Skip to content

Conversation

@szetszwo
Copy link
Contributor

@szetszwo szetszwo commented Nov 25, 2025

What changes were proposed in this pull request?

In HDDS-13973, sendCommandReadBlock(..) creates a new gRPC stream per read(..) call but not a single stream per block. If the application code uses small (e.g. 64KB) buffer, then it is going to create a lot of gRPC streams.

It should keep using the same gRPC connection for streaming a block.

Note that the code is merged from #6613 up to 726f8e4 by @sodonnel

What is the link to the Apache JIRA

HDDS-13974

How was this patch tested?

Updating existing tests

chungen0126 and others added 29 commits November 25, 2025 13:14
… offsets to be calculated correctly on the client
…w checksum offsets to be calculated correctly on the client"

This reverts commit 1f90a0a.
Copy link
Contributor Author

@szetszwo szetszwo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1 the change looks good.

@szetszwo szetszwo merged commit 58d1b65 into apache:master Nov 26, 2025
45 checks passed
@chungen0126
Copy link
Contributor

Thanks @szetszwo for the patch. A commit is missing here (#6613 (comment)). It's about releasing resources properly. Created a jira ticket for this (https://issues.apache.org/jira/browse/HDDS-14013).

@szetszwo
Copy link
Contributor Author

szetszwo commented Nov 26, 2025

@chungen0126 , I have merged only up to 726f8e4 as mentioned in #6613 (comment)

For further works, please wait for HDDS-13975. It is bigger change that we should merge first. Otherwise, it will keep getting conflicts.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants