Stop being so aggressive about fetching dir info if your DirPort is

on but your ORPort is off.

Add a new config option BridgeRelay that specifies you want to
be a bridge relay. Right now the only difference is that it makes
you answer begin_dir requests, and it makes you cache dir info,
even if your DirPort isn't on.

Refactor directory_caches_dir_info() into some more functions.


svn:r12668

9 files changed, 97 insertions, 43 deletions

diff --git a/ChangeLog b/ChangeLog
index 3567d07c4b..53afd291a2 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -14,6 +14,8 @@ Changes in version 0.2.0.13-alpha - 2007-12-??
     - Stop thinking that 0.1.2.x directory servers can handle "begin_dir"
       requests. Should ease bugs 406 and 419 where 0.1.2.x relays are
       crashing or mis-answering these requests.
+    - Stop being so aggressive about fetching dir info if your DirPort is
+      on but your ORPort is off.
 
   o Minor bugfixes:
     - The fix in 0.2.0.12-alpha cleared the "hsdir" flag in v3 network
@@ -37,9 +39,13 @@ Changes in version 0.2.0.13-alpha - 2007-12-??
       consumers. (We already do this on HUP.)
     - Authorities and caches fetch the v2 networkstatus documents
       less often, now that v3 is encouraged.
+    - Add a new config option BridgeRelay that specifies you want to
+      be a bridge relay. Right now the only difference is that it makes
+      you answer begin_dir requests, and it makes you cache dir info,
+      even if your DirPort isn't on.
 
   o Code simplifications:
-    - 
+    - ????
 
 
 Changes in version 0.2.0.12-alpha - 2007-11-16
diff --git a/src/or/command.c b/src/or/command.c
index 41c0c1112b..aa060faf3c 100644
--- a/src/or/command.c
+++ b/src/or/command.c
@@ -375,7 +375,7 @@ command_process_relay_cell(cell_t *cell, or_connection_t *conn)
   }
 
   if (CIRCUIT_IS_ORIGIN(circ)) {
-    /* if we're a server and treating connections with recent local
+    /* if we're a relay and treating connections with recent local
      * traffic better, then this is one of them. */
     conn->client_used = time(NULL);
   }
diff --git a/src/or/config.c b/src/or/config.c
index a119644b98..a8060f2269 100644
--- a/src/or/config.c
+++ b/src/or/config.c
@@ -144,6 +144,7 @@ static config_var_t _option_vars[] = {
   V(BandwidthRate,               MEMUNIT,  "5 MB"),
   V(BridgeAuthoritativeDir,      BOOL,     "0"),
   VAR("Bridge",                  LINELIST, Bridges,    NULL),
+  V(BridgeRelay,                 BOOL,     "0"),
   V(CircuitBuildTimeout,         INTERVAL, "1 minute"),
   V(CircuitIdleTimeout,          INTERVAL, "1 hour"),
   V(ClientDNSRejectInternalAddresses, BOOL,"1"),
@@ -1132,8 +1133,10 @@ options_act(or_options_t *old_options)
   if (old_options) {
     if (authdir_mode_v3(options) && !authdir_mode_v3(old_options))
       dirvote_recalculate_timing(options, time(NULL));
-    if (!bool_eq(directory_caches_dir_info(options),
-                 directory_caches_dir_info(old_options))) {
+    if (!bool_eq(directory_fetches_dir_info_like_mirror(options),
+                 directory_fetches_dir_info_like_mirror(old_options)) ||
+        !bool_eq(directory_fetches_dir_info_like_bridge_user(options),
+                 directory_fetches_dir_info_like_bridge_user(old_options))) {
       /* Make sure update_router_have_min_dir_info gets called. */
       router_dir_info_changed();
       /* We might need to download a new consensus status later or sooner than
diff --git a/src/or/connection_edge.c b/src/or/connection_edge.c
index 8335a42a0d..f62984259e 100644
--- a/src/or/connection_edge.c
+++ b/src/or/connection_edge.c
@@ -2292,8 +2292,7 @@ connection_exit_begin_conn(cell_t *cell, circuit_t *circ)
       return 0;
     }
   } else if (rh.command == RELAY_COMMAND_BEGIN_DIR) {
-    or_options_t *options = get_options();
-    if (!directory_permits_begindir_requests(options) ||
+    if (!directory_permits_begindir_requests(get_options()) ||
         circ->purpose != CIRCUIT_PURPOSE_OR) {
       end_payload[0] = END_STREAM_REASON_NOTDIRECTORY;
       relay_send_command_from_edge(rh.stream_id, circ, RELAY_COMMAND_END,
diff --git a/src/or/dirserv.c b/src/or/dirserv.c
index 03a0c2791b..f36ac55b67 100644
--- a/src/or/dirserv.c
+++ b/src/or/dirserv.c
@@ -1089,25 +1089,50 @@ dirserv_dump_directory_to_string(char **dir_out,
 /********************************************************************/
 
