back out r12509 and fix a better bug instead:

When authorities detected more than two relays running on the same
IP address, they were clearing all the status flags but forgetting
to clear the "hsdir" flag. So clients were being told that a
given relay was the right choice for a v2 hsdir lookup, yet they
never had its descriptor because it was marked as 'not running'
in the consensus.


svn:r12515

3 files changed, 11 insertions, 4 deletions

diff --git a/ChangeLog b/ChangeLog
index 6eb820ec50..c4f18ba8b0 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -10,8 +10,12 @@ Changes in version 0.2.0.12-alpha - 2007-11-??
       newly picked entry guard. Reported by Mike Perry.
     - Changing the ExitPolicyRejectPrivate setting should cause us to
       rebuild the descriptor.
-    - When picking v2 hidden service directories, don't pick ones that
-      aren't listed as Running.
+    - When authorities detected more than two relays running on the same
+      IP address, they were clearing all the status flags but forgetting
+      to clear the "hsdir" flag. So clients were being told that a
+      given relay was the right choice for a v2 hsdir lookup, yet they
+      never had its descriptor because it was marked as 'not running'
+      in the consensus.
     - If we're trying to fetch a bridge descriptor and there's no way
       the bridge authority could help us (for example, we don't know
       a digest, or there is no bridge authority), don't be so eager to
diff --git a/src/or/dirserv.c b/src/or/dirserv.c
index 36f661a4ca..39263467d3 100644
--- a/src/or/dirserv.c
+++ b/src/or/dirserv.c
@@ -2022,7 +2022,10 @@ dirserv_generate_networkstatus_vote_obj(crypto_pk_env_t *private_key,
       if (digestmap_get(omit_as_sybil, ri->cache_info.identity_digest)) {
         rs->is_authority = rs->is_exit = rs->is_stable = rs->is_fast =
           rs->is_running = rs->is_named = rs->is_valid = rs->is_v2_dir =
-          rs->is_possible_guard = 0;
+          rs->is_hs_dir = rs->is_possible_guard = 0;
+        /* FFFF we might want some mechanism to check later on if we
+         * missed zeroing any flags: it's easy to add a new flag but
+         * forget to add it to this clause. */
       }
       if (!vote_on_reachability)
         rs->is_running = 0;
diff --git a/src/or/routerlist.c b/src/or/routerlist.c
index eeb4909649..dce42b5731 100644
--- a/src/or/routerlist.c
+++ b/src/or/routerlist.c
@@ -4480,7 +4480,7 @@ hid_serv_get_responsible_directories(smartlist_t *responsible_dirs,
   i = start;
   do {
     routerstatus_t *r = smartlist_get(c->routerstatus_list, i);
-    if (r->is_hs_dir && r->is_running) {
+    if (r->is_hs_dir) {
       smartlist_add(responsible_dirs, r);
       if (++n_added == REND_NUMBER_OF_CONSECUTIVE_REPLICAS)
         return 0;