X-Git-Url: https://git.octo.it/?p=rrdtool.git;a=blobdiff_plain;f=src%2Frrd_daemon.c;h=9fbbc1133af5a79869a53e84657086b91d4f0b60;hp=03dd181f76189a60945c89c7aaf59de8f6c222ec;hb=607eea59c2ba85c63e2de1b665b37a6093e1575c;hpb=1164122dace49986850172ef2cb198dc301fe750

diff --git a/src/rrd_daemon.c b/src/rrd_daemon.c
index 03dd181..9fbbc11 100644
--- a/src/rrd_daemon.c
+++ b/src/rrd_daemon.c
@@ -116,6 +116,10 @@ struct listen_socket_s
   int family;
   socket_privilege privilege;
 
+  /* state for BATCH processing */
+  time_t batch_start;
+  int batch_cmd;
+
   /* buffered IO */
   char *rbuf;
   off_t next_cmd;
@@ -134,6 +138,7 @@ struct cache_item_s
   char **values;
   int values_num;
   time_t last_flush_time;
+  time_t last_update_stamp;
 #define CI_FLAGS_IN_TREE  (1<<0)
 #define CI_FLAGS_IN_QUEUE (1<<1)
   int flags;
@@ -166,6 +171,7 @@ typedef enum queue_side_e queue_side_t;
  * Variables
  */
 static int stay_foreground = 0;
+static uid_t daemon_uid;
 
 static listen_socket_t *listen_fds = NULL;
 static size_t listen_fds_num = 0;
@@ -280,7 +286,7 @@ static void install_signal_handlers(void) /* {{{ */
 
 } /* }}} void install_signal_handlers */
 
-static int open_pidfile(void) /* {{{ */
+static int open_pidfile(char *action, int oflag) /* {{{ */
 {
   int fd;
   char *file;
@@ -289,14 +295,52 @@ static int open_pidfile(void) /* {{{ */
     ? config_pid_file
     : LOCALSTATEDIR "/run/rrdcached.pid";
 
-  fd = open(file, O_CREAT|O_EXCL|O_WRONLY, S_IRUSR|S_IRGRP|S_IROTH);
+  fd = open(file, oflag, S_IWUSR|S_IRUSR|S_IRGRP|S_IROTH);
   if (fd < 0)
-    fprintf(stderr, "FATAL: cannot create '%s' (%s)\n",
-            file, rrd_strerror(errno));
+    fprintf(stderr, "rrdcached: can't %s pid file '%s' (%s)\n",
+            action, file, rrd_strerror(errno));
 
   return(fd);
 } /* }}} static int open_pidfile */
 
+/* check existing pid file to see whether a daemon is running */
+static int check_pidfile(void)
+{
+  int pid_fd;
+  pid_t pid;
+  char pid_str[16];
+
+  pid_fd = open_pidfile("open", O_RDWR);
+  if (pid_fd < 0)
+    return pid_fd;
+
+  if (read(pid_fd, pid_str, sizeof(pid_str)) <= 0)
+    return -1;
+
+  pid = atoi(pid_str);
+  if (pid <= 0)
+    return -1;
+
+  /* another running process that we can signal COULD be
+   * a competing rrdcached */
+  if (pid != getpid() && kill(pid, 0) == 0)
+  {
+    fprintf(stderr,
+            "FATAL: Another rrdcached daemon is running?? (pid %d)\n", pid);
+    close(pid_fd);
+    return -1;
+  }
+
+  lseek(pid_fd, 0, SEEK_SET);
+  ftruncate(pid_fd, 0);
+
+  fprintf(stderr,
+          "rrdcached: removed stale PID file (no rrdcached on pid %d)\n"
+          "rrdcached: starting normally.\n", pid);
+
+  return pid_fd;
+} /* }}} static int check_pidfile */
+
 static int write_pidfile (int fd) /* {{{ */
 {
   pid_t pid;
@@ -399,6 +443,7 @@ static int add_response_info(listen_socket_t *sock, char *fmt, ...) /* {{{ */
   int len;
 
   if (sock == NULL) return 0; /* journal replay mode */
+  if (sock->batch_start) return 0; /* no extra info returned when in BATCH */
 
   va_start(argp, fmt);
 #ifdef HAVE_VSNPRINTF
@@ -446,10 +491,14 @@ static int send_response (listen_socket_t *sock, response_code rc,
 
   if (sock == NULL) return rc;  /* journal replay mode */
 
-  if (rc == RESP_OK)
+  if (sock->batch_start)
   {
-    lines = count_lines(sock->wbuf);
+    if (rc == RESP_OK)
+      return rc; /* no response on success during BATCH */
+    lines = sock->batch_cmd;
   }
+  else if (rc == RESP_OK)
+    lines = count_lines(sock->wbuf);
   else
     lines = -1;
 