 /* A set of functions to answer questions about how we'd like to behave
- * as a directory cache/client. */
+ * as a directory mirror/client. */
 
-/** Return 1 if we want to keep descriptors, networkstatuses, etc around
- * and serve them to others, or 0 otherwise.
- * Also causes us to fetch new networkstatuses, descriptors, etc on the
- * "mirror" schedule rather than the "client" schedule.
+/** Return 1 if we fetch our directory material directly from the
+ * authorities, rather than from a mirror. */
+int
+directory_fetches_from_authorities(or_options_t *options)
+{
+  if (options->DirPort == 0)
+    return 0;
+  /* XXX if dirport not advertised, return 0 too */
+  if (!server_mode(options))
+    return 0;
+  /* XXX if orport or dirport not reachable, return 0 too */
+  return 1;
+}
+
+/* Return 1 if we should fetch new networkstatuses, descriptors, etc
+ * on the "mirror" schedule rather than the "client" schedule.
  */
 int
-directory_caches_dir_info(or_options_t *options)
+directory_fetches_dir_info_like_mirror(or_options_t *options)
 {
-  return options->DirPort != 0;
+  return directory_fetches_from_authorities(options);
 }
 
-/** Return 1 if we fetch our directory material directly from the
- * authorities, rather than some other cache. */
+/* Return 1 if we should fetch new networkstatuses, descriptors, etc
+ * on a very passive schedule -- waiting long enough for ordinary clients
+ * to probably have the info we want. These would include bridge users,
+ * and maybe others in the future e.g. if a Tor client uses another Tor
+ * client as a directory guard.
+ */
 int
-directory_fetches_from_authorities(or_options_t *options)
+directory_fetches_dir_info_like_bridge_user(or_options_t *options)
+{
+  return options->UseBridges != 0;
+}
+
+/** Return 1 if we want to keep descriptors, networkstatuses, etc around
+ * and we're willing to serve them to others. Else return 0.
+ */
+int
+directory_caches_dir_info(or_options_t *options)
 {
-  return server_mode(options) && options->DirPort != 0;
+  return options->BridgeRelay != 0 || options->DirPort != 0;
 }
 
 /** Return 1 if we want to allow remote people to ask us directory
@@ -1116,7 +1141,7 @@ directory_fetches_from_authorities(or_options_t *options)
 int
 directory_permits_begindir_requests(or_options_t *options)
 {
-  return options->DirPort != 0;
+  return options->BridgeRelay != 0 || options->DirPort != 0;
 }
 
 /** Return 1 if we want to allow controllers to ask us directory
@@ -1128,6 +1153,17 @@ directory_permits_controller_requests(or_options_t *options)
   return options->DirPort != 0;
 }
 
+/** Return 1 if we have no need to fetch new descriptors. This generally
+ * happens when we're not a dir cache and we haven't built any circuits
+ * lately.
+ */
+int
+directory_too_idle_to_fetch_descriptors(or_options_t *options, time_t now)
+{
+  return !options->DirPort && !options->FetchUselessDescriptors &&
+         rep_hist_circbuilding_dormant(now);
+}
+
 /********************************************************************/
 
