Search thread pool active and queue not full but rejected happened

[casterQ (caster)]

i found a datanode search thread pool has rejected，but grafana monitor show that moment search threap_pool ’s active(size=49) and queue(size=2000) is not full，I can't understand why ，please help me。thanks a lot...

[warkolm (Mark Walkom)]

Welcome to our community!

What is the output from the _cluster/stats?pretty&human API?

[casterQ (caster)]

@warkolm

_cluster/stats?pretty&human

Thank you for your reply ，i can't get the stats of that moment now and do you want current stats？which part do you need？
And i found that moment an index has both write and search tasks ，Is that a problem？

[warkolm (Mark Walkom)]

We need to see the whole thing please.

[warkolm (Mark Walkom)]

You have a pretty large cluster there!

What is the output from _cat/threadpools?v and hot threads when you get these rejections?

@casterQ

"versions" : [
      "7.10.2"
    ],

You should upgrade, 7.10 is relatively old and latest is 7.14.

[casterQ (caster)]

@casterQ

From grafana i can see active and pool is not full but rejected happened。i'm sorry for i can‘t get hot thread info in that moment。 and is there some thing can cause this problem？

From grafana i can see active and pool is not full but rejected happened。i'm sorry for i can‘t get hot thread info in that moment。 and is there some thing can cause this problem？

[warkolm (Mark Walkom)]

That's hard to say without more info.

[casterQ (caster)]

OK，Thank you for your reply anyway。