Mongo进程丢失后,countly运行报错,求解答


#1

报错如下,05:01分mongo primary挂了,但是另外一台被选举为primary并运行正常。05:03分后countly几台机器的CPU一直100%,没有下降。持续半小时后重启countly服务后才正常。

2017-05-18T02:00:00.169Z: INFO [jobs:manager] Job 59151ec8c4d09b1cdc3209d4 is running on another server, won't start it here
2017-05-18T02:11:39.096Z: ERROR [db:read] Error reading jobs {"name":"find","args":[null]} MongoError: read ECONNRESET {"name":"MongoError","message":"read ECONNRESET"}
2017-05-18T02:11:39.097Z: ERROR [jobs:manager] Error while looking for jobs: {"name":"MongoError","message":"read ECONNRESET"}
2017-05-18T03:00:00.555Z: INFO [jobs:manager] Trying to start job {"_id":"59151ec8c4d09b1cdc3209d4","name":"reports:send","created":1494556360744,"status":0,"started":1495072800166,"finished":1495072800187,"duration":21,"schedule":"every 1 hour starting on the 0 min","next":1495076400000,"modified":1495072800208,"error":null}
2017-05-18T04:00:00.118Z: INFO [jobs:manager] Trying to start job {"_id":"59151ec8c4d09b1cdc3209d4","name":"reports:send","created":1494556360744,"status":0,"started":1495076400556,"finished":1495076400965,"duration":409,"schedule":"every 1 hour starting on the 0 min","next":1495080000000,"modified":1495076401208,"error":null}
2017-05-18T05:00:00.973Z: INFO [jobs:manager] Trying to start job {"_id":"59151ec8c4d09b1cdc3209d4","name":"reports:send","created":1494556360744,"status":0,"started":1495080000120,"finished":1495080000498,"duration":378,"schedule":"every 1 hour starting on the 0 min","next":1495083600000,"modified":1495080000716,"error":null}
2017-05-18T05:00:01.539Z: INFO [jobs:manager] Job 59151ec8c4d09b1cdc3209d4 is running on another server, won't start it here


#2

你好,你们使用什么版本的MongoDB? 目前机器部署架构是怎样的? 每天采集的事件量大概多少?


#3

mongodb:3.2.12, mongo三台两台做副本集,一台做仲裁。三台countly,外挂一台nginx做负载接入。目前在做压力测试,TPS在1000左右。机器的配置是4C16G,redhat 7.3。


#4

可能是数据库方面的问题,社区版我们目前不提供这方面的技术支持,如果有意向欢迎升级企业版,我们可以提供更可靠的部署架构方案,生产环境我们的客户中最高有50亿的月度数据采集量。