grafana influx quering

analytics/analytic_unit_worker.py

@@ -44,12 +44,13 @@ class AnalyticUnitWorker(object):
     async def do_learn(self, analytic_unit_id, payload):
         pattern = payload['pattern']
         segments = payload['segments']
+        data = payload['data'] # [time, value][]
 
-        model = self.get_detector(analytic_unit_id, pattern)
-        model.synchronize_data()
-        last_prediction_time = await model.learn(segments)
+        detector = self.get_detector(analytic_unit_id, pattern)
+        detector.synchronize_data()
+        last_prediction_time = await detector.learn(segments)
         # TODO: we should not do predict before labeling in all models, not just in drops
-        
+
         if pattern == 'DROP' and len(segments) == 0:
             # TODO: move result to a class which renders to json for messaging to analytics
             result = {
@@ -68,9 +69,9 @@ class AnalyticUnitWorker(object):
         pattern = payload['pattern']
         last_prediction_time = payload['lastPredictionTime']
 
-        model = self.get_detector(analytic_unit_id, pattern)
-        model.synchronize_data()
-        segments, last_prediction_time = await model.predict(last_prediction_time)
+        detector = self.get_detector(analytic_unit_id, pattern)
+        detector.synchronize_data()
+        segments, last_prediction_time = await detector.predict(last_prediction_time)
         return {
             'task': 'PREDICT',
             'status': 'SUCCESS',

analytics/detectors/general_detector/general_detector.py

@@ -84,10 +84,6 @@ class GeneralDetector:
         logger.info("Predicting is finished for anomaly type='%s'" % self.anomaly_name)
         return predicted_anomalies, last_prediction_time
 
-    def synchronize_data(self):
-        self.data_prov.synchronize()
-        self.preprocessor.set_data_provider(self.data_prov)
-        self.preprocessor.synchronize()
 
     def create_algorithm(self):
         return SupervisedAlgorithm()

analytics/detectors/pattern_detector.py

@@ -37,19 +37,16 @@ class PatternDetector:
         self.model = None
         self.__load_model(pattern_type)
 
-    async def learn(self, segments):
+    async def learn(self, segments, data):
         self.model = resolve_model_by_pattern(self.pattern_type)
         window_size = 200
 
-        dataframe = self.data_prov.get_dataframe()
-
-        segments = self.data_prov.transform_anomalies(segments)
         # TODO: pass only part of dataframe that has segments
-        self.model.fit(dataframe, segments)
+        self.model.fit(dataframe, segments, data)
         self.__save_model()
         return 0
 
-    async def predict(self, last_prediction_time):
+    async def predict(self, last_prediction_time, data):
         if self.model is None:
             return [], last_prediction_time
 
@@ -78,9 +75,6 @@ class PatternDetector:
         return segments, last_prediction_time
         # return predicted_anomalies, last_prediction_time
 
-    def synchronize_data(self):
-        self.data_prov.synchronize()
-
     def __save_model(self):
         pass
         # TODO: use data_service to save anything

server/package.json

@@ -24,6 +24,7 @@
     "@types/koa": "^2.0.45",
     "@types/koa-bodyparser": "^4.2.0",
     "@types/koa-router": "^7.0.28",
+    "@types/lodash": "^4.14.116",
     "@types/nedb": "^1.8.0",
     "axios": "^0.18.0",
     "babel-core": "^6.26.3",
@@ -37,6 +38,7 @@
     "koa": "^2.5.1",
     "koa-bodyparser": "^4.2.1",
     "koa-router": "^7.4.0",
+    "lodash": "^4.17.10",
     "nedb": "^1.8.0",
     "node-loader": "^0.6.0",
     "nodemon": "^1.17.5",

server/src/controllers/analytics_controller.ts

@@ -5,6 +5,8 @@ import * as AnalyticUnit from '../models/analytic_unit_model';
 import { AnalyticsService } from '../services/analytics_service';
 import { queryByMetric } from '../services/grafana_service';
 
+import * as _ from 'lodash';
+
 
 type TaskResult = any;
 export type TaskResolver = (taskResult: TaskResult) => void;
@@ -60,43 +62,59 @@ export function terminate() {
 }
 
 async function runTask(task: AnalyticsTask): Promise<TaskResult> {
-  // task.metric = {
-  //   datasource: anomaly.metric.datasource,
-  //   targets: anomaly.metric.targets.map(getTarget)
-  // };
-  
   return new Promise<TaskResult>((resolver: TaskResolver) => {
     taskResolvers.set(task.id, resolver); // it will be resolved in onTaskResult()
     analyticsService.sendTask(task);      // we dont wait for result here
   });
 }
 
+/**
+ * Finds range for selecting subset for learning
+ * @param segments labeled segments
+ */
+function getQueryRangeForLearningBySegments(segments: Segment.Segment[]) {
+  if(segments.length < 2) {
+    throw new Error('Need at least 2 labeled segments');
+  }
+
+  let from = _.minBy(segments, s => s.from).from;
+  let to = _.maxBy(segments, s => s.to).to;
+  let diff = to - from;
+  from -= Math.round(diff * 0.1);
+  to += Math.round(diff * 0.1);
+
+  return { from, to };
+}
+
 export async function runLearning(id: AnalyticUnit.AnalyticUnitId) {
-  
+
   let previousLastPredictionTime: number = undefined;
 
