Hi Alain,

Thanks for your response :)



> A replication factor of 3 for a 3 node cluster does not balance the load:
> since you ask for 3 copies of the data (rf=3) on 3 nodes cluster,
> each node will have a copy of the data and you are overloading all nodes.
> May be you should try with a rf = 2 or add nodes to your cluster ?
>

I agree that with a RF=3 and 3 nodes, all nodes should have almost the same
load and i'm fine with that, this is what I want.
My problem is that this isn't the case, one node get a lot of CPU usage
compared to the others :
node 1 : load average = 17
node 2 : load average = 3
node 3 : load average = 3


Funny fact :
If I stop node 1, the two others nodes have the same load as before.
So this seems not to be a problem of "load is big on node1 because all
request goes to this node" : requests are now spread only amongst node 2
and 3 and they are not overwhelm as node 1 was..
Instead, I suspect hardware problem on node 1..



> "working" ? How can I list connections on a node ?
>>
>> For a 3.x (I think also 2.x) you can trace  requests at the query level
> with enableTracing() method.
> something like : (uncomment the line with .enableTracing() )
>
> session.execute( boundedInsertEventStatement.bind( aggregateId,
> aggregateType, eventType, payload )
>         .setConsistencyLevel(ConsistencyLevel.ONE)
>         //.enableTracing()
>         );
> see the doc for other classes and tracing or consistency options,
> and have a look at nodetool settraceprobability if you cannot change the
> code
>
> The queries and query plans appear in the system_traces.sessions and
> system_traces.events tables. It can be very verbose for query plans (events
> table), may be you should truncate the sessions and events tables before
> running your load (on 3.x tables are  truncated on startup)
>
>

I'll take a look, this might be verbose because of the OPS rate (4k) but it
can give me info

Thanks again



> Regards,
>> Bastien
>>
>>
> HTH,
> --
> best,
> Alain
>

Reply via email to