Flink standalone cluster load average is too high.

[weilongxing]

I am using flink 1.5.2 in standalone mode. The machine is 8 cores with 32GB memory. I set slots to 4 or 8. I set most jobs parallelism to 1. And each machine has jobs less than the slot number.

The load average is too high as the picture shows below. Each line presents load average 1 on one machine. CPU usage on each machine is 100%. 

All jobs are flink stream SQL. 

  

Can anybody help?

Thanks.

[Congxian Qiu]

Hi

In my opinion, you can use jstack to find what are the processes doing, If you still have problem after that, you could share the code or some demo code you are running.

weilongxing <[hidden email]> 于2018年9月29日周六 下午1:43写道：

I am using flink 1.5.2 in standalone mode. The machine is 8 cores with 32GB memory. I set slots to 4 or 8. I set most jobs parallelism to 1. And each machine has jobs less than the slot number.

The load average is too high as the picture shows below. Each line presents load average 1 on one machine. CPU usage on each machine is 100%. 

All jobs are flink stream SQL. 

  

Can anybody help?

Thanks.

--

[weilongxing]

It works. Thanks.

在 2018年9月29日，下午2:14，Congxian Qiu <[hidden email]> 写道：

Hi

In my opinion, you can use jstack to find what are the processes doing, If you still have problem after that, you could share the code or some demo code you are running.

weilongxing <[hidden email]> 于2018年9月29日周六 下午1:43写道：

I am using flink 1.5.2 in standalone mode. The machine is 8 cores with 32GB memory. I set slots to 4 or 8. I set most jobs parallelism to 1. And each machine has jobs less than the slot number.

The load average is too high as the picture shows below. Each line presents load average 1 on one machine. CPU usage on each machine is 100%. 

All jobs are flink stream SQL. 

  

Can anybody help?

Thanks.

--
<PastedGraphic-16.png><PastedGraphic-16.png>