[
https://issues.apache.org/jira/browse/THRIFT-5499?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17477139#comment-17477139
]
Plumet commented on THRIFT-5499:
--------------------------------
I've done a little research and here is my conclusion:
The underlying operation that's used to read from the stream is ReadAsync.
Reading asynchronously does not trigger the ReceiveTimeout.
cfr. The .Net docs
[https://docs.microsoft.com/en-us/dotnet/api/system.net.sockets.socket.receivetimeout?view=net-6.0]
Gets or sets a value that specifies the amount of time after which a
_synchronous_ Receive call will time out.
cfr. The underlying winsock docs
[https://docs.microsoft.com/en-us/windows/win32/winsock/sol-socket-socket-options:]
The timeout, in milliseconds, for _blocking_ receive calls.
Point is, when this value is set in the Thrift client, the client should be
using blocking calls in a separate task/thread.
> Receive timeout not triggered when response exceeds timeout
> -----------------------------------------------------------
>
> Key: THRIFT-5499
> URL: https://issues.apache.org/jira/browse/THRIFT-5499
> Project: Thrift
> Issue Type: Bug
> Components: netstd - Library
> Affects Versions: 0.15.0, 0.16.0
> Reporter: Plumet
> Priority: Major
> Fix For: 0.16.0
>
> Attachments: THRIFT-ReceiveTimeout.zip
>
>
> When the client's timeout is set to valid value and the response exceeds this
> timeout there is no exception thrown.
> I've provided a reproducible testcase[^THRIFT-ReceiveTimeout.zip].
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