@@ -466,6 +515,10 @@ static int send_response (listen_socket_t *sock, response_code rc,
 
   len += rclen;
 
+  /* append the result to the wbuf, don't write to the user */
+  if (sock->batch_start)
+    return add_to_wbuf(sock, buffer, len);
+
   /* first write must be complete */
   if (len != write(sock->fd, buffer, len))
   {
@@ -473,7 +526,7 @@ static int send_response (listen_socket_t *sock, response_code rc,
     return -1;
   }
 
-  if (sock->wbuf != NULL)
+  if (sock->wbuf != NULL && rc == RESP_OK)
   {
     wrote = 0;
     while (wrote < sock->wbuf_len)
@@ -526,6 +579,34 @@ static void remove_from_queue(cache_item_t *ci) /* {{{ */
   ci->flags &= ~CI_FLAGS_IN_QUEUE;
 } /* }}} static void remove_from_queue */
 
+/* remove an entry from the tree and free all its resources.
+ * must hold 'cache lock' while calling this.
+ * returns 0 on success, otherwise errno */
+static int forget_file(const char *file)
+{
+  cache_item_t *ci;
+
+  ci = g_tree_lookup(cache_tree, file);
+  if (ci == NULL)
+    return ENOENT;
+
+  g_tree_remove (cache_tree, file);
+  remove_from_queue(ci);
+
+  for (int i=0; i < ci->values_num; i++)
+    free(ci->values[i]);
+
+  free (ci->values);
+  free (ci->file);
+
+  /* in case anyone is waiting */
+  pthread_cond_broadcast(&ci->flushed);
+
+  free (ci);
+
+  return 0;
+} /* }}} static int forget_file */
+
 /*
  * enqueue_cache_item:
  * `cache_lock' must be acquired before calling this function!
@@ -660,26 +741,10 @@ static int flush_old_values (int max_age)
 
   for (k = 0; k < cfd.keys_num; k++)
   {
-    cache_item_t *ci;
-
-    /* This must not fail. */
-    ci = (cache_item_t *) g_tree_lookup (cache_tree, cfd.keys[k]);
-    assert (ci != NULL);
-
-    /* If we end up here with values available, something's seriously
-     * messed up. */
-    assert (ci->values_num == 0);
-
-    /* Remove the node from the tree */
-    g_tree_remove (cache_tree, cfd.keys[k]);
-    cfd.keys[k] = NULL;
-
-    /* Now free and clean up `ci'. */
-    free (ci->file);
-    ci->file = NULL;
-    free (ci);
-    ci = NULL;
-  } /* for (k = 0; k < cfd.keys_num; k++) */
+    /* should never fail, since we have held the cache_lock
+     * the entire time */
+    assert( forget_file(cfd.keys[k]) == 0 );
+  }
 
   if (cfd.keys != NULL)
   {
@@ -721,8 +786,8 @@ static void *queue_thread_main (void *args __attribute__((unused))) /* {{{ */
       flush_old_values (config_write_interval);
 
       /* Determine the time of the next cache flush. */
-      while (next_flush.tv_sec <= now.tv_sec)
-        next_flush.tv_sec += config_flush_interval;
+      next_flush.tv_sec =
+        now.tv_sec + next_flush.tv_sec % config_flush_interval;
 
       /* unlock the cache while we rotate so we don't block incoming
        * updates if the fsync() blocks on disk I/O */
@@ -929,6 +994,40 @@ err:
   return 0;
 } /* }}} static int check_file_access */
 
