import time import gevent import hashlib import six from six.moves import xrange from . import events from .exception import StopLocust from .log import console_logger STATS_NAME_WIDTH = 60 class RequestStatsAdditionError(Exception): pass class RequestStats(object): def __init__(self): self.entries = {} self.errors = {} self.num_requests = 0 self.num_failures = 0 self.max_requests = None self.last_request_timestamp = None self.start_time = None def get(self, name, method): """ Retrieve a StatsEntry instance by name and method """ entry = self.entries.get((name, method)) if not entry: entry = StatsEntry(self, name, method) self.entries[(name, method)] = entry return entry def aggregated_stats(self, name="Total", full_request_history=False): """ Returns a StatsEntry which is an aggregate of all stats entries within entries. """ total = StatsEntry(self, name, method=None) for r in six.itervalues(self.entries): total.extend(r, full_request_history=full_request_history) return total def reset_all(self): """ Go through all stats entries and reset them to zero """ self.start_time = time.time() self.num_requests = 0 self.num_failures = 0 for r in six.itervalues(self.entries): r.reset() def clear_all(self): """ Remove all stats entries and errors """ self.num_requests = 0 self.num_failures = 0 self.entries = {} self.errors = {} self.max_requests = None self.last_request_timestamp = None self.start_time = None class StatsEntry(object): """ Represents a single stats entry (name and method) """ name = None """ Name (URL) of this stats entry """ method = None """ Method (GET, POST, PUT, etc.) """ num_requests = None """ The number of requests made """ num_failures = None """ Number of failed request """ total_response_time = None """ Total sum of the response times """ min_response_time = None """ Minimum response time """ max_response_time = None """ Maximum response time """ num_reqs_per_sec = None """ A {second => request_count} dict that holds the number of requests made per second """ response_times = None """ A {response_time => count} dict that holds the response time distribution of all the requests. The keys (the response time in ms) are rounded to store 1, 2, ... 9, 10, 20. .. 90, 100, 200 .. 900, 1000, 2000 ... 9000, in order to save memory. This dict is used to calculate the median and percentile response times. """ total_content_length = None """ The sum of the content length of all the requests for this entry """ start_time = None """ Time of the first request for this entry """ last_request_timestamp = None """ Time of the last request for this entry """ def __init__(self, stats, name, method): self.stats = stats self.name = name self.method = method self.reset() def reset(self): self.start_time = time.time() self.num_requests = 0 self.num_failures = 0 self.total_response_time = 0 self.response_times = {} self.min_response_time = None self.max_response_time = 0 self.last_request_timestamp = int(time.time()) self.num_reqs_per_sec = {} self.total_content_length = 0 def log(self, response_time, content_length): self.stats.num_requests += 1 self.num_requests += 1 self._log_time_of_request() self._log_response_time(response_time) # increase total content-length self.total_content_length += content_length def _log_time_of_request(self): t = int(time.time()) self.num_reqs_per_sec[t] = self.num_reqs_per_sec.setdefault(t, 0) + 1 self.last_request_timestamp = t self.stats.last_request_timestamp = t def _log_response_time(self, response_time): self.total_response_time += response_time if self.min_response_time is None: self.min_response_time = response_time self.min_response_time = min(self.min_response_time, response_time) self.max_response_time = max(self.max_response_time, response_time) # to avoid to much data that has to be transfered to the master node when # running in distributed mode, we save the response time rounded in a dict # so that 147 becomes 150, 3432 becomes 3400 and 58760 becomes 59000 if response_time < 100: rounded_response_time = response_time elif response_time < 1000: rounded_response_time = int(round(response_time, -1)) elif response_time < 10000: rounded_response_time = int(round(response_time, -2)) else: rounded_response_time = int(round(response_time, -3)) # increase request count for the rounded key in response time dict self.response_times.setdefault(rounded_response_time, 0) self.response_times[rounded_response_time] += 1 def log_error(self, error): self.num_failures += 1 self.stats.num_failures += 1 key = StatsError.create_key(self.method, self.name, error) entry = self.stats.errors.get(key) if not entry: entry = StatsError(self.method, self.name, error) self.stats.errors[key] = entry entry.occured() @property def fail_ratio(self): try: return float(self.num_failures) / (self.num_requests + self.num_failures) except ZeroDivisionError: if self.num_failures > 0: return 1.0 else: return 0.0 @property def avg_response_time(self): try: return float(self.total_response_time) / self.num_requests except ZeroDivisionError: return 0 @property def median_response_time(self): if not self.response_times: return 0 return median_from_dict(self.num_requests, self.response_times) @property