BlockingBatchedESOutput

Hi All,

I’m trying tune my graylog/es setup, but even though i read everything about it I cannot come across the exact tuning I need to make. I cannot get more than 4K on the following metric

org.graylog2.outputs.BlockingBatchedESOutput.batchSize

I’m testing with:

output_batch_size = 7000
processbuffer_processors = 10
outputbuffer_processors = 4
inputbuffer_processors = 3

No number on output_batch_size seems to make any effect.
What else should I check?

Regards,
Mitch

Doing a few more testings I came across the following, only the Master Graylog server could not ramp up to more than 4000, the others two graylog nodes have 8K

org.graylog2.outputs.BlockingBatchedESOutput.batchSize

Even though all three servers have the same config

output_batch_size = 10000
processbuffer_processors = 10
outputbuffer_processors = 4
inputbuffer_processors = 3

Any Clue?

the question is, how much cores does your Graylog server has, how many Graylog servers you have in total, how many elasticsearch servers you have?

Did you monitor the threads in elasticsearch?

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.