+/* when using a base dir, convert relative paths to absolute paths.
+ * if necessary, modifies the "filename" pointer to point
+ * to the new path created in "tmp".  "tmp" is provided
+ * by the caller and sizeof(tmp) must be >= PATH_MAX.
+ *
+ * this allows us to optimize for the expected case (absolute path)
+ * with a no-op.
+ */
+static void get_abs_path(char **filename, char *tmp)
+{
+  assert(tmp != NULL);
+  assert(filename != NULL && *filename != NULL);
+
+  if (config_base_dir == NULL || **filename == '/')
+    return;
+
+  snprintf(tmp, PATH_MAX, "%s/%s", config_base_dir, *filename);
+  *filename = tmp;
+} /* }}} static int get_abs_path */
+
+/* returns 1 if we have the required privilege level,
+ * otherwise issue an error to the user on sock */
+static int has_privilege (listen_socket_t *sock, /* {{{ */
+                          socket_privilege priv)
+{
+  if (sock == NULL) /* journal replay */
+    return 1;
+
+  if (sock->privilege >= priv)
+    return 1;
+
+  return send_response(sock, RESP_ERR, "%s\n", rrd_strerror(EACCES));
+} /* }}} static int has_privilege */
+
 static int flush_file (const char *filename) /* {{{ */
 {
   cache_item_t *ci;
@@ -949,6 +1048,9 @@ static int flush_file (const char *filename) /* {{{ */
     pthread_cond_wait(&ci->flushed, &cache_lock);
   }
 
+  /* DO NOT DO ANYTHING WITH ci HERE!!  The entry
+   * may have been purged during our cond_wait() */
+
   pthread_mutex_unlock(&cache_lock);
 
   return (0);
@@ -965,10 +1067,13 @@ static int handle_request_help (listen_socket_t *sock, /* {{{ */
   {
     "Command overview\n"
     ,
+    "HELP [<command>]\n"
     "FLUSH <filename>\n"
     "FLUSHALL\n"
-    "HELP [<command>]\n"
+    "PENDING <filename>\n"
+    "FORGET <filename>\n"
     "UPDATE <filename> <values> [<values> ...]\n"
+    "BATCH\n"
     "STATS\n"
   };
 
@@ -991,6 +1096,26 @@ static int handle_request_help (listen_socket_t *sock, /* {{{ */
     "Triggers writing of all pending updates.  Returns immediately.\n"
   };
 
+  char *help_pending[2] =
+  {
+    "Help for PENDING\n"
+    ,
+    "Usage: PENDING <filename>\n"
+    "\n"
+    "Shows any 'pending' updates for a file, in order.\n"
+    "The updates shown have not yet been written to the underlying RRD file.\n"
+  };
+
+  char *help_forget[2] =
+  {
+    "Help for FORGET\n"
+    ,
+    "Usage: FORGET <filename>\n"
+    "\n"
+    "Removes the file completely from the cache.\n"
+    "Any pending updates for the file will be lost.\n"
+  };
+
   char *help_update[2] =
   {
     "Help for UPDATE\n"
@@ -1016,6 +1141,28 @@ static int handle_request_help (listen_socket_t *sock, /* {{{ */
     "a description of the values.\n"
   };
 
+  char *help_batch[2] =
+  {
+    "Help for BATCH\n"
+    ,
+    "The 'BATCH' command permits the client to initiate a bulk load\n"
+    "   of commands to rrdcached.\n"
+    "\n"
+    "Usage:\n"
+    "\n"
+    "    client: BATCH\n"
+    "    server: 0 Go ahead.  End with dot '.' on its own line.\n"
+    "    client: command #1\n"
+    "    client: command #2\n"
+    "    client: ... and so on\n"
+    "    client: .\n"
+    "    server: 2 errors\n"
+    "    server: 7 message for command #7\n"
+    "    server: 9 message for command #9\n"
+    "\n"
+    "For more information, consult the rrdcached(1) documentation.\n"
+  };
+
   status = buffer_get_field (&buffer, &buffer_size, &command);
   if (status != 0)
     help_text = help_help;
@@ -1027,8 +1174,14 @@ static int handle_request_help (listen_socket_t *sock, /* {{{ */
       help_text = help_flush;
     else if (strcasecmp (command, "flushall") == 0)
       help_text = help_flushall;
+    else if (strcasecmp (command, "pending") == 0)
+      help_text = help_pending;
+    else if (strcasecmp (command, "forget") == 0)
+      help_text = help_forget;
     else if (strcasecmp (command, "stats") == 0)
       help_text = help_stats;
+    else if (strcasecmp (command, "batch") == 0)
+      help_text = help_batch;
     else
       help_text = help_help;
   }
@@ -1088,7 +1241,7 @@ static int handle_request_stats (listen_socket_t *sock) /* {{{ */
 static int handle_request_flush (listen_socket_t *sock, /* {{{ */
     char *buffer, size_t buffer_size)
 {
-  char *file;
+  char *file, file_tmp[PATH_MAX];
   int status;
 
