lordgamez commented on code in PR #1340:
URL: https://github.com/apache/nifi-minifi-cpp/pull/1340#discussion_r905120725


##########
libminifi/include/core/state/nodes/RepositoryMetrics.h:
##########
@@ -90,15 +87,18 @@ class RepositoryMetrics : public ResponseNode {
     return serialized;
   }
 
+  std::vector<PublishedMetric> calculateMetrics() override {
+    std::vector<PublishedMetric> metrics;
+    for (const auto& [_, repo] : repositories_) {
+      metrics.push_back({"is_running", (repo->isRunning() ? 1.0 : 0.0), 
{{"metric_class", getName()}, {"repository_name", repo->getName()}}});
+      metrics.push_back({"is_full", (repo->isFull() ? 1.0 : 0.0), 
{{"metric_class", getName()}, {"repository_name", repo->getName()}}});
+      metrics.push_back({"repository_size", 
static_cast<double>(repo->getRepoSize()), {{"metric_class", getName()}, 
{"repository_name", repo->getName()}}});
+    }
+    return metrics;
+  }

Review Comment:
   Could you elaborate a bit more please? What do you think is currently too 
Prometheus specific format here that should be changed and transformed later? 
Also what do you mean by unifying `serialize`? The serialized nodes are not 
used by Prometheus only on the C2 protocol.



##########
libminifi/include/properties/Configuration.h:
##########
@@ -156,12 +156,17 @@ class Configuration : public Properties {
 
   static constexpr const char *nifi_asset_directory = "nifi.asset.directory";
 
+  // Metrics publisher options
+  static constexpr const char *nifi_metrics_publisher_class = 
"nifi.metrics.publisher.class";
+  static constexpr const char 
*nifi_metrics_publisher_prometheus_metrics_publisher_port = 
"nifi.metrics.publisher.PrometheusMetricsPublisher.port";

Review Comment:
   The problem is that all possible configuration options sent through C2 are 
retrieved from here, so if we want to advertise the prometheus options as well, 
they should be added here.



##########
METRICS.md:
##########
@@ -0,0 +1,155 @@
+<!--
+  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.
+-->
+
+# Apache NiFi - MiNiFi - C++ Metrics Readme.
+
+
+This readme defines the metrics published by Apache NiFi. All options defined 
are located in minifi.properties.
+
+## Table of Contents
+
+- [Description](#description)
+- [Configuration](#configuration)
+- [Metrics](#metrics)
+
+## Description
+
+Apache NiFi MiNiFi C++ can communicate metrics about the agent's status, that 
can be a system level or component level metric.
+These metrics are exposed through the agent implemented metric publishers that 
can be configured in the minifi.properties.
+Aside from the publisher exposed metrics, metrics are also sent through C2 
protocol of which there is more information in the
+[C2 documentation](C2.md#metrics).
+
+## Configuration
+
+To configure the a metrics publisher first we have to set which publisher 
class should be used:
+
+       # in minifi.properties
+
+       nifi.metrics.publisher.class=PrometheusMetricsPublisher
+
+Currently PrometheusMetricsPublisher is the only available publisher in MiNiFi 
C++ which publishes metrics to a Prometheus server.
+To use the publisher a port should also be configured where the metrics will 
be available to be scraped through:
+
+       # in minifi.properties
+
+       nifi.metrics.publisher.PrometheusMetricsPublisher.port=9936
+
+The last option defines which metric classes should be exposed through the 
metrics publisher in configured with a comma separated value:
+
+       # in minifi.properties
+
+       
nifi.metrics.publisher.metrics=QueueMetrics,RepositