dbot-metrics 0.1.5

ATN atn dbot metrics package

DBot-server metrics

The metrics client for dbot-server.

It supports collecting and transporting metrics for each dbot-server.

• Currently we have some inner output-plugins, and more, you can define your own ouput-plugins.
• Output-plugins can work at the same time, but we recommand only one.
• Also the input metric-plugin can be defined by yourself.
• Only you need to do is implementing the Plugin-Interface and configuring the file collector.conf.

Installation

pip install dbot_metrics

If you used default collector.conf file from system, you should copy it to /etc

Dependencies

Install leveldb if you want to use the recommand output-plugin leveldb-client

pip install leveldb

Install influxdb-python first if you want to use the inner output-plugin influxdb-client.

$ sudo apt-get install python-influxdb

Get more information, see https://github.com/influxdata/influxdb-python.

Install tinydb if you want to use the inner output-plugin tinydb-client.

pip install tinydb

Install unqlite if you want to use output-plugin unqlite-client

pip install unqlite

Documentation

See https://github.com/ATNIO/AI_market_plan/wiki/metric-heartbeat.

Interface

Metrics support query interface, it receive a json-type request parameter like this:

{ 'range': { 'from': 1532023625, 'to': 1532025313 },
  'targets': [{'table':'student', 'type':'all', 'sentence':''},
              {'table':'teacher', 'type':'sql', 'sentence': 'select * from teacher'},
              {'table':'others', 'type':'columns', 'sentence': 'age|name|no'}]}

and the response format is like this:

[ {
    'name': 'influxdb',
    'content':[ {
                  'time':1532023688,
                  'tags':{'name':'John', 'No':10023},
                  'fields': {'age': 23, 'gender': 'male'} 
                },
                {
                  'time':1532023798,
                  'tags':{'name':'Tom', 'No':10024},
                  'fields': {'age': 25, 'gender': 'male'}
                }
              ]
   }
]

Examples

from dbot_metrics import DBotMetricsCollector
# DBotApiMetric is a special metric in dbot_service_metric defined by yourself
from dbot_service_metric import DBotApiMetric

DBotMetricsCollector().Start("./collector.conf")
m = DBotApiMetric()
DBotMetricsCollector().RegisterMetrics(m)
m.EnableDetailRecord(False)

apiinfo = m.CallBegin("/api/dbot_server/api_call_test", "user")
# ... ... processing
m.CallEnd(apiinfo, 0)

DBotMetricsCollector().Stop()