   status = buffer_get_field (&buffer, &buffer_size, &file);
@@ -1102,6 +1255,7 @@ static int handle_request_flush (listen_socket_t *sock, /* {{{ */
     stats_flush_received++;
     pthread_mutex_unlock(&stats_lock);
 
+    get_abs_path(&file, file_tmp);
     if (!check_file_access(file, sock)) return 0;
 
     status = flush_file (file);
@@ -1126,10 +1280,15 @@ static int handle_request_flush (listen_socket_t *sock, /* {{{ */
 
   /* NOTREACHED */
   assert(1==0);
-} /* }}} int handle_request_slurp */
+} /* }}} int handle_request_flush */
 
 static int handle_request_flushall(listen_socket_t *sock) /* {{{ */
 {
+  int status;
+
+  status = has_privilege(sock, PRIV_HIGH);
+  if (status <= 0)
+    return status;
 
   RRDD_LOG(LOG_DEBUG, "Received FLUSHALL");
 
@@ -1140,17 +1299,94 @@ static int handle_request_flushall(listen_socket_t *sock) /* {{{ */
   return send_response(sock, RESP_OK, "Started flush.\n");
 } /* }}} static int handle_request_flushall */
 
+static int handle_request_pending(listen_socket_t *sock, /* {{{ */
+                                  char *buffer, size_t buffer_size)
+{
+  int status;
+  char *file, file_tmp[PATH_MAX];
+  cache_item_t *ci;
+
+  status = buffer_get_field(&buffer, &buffer_size, &file);
+  if (status != 0)
+    return send_response(sock, RESP_ERR,
+                         "Usage: PENDING <filename>\n");
+
+  status = has_privilege(sock, PRIV_HIGH);
+  if (status <= 0)
+    return status;
+
+  get_abs_path(&file, file_tmp);
+
+  pthread_mutex_lock(&cache_lock);
+  ci = g_tree_lookup(cache_tree, file);
+  if (ci == NULL)
+  {
+    pthread_mutex_unlock(&cache_lock);
+    return send_response(sock, RESP_ERR, "%s\n", rrd_strerror(ENOENT));
+  }
+
+  for (int i=0; i < ci->values_num; i++)
+    add_response_info(sock, "%s\n", ci->values[i]);
+
+  pthread_mutex_unlock(&cache_lock);
+  return send_response(sock, RESP_OK, "updates pending\n");
+} /* }}} static int handle_request_pending */
+
+static int handle_request_forget(listen_socket_t *sock, /* {{{ */
+                                 char *buffer, size_t buffer_size)
+{
+  int status;
+  char *file, file_tmp[PATH_MAX];
+
+  status = buffer_get_field(&buffer, &buffer_size, &file);
+  if (status != 0)
+    return send_response(sock, RESP_ERR,
+                         "Usage: FORGET <filename>\n");
+
+  status = has_privilege(sock, PRIV_HIGH);
+  if (status <= 0)
+    return status;
+
+  get_abs_path(&file, file_tmp);
+  if (!check_file_access(file, sock)) return 0;
+
+  pthread_mutex_lock(&cache_lock);
+  status = forget_file(file);
+  pthread_mutex_unlock(&cache_lock);
+
+  if (status == 0)
+  {
+    if (sock != NULL)
+      journal_write("forget", file);
+
+    return send_response(sock, RESP_OK, "Gone!\n");
+  }
+  else
+    return send_response(sock, RESP_ERR, "cannot forget: %s\n",
+                         status < 0 ? "Internal error" : rrd_strerror(status));
+
+  /* NOTREACHED */
+  assert(1==0);
+} /* }}} static int handle_request_forget */
+
 static int handle_request_update (listen_socket_t *sock, /* {{{ */
-    char *buffer, size_t buffer_size)
+                                  time_t now,
+                                  char *buffer, size_t buffer_size)
 {
-  char *file;
+  char *file, file_tmp[PATH_MAX];
   int values_num = 0;
+  int bad_timestamps = 0;
   int status;
+  char orig_buf[CMD_MAX];
 
