Commits on Mar 25, 2013
  1. Sockets were being leaked - the WB_ALIVE flags is cleared in various …

    …places, but without calling wb_close_socket. When the connection was finally torn down, wb_close_client would only close the socket when WB_ALIVE was still set. Seems more robust to check socket != WB_INVALID_SOCKET?
    hulkholden committed Mar 25, 2013
Commits on Mar 24, 2013
  1. If we don't return after clearing WB_ALIVE, we can spin forever in wb…

    …_update_client. (Seen when specifying content-length and WB_CLOSE_AFTER_RESPONSE was set)
    hulkholden committed Mar 24, 2013
  2. Fix a stray argument to dbg()

    hulkholden committed Mar 24, 2013
  3. [~] Fix memory corruption - N connections are initialised, but memory…

    … for just N-1 connections is reserved. Reading into the first connection header_buf was corrupting srv->connections[3]. Wouldn't cause any problems until 4 connections came in simultaneously
    hulkholden committed Mar 24, 2013