forked from TencentBlueKing/bk-itsm
-
Notifications
You must be signed in to change notification settings - Fork 0
/
celery_statsd.py
196 lines (159 loc) · 7.02 KB
/
celery_statsd.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
# -*- coding: utf-8 -*-
"""
参考：
https://docs.celeryproject.org/en/latest/userguide/monitoring.html
https://lxkaka.wang/2018/11/21/celery-monitor/#&gid=1&pid=5
https://www.infoq.com/news/2018/02/monitoring-queues-meilleursagent/
使用方法:
1.启动graphite和grafana: cd monitors && docker-compose up
2.修改`celery_statsd.py`的参数`http_api`为rabbitmq的api地址，默认为：
'http://guest:guest@localhost:15672/api/'
3.启动celery
4.监测worker和队列：python manage.py celery events -c celery_statsd.Camera -F 2.0
5.监测celery事件（任务、心跳等）：python celery_statsd.py
6.打开grafana，并导入`monitors/grafana_dashboard.json`
可选：python flowerd.py --persistent=true --port=6666 --url_prefix=flower --basic_auth=admin:admin
--broker_api=http://guest:guest@localhost:15672/api/
启动flower作为实时监控方案，并提供一定的实时干预能力
statsd上报目录：
celery.workers.{worker_hostname}.*
celery.queues.{queue_name}.tasks
celery.{task.name}.*
任务监控维度：
统计维度 - count
总任务数
失败任务数【主要】
成功任务数
单任务维度 - count
成功数
失败数【主要】
平均耗时/最大耗时【主要】
异常监控维度：
worker
心跳监测
离线监测
queue
堵塞监测
flow
启动异常【start_pipeline/start】
分支执行异常
节点失败异常【service_schedule】
"""
import os
from collections import defaultdict
import django
from celery import states
from celery.events.snapshot import Polaroid
from monitors.utils import Broker
class Camera(Polaroid):
clear_after = True # clear after flush (incl, state.event_count).
def on_shutter(self, state):
if not state.event_count:
# No new events since last snapshot.
return
# print('Workers: {0}'.format(pformat(state.workers, indent=4)))
# print('Tasks: {0}'.format(pformat(state.tasks, indent=4)))
# print('Total: {0.event_count} events, {0.task_count} tasks'.format(state))
proceed_success, proceed_failure = defaultdict(int), defaultdict(int)
for _, task in state.itertasks():
if task.state == states.SUCCESS:
proceed_success[task.worker.hostname] += 1
if task.state == states.FAILURE:
proceed_failure[task.worker.hostname] += 1
for worker in state.workers:
print(worker)
if worker not in proceed_success:
proceed_success[worker] = 0
if worker not in proceed_failure:
proceed_failure[worker] = 0
from statsd.defaults.django import statsd as statsd_client
for hostname, worker in state.workers.items():
worker_name = worker.hostname
statsd_client.gauge(f"celery.workers.{worker_name}.active", worker.active)
statsd_client.gauge(f"celery.workers.{worker_name}.alive", worker.alive)
statsd_client.gauge(f"celery.workers.{worker_name}.processed", worker.processed)
statsd_client.gauge(
f"celery.workers.{worker_name}.processed_succeeded", proceed_success[worker_name],
)
statsd_client.gauge(
f"celery.workers.{worker_name}.processed_failed", proceed_failure[worker_name],
)
# statsd_client.gauge(
# f"celery.workers.{worker_name}.status_string", worker.status_string
# )
from blueapps.core.celery.celery import app as celery_app
try:
broker = Broker(
celery_app.connection().as_uri(include_password=True),
http_api="http://guest:guest@localhost:15672/api/",
)
except NotImplementedError:
return
# inspect active queue names
inspect = celery_app.control.inspect()
queue_names = set()
for worker_name, queues in inspect.active_queues().items():
for q in queues:
queue_names.add(q["name"])
# query broker: rq/redis...
data = defaultdict(int)
for queue in broker.queues(queue_names):
data[queue["name"]] = queue.get("messages", 0)
statsd_client.gauge(f"celery.queues.{queue['name']}.tasks", queue.get("messages", 0))
# print("Queues: {0}".format(pformat(data, indent=4)))
def monitor(celery_app, statsd_client):
state = celery_app.events.State()
def _get_task(event):
# task name is sent only with -received event,
# and state will keep track of this for us.
state.event(event)
return state.tasks.get(event["uuid"])
def on_task_received(event):
task = _get_task(event)
statsd_client.incr(f"celery.{task.name}.received")
# print('TASK RECEIVED: %s[%s] %s' % (task.name, task.uuid, task.info(),))
def on_task_revoked(event):
task = _get_task(event)
statsd_client.incr(f"celery.{task.name}.revoked")
# print('TASK REVOKED: %s[%s] %s' % (task.name, task.uuid, task.info(),))
def on_task_started(event):
task = _get_task(event)
statsd_client.incr(f"celery.{task.name}.started")
# print('TASK STARTED: %s[%s] %s' % (task.name, task.uuid, task.info(),))
def on_task_failed(event):
task = _get_task(event)
statsd_client.incr(f"celery.{task.name}.failed")
# print('TASK FAILED: %s[%s] %s' % (task.name, task.uuid, task.info(),))
def on_task_succeeded(event):
task = _get_task(event)
task_info = task.info()
statsd_client.incr(f"celery.{task.name}.succeeded")
statsd_client.timing(f"celery.{task.name}.runtime", int(task_info["runtime"] * 1000))
# print('TASK SUCCEEDED: %s[%s] %s' % (task.name, task.uuid, task_info,))
def on_worker_online(event):
statsd_client.incr(f'celery.workers.{event["hostname"]}.online')
def on_worker_offline(event):
statsd_client.incr(f'celery.workers.{event["hostname"]}.offline')
def on_worker_heartbeat(event):
statsd_client.incr(f'celery.workers.{event["hostname"]}.heartbeat')
with celery_app.connection() as connection:
recv = celery_app.events.Receiver(
connection,
handlers={
"task-started": on_task_started,
"task-revoked": on_task_revoked,
"task-received": on_task_received,
"task-failed": on_task_failed,
"task-succeeded": on_task_succeeded,
"worker-online": on_worker_online,
"worker-offline": on_worker_offline,
"worker-heartbeat": on_worker_heartbeat,
},
)
recv.capture(limit=None, timeout=None, wakeup=True)
if __name__ == "__main__":
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "settings")
django.setup()
from blueapps.core.celery.celery import app as celery_app
from statsd.defaults.django import statsd as statsd_client
monitor(celery_app, statsd_client)