kocolosk commented on issue #3939: URL: https://github.com/apache/couchdb/issues/3939#issuecomment-1049102965
I tried adding an integration test for this in e85e34359 but it doesn't do quite what you'd expect. I tried writing a doc with attachments and then saving that same revision again using `new_edits=false`. The client does receive a rapid response to **both** requests, but (I think) the attachment bytes are never consumed off of the wire and so the _next_ request on the connection from a different part of the test suite hangs. The second `new_edits=false` multipart request never shows up in the logs because the test suite times out first and destroys the DB. Perhaps this is also what you were observing in production? At any rate, I added the test in part to demonstrate that the data race with the quorum logic might be a red herring. _Anytime_ a `multipart/related` request with `new_edits=false` is made with a revision that already exists in the database the server does not properly clean up the connection. -- 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]
