Hi,

If immideate data precence in GL is not your case, but you coul do with 
some delay,
the setting "index.refresh_interval: 5s"  (or bigger) can improve the 
performance of GL.

Arie

On Wednesday, September 28, 2016 at 9:29:14 PM UTC+2, Evgueni Gordienko 
wrote:
>
> We have GL server suppliing data from filebeat clients to three node ES 
> server.
> We tuned heap-size of GL jvm (Xmx/Xms are now 16g) and quite a few 
> parameters on server.conf
> such as 
> processbuffer_processors = 15
> outputbuffer_processors = 10
> rring_size = 131072
> inputbuffer_ring_size = 131072
> inputbuffer_processors = 6
>
> In our stress test we did no loose data but we see GL continues to push 
> buffered data for more than 20 minutes after 
> last filebeat uplads data to GL.
>
> Any recommendations to improve performance so we could avoid these 20 
> minutes of delay data delivery to ES from GL?
>
> Thanks,
> Evgueni 
>

-- 
You received this message because you are subscribed to the Google Groups 
"Graylog Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/graylog2/9a5a39e4-a001-4591-a4c9-fb4bfb314d99%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to