 /* Used only by non-v1-auth dirservers: The v1 directory and
diff --git a/src/or/main.c b/src/or/main.c
index acb7cac3f4..ccf1e1c637 100644
--- a/src/or/main.c
+++ b/src/or/main.c
@@ -952,9 +952,8 @@ run_scheduled_events(time_t now)
    * (if we've passed our internal checks). */
   if (time_to_fetch_directory < now) {
     /* Only caches actually need to fetch directories now. */
-    if (directory_caches_dir_info(options) && !authdir_mode_v1(options)) {
-      /* XXX020 actually, we should only do this if we want to advertise
-       * our dirport. not simply if we configured one. -RD */
+    if (directory_fetches_dir_info_like_mirror(options) &&
+        !authdir_mode_v1(options)) {
       if (any_trusted_dir_is_v1_authority() &&
           !should_delay_dir_fetches(options))
         directory_get_from_dirserver(DIR_PURPOSE_FETCH_DIR,
@@ -966,7 +965,7 @@ run_scheduled_events(time_t now)
   }
 
   /* Caches need to fetch running_routers; directory clients don't. */
-  if (directory_caches_dir_info(options) &&
+  if (directory_fetches_dir_info_like_mirror(options) &&
       time_to_fetch_running_routers < now) {
     if (!authdir_mode_v1(options) && !should_delay_dir_fetches(options)) {
       directory_get_from_dirserver(DIR_PURPOSE_FETCH_RUNNING_LIST,
diff --git a/src/or/networkstatus.c b/src/or/networkstatus.c
index 6472576f52..7051e82158 100644
--- a/src/or/networkstatus.c
+++ b/src/or/networkstatus.c
@@ -1051,7 +1051,7 @@ update_consensus_networkstatus_fetch_time(time_t now)
     long dl_interval;
     long interval = c->fresh_until - c->valid_after;
     time_t start;
-    if (directory_caches_dir_info(options)) {
+    if (directory_fetches_dir_info_like_mirror(options)) {
       /* We want to cache the next one at some point after this one
        * is no longer fresh... */
       start = c->fresh_until + CONSENSUS_MIN_SECONDS_BEFORE_CACHING;
@@ -1062,6 +1062,8 @@ update_consensus_networkstatus_fetch_time(time_t now)
       start = c->fresh_until + (interval*3)/4;
       /* But download the next one before this one is expired. */
       dl_interval = ((c->valid_until - start) * 7 )/ 8;
+    /* XXX020 do something different if
+     * directory_fetches_dir_info_like_bridge_user() */
     }
     if (dl_interval < 1)
       dl_interval = 1;
@@ -1110,7 +1112,7 @@ update_networkstatus_downloads(time_t now)
   or_options_t *options = get_options();
   if (should_delay_dir_fetches(options))
     return;
-  if (directory_caches_dir_info(options))
+  if (directory_fetches_dir_info_like_mirror(options))
     update_v2_networkstatus_cache_downloads(now);
   update_consensus_networkstatus_downloads(now);
   update_certificate_downloads(now);
diff --git a/src/or/or.h b/src/or/or.h
index d992d7cdc6..ddafcd3d20 100644
--- a/src/or/or.h
+++ b/src/or/or.h
@@ -2084,6 +2084,10 @@ typedef struct {
   int UseBridges; /**< Boolean: should we start all circuits with a bridge? */
   config_line_t *Bridges; /**< List of bootstrap bridge addresses. */
 
+  int BridgeRelay; /**< Boolean: are we acting as a bridge relay? We make
+                    * this explicit so we can change how we behave in the
+                    * future. */
+
   /** Boolean: if we know the bridge's digest, should we get new
    * descriptors from the bridge authorities or from the bridge itself? */
   int UpdateBridgesFromAuthority;
@@ -3025,10 +3029,13 @@ int list_server_status(smartlist_t *routers, char **router_status_out,
 int dirserv_dump_directory_to_string(char **dir_out,
                                      crypto_pk_env_t *private_key);
 
