Add new stats type: descriptor fetch stats

This is used for the bridge authority currently, to get a better
intuition on how many descriptors are actually fetched from it and how
many fetches happen in total.

Implements ticket 4200.

changes/ticket4200

@@ -0,0 +1,5 @@
+  o Minor features:
+    - The bridge authority now writes statistics on how many bridge
+      descriptors it gave out in total, and how unique descriptors
+      it gave out. Implements ticket 4200.
+

src/or/config.c

@@ -1532,7 +1532,8 @@ options_act(const or_options_t *old_options)
 
   if (options->CellStatistics || options->DirReqStatistics ||
       options->EntryStatistics || options->ExitPortStatistics ||
-      options->ConnDirectionStatistics) {
+      options->ConnDirectionStatistics ||
+      options->BridgeAuthoritativeDir) {
     time_t now = time(NULL);
     int print_notice = 0;
     if ((!old_options || !old_options->CellStatistics) &&
@@ -1577,6 +1578,10 @@ options_act(const or_options_t *old_options)
         options->ConnDirectionStatistics) {
       rep_hist_conn_stats_init(now);
     }
+    if (!old_options || !old_options->BridgeAuthoritativeDir) {
+      rep_hist_desc_stats_init(now);
+      print_notice = 1;
+    }
     if (print_notice)
       log_notice(LD_CONFIG, "Configured to measure statistics. Look for "
                  "the *-stats files that will first be written to the "
@@ -1598,6 +1603,9 @@ options_act(const or_options_t *old_options)
   if (old_options && old_options->ConnDirectionStatistics &&
       !options->ConnDirectionStatistics)
     rep_hist_conn_stats_term();
+  if (old_options && old_options->BridgeAuthoritativeDir &&
+      !options->BridgeAuthoritativeDir)
+    rep_hist_desc_stats_term();
 
   /* Check if we need to parse and add the EntryNodes config option. */
   if (options->EntryNodes &&

src/or/dirserv.c

@@ -3509,6 +3509,7 @@ 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);
     body = signed_descriptor_get_body(sd);
     if (conn->zlib_state) {
       /* XXXX022 This 'last' business should actually happen on the last

src/or/main.c

@@ -1268,6 +1268,11 @@ run_scheduled_events(time_t now)
       if (next_write && next_write < next_time_to_write_stats_files)
         next_time_to_write_stats_files = next_write;
     }
+    if (options->BridgeAuthoritativeDir) {
+      time_t next_write = rep_hist_desc_stats_write(time_to_write_stats_files);
+      if (next_write && next_write < next_time_to_write_stats_files)
+        next_time_to_write_stats_files = next_write;
+    }
     time_to_write_stats_files = next_time_to_write_stats_files;
   }
 

src/or/rephist.c

@@ -2596,6 +2596,120 @@ rep_hist_buffer_stats_write(time_t now)
   return start_of_buffer_stats_interval + WRITE_STATS_INTERVAL;
 }
 
