[
https://issues.apache.org/jira/browse/HDDS-16208?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18106382#comment-18106382
]
Sergey Soldatov commented on HDDS-16208:
----------------------------------------
The behavior that described here was a part of another task that was never
landed. So, it's not relevant to the master branch.
> Cache out-of-order streaming ReadBlock responses instead of discarding them
> to avoid client poll timeout
> --------------------------------------------------------------------------------------------------------
>
> Key: HDDS-16208
> URL: https://issues.apache.org/jira/browse/HDDS-16208
> Project: Apache Ozone
> Issue Type: Sub-task
> Components: Ozone Client
> Affects Versions: 2.3.0
> Reporter: Sergey Soldatov
> Assignee: Sergey Soldatov
> Priority: Major
> Labels: pull-request-available
>
> Streaming read responses can arrive out of order because the datanode serves
> pre-read requests on concurrent chunk reader threads. When a response arrives
> ahead of the current read position, the client drops it and never re-requests
> that range, so the read eventually fails with a poll timeout (default 10s).
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
