Open
Conversation
* Keeping track of retained buffers, always releasing them after processing * Removed logging, added cl * Removed system.out.println from test * Try to keep track of as few ByteBufs as possible by removing the reference to already freed ones while reading from an open channel. * alternate approach to buffer release * unit tests for edge cases * CL --------- Co-authored-by: Florian Petersen <188503754+fpetersen-gl@users.noreply.github.com> (cherry picked from commit 41f67ba)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Note: This is a backport of ##25120 to
7.0.Resolves #24336
This PR adds some unit tests to #24525 and proposes a simpler memory management approach to handling issues due to incomplete Syslog frames.
Details
SimpleChannelInboundHandler(boolean autoRelease) — The constructor accepts a boolean to control auto-release behavior. When false, the handler does NOT release the message after channelRead0() returns. This transfers buffer ownership entirely to whoever channelRead0() passes it to. Since ByteToMessageDecoder.channelRead() already manages buffer lifecycle correctly (its cumulator absorbs or copies-and-releases the input), there's no need for external tracking at all.
When a Netty handler needs to pass a buffer to a downstream ByteToMessageDecoder that isn't in the pipeline, the cleanest pattern is to disable SimpleChannelInboundHandler's auto-release via super(false) rather than fighting it with retain()/tracking. The downstream decoder already manages buffer lifecycle through its cumulator. The only responsibility left is forwarding channelInactive so the decoder releases its internal cumulation on connection close.