+/*** Descriptor serving statistics ***/
+
+/** Digestmap to track which descriptors were downloaded this stats
+ *  collection interval. It maps descriptor digest to pointers to 1,
+ *  effectively turning this into a list. */
+static digestmap_t *served_descs = NULL;
+
+/** Number of how many descriptors were downloaded in total during this
+ * interval. */
+static unsigned long total_descriptor_downloads;
+
+/** Start time of served descs stats or 0 if we're not collecting those. */
+static time_t start_of_served_descs_stats_interval;
+
+/** Initialize descriptor stats. */
+void
+rep_hist_desc_stats_init(time_t now)
+{
+  if (served_descs) {
+    log_warn(LD_BUG, "Called rep_hist_desc_stats_init() when desc stats were "
+             "already initialized. This is probably harmless.");
+    return; // Already initialized
+  }
+  served_descs = digestmap_new();
+  total_descriptor_downloads = 0;
+  start_of_served_descs_stats_interval = now;
+}
+
+/** Reset served descs stats to empty, starting a new interval <b>now</b>. */
+static void
+rep_hist_reset_desc_stats(time_t now)
+{
+  rep_hist_desc_stats_term();
+  rep_hist_desc_stats_init(now);
+}
+
+/** Stop collecting served descs stats, so that rep_hist_desc_stats_init() is
+ * safe to be called again. */
+void
+rep_hist_desc_stats_term(void)
+{
+  digestmap_free(served_descs, NULL);
+  served_descs = NULL;
+  start_of_served_descs_stats_interval = 0;
+  total_descriptor_downloads = 0;
+}
+
+/** Helper for rep_hist_desc_stats_write(). Return a newly allocated string
+ * containing the served desc statistics until now, or NULL if we're not
+ * collecting served desc stats. Caller must ensure that now is not before
+ * start_of_served_descs_stats_interval. */
+static char *
+rep_hist_format_desc_stats(time_t now)
+{
+  char t[ISO_TIME_LEN+1];
+  char *result;
+
+  if (!start_of_served_descs_stats_interval)
+    return NULL;
+
+  format_iso_time(t, now);
+
+  tor_asprintf(&result,
+               "served-descs-stats-end %s (%d s) total=%lu unique=%u\n",
+               t,
+               (unsigned) (now - start_of_served_descs_stats_interval),
+               total_descriptor_downloads,
+               digestmap_size(served_descs));
+
+  return result;
+}
+
+/** If WRITE_STATS_INTERVAL seconds have passed since the beginning of
+ * the current served desc stats interval, write the stats to
+ * $DATADIR/stats/served-desc-stats (possibly appending to an existing file)
+ * and reset the state for the next interval. Return when we would next want
+ * to write served desc stats or 0 if we won't want to write. */
+time_t
+rep_hist_desc_stats_write(time_t now)
+{
+  char *statsdir = NULL, *filename = NULL, *str = NULL;
+
+  if (!start_of_served_descs_stats_interval)
+    return 0; /* We're not collecting stats. */
+  if (start_of_served_descs_stats_interval + WRITE_STATS_INTERVAL > now)
+    return start_of_served_descs_stats_interval + WRITE_STATS_INTERVAL;
+
+  str = rep_hist_format_desc_stats(now);
+
+  statsdir = get_datadir_fname("stats");
+  if (check_private_dir(statsdir, CPD_CREATE, get_options()->User) < 0) {
+    log_warn(LD_HIST, "Unable to create stats/ directory!");
+      goto done;
+  }
+  filename = get_datadir_fname2("stats", "served-desc-stats");
+  if (append_bytes_to_file(filename, str, strlen(str), 0) < 0)
+    log_warn(LD_HIST, "Unable to write served descs statistics to disk!");
+
+  rep_hist_reset_desc_stats(now);
+
+ done:
+  tor_free(statsdir);
+  tor_free(filename);
+  tor_free(str);
+  return start_of_served_descs_stats_interval + WRITE_STATS_INTERVAL;
+}
+
+void
+rep_hist_note_desc_served(const char * desc)
+{
+  digestmap_set(served_descs, desc, (void *)1);
+  total_descriptor_downloads++;
+}
+
 /*** Connection statistics ***/
 
 /** Start of the current connection stats interval or 0 if we're not
@@ -2839,5 +2953,7 @@ rep_hist_free_all(void)
     smartlist_free(circuits_for_buffer_stats);
     circuits_for_buffer_stats = NULL;
   }
+  rep_hist_desc_stats_term();
+  total_descriptor_downloads = 0;
 }
 

src/or/rephist.h

@@ -82,6 +82,11 @@ void rep_hist_add_buffer_stats(double mean_num_cells_in_queue,
 char *rep_hist_format_buffer_stats(time_t now);
 void rep_hist_reset_buffer_stats(time_t now);
 
+void rep_hist_desc_stats_init(time_t now);
+void rep_hist_note_desc_served(const char * desc);
+void rep_hist_desc_stats_term(void);
+time_t rep_hist_desc_stats_write(time_t now);
+
 void rep_hist_conn_stats_init(time_t now);
 void rep_hist_note_or_conn_bytes(uint64_t conn_id, size_t num_read,
                                  size_t num_written, time_t when);