Do not wait for request data after decode failure #6941
Merged
+71
−2
Conversation
This file contains 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
When there is a failure during decoding of header values, netty will consume the entire input and not forward it. However, if we expect content data e.g. due to a Content-Length header, RoutingInBoundHandler would wait indefinitely for that data before sending the error response. This patch changes HttpStreamsServerHandler to treat a decode failure request as if it had no body. This way a StreamedHttpRequest is never created. I also considered just bypassing the StreamedHttpRequest wait in RoutingInBoundHandler, but that risks unexpected behavior from the StreamedHttpRequest publisher never finishing (see #6935), so I tried to avoid StreamedHttpRequest entirely. There is also a change to force the connection to close, to prevent possible downstream errors from connection reuse. Fixes #6925
jameskleeh
approved these changes
Feb 21, 2022
graemerocher
approved these changes
Feb 21, 2022
yawkat
added a commit
that referenced
this pull request
Mar 21, 2022
sdelamo
pushed a commit
that referenced
this pull request
Mar 24, 2022
* Add netty buffer leak detection for all tests in http-server-netty * fix HttpPipeliningSpec leak * fix leak for cancelled FormDataHttpContentProcessor * touch content in NettyHttpRequest to make debugging easier * stop forwarding messages in HandlerPublisher immediately, because downstream CompletionAwareSubscriber might discard and leak them * Don't use ByteBuffer in TextStreamCodec because we can't release it in http-server * Better buffer ownership for form data / routing handlers * log discarded errors in CompletionAwareSubscriber * don't add more data to already-destroyed http requests * leak detection for NettyCompletedFileUpload * discard populated parameters when route match fails * discard uploads in test * fix leak introduced by #6941 * Add getInitialHint support to BufferLeakDetection Co-authored-by: jameskleeh <james.kleeh@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
When there is a failure during decoding of header values, netty will consume the entire input and not forward it. However, if we expect content data e.g. due to a Content-Length header, RoutingInBoundHandler would wait indefinitely for that data before sending the error response.
This patch changes HttpStreamsServerHandler to treat a decode failure request as if it had no body. This way a StreamedHttpRequest is never created. I also considered just bypassing the StreamedHttpRequest wait in RoutingInBoundHandler, but that risks unexpected behavior from the StreamedHttpRequest publisher never finishing (see #6935), so I tried to avoid StreamedHttpRequest entirely.
There is also a change to force the connection to close, to prevent possible downstream errors from connection reuse.
Fixes #6925