Github user joshelser commented on a diff in the pull request:

    https://github.com/apache/phoenix/pull/242#discussion_r117042628
  
    --- Diff: 
phoenix-queryserver/src/main/java/org/apache/phoenix/queryserver/metrics/PqsMetricsSystem.java
 ---
    @@ -0,0 +1,105 @@
    +/*
    + * Licensed to the Apache Software Foundation (ASF) under one
    + * or more contributor license agreements.  See the NOTICE file
    + * distributed with this work for additional information
    + * regarding copyright ownership.  The ASF licenses this file
    + * to you under the Apache License, Version 2.0 (the
    + * "License"); you may not use this file except in compliance
    + * with the License.  You may obtain a copy of the License at
    + *
    + * http://www.apache.org/licenses/LICENSE-2.0
    + *
    + * Unless required by applicable law or agreed to in writing, software
    + * distributed under the License is distributed on an "AS IS" BASIS,
    + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
    + * See the License for the specific language governing permissions and
    + * limitations under the License.
    + */
    +
    +package org.apache.phoenix.queryserver.metrics;
    +
    +
    +import org.apache.hadoop.conf.Configuration;
    +import org.apache.phoenix.query.QueryServicesOptions;
    +import org.apache.phoenix.queryserver.metrics.sink.PqsFileSink;
    +import org.apache.phoenix.queryserver.metrics.sink.PqsSink;
    +import org.apache.phoenix.queryserver.metrics.sink.PqsSlf4jSink;
    +import org.slf4j.Logger;
    +import org.slf4j.LoggerFactory;
    +
    +import static 
org.apache.phoenix.query.QueryServices.PHOENIX_PQS_FILE_SINK_FILENAME;
    +import static 
org.apache.phoenix.query.QueryServices.PHOENIX_PQS_METRIC_REPORTING_INTERVAL_MS;
    +import static 
org.apache.phoenix.query.QueryServices.PHOENIX_PQS_TYPE_OF_SINK;
    +
    +
    +public class PqsMetricsSystem {
    +
    +    public final static String statementReadMetrics = 
"Statement-RequestReadMetrics";
    +    public final static String overAllReadRequestMetrics = 
"Statement-OverAllReadRequestMetrics";
    +    public final static String connectionWriteMetricsForMutations = 
"Connection-WriteMetricsForMutations";
    +    public final static String connectionReadMetricsForMutations = 
"Connection-ReadMetricsForMutations";
    +
    +    public enum MetricType {
    +        global,
    +        request
    +    }
    +
    +    protected static final Logger LOG = 
LoggerFactory.getLogger(PqsMetricsSystem.class);
    +
    +    public Thread getGlobalMetricThread() {
    +        return globalMetricThread;
    +    }
    +
    +    private Thread globalMetricThread = null;
    +
    +
    +    public PqsMetricsSystem(String sinkType,String fileName, Integer 
reportingInterval){
    +        PqsGlobalMetrics pqsGlobalMetricsToJMX = null;
    +        try {
    +            pqsGlobalMetricsToJMX = new PqsGlobalMetrics(sinkType, 
fileName,reportingInterval);
    +            globalMetricThread = new Thread(pqsGlobalMetricsToJMX);
    +            globalMetricThread.setName("globalMetricsThread");
    +            globalMetricThread.start();
    +        }catch (Exception ex){
    +            LOG.error(" could not instantiate the PQS Metrics System");
    +            if (globalMetricThread!=null) {
    +                try {
    +                    globalMetricThread.interrupt();
    +                } catch (Exception ine) {
    +                    LOG.error(" unable to interrupt the global metrics 
thread",ine);
    +                }
    +            }
    +
    +        }
    +     }
    +
    +    public static PqsSink getSinkObject(String typeOfSink,String filename){
    +        PqsSink pqsSink;
    +        switch(typeOfSink.toLowerCase()) {
    +            case "file":
    +                pqsSink = new PqsFileSink(filename);
    --- End diff --
    
    > But the only interface Phoenix driver exposes is via static methods in 
PhoenixRuntime, which provides aggregate values for various metrics. So, it is 
difficult to use Hadoop metrics or Codahale.
    
    I can understand why you can't use it directly, but I'm not seeing why you 
couldn't write a layer to convert from the Phoenix internal representation to 
the metrics system of choice (a shim).
    
    With Phoenix, I don't think we should be in the business of writing our own 
metrics reporting code. It's debt for the project as well as more "tribal 
knowledge" that our users have to learn to access these metrics instead of 
metrics automatically showing up in aggregated stores (such as Ambari Metrics 
System via Hadoop Metrics2).
    
    Again, sorry I did not speak up about this when you put up a design doc. 
Curious what others think.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---

Reply via email to