wwbmmm opened a new pull request, #3449:
URL: https://github.com/apache/brpc/pull/3449
### What problem does this PR solve?
Issue Number: null
Problem Summary:
The SOFA PBRPC parser (`ParseSofaMessage`) enforces `FLAGS_max_body_size`
only on the frame's `body_size` field. The `meta_size` field, and
therefore the whole frame size, is not bounded. A peer can send a frame
with a large `meta_size` and `body_size = 0`, which passes the current
check, so the connection keeps reading and buffering the declared frame
until the metadata is finally parsed (and rejected). This makes the
configured per-message limit ineffective and can consume much more memory
than expected.
### What is changed and the side effects?
Changed:
- In `ParseSofaMessage`, reject frames whose `meta_size` exceeds
`FLAGS_max_body_size`, in addition to the existing `body_size` check.
- Add unit tests in `test/brpc_sofa_pbrpc_protocol_unittest.cpp` covering
an oversized body and an oversized metadata block.
Side effects:
- Performance effects: none.
- Breaking backward compatibility: SOFA frames whose metadata exceeds
`max_body_size` (previously buffered and then rejected at protobuf
parsing time) are now rejected earlier with a "too big data" error and
the connection is closed.
---
### Check List:
- Please make sure your changes are compilable.
- When providing us with a new feature, it is best to add related tests.
- Please follow [Contributor Covenant Code of
Conduct](https://github.com/apache/brpc/blob/master/CODE_OF_CONDUCT.md).
Tests run:
- `test/brpc_sofa_pbrpc_protocol_unittest` (all cases passed)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