-  time_t now;
   cache_item_t *ci;
 
-  now = time (NULL);
+  status = has_privilege(sock, PRIV_HIGH);
+  if (status <= 0)
+    return status;
+
+  /* save it for the journal later */
+  strncpy(orig_buf, buffer, sizeof(orig_buf)-1);
 
   status = buffer_get_field (&buffer, &buffer_size, &file);
   if (status != 0)
@@ -1161,6 +1397,7 @@ static int handle_request_update (listen_socket_t *sock, /* {{{ */
   stats_updates_received++;
   pthread_mutex_unlock(&stats_lock);
 
+  get_abs_path(&file, file_tmp);
   if (!check_file_access(file, sock)) return 0;
 
   pthread_mutex_lock (&cache_lock);
@@ -1219,10 +1456,16 @@ static int handle_request_update (listen_socket_t *sock, /* {{{ */
   } /* }}} */
   assert (ci != NULL);
 
+  /* don't re-write updates in replay mode */
+  if (sock != NULL)
+    journal_write("update", orig_buf);
+
   while (buffer_size > 0)
   {
     char **temp;
     char *value;
+    time_t stamp;
+    char *eostamp;
 
     status = buffer_get_field (&buffer, &buffer_size, &value);
     if (status != 0)
@@ -1231,6 +1474,26 @@ static int handle_request_update (listen_socket_t *sock, /* {{{ */
       break;
     }
 
+    /* make sure update time is always moving forward */
+    stamp = strtol(value, &eostamp, 10);
+    if (eostamp == value || eostamp == NULL || *eostamp != ':')
+    {
+      ++bad_timestamps;
+      add_response_info(sock, "Cannot find timestamp in '%s'!\n", value);
+      continue;
+    }
+    else if (stamp <= ci->last_update_stamp)
+    {
+      ++bad_timestamps;
+      add_response_info(sock,
+                        "illegal attempt to update using time %ld when"
+                        " last update time is %ld (minimum one second step)\n",
+                        stamp, ci->last_update_stamp);
+      continue;
+    }
+    else
+      ci->last_update_stamp = stamp;
+
     temp = (char **) realloc (ci->values,
         sizeof (char *) * (ci->values_num + 1));
     if (temp == NULL)
@@ -1261,9 +1524,21 @@ static int handle_request_update (listen_socket_t *sock, /* {{{ */
   pthread_mutex_unlock (&cache_lock);
 
   if (values_num < 1)
-    return send_response(sock, RESP_ERR, "No values updated.\n");
+  {
+    /* if we had only one update attempt, then return the full
+       error message... try to get the most information out
+       of the limited error space allowed by the protocol
+    */
+    if (bad_timestamps == 1)
+      return send_response(sock, RESP_ERR, "%s", sock->wbuf);
+    else
+      return send_response(sock, RESP_ERR,
+                           "No values updated (%d bad timestamps).\n",
+                           bad_timestamps);
+  }
   else
-    return send_response(sock, RESP_OK, "Enqueued %i value(s).\n", values_num);
+    return send_response(sock, RESP_OK,
+                         "errors, enqueued %i value(s).\n", values_num);
 
   /* NOTREACHED */
   assert(1==0);
@@ -1273,7 +1548,7 @@ static int handle_request_update (listen_socket_t *sock, /* {{{ */
 /* we came across a "WROTE" entry during journal replay.
  * throw away any values that we have accumulated for this file
  */
-static int handle_request_wrote (const char *buffer) /* {{{ */
+static int handle_request_wrote (const char *buffer, time_t now) /* {{{ */
 {
   int i;
   cache_item_t *ci;
@@ -1296,28 +1571,40 @@ static int handle_request_wrote (const char *buffer) /* {{{ */
     free(ci->values);
   }
 
-  wipe_ci_values(ci, time(NULL));
+  wipe_ci_values(ci, now);
   remove_from_queue(ci);
 
   pthread_mutex_unlock(&cache_lock);
   return (0);
 } /* }}} int handle_request_wrote */
 