   try {
 
-    let segments = await Segment.findMany(id, { labeled: true });
     let analyticUnit = await AnalyticUnit.findById(id);
+    if(analyticUnit.status === AnalyticUnit.AnalyticUnitStatus.LEARNING) {
+      throw new Error('Can`t starn learning when it`s already started [' + id + ']');
+    }
+
+    let segments = await Segment.findMany(id, { labeled: true });
+    if(segments.length < 2) {
+      throw new Error('Need at least 2 labeled segments');
+    }
 
     let segmentObjs = segments.map(s => s.toObject());
-    let data = await queryByMetric(analyticUnit.metric, analyticUnit.panelUrl);
-    
+
+    let { from, to } = getQueryRangeForLearningBySegments(segments);
+    let data = await queryByMetric(analyticUnit.metric, analyticUnit.panelUrl, from, to);
     if(data.length === 0) {
       throw new Error('Empty data to learn on');
     }
 
-    if(analyticUnit.status === AnalyticUnit.AnalyticUnitStatus.LEARNING) {
-      throw new Error('Can`t starn learning when it`s already started [' + id + ']');
-    }
-
-    AnalyticUnit.setStatus(id, AnalyticUnit.AnalyticUnitStatus.LEARNING);
-
     let pattern = analyticUnit.type;
     let task = new AnalyticsTask(
-      id, AnalyticsTaskType.LEARN, { pattern, segments: segmentObjs }
+      id, AnalyticsTaskType.LEARN, { pattern, segments: segmentObjs, data }
     );
+    AnalyticUnit.setStatus(id, AnalyticUnit.AnalyticUnitStatus.LEARNING);
     let result = await runTask(task);
     let { lastPredictionTime, segments: predictedSegments } = await processLearningResult(result);
     previousLastPredictionTime = analyticUnit.lastPredictionTime;

server/src/models/grafana_metric_model.ts

@@ -62,9 +62,11 @@ export class MetricQuery {
   private static INFLUX_QUERY_TIME_REGEX = /time >[^A-Z]+/;
 
   private _queryParts: string[];
+  private _type: string;
 
   constructor(metric: GrafanaMetric) {
-    if (metric.datasource.type !== 'influxdb') {
+    this._type = metric.datasource.type;
+    if (this._type !== 'influxdb') {
       throw new Error(`Queries of type "${metric.datasource.type}" are not supported yet.`);
     }
     var queryStr = metric.datasource.params.q;
@@ -79,7 +81,8 @@ export class MetricQuery {
     }
   }
 
-  getQuery(limit: number, offset: number): string {
-    return `${this._queryParts[0]} TRUE ${this._queryParts[1]} LIMIT ${limit} OFFSET ${offset}`;
+  getQuery(from: number, to: number, limit: number, offset: number): string {
+    let timeClause = `time >= ${from}ms AND time <= ${to}ms`;
+    return `${this._queryParts[0]} ${timeClause} ${this._queryParts[1]} LIMIT ${limit} OFFSET ${offset}`;
   }
 }

server/src/services/grafana_service.ts

@@ -11,47 +11,37 @@ const CHUNK_SIZE = 50000;
  * @param metric to query to Grafana
  * @returns [time, value][] array
  */
-export async function queryByMetric(metric: GrafanaMetric, panelUrl: string): Promise<[number, number][]> {
+export async function queryByMetric(
+  metric: GrafanaMetric, panelUrl: string, from: number, to: number
+): Promise<[number, number][]> {
+
   let datasource = metric.datasource;
 
   let origin = new URL(panelUrl).origin;
   let url = `${origin}/${datasource.url}`;
 
   let params = datasource.params
-  let records = await getRecordsCount(url, params);
-
-  let limit = Math.min(records, CHUNK_SIZE);
-  let offset = 0;
-
   let data = [];
-  while (offset <= records) {
-    let paramsClone = Object.assign({}, params);
-    paramsClone.q = metric.metricQuery.getQuery(limit, offset);
-    let chunk = await queryGrafana(url, paramsClone);
+
+  let chunkParams = Object.assign({}, params);
+  while(true) {
+    chunkParams.q = metric.metricQuery.getQuery(from, to, CHUNK_SIZE, data.length);
+    var chunk = await queryGrafana(url, chunkParams);
     data = data.concat(chunk);
-    offset += CHUNK_SIZE;
+    if(chunk.length < CHUNK_SIZE) {
+      // because if we get less that we could, then there is nothing more
+      break;
+    }
   }
 
   return data;
 }
 
-async function getRecordsCount(url: string, params: any) {
-  let paramsClone = Object.assign({}, params);
-  let query = paramsClone.q;
-
-  let field = query.match(/"(\w+)"\)*\sFROM/)[1];
-  let measurement = query.match(/FROM\s"(\w+)"/)[1];
-  paramsClone.q = `SELECT COUNT(${field}) FROM ${measurement}`;
-  let result = await queryGrafana(url, paramsClone);
-  return result[0][1];
-}
-
 async function queryGrafana(url: string, params: any) {
   let headers = { Authorization: `Bearer ${HASTIC_API_KEY}` };
 
-  let res;
   try {
-    res = await axios.get(url, { params, headers });
+    var res = await axios.get(url, { params, headers });
   } catch (e) {
     if(e.response.status === 401) {
       throw new Error('Unauthorized. Check the $HASTIC_API_KEY.');
@@ -69,5 +59,5 @@ async function queryGrafana(url: string, params: any) {
     return [];
   }
 
-  return results.series[0].values;
+  return results.series[0].values as [number, number][];
 }