Browse Source

Don't cannibalize one-hop circuits

In rare cases, we could cannibalize a one-hop circuit, ending up
with a two-hop circuit. This circuit would not be actually used,
but we should prevent its creation in the first place.

Thanks to outofwords and swissknife for helping to analyse this.
Sebastian Hahn 14 years ago
parent
commit
faf51fa52a
2 changed files with 12 additions and 1 deletions
  1. 6 0
      changes/dont_cannibalize_onehop_circuits
  2. 6 1
      src/or/circuitlist.c

+ 6 - 0
changes/dont_cannibalize_onehop_circuits

@@ -0,0 +1,6 @@
+ o Refactorings:
+   - Make it explicit that we don't cannibalize one-hop circuits. This happens
+     in the wild, but doesn't turn out to be a problem because we fortunately
+     don't use those circuits. Many thanks to outofwords for the initial
+     analysis and to swissknife who confirmed that two-hop circuits are
+     actually created.

+ 6 - 1
src/or/circuitlist.c

@@ -905,6 +905,10 @@ circuit_find_to_cannibalize(uint8_t purpose, extend_info_t *info,
   int need_capacity = (flags & CIRCLAUNCH_NEED_CAPACITY) != 0;
   int internal = (flags & CIRCLAUNCH_IS_INTERNAL) != 0;
 
+  /* Make sure we're not trying to create a onehop circ by
+   * cannibalization. */
+  tor_assert(!(flags & CIRCLAUNCH_ONEHOP_TUNNEL));
+
   log_debug(LD_CIRC,
             "Hunting for a circ to cannibalize: purpose %d, uptime %d, "
             "capacity %d, internal %d",
@@ -920,7 +924,8 @@ circuit_find_to_cannibalize(uint8_t purpose, extend_info_t *info,
       if ((!need_uptime || circ->build_state->need_uptime) &&
           (!need_capacity || circ->build_state->need_capacity) &&
           (internal == circ->build_state->is_internal) &&
-          circ->remaining_relay_early_cells) {
+          circ->remaining_relay_early_cells &&
+          !circ->build_state->onehop_tunnel) {
         if (info) {
           /* need to make sure we don't duplicate hops */
           crypt_path_t *hop = circ->cpath;