46 lines
2.4 KiB
Python
46 lines
2.4 KiB
Python
from huaweicloudsdkcore.auth.credentials import BasicCredentials
|
|
from huaweicloudsdkces.v1.region.ces_region import CesRegion
|
|
from huaweicloudsdkces.v1 import *
|
|
from datetime import datetime
|
|
from units import consul_kv
|
|
|
|
def exporter(vendor,account,region):
|
|
ak,sk = consul_kv.get_aksk(vendor,account)
|
|
credentials = BasicCredentials(ak, sk)
|
|
client = CesClient.new_builder() \
|
|
.with_credentials(credentials) \
|
|
.with_region(CesRegion.value_of(region)) \
|
|
.build()
|
|
metric_name_dict = {"cpu_usage":["# HELP redis_cpu_util CPU使用率","# TYPE redis_cpu_util gauge"],
|
|
"memory_usage":["# HELP redis_mem_util 内存使用率","# TYPE redis_mem_util gauge"],
|
|
"keyspace_hits_perc":["# HELP redis_hits_util 缓存命中率","# TYPE redis_hits_util gauge"],
|
|
"total_connections_received":["# HELP redis_newconn_count 每分钟新建的连接数","# TYPE redis_newconn_count gauge"],
|
|
"rx_controlled":["# HELP redis_rx_controlled 每分钟被流控的次数","# TYPE redis_rx_controlled gauge"],
|
|
"is_slow_log_exist":["# HELP redis_slow_log 慢日志情况","# TYPE redis_slow_log gauge"],
|
|
}
|
|
metric_body_list = []
|
|
now = int(datetime.now().timestamp()*1000)
|
|
redis_list = consul_kv.get_services_list_by_region(f'{vendor}_{account}_redis',region)
|
|
for i in metric_name_dict.keys():
|
|
for id in redis_list:
|
|
metric_body_list.append(MetricInfo(namespace="SYS.DCS",metric_name=i,dimensions=[MetricsDimension(name="dcs_instance_id",value=id)]))
|
|
|
|
request = BatchListMetricDataRequest()
|
|
request.body = BatchListMetricDataRequestBody(to=now,_from=now-180000,filter="max",period="1",metrics=metric_body_list)
|
|
response = client.batch_list_metric_data(request).to_dict()
|
|
for i in response['metrics']:
|
|
id= i['dimensions'][0]['value']
|
|
try:
|
|
value = i['datapoints'][-1]['max']
|
|
ts = i['datapoints'][-1]['timestamp']
|
|
except:
|
|
value = -1
|
|
ts = now
|
|
metric = i['metric_name']
|
|
prom_metric_name = metric_name_dict[metric][0].split()[2]
|
|
metric_name_dict[metric].append(f'{prom_metric_name}{{iid="{id}"}} {float(value)} {ts}')
|
|
prom_metric_list = []
|
|
for x in metric_name_dict.values():
|
|
prom_metric_list = prom_metric_list + x
|
|
return prom_metric_list
|