def current_rps(self): if self.stats.last_request_timestamp is None: return 0 slice_start_time = max(self.stats.last_request_timestamp - 12, int(self.stats.start_time or 0)) reqs = [self.num_reqs_per_sec.get(t, 0) for t in range(slice_start_time, self.stats.last_request_timestamp-2)] return avg(reqs) @property def total_rps(self): if not self.stats.last_request_timestamp or not self.stats.start_time: return 0.0 return self.num_requests / max(self.stats.last_request_timestamp - self.stats.start_time, 1) @property def avg_content_length(self): try: return self.total_content_length / self.num_requests except ZeroDivisionError: return 0 def extend(self, other, full_request_history=False): """ Extend the data fro the current StatsEntry with the stats from another StatsEntry instance. If full_request_history is False, we'll only care to add the data from the last 20 seconds of other's stats. The reason for this argument is that extend can be used to generate an aggregate of multiple different StatsEntry instances on the fly, in order to get the *total* current RPS, average response time, etc. """ self.last_request_timestamp = max(self.last_request_timestamp, other.last_request_timestamp) self.start_time = min(self.start_time, other.start_time) self.num_requests = self.num_requests + other.num_requests self.num_failures = self.num_failures + other.num_failures self.total_response_time = self.total_response_time + other.total_response_time self.max_response_time = max(self.max_response_time, other.max_response_time) self.min_response_time = min(self.min_response_time or 0, other.min_response_time or 0) or other.min_response_time self.total_content_length = self.total_content_length + other.total_content_length if full_request_history: for key in other.response_times: self.response_times[key] = self.response_times.get(key, 0) + other.response_times[key] for key in other.num_reqs_per_sec: self.num_reqs_per_sec[key] = self.num_reqs_per_sec.get(key, 0) + other.num_reqs_per_sec[key] else: # still add the number of reqs per seconds the last 20 seconds for i in xrange(other.last_request_timestamp-20, other.last_request_timestamp+1): if i in other.num_reqs_per_sec: self.num_reqs_per_sec[i] = self.num_reqs_per_sec.get(i, 0) + other.num_reqs_per_sec[i] def serialize(self): return { "name": self.name, "method": self.method, "last_request_timestamp": self.last_request_timestamp, "start_time": self.start_time, "num_requests": self.num_requests, "num_failures": self.num_failures, "total_response_time": self.total_response_time, "max_response_time": self.max_response_time, "min_response_time": self.min_response_time, "total_content_length": self.total_content_length, "response_times": self.response_times, "num_reqs_per_sec": self.num_reqs_per_sec, } @classmethod def unserialize(cls, data): obj = cls(None, data["name"], data["method"]) for key in [ "last_request_timestamp", "start_time", "num_requests", "num_failures", "total_response_time", "max_response_time", "min_response_time", "total_content_length", "response_times", "num_reqs_per_sec", ]: setattr(obj, key, data[key]) return obj def get_stripped_report(self): """ Return the serialized version of this StatsEntry, and then clear the current stats. """ report = self.serialize() self.reset() return report def __str__(self): try: fail_percent = (self.num_failures/float(self.num_requests + self.num_failures))*100 except ZeroDivisionError: fail_percent = 0 return (" %-" + str(STATS_NAME_WIDTH) + "s %7d %12s %7d %7d %7d | %7d %7.2f") % ( self.method + " " + self.name, self.num_requests, "%d(%.2f%%)" % (self.num_failures, fail_percent), self.avg_response_time, self.min_response_time or 0, self.max_response_time, self.median_response_time or 0, self.current_rps or 0 ) def get_response_time_percentile(self, percent): """ Get the response time that a certain number of percent of the requests finished within. Percent specified in range: 0.0 - 1.0 """ num_of_request = int((self.num_requests * percent)) processed_count = 0 for response_time in sorted(six.iterkeys(self.response_times), reverse=True): processed_count += self.response_times[response_time] if((self.num_requests - processed_count) <= num_of_request): return response_time def percentile(self, tpl=" %-" + str(STATS_NAME_WIDTH) + "s %8d %6d %6d %6d %6d %6d %6d %6d %6d %6d"): if not self.num_requests: raise ValueError("Can't calculate percentile on url with no successful requests") return tpl % ( str(self.method) + " " + self.name, self.num_requests, self.get_response_time_percentile(0.5), self.get_response_time_percentile(0.66), self.get_response_time_percentile(0.75), self.get_response_time_percentile(0.80), self.get_response_time_percentile(0.90), self.get_response_time_percentile(0.95), self.get_response_time_percentile(0.98), self.get_response_time_percentile(0.99), self.max_response_time ) class StatsError(object): def __init__(self, method, name, error, occurences=0): self.method = method self.name = name self.error = error self.occurences = occurences @classmethod def create_key(cls, method, name, error): key = "%s.%s.