Sign Up to our social questions and Answers Engine to ask questions, answer people’s questions, and connect with other people.
Login to our social questions & Answers Engine to ask questions answer people’s questions & connect with other people.
Lost your password? Please enter your email address. You will receive a link and will create a new password via email.
Please briefly explain why you feel this question should be reported.
Please briefly explain why you feel this answer should be reported.
Please briefly explain why you feel this user should be reported.
(Linux 3.0, x86_64)
If I stop calling recv(2) for a while on a connected tcp socket, and the other host keeps sending data, than presumably this data is received and buffered by the kernel, so the next time I call recv it will return quickly with all the waiting data.
My question is how big is this receive buffer, and how does it work? Are there any relevant kernel or socket settings I can use to tune its behavior? What does the tcp stack do when it fills up?
The receive buffer can be changed with
setsockopt(SO_RCVBUF). The default size is controlled by the tcp_rmem sysctl. If the buffer fills up, the stack will stop extending the receive window it offers to the other side, forcing the other size to stop sending data until there’s space in the buffer.