More than 80k index failures

Graylog Central (peer support)
1

Hi guys,

i’m having plenty of index failures like

{"type":"es_rejected_execution_exception","reason":"rejected execution of processing of [1837477][indices:data/write/bulk[s][p]]: request: BulkShardRequest [[firepower_555][1]] containing [259] requests, target allocation id: 3GXnrmVsTh-1SyvC7u2fxQ, primary term: 1 on EsThreadPoolExecutor[name = data-node-5/write, queue capacity = 200, org.elasticsearch.common.util.concurrent.EsThreadPoolExecutor@3a422700[Running, pool size = 10, active threads = 10, queued tasks = 200, completed tasks = 832379]]"}

questions:

  1. do i loose the logs or they are getting still written to elasticsearch?
  2. any indea what i can do to prevent this thing from happening?

Thanks,
Marius.

2
  1. yes you loose messages (IMHO) because it is rejected
  2. it looks like your elasticsearch is having to much to work (no threads left)
3

how do i make room for more threads for elasticsearch?

4

add move CPUs to elasticsearch …

( just a quick search revealed this https://qbox.io/blog/authoritative-guide-elasticsearch-performance-tuning-part-3 )

5

i have 1.5 /data node with 10 cpu cores which are not doing anything max they get to 20-30%…

6

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