-/* returns 1 if we have the required privilege level */
-static int has_privilege (listen_socket_t *sock, /* {{{ */
-                          socket_privilege priv)
+/* start "BATCH" processing */
+static int batch_start (listen_socket_t *sock) /* {{{ */
 {
-  if (sock == NULL) /* journal replay */
-    return 1;
+  int status;
+  if (sock->batch_start)
+    return send_response(sock, RESP_ERR, "Already in BATCH\n");
 
-  if (sock->privilege >= priv)
-    return 1;
+  status = send_response(sock, RESP_OK,
+                         "Go ahead.  End with dot '.' on its own line.\n");
+  sock->batch_start = time(NULL);
+  sock->batch_cmd = 0;
 
-  return send_response(sock, RESP_ERR, "%s\n", rrd_strerror(EACCES));
-} /* }}} static int has_privilege */
+  return status;
+} /* }}} static int batch_start */
+
+/* finish "BATCH" processing and return results to the client */
+static int batch_done (listen_socket_t *sock) /* {{{ */
+{
+  assert(sock->batch_start);
+  sock->batch_start = 0;
+  sock->batch_cmd  = 0;
+  return send_response(sock, RESP_OK, "errors\n");
+} /* }}} static int batch_done */
 
 /* if sock==NULL, we are in journal replay mode */
 static int handle_request (listen_socket_t *sock, /* {{{ */
+                           time_t now,
                            char *buffer, size_t buffer_size)
 {
   char *buffer_ptr;
@@ -1335,37 +1622,32 @@ static int handle_request (listen_socket_t *sock, /* {{{ */
     return (-1);
   }
 
-  if (strcasecmp (command, "update") == 0)
-  {
-    status = has_privilege(sock, PRIV_HIGH);
-    if (status <= 0)
-      return status;
-
-    /* don't re-write updates in replay mode */
-    if (sock != NULL)
-      journal_write(command, buffer_ptr);
+  if (sock != NULL && sock->batch_start)
+    sock->batch_cmd++;
 
-    return (handle_request_update (sock, buffer_ptr, buffer_size));
-  }
+  if (strcasecmp (command, "update") == 0)
+    return (handle_request_update (sock, now, buffer_ptr, buffer_size));
   else if (strcasecmp (command, "wrote") == 0 && sock == NULL)
   {
     /* this is only valid in replay mode */
-    return (handle_request_wrote (buffer_ptr));
+    return (handle_request_wrote (buffer_ptr, now));
   }
   else if (strcasecmp (command, "flush") == 0)
     return (handle_request_flush (sock, buffer_ptr, buffer_size));
   else if (strcasecmp (command, "flushall") == 0)
-  {
-    status = has_privilege(sock, PRIV_HIGH);
-    if (status <= 0)
-      return status;
-
     return (handle_request_flushall(sock));
-  }
+  else if (strcasecmp (command, "pending") == 0)
+    return (handle_request_pending(sock, buffer_ptr, buffer_size));
+  else if (strcasecmp (command, "forget") == 0)
+    return (handle_request_forget(sock, buffer_ptr, buffer_size));
   else if (strcasecmp (command, "stats") == 0)
     return (handle_request_stats (sock));
   else if (strcasecmp (command, "help") == 0)
     return (handle_request_help (sock, buffer_ptr, buffer_size));
+  else if (strcasecmp (command, "batch") == 0 && sock != NULL)
+    return batch_start(sock);
+  else if (strcasecmp (command, ".") == 0 && sock != NULL && sock->batch_start)
+    return batch_done(sock);
   else
     return send_response(sock, RESP_ERR, "Unknown command: %s\n", command);
 
@@ -1377,6 +1659,7 @@ static int handle_request (listen_socket_t *sock, /* {{{ */
 static void journal_rotate(void) /* {{{ */
 {
   FILE *old_fh = NULL;
+  int new_fd;
 
   if (journal_cur == NULL || journal_old == NULL)
     return;
@@ -1391,11 +1674,20 @@ static void journal_rotate(void) /* {{{ */
   if (journal_fh != NULL)
   {
     old_fh = journal_fh;
+    journal_fh = NULL;
     rename(journal_cur, journal_old);
     ++stats_journal_rotate;
   }
 
