validate the query an
Dear all,
How I can checking syntax error of Cypher queries before execution them on
Graphdb ?
Thanks all
--
You received this message because you are subscribed to the Google Groups
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to neo4j+unsu
See: http://jexp.de/blog/2014/06/styling-neo4j-server-visualisation/
On Mon, Jun 30, 2014 at 8:42 AM, Pavan Kumar
wrote:
> I got it...
>
>
> On Mon, Jun 30, 2014 at 11:49 AM, Pavan Kumar
> wrote:
>
>> Hello Everybody,
>> Can i make a name available inside the node instead of ID's
>> Please see
12 matches
Mail list logo
