Date: Fri, 22 Mar 2002 18:53:52 -0800 (PST) From: John Polstra <jdp@polstra.com> To: net@freebsd.org Cc: dinesh@verniernetworks.com Subject: Re: Socket buffer size limitation Message-ID: <200203230253.g2N2rqj07849@vashon.polstra.com> In-Reply-To: <20020322172352.R60156-100000@getafix.verniernetworks.com> References: <20020322172352.R60156-100000@getafix.verniernetworks.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <20020322172352.R60156-100000@getafix.verniernetworks.com>, Dinesh Nambisan <dinesh@verniernetworks.com> wrote: > Hello, > I have a situation where I'm trying to set the send and receive buffer > sizes on a socket to large values like 1 MB or so (dont ask why :-), > simple reason being we need it); and setsockopt for setting the buffer > sizes seems to fail with ENOBUFS. On looking at the code in uipc_socket.c > where it handles the SO_SNDBUF & SO_RCVBUF options, and the code in > uipc_socket2.c where it actually tries to reserve space (in sbreserve > function_, the issue seems to be a limit imposed by the code to > approx 227K or so [ sb_max(=256) * MCLBYTES / (MSIZE + MCLBYTES) ; which > works out to approx 227 K on my platform] > Was wondering if anyone could explain as to why this limit is there ? It's probably there to prevent a hostile or broken process from consuming all of the mbufs and/or mbuf clusters. You can change the limit via the sysctl variable "kern.ipc.maxsockbuf". John -- John Polstra John D. Polstra & Co., Inc. Seattle, Washington USA "Disappointment is a good sign of basic intelligence." -- Chögyam Trungpa To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200203230253.g2N2rqj07849>