YAO/YAO-agent
1
0
Fork 0
mirror of https://github.com/newnius/YAO-agent.git synced 2025-06-06 05:21:55 +00:00
Code Issues Releases Wiki Activity

remove docker files

Browse Source
This commit is contained in:
Newnius 2020-05-24 00:09:02 +08:00
parent d606f5989b
commit 660fe82449
2 changed files with 16 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
agent.py
View File

@ -286,8 +286,7 @@ def report():
print("execute failed, ", msg_gpu) print("execute failed, ", msg_gpu)
stats = get_gpu_status() stats = get_gpu_status()
report_msg(stats) report_msg(stats)
t = Thread(target=launch_tasks, name='launch_tasks', args=(stats,)) Thread(target=launch_tasks, name='launch_tasks', args=(stats,)).start()
t.start()
except Exception as e: except Exception as e:
print(e) print(e)
@ -385,19 +384,12 @@ if __name__ == '__main__':
os.environ["TZ"] = 'Asia/Shanghai' os.environ["TZ"] = 'Asia/Shanghai'
if hasattr(time, 'tzset'): if hasattr(time, 'tzset'):
time.tzset() time.tzset()
threads = []
t1 = Thread(target=reporter) Thread(target=reporter).start()
threads.append(t1) Thread(target=listener).start()
t2 = Thread(target=listener)
threads.append(t2)
if EnableEventTrigger == 'true': if EnableEventTrigger == 'true':
t3 = Thread(target=event_trigger) print('start event trigger')
threads.append(t3) Thread(target=event_trigger).start()
# Start all threads while True:
for t in threads: time.sleep(5)
t.start()
# Wait for all of them to finish
for t in threads:
t.join()

26
monitor.py
View File

@ -1,32 +1,29 @@
import os import os
import threading
import time import time
import subprocess import subprocess
import json import json
from xml.dom.minidom import parse from xml.dom.minidom import parse
import xml.dom.minidom import xml.dom.minidom
from kafka import KafkaProducer
import multiprocessing import multiprocessing
import psutil import psutil
import math import math
from executor import launch_tasks import datetime
ClientID = os.getenv('ClientID', 1)
ClientHost = os.getenv('ClientHost', "localhost")
KafkaBrokers = os.getenv('KafkaBrokers', 'localhost:9092').split(',')
def main(): def main():
interval = 5 interval = 1
while True: while True:
try: try:
status, msg_gpu = execute(['nvidia-smi', '-q', '-x', '-f', 'status.xml']) status, msg_gpu = execute(['nvidia-smi', '-q', '-x', '-f', 'status.xml'])
if not status: if not status:
print("execute failed, ", msg_gpu) print("execute failed, ", msg_gpu)
stats = get_gpu_status() stats = get_gpu_status()
report_msg(stats) utils = []
t = threading.Thread(target=launch_tasks, name='launch_tasks',args=(stats,)) for stat in stats:
t.start() utils.append(str(stat['utilization_gpu']))
# report_msg(stats)
t = datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")
print(str(t) + ',' + ','.join(utils))
time.sleep(interval) time.sleep(interval)
except Exception as e: except Exception as e:
print(e) print(e)
@ -85,8 +82,6 @@ def report_msg(stats):
mem = psutil.virtual_memory() mem = psutil.virtual_memory()
post_fields = { post_fields = {
'id': ClientID,
'host': ClientHost,
'status': stats, 'status': stats,
'cpu_num': multiprocessing.cpu_count(), 'cpu_num': multiprocessing.cpu_count(),
'cpu_load': os.getloadavg()[0], 'cpu_load': os.getloadavg()[0],
@ -95,11 +90,6 @@ def report_msg(stats):
} }
data = json.dumps(post_fields) data = json.dumps(post_fields)
producer = KafkaProducer(bootstrap_servers=KafkaBrokers)
future = producer.send('yao', value=data.encode(), partition=0)
result = future.get(timeout=10)
print(result)
if __name__ == '__main__': if __name__ == '__main__':
os.environ["TZ"] = 'Asia/Shanghai' os.environ["TZ"] = 'Asia/Shanghai'