-  journal_fh = fopen(journal_cur, "a");
+  new_fd = open(journal_cur, O_WRONLY|O_CREAT|O_APPEND,
+                S_IRUSR|S_IWUSR|S_IRGRP|S_IROTH);
+  if (new_fd >= 0)
+  {
+    journal_fh = fdopen(new_fd, "a");
+    if (journal_fh == NULL)
+      close(new_fd);
+  }
+
   pthread_mutex_unlock(&journal_lock);
 
   if (old_fh != NULL)
@@ -1470,9 +1762,48 @@ static int journal_replay (const char *file) /* {{{ */
   int fail_cnt = 0;
   uint64_t line = 0;
   char entry[CMD_MAX];
+  time_t now;
 
   if (file == NULL) return 0;
 
+  {
+    char *reason;
+    int status = 0;
+    struct stat statbuf;
+
+    memset(&statbuf, 0, sizeof(statbuf));
+    if (stat(file, &statbuf) != 0)
+    {
+      if (errno == ENOENT)
+        return 0;
+
+      reason = "stat error";
+      status = errno;
+    }
+    else if (!S_ISREG(statbuf.st_mode))
+    {
+      reason = "not a regular file";
+      status = EPERM;
+    }
+    if (statbuf.st_uid != daemon_uid)
+    {
+      reason = "not owned by daemon user";
+      status = EACCES;
+    }
+    if (statbuf.st_mode & (S_IWGRP|S_IWOTH))
+    {
+      reason = "must not be user/group writable";
+      status = EACCES;
+    }
+
+    if (status != 0)
+    {
+      RRDD_LOG(LOG_ERR, "journal_replay: %s : %s (%s)",
+               file, rrd_strerror(status), reason);
+      return 0;
+    }
+  }
+
   fh = fopen(file, "r");
   if (fh == NULL)
   {
@@ -1484,6 +1815,8 @@ static int journal_replay (const char *file) /* {{{ */
   else
     RRDD_LOG(LOG_NOTICE, "replaying from journal: %s", file);
 
+  now = time(NULL);
+
   while(!feof(fh))
   {
     size_t entry_len;
@@ -1505,7 +1838,7 @@ static int journal_replay (const char *file) /* {{{ */
 
     entry[entry_len - 1] = '\0';
 
-    if (handle_request(NULL, entry, entry_len) == 0)
+    if (handle_request(NULL, now, entry, entry_len) == 0)
       ++entry_cnt;
     else
       ++fail_cnt;
@@ -1513,17 +1846,36 @@ static int journal_replay (const char *file) /* {{{ */
 
   fclose(fh);
 
-  if (entry_cnt > 0)
-  {
-    RRDD_LOG(LOG_INFO, "Replayed %d entries (%d failures)",
-             entry_cnt, fail_cnt);
-    return 1;
-  }
-  else
-    return 0;
+  RRDD_LOG(LOG_INFO, "Replayed %d entries (%d failures)",
+           entry_cnt, fail_cnt);
 
+  return entry_cnt > 0 ? 1 : 0;
 } /* }}} static int journal_replay */
 
+static void journal_init(void) /* {{{ */
+{
+  int had_journal = 0;
+
+  if (journal_cur == NULL) return;
+
+  pthread_mutex_lock(&journal_lock);
+
+  RRDD_LOG(LOG_INFO, "checking for journal files");
+
+  had_journal += journal_replay(journal_old);
+  had_journal += journal_replay(journal_cur);
+
+  /* it must have been a crash.  start a flush */
+  if (had_journal && config_flush_at_shutdown)
+    flush_old_values(-1);
+
+  pthread_mutex_unlock(&journal_lock);
+  journal_rotate();
+
+  RRDD_LOG(LOG_INFO, "journal processing complete");
+
+} /* }}} static void journal_init */
+
 static void close_connection(listen_socket_t *sock)
 {
   close(sock->fd) ;  sock->fd   = -1;
@@ -1577,6 +1929,7 @@ static void *connection_thread_main (void *args) /* {{{ */
     char *cmd;
     ssize_t cmd_len;
     ssize_t rbytes;
+    time_t now;
 
