Allow any valid 'end' cell to mean a circuit was used successfully.

Also improve some log messages.

1 files changed, 11 insertions, 6 deletions

diff --git a/src/or/connection_edge.c b/src/or/connection_edge.c
index 79bb54cbb7..ca6060ca57 100644
--- a/src/or/connection_edge.c
+++ b/src/or/connection_edge.c
@@ -2178,14 +2178,19 @@ connection_ap_handshake_socks_reply(entry_connection_t *conn, char *reply,
 
   /* Flag this stream's circuit as having completed a stream successfully
    * (for path bias) */
-  if (status == SOCKS5_SUCCEEDED) {
+  if (status == SOCKS5_SUCCEEDED ||
+      endreason == END_STREAM_REASON_RESOLVEFAILED ||
+      endreason == END_STREAM_REASON_CONNECTREFUSED ||
+      endreason == END_STREAM_REASON_CONNRESET ||
+      endreason == END_STREAM_REASON_NOROUTE ||
+      endreason == END_STREAM_REASON_RESOURCELIMIT) {
     if(!conn->edge_.on_circuit ||
        !CIRCUIT_IS_ORIGIN(conn->edge_.on_circuit)) {
-      // XXX: Weird. We hit this a lot, and yet have no unusable_circs.
-      // Maybe during addrmaps/resolves?
-      log_warn(LD_BUG,
-               "(Harmless.) No origin circuit for successful SOCKS stream. "
-               "Reason: %d", endreason);
+      // DNS remaps can trigger this. So can failed hidden service
+      // lookups.
+      log_info(LD_BUG,
+               "(Harmless.) No origin circuit for successful SOCKS stream %ld. "
+               "Reason: %d", ENTRY_TO_CONN(conn)->global_identifier, endreason);
     } else {
       TO_ORIGIN_CIRCUIT(conn->edge_.on_circuit)->path_state
           = PATH_STATE_USE_SUCCEEDED;