[jira] [Commented] (RATIS-569) StatusRuntimeException on the datanode because clients do not shutdown the observer cleanly.

[Tsz Wo Nicholas Sze (Jira)]

    [ 
https://issues.apache.org/jira/browse/RATIS-569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16916148#comment-16916148
 ] 

Tsz Wo Nicholas Sze commented on RATIS-569:
-------------------------------------------

Thanks [~swagle]. 
- The [gRPC 
example|https://github.com/grpc/grpc-java/blob/master/examples/src/main/java/io/grpc/examples/routeguide/RouteGuideClient.java#L63]
 uses 5s timeout.  Let's do the same?
- Let's re-throw the example, if there is any.
{code}
  public void close() throws InterruptedIOException {
    try {
      channel.shutdownNow().awaitTermination(5, TimeUnit.SECONDS);
    } catch (InterruptedException e) {
      throw IOUtils.toInterruptedIOException("Failed to close.", e);
    }
  }
{code}

> StatusRuntimeException on the datanode because clients do not shutdown the 
> observer cleanly.
> --------------------------------------------------------------------------------------------
>
>                 Key: RATIS-569
>                 URL: https://issues.apache.org/jira/browse/RATIS-569
>             Project: Ratis
>          Issue Type: Bug
>          Components: server
>    Affects Versions: 0.3.0
>            Reporter: Mukul Kumar Singh
>            Assignee: Siddharth Wagle
>            Priority: Blocker
>              Labels: ozone
>         Attachments: RATIS-569.01.patch
>
>
> Running TestDataValidate in Ozone leads to StatusRuntimeException on the 
> datanode frequently.
> This causes an unclean shutdown on the stream on the datanode.
> In GrpcClientProtocolClient, shutdownNow should be followed by a 
> awaitTermination to wait for a clean shutdown.



--
This message was sent by Atlassian Jira
(v8.3.2#803003)