aboutsummaryrefslogtreecommitdiff
path: root/src
diff options
context:
space:
mode:
authorRoger Dingledine <arma@torproject.org>2003-11-11 04:09:34 +0000
committerRoger Dingledine <arma@torproject.org>2003-11-11 04:09:34 +0000
commit2a3479af5f995a23e8d6830ac60aa8bed19dfaed (patch)
treea83e4056e3907fe3810451feba00416f2b0f4480 /src
parent3d19a9b514afc65701e0d59820e344b696284a21 (diff)
downloadtor-2a3479af5f995a23e8d6830ac60aa8bed19dfaed.tar
tor-2a3479af5f995a23e8d6830ac60aa8bed19dfaed.tar.gz
respond to nick's questions
svn:r786
Diffstat (limited to 'src')
-rw-r--r--src/or/connection_edge.c19
1 files changed, 12 insertions, 7 deletions
diff --git a/src/or/connection_edge.c b/src/or/connection_edge.c
index 3400e37e9..b563844b6 100644
--- a/src/or/connection_edge.c
+++ b/src/or/connection_edge.c
@@ -483,9 +483,9 @@ void connection_ap_attach_pending(void)
/* r<0: There was an error sending the begin cell; other pending
* AP connections may succeed.
*/
- /* XXX Is this right? How do we say that the connection failed?
- * Should I close it? mark it for close? -NM */
connection_ap_handshake_socks_reply(conn, NULL, 0, 0);
+ conn->marked_for_close = 1;
+ conn->has_sent_end = 1; /* if the begin failed, don't try an end */
}
}
}
@@ -551,7 +551,7 @@ static int connection_ap_handshake_process_socks(connection_t *conn) {
/* Try to find a live circuit. If we don't find one, tell 'conn' to
* stop reading and return 1. Otherwise, associate the CONN_TYPE_AP
* connection 'conn' with the newest live circuit, and start sending a
- * BEGIN cell down the circuit. Returns 0 on success, and -1 on
+ * BEGIN cell down the circuit. Return 0 on success, and -1 on
* error.
*/
static int connection_ap_handshake_attach_circuit(connection_t *conn) {
@@ -567,10 +567,15 @@ static int connection_ap_handshake_attach_circuit(connection_t *conn) {
if(!circ) {
log_fn(LOG_INFO,"No circuit ready for edge connection; delaying.");
- connection_stop_reading(conn); /* XXX Is this correct? -NM */
+ connection_stop_reading(conn);
+ /* XXX both this and the start_reading below can go away if we
+ * remove our notion that we shouldn't read from a socks
+ * client until we're connected. the socks spec promises that it
+ * won't write. is that good enough?
+ */
return 1;
}
- connection_start_reading(conn); /* XXX Is this correct? -NM */
+ connection_start_reading(conn);
circ->dirty = 1;
@@ -600,6 +605,7 @@ static int connection_ap_handshake_send_begin(connection_t *ap_conn, circuit_t *
assert(ap_conn->type == CONN_TYPE_AP);
assert(ap_conn->state == AP_CONN_STATE_CIRCUIT_WAIT);
assert(ap_conn->socks_request);
+ assert(ap_conn->socks_request->addr);
if(crypto_pseudo_rand(STREAM_ID_SIZE, ap_conn->stream_id) < 0)
return -1;
@@ -619,8 +625,7 @@ static int connection_ap_handshake_send_begin(connection_t *ap_conn, circuit_t *
ap_conn->package_window = STREAMWINDOW_START;
ap_conn->deliver_window = STREAMWINDOW_START;
ap_conn->state = AP_CONN_STATE_OPEN;
- tor_free(ap_conn->socks_request);
- ap_conn->socks_request = NULL;
+ tor_free(ap_conn->socks_request); /* this also NULLs it out */
log_fn(LOG_INFO,"Address/port sent, ap socket %d, n_circ_id %d",ap_conn->s,circ->n_circ_id);
return 0;
}