%r" % (method, name, error) return hashlib.md5(key.encode('utf-8')).hexdigest() def occured(self): self.occurences += 1 def to_name(self): return "%s %s: %r" % (self.method, self.name, repr(self.error)) def to_dict(self): return { "method": self.method, "name": self.name, "error": repr(self.error), "occurences": self.occurences } @classmethod def from_dict(cls, data): return cls( data["method"], data["name"], data["error"], data["occurences"] ) def avg(values): return sum(values, 0.0) / max(len(values), 1) def median_from_dict(total, count): """ total is the number of requests made count is a dict {response_time: count} """ pos = (total - 1) / 2 for k in sorted(six.iterkeys(count)): if pos < count[k]: return k pos -= count[k] global_stats = RequestStats() """ A global instance for holding the statistics. Should be removed eventually. """ def on_request_success(request_type, name, response_time, response_length): if global_stats.max_requests is not None and (global_stats.num_requests + global_stats.num_failures) >= global_stats.max_requests: raise StopLocust("Maximum number of requests reached") global_stats.get(name, request_type).log(response_time, response_length) def on_request_failure(request_type, name, response_time, exception): if global_stats.max_requests is not None and (global_stats.num_requests + global_stats.num_failures) >= global_stats.max_requests: raise StopLocust("Maximum number of requests reached") global_stats.get(name, request_type).log_error(exception) def on_report_to_master(client_id, data): data["stats"] = [global_stats.entries[key].get_stripped_report() for key in six.iterkeys(global_stats.entries) if not (global_stats.entries[key].num_requests == 0 and global_stats.entries[key].num_failures == 0)] data["errors"] = dict([(k, e.to_dict()) for k, e in six.iteritems(global_stats.errors)]) global_stats.errors = {} def on_slave_report(client_id, data): for stats_data in data["stats"]: entry = StatsEntry.unserialize(stats_data) request_key = (entry.name, entry.method) if not request_key in global_stats.entries: global_stats.entries[request_key] = StatsEntry(global_stats, entry.name, entry.method) global_stats.entries[request_key].extend(entry, full_request_history=True) global_stats.last_request_timestamp = max(global_stats.last_request_timestamp or 0, entry.last_request_timestamp) for error_key, error in six.iteritems(data["errors"]): if error_key not in global_stats.errors: global_stats.errors[error_key] = StatsError.from_dict(error) else: global_stats.errors[error_key].occurences += error["occurences"] events.request_success += on_request_success events.request_failure += on_request_failure events.report_to_master += on_report_to_master events.slave_report += on_slave_report def print_stats(stats): console_logger.info((" %-" + str(STATS_NAME_WIDTH) + "s %7s %12s %7s %7s %7s | %7s %7s") % ('Name', '# reqs', '# fails', 'Avg', 'Min', 'Max', 'Median', 'req/s')) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) total_rps = 0 total_reqs = 0 total_failures = 0 for key in sorted(six.iterkeys(stats)): r = stats[key] total_rps += r.current_rps total_reqs += r.num_requests total_failures += r.num_failures console_logger.info(r) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) try: fail_percent = (total_failures/float(total_reqs))*100 except ZeroDivisionError: fail_percent = 0 console_logger.info((" %-" + str(STATS_NAME_WIDTH) + "s %7d %12s %42.2f") % ('Total', total_reqs, "%d(%.2f%%)" % (total_failures, fail_percent), total_rps)) console_logger.info("") def print_percentile_stats(stats): console_logger.info("Percentage of the requests completed within given times") console_logger.info((" %-" + str(STATS_NAME_WIDTH) + "s %8s %6s %6s %6s %6s %6s %6s %6s %6s %6s") % ('Name', '# reqs', '50%', '66%', '75%', '80%', '90%', '95%', '98%', '99%', '100%')) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) for key in sorted(six.iterkeys(stats)): r = stats[key] if r.response_times: console_logger.info(r.percentile()) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) total_stats = global_stats.aggregated_stats() if total_stats.response_times: console_logger.info(total_stats.percentile()) console_logger.info("") def print_error_report(): if not len(global_stats.errors): return console_logger.info("Error report") console_logger.info(" %-18s %-100s" % ("# occurences", "Error")) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) for error in six.itervalues(global_stats.errors): console_logger.info(" %-18i %-100s" % (error.occurences, error.to_name())) console_logger.info("-" * (80 + STATS_NAME_WIDTH)) console_logger.info("") def store_stats(filename, stats): with open(filename, "w") as f: f.write("path,method,num_requests,num_failures,min_response_time,max_response_time,avg_response_time\n") for k in stats: r = stats[k] try: f.write("%s,%s,%d,%d,%d,%d,%d\n" % (r.name, r.method,r .num_requests, r.num_failures, r.min_response_time, r.max_response_time, r.avg_response_time)) except: pass def stats_printer(): from runners import locust_runner while True: print_stats(locust_runner.request_stats) gevent.sleep(2) def stats_persist(filename): from runners import locust_runner while True: store_stats(filename, locust_runner.request_stats); gevent.sleep(2)