-int directory_caches_dir_info(or_options_t *options);
 int directory_fetches_from_authorities(or_options_t *options);
+int directory_fetches_dir_info_like_mirror(or_options_t *options);
+int directory_fetches_dir_info_like_bridge_user(or_options_t *options);
+int directory_caches_dir_info(or_options_t *options);
 int directory_permits_begindir_requests(or_options_t *options);
 int directory_permits_controller_requests(or_options_t *options);
+int directory_too_idle_to_fetch_descriptors(or_options_t *options, time_t now);
 
 void directory_set_dirty(void);
 cached_dir_t *dirserv_get_directory(void);
diff --git a/src/or/routerlist.c b/src/or/routerlist.c
index d7d33305bd..524113c4bf 100644
--- a/src/or/routerlist.c
+++ b/src/or/routerlist.c
@@ -2909,7 +2909,7 @@ routerlist_remove_old_routers(void)
   routerinfo_t *router;
   signed_descriptor_t *sd;
   digestmap_t *retain;
-  int dirserv = directory_caches_dir_info(get_options());
+  int caches = directory_caches_dir_info(get_options());
   const networkstatus_vote_t *consensus = networkstatus_get_latest_consensus();
   const smartlist_t *networkstatus_v2_list = networkstatus_get_v2_list();
 
@@ -2923,7 +2923,7 @@ routerlist_remove_old_routers(void)
   retain = digestmap_new();
   cutoff = now - OLD_ROUTER_DESC_MAX_AGE;
   /* Build a list of all the descriptors that _anybody_ lists. */
-  if (dirserv) {
+  if (caches) {
     SMARTLIST_FOREACH(networkstatus_v2_list, networkstatus_v2_t *, ns,
     {
       /* XXXX The inner loop here gets pretty expensive, and actually shows up
@@ -2951,7 +2951,7 @@ routerlist_remove_old_routers(void)
    * routers that are too old and that nobody recommends.  (If we don't have
    * enough networkstatuses, then we should get more before we decide to kill
    * routers.) */
-  if (!dirserv ||
+  if (!caches ||
       smartlist_len(networkstatus_v2_list) > get_n_v2_authorities() / 2) {
     cutoff = now - ROUTER_MAX_AGE;
     /* Remove too-old unrecommended members of routerlist->routers. */
@@ -3000,7 +3000,7 @@ routerlist_remove_old_routers(void)
    * total number doesn't approach max_descriptors_per_router()*len(router).
    */
   if (smartlist_len(routerlist->old_routers) <
-      smartlist_len(routerlist->routers) * (dirserv?4:2))
+      smartlist_len(routerlist->routers) * (caches?4:2))
     goto done;
 
   smartlist_sort(routerlist->old_routers, _compare_old_routers_by_identity);
@@ -3224,7 +3224,7 @@ signed_desc_digest_is_recognized(signed_descriptor_t *desc)
 {
   routerstatus_t *rs;
   networkstatus_vote_t *consensus = networkstatus_get_latest_consensus();
-  int dirserv = directory_caches_dir_info(get_options());
+  int caches = directory_caches_dir_info(get_options());
   const smartlist_t *networkstatus_v2_list = networkstatus_get_v2_list();
 
   if (consensus) {
@@ -3233,7 +3233,7 @@ signed_desc_digest_is_recognized(signed_descriptor_t *desc)
                       desc->signed_descriptor_digest, DIGEST_LEN))
       return 1;
   }
-  if (dirserv && networkstatus_v2_list) {
+  if (caches && networkstatus_v2_list) {
     SMARTLIST_FOREACH(networkstatus_v2_list, networkstatus_v2_t *, ns,
     {
       if (!(rs = networkstatus_v2_find_entry(ns, desc->identity_digest)))
@@ -3560,7 +3560,7 @@ launch_router_descriptor_downloads(smartlist_t *downloadable, time_t now)
   or_options_t *options = get_options();
 
