Try this on elasticsearch:

index.refresh_interval: 5s (or a lager interval)
This enhances the throughput capability of ES. You can even so try 10 or 15 
seconds.

This means you haver your results in ES after 5 secs and not one, what is 
the default.


On Wednesday, December 3, 2014 2:24:57 PM UTC+1, Dmitri Stoljarov wrote:
>
> Hi,
>
> Currently i have following graylog server settings:
>
> output_batch_size = 1000
> output_flush_interval = 1
> processbuffer_processors = 20
> outputbuffer_processors = 10
> processor_wait_strategy = blocking
> ring_size = 2048
>
>
> And radio settings:
> processbuffer_processors = 20
> processor_wait_strategy = blocking
> ring_size = 2048
>
>
> Average message rate is 3000-4000 events per second.
> Radio and server node input and output buffer utilization is 0.0%.
>
> But master caches: *0* IMC / *1018* OMC. (1018 has been like this for 
> several hours and not changing anymore).
>
> What are the reasons of increasing output master cache?
>
> What setting can be tweaked?
>
> p.s. i'm running graylog2-radio-0.93.0-SNAPSHOT-20141202145626 + 
> graylog2-server-0.93.0-SNAPSHOT-20141202145626 + 
> graylog2-web-interface-0.93.0-SNAPSHOT-20141202133029.
>

-- 
You received this message because you are subscribed to the Google Groups 
"graylog2" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to