[
https://issues.apache.org/jira/browse/HDFS-12163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16093842#comment-16093842
]
Tsz Wo Nicholas Sze commented on HDFS-12163:
--------------------------------------------
1-datanode tests
- Ratis-gRPC with 1 datanode
| (GRPC,1) | init | 6 |
| (GRPC,1) | MiniOzoneCluster | 404 |
| (GRPC,1) | createRatisCluster | 413 |
| (GRPC,1) | shutdown | 65 |
| (GRPC,1) | sleep | 38 |
- Ratis-Netty with 1 datanode
| (NETTY,1) | init | 6 |
| (NETTY,1) | MiniOzoneCluster | 404 |
| (NETTY,1) | createRatisCluster | 424 |
| (NETTY,1) | shutdown | 76 |
| (NETTY,1) | sleep | 33 |
- Distributed handlers with 1 datanode
| (distributed,1) | init | 6 |
| (distributed,1) | MiniOzoneCluster | 402 |
| (distributed,1) | shutdown | 54 |
| (distributed,1) | sleep | 27 |
- Local handlers with 1 datanode
| (local,1) | init | 6 |
| (local,1) | MiniOzoneCluster | 402 |
| (local,1) | shutdown | 54 |
| (local,1) | sleep | 27 |
All the results above are from separated runs. Below is the result running
them together.
- All 4 cases above running consecutively.
| (distributed,1) | init | 6 |
| (distributed,1) | MiniOzoneCluster | 402 |
| (distributed,1) | shutdown | 54 |
| (distributed,1) | sleep | 27 |
| (GRPC,1) | init | 27 |
| (GRPC,1) | MiniOzoneCluster | 423 |
| (GRPC,1) | createRatisCluster | 434 |
| (GRPC,1) | shutdown | 85 |
| (GRPC,1) | sleep | 67 |
| (local,1) | init | 67 |
| (local,1) | MiniOzoneCluster | 463 |
| (local,1) | shutdown | 115 |
| (local,1) | sleep | 106 |
| (NETTY,1) | init | 106 |
| (NETTY,1) | MiniOzoneCluster | 507 |
| (NETTY,1) | createRatisCluster | 530 |
| (NETTY,1) | shutdown | 182 |
| (NETTY,1) | sleep | 166 |
As shown in the table, some threads are not terminated after MiniOzoneCluster
shutdown and a 10-second sleep. For example, in (distributed,1), it has 6
threads initially but there are 27 threads after sleep.
> Ozone: MiniOzoneCluster uses 400+ threads
> -----------------------------------------
>
> Key: HDFS-12163
> URL: https://issues.apache.org/jira/browse/HDFS-12163
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: ozone, test
> Reporter: Tsz Wo Nicholas Sze
> Attachments: TestOzoneThreadCount20170719.patch
>
>
> Checked the number of active threads used in MiniOzoneCluster with various
> settings:
> - Local handlers
> - Distributed handlers
> - Ratis-Netty
> - Ratis-gRPC
> The results are similar for all the settings. It uses 400+ threads for an
> 1-datanode MiniOzoneCluster.
> Moreover, there is a thread leak -- a number of the threads do not shutdown
> after the test is finished. Therefore, when tests run consecutively, the
> later tests use more threads.
> Will post the details in comments.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