     struct pollfd pollfd;
     int status;
@@ -1593,23 +1946,18 @@ static void *connection_thread_main (void *args) /* {{{ */
     else if (status < 0) /* error */
     {
       status = errno;
-      if (status == EINTR)
-        continue;
-      RRDD_LOG (LOG_ERR, "connection_thread_main: poll(2) failed.");
+      if (status != EINTR)
+        RRDD_LOG (LOG_ERR, "connection_thread_main: poll(2) failed.");
       continue;
     }
 
     if ((pollfd.revents & POLLHUP) != 0) /* normal shutdown */
-    {
-      close_connection(sock);
       break;
-    }
     else if ((pollfd.revents & (POLLIN | POLLPRI)) == 0)
     {
       RRDD_LOG (LOG_WARNING, "connection_thread_main: "
           "poll(2) returned something unexpected: %#04hx",
           pollfd.revents);
-      close_connection(sock);
       break;
     }
 
@@ -1625,9 +1973,14 @@ static void *connection_thread_main (void *args) /* {{{ */
 
     sock->next_read += rbytes;
 
+    if (sock->batch_start)
+      now = sock->batch_start;
+    else
+      now = time(NULL);
+
     while ((cmd = next_cmd(sock, &cmd_len)) != NULL)
     {
-      status = handle_request (sock, cmd, cmd_len+1);
+      status = handle_request (sock, now, cmd, cmd_len+1);
       if (status != 0)
         goto out_close;
     }
@@ -2003,11 +2356,16 @@ static void *listen_thread_main (void *args __attribute__((unused))) /* {{{ */
 static int daemonize (void) /* {{{ */
 {
   int status;
-  int fd;
+  int pid_fd;
   char *base_dir;
 
-  fd = open_pidfile();
-  if (fd < 0) return fd;
+  daemon_uid = geteuid();
+
+  pid_fd = open_pidfile("create", O_CREAT|O_EXCL|O_WRONLY);
+  if (pid_fd < 0)
+    pid_fd = check_pidfile();
+  if (pid_fd < 0)
+    return pid_fd;
 
   if (!stay_foreground)
   {
@@ -2060,7 +2418,7 @@ static int daemonize (void) /* {{{ */
     return (-1);
   }
 
-  status = write_pidfile (fd);
+  status = write_pidfile (pid_fd);
   return status;
 } /* }}} int daemonize */
 
@@ -2176,6 +2534,7 @@ static int read_options (int argc, char **argv) /* {{{ */
       case 'b':
       {
         size_t len;
+        char base_realpath[PATH_MAX];
 
         if (config_base_dir != NULL)
           free (config_base_dir);
@@ -2186,6 +2545,27 @@ static int read_options (int argc, char **argv) /* {{{ */
           return (3);
         }
 
+        /* make sure that the base directory is not resolved via
+         * symbolic links.  this makes some performance-enhancing
+         * assumptions possible (we don't have to resolve paths
+         * that start with a "/")
+         */
+        if (realpath(config_base_dir, base_realpath) == NULL)
+        {
+          fprintf (stderr, "Invalid base directory '%s'.\n", config_base_dir);
+          return 5;
+        }
+        else if (strncmp(config_base_dir,
+                         base_realpath, sizeof(base_realpath)) != 0)
+        {
+          fprintf(stderr,
+                  "Base directory (-b) resolved via file system links!\n"
+                  "Please consult rrdcached '-b' documentation!\n"
+                  "Consider specifying the real directory (%s)\n",
+                  base_realpath);
+          return 5;
+        }
+
         len = strlen (config_base_dir);
         while ((len > 0) && (config_base_dir[len - 1] == '/'))
         {
@@ -2330,25 +2710,7 @@ int main (int argc, char **argv)
     return (1);
   }
 
-  if (journal_cur != NULL)
-  {
-    int had_journal = 0;
-
-    pthread_mutex_lock(&journal_lock);
-
-    RRDD_LOG(LOG_INFO, "checking for journal files");
-
-    had_journal += journal_replay(journal_old);
-    had_journal += journal_replay(journal_cur);
-
-    if (had_journal)
-      flush_old_values(-1);
-
-    pthread_mutex_unlock(&journal_lock);
-    journal_rotate();
-
-    RRDD_LOG(LOG_INFO, "journal processing complete");
-  }
+  journal_init();
 
   /* start the queue thread */
   memset (&queue_thread, 0, sizeof (queue_thread));