Repository navigation
Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
This branch has not been deployed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
TL;DR
This change sets the
EPOLLOUTregistration after a successfulsend()to match whether data is still pending, instead of only clearing it once the buffer is empty.Summary
A TCP transfer from guest to host can lose its tail when the host socket accepts only part of the pending data.
flush_to_host()registersEPOLLOUTonly whensend()fails withEAGAIN. Whensend()succeeds but writes less than what is buffered, the remainder stays inrx_bufferwith no writable event armed. It is only flushed if another guest packet arrives and triggers the next call. If that partial write happens on the guest's last packets, nothing flushes it. The guest has already been ACKed for all of it, so from inside the jail the request looks fully sent and the client waits for a response that never comes.Before
sequenceDiagram participant G as Guest participant N as nstun participant H as Host socket participant S as Server G->>N: last data segment N->>G: ACK for all of it N->>H: send() H-->>N: short write H->>S: first part only Note over N: tail kept in rx_buffer,<br/>EPOLLOUT not armed Note over H: becomes writable,<br/>nobody asked to be told Note over G: nothing left to send,<br/>waits for response Note over S: waits for rest of bodyAfter
sequenceDiagram participant G as Guest participant N as nstun participant H as Host socket participant S as Server G->>N: last data segment N->>G: ACK for all of it N->>H: send() H-->>N: short write H->>S: first part rect rgba(46, 160, 67, 0.15) Note over N: tail kept in rx_buffer,<br/>EPOLLOUT armed H-->>N: EPOLLOUT (writable) N->>H: send() the tail H->>S: rest of body Note over N: rx_buffer empty,<br/>EPOLLOUT disarmed end S->>H: response H->>N: response N->>G: responseI have not added a test. I'm happy to add one if there is a preferred way to drive the host socket into a short write.