Add percentiles to the desc stats reporting

To get a better idea what's going on on Tonga, add some code to report
how often the most and least frequently fetched descriptor was fetched,
as well as 25, 50, 75 percentile.

Also ensure we only count bridge descriptors here.

3 files changed, 58 insertions, 6 deletions

diff --git a/changes/ticket4294 b/changes/ticket4294
new file mode 100644
index 0000000000..5041495542
--- /dev/null
+++ b/changes/ticket4294
@@ -0,0 +1,8 @@
+(NOTE: THIS SHOULD REPLACE THE TICKET 4200 ENTRY IN THE CHANGELOG)
+  o Minor features:
+    - The Bridge Authority now writes statistics on how many bridge
+      descriptors it gave out in total, and how many unique descriptors it
+      gave out. It also lists how often the most and least commonly fetched
+      descriptors were given out, as well as the median and 25/75 percentile.
+      Implements tickets 4200 and 4294.
+
diff --git a/src/or/dirserv.c b/src/or/dirserv.c
index 4b9587b0fc..5cb4aba5aa 100644
--- a/src/or/dirserv.c
+++ b/src/or/dirserv.c
@@ -3488,6 +3488,8 @@ connection_dirserv_add_servers_to_outbuf(dir_connection_t *conn)
                conn->dir_spool_src == DIR_SPOOL_EXTRA_BY_DIGEST);
   time_t publish_cutoff = time(NULL)-ROUTER_MAX_AGE_TO_PUBLISH;
 
+  const or_options_t *options = get_options();
+
   while (smartlist_len(conn->fingerprint_stack) &&
          connection_get_outbuf_len(TO_CONN(conn)) < DIRSERV_BUFFER_MIN) {
     const char *body;
@@ -3509,7 +3511,16 @@ connection_dirserv_add_servers_to_outbuf(dir_connection_t *conn)
        * unknown bridge descriptor has shown up between then and now. */
       continue;
     }
-    rep_hist_note_desc_served(sd->identity_digest);
+
+    /** If we are the bridge authority and the descriptor is a bridge
+     * descriptor, remember that we served this descriptor for desc stats. */
+    if (options->BridgeAuthoritativeDir && by_fp) {
+      const routerinfo_t *router =
+          router_get_by_id_digest(sd->identity_digest);
+      tor_assert(router);
+      if (router->purpose == ROUTER_PURPOSE_BRIDGE)
+        rep_hist_note_desc_served(sd->identity_digest);
+    }
     body = signed_descriptor_get_body(sd);
     if (conn->zlib_state) {
       /* XXXX022 This 'last' business should actually happen on the last
diff --git a/src/or/rephist.c b/src/or/rephist.c
index f8ebe644fc..359e4886a3 100644
--- a/src/or/rephist.c
+++ b/src/or/rephist.c
@@ -2643,7 +2643,7 @@ rep_hist_reset_desc_stats(time_t now)
 void
 rep_hist_desc_stats_term(void)
 {
-  digestmap_free(served_descs, NULL);
+  digestmap_free(served_descs, _tor_free);
   served_descs = NULL;
   start_of_served_descs_stats_interval = 0;
   total_descriptor_downloads = 0;
@@ -2659,18 +2659,43 @@ rep_hist_format_desc_stats(time_t now)
   char t[ISO_TIME_LEN+1];
   char *result;
 
+  digestmap_iter_t *iter;
+  const char *key;
+  void *val;
+  unsigned size;
+  int *vals;
+  int n = 0, *count;
+
   if (!start_of_served_descs_stats_interval)
     return NULL;
 
+  size =  digestmap_size(served_descs);
+  vals = tor_malloc(size * sizeof(int));
+
+  for (iter = digestmap_iter_init(served_descs); !digestmap_iter_done(iter);
+      iter = digestmap_iter_next(served_descs, iter) ) {
+    digestmap_iter_get(iter, &key, &val);
+    count = val;
+    vals[n++] = *count;
+      (void)key;
+  }
+
   format_iso_time(t, now);
 
   tor_asprintf(&result,
-               "served-descs-stats-end %s (%d s) total=%lu unique=%u\n",
+               "served-descs-stats-end %s (%d s) total=%lu unique=%u "
+               "max=%d q3=%d md=%d q1=%d min=%d\n",
                t,
                (unsigned) (now - start_of_served_descs_stats_interval),
                total_descriptor_downloads,
-               digestmap_size(served_descs));
-
+               size,
+               find_nth_int(vals, size, size-1),
+               find_nth_int(vals, size, (3*size-1)/4),
+               find_nth_int(vals, size, (size-1)/2),
+               find_nth_int(vals, size, (size-1)/4),
+               find_nth_int(vals, size, 0));
+
+  tor_free(vals);
   return result;
 }
 
@@ -2712,9 +2737,17 @@ rep_hist_desc_stats_write(time_t now)
 void
 rep_hist_note_desc_served(const char * desc)
 {
+  int *val;
   if (!served_descs)
     return; // We're not collecting stats
-  digestmap_set(served_descs, desc, (void *)1);
+  val = digestmap_get(served_descs, desc);
+  if (!val) {
+    val = tor_malloc(sizeof(int));
+    *val = 1;
+    digestmap_set(served_descs, desc, val);
+  } else {
+    (*val)++;
+  }
   total_descriptor_downloads++;
 }