   n_downloadable = smartlist_len(downloadable);
-  if (!directory_caches_dir_info(options)) {
+  if (!directory_fetches_dir_info_like_mirror(options)) {
     if (n_downloadable >= MAX_DL_TO_DELAY) {
       log_debug(LD_DIR,
              "There are enough downloadable routerdescs to launch requests.");
@@ -3582,6 +3582,8 @@ launch_router_descriptor_downloads(smartlist_t *downloadable, time_t now)
       }
     }
   }
+  /* XXX020 should we consider having even the dir mirrors delay
+   * a little bit, so we don't load the authorities as much? -RD */
 
   if (! should_delay && n_downloadable) {
     int i, n_per_request;
@@ -3611,10 +3613,11 @@ launch_router_descriptor_downloads(smartlist_t *downloadable, time_t now)
 }
 
 /** Launch downloads for router status as needed, using the strategy used by
- * authorities and caches: download every descriptor we don't have but would
- * serve, from a random authority that lists it. */
+ * authorities and caches: based on the v2 networkstatuses we have, download
+ * every descriptor we don't have but would serve, from a random authority
+ * that lists it. */
 static void
-update_router_descriptor_cache_downloads(time_t now)
+update_router_descriptor_cache_downloads_v2(time_t now)
 {
   smartlist_t **downloadable; /* For each authority, what can we dl from it? */
   smartlist_t **download_from; /*          ... and, what will we dl from it? */
@@ -3624,8 +3627,8 @@ update_router_descriptor_cache_downloads(time_t now)
   or_options_t *options = get_options();
   const smartlist_t *networkstatus_v2_list = networkstatus_get_v2_list();
 
-  if (! directory_caches_dir_info(options)) {
-    log_warn(LD_BUG, "Called update_router_descriptor_cache_downloads() "
+  if (! directory_fetches_dir_info_like_mirror(options)) {
+    log_warn(LD_BUG, "Called update_router_descriptor_cache_downloads_v2() "
              "on a non-dir-mirror?");
   }
 
@@ -3763,14 +3766,12 @@ update_consensus_router_descriptor_downloads(time_t now)
   smartlist_t *no_longer_old = smartlist_create();
   smartlist_t *downloadable = smartlist_create();
   int authdir = authdir_mode(options);
-  int dirserver = directory_caches_dir_info(options);
   networkstatus_vote_t *consensus =
     networkstatus_get_reasonably_live_consensus(now);
   int n_delayed=0, n_have=0, n_would_reject=0, n_wouldnt_use=0,
     n_inprogress=0, n_in_oldrouters=0;
 
-  if (!dirserver && !options->FetchUselessDescriptors &&
-      rep_hist_circbuilding_dormant(now))
+  if (directory_too_idle_to_fetch_descriptors(options, now))
     goto done;
   if (!consensus)
     goto done;
@@ -3804,7 +3805,8 @@ update_consensus_router_descriptor_downloads(time_t now)
         ++n_would_reject;
         continue; /* We would throw it out immediately. */
       }
-      if (!dirserver && !client_would_use_router(rs, now, options)) {
+      if (!directory_caches_dir_info(options) &&
+          !client_would_use_router(rs, now, options)) {
         ++n_wouldnt_use;
         continue; /* We would never use it ourself. */
       }
@@ -3859,8 +3861,8 @@ update_router_descriptor_downloads(time_t now)
   or_options_t *options = get_options();
   if (should_delay_dir_fetches(options))
     return;
-  if (directory_caches_dir_info(options)) {
-    update_router_descriptor_cache_downloads(now);
+  if (directory_fetches_dir_info_like_mirror(options)) {
+    update_router_descriptor_cache_downloads_v2(now);
   }
   update_consensus_router_descriptor_downloads(now);
 }