On Fri, Sep 08, 2017 at 10:38:00AM -0700, Brandon Carpenter wrote:
> Keep pings and gratuitous pongs generated by web browsers from killing
> websocket connections.
>
> Signed-off-by: Brandon Carpenter <brandon.carpen...@cypherpath.com>
> ---
> io/channel-websock.c | 21 +++++++++++++++++----
> 1 file changed, 17 insertions(+), 4 deletions(-)
>
> diff --git a/io/channel-websock.c b/io/channel-websock.c
> index 3183aeff77..50387050d5 100644
> --- a/io/channel-websock.c
> +++ b/io/channel-websock.c
> @@ -86,6 +86,7 @@
> #define QIO_CHANNEL_WEBSOCK_HEADER_FIELD_OPCODE 0x0f
> #define QIO_CHANNEL_WEBSOCK_HEADER_FIELD_HAS_MASK 0x80
> #define QIO_CHANNEL_WEBSOCK_HEADER_FIELD_PAYLOAD_LEN 0x7f
> +#define QIO_CHANNEL_WEBSOCK_CONTROL_OPCODE_MASK 0x8
>
> typedef struct QIOChannelWebsockHeader QIOChannelWebsockHeader;
>
> @@ -565,8 +566,11 @@ static int
> qio_channel_websock_decode_header(QIOChannelWebsock *ioc,
> return -1;
> }
> } else {
> - if (opcode != QIO_CHANNEL_WEBSOCK_OPCODE_BINARY_FRAME) {
> - error_setg(errp, "only binary websocket frames are supported");
> + if (opcode != QIO_CHANNEL_WEBSOCK_OPCODE_BINARY_FRAME &&
> + opcode != QIO_CHANNEL_WEBSOCK_OPCODE_PING &&
> + opcode != QIO_CHANNEL_WEBSOCK_OPCODE_PONG) {
Why would we need to ignore PONG ? A client should only send a PONG in
response to a PING that we send, and we never send PINGs. So if we
received a PONG that would be a serious error by the client, which should
cause us to close the connection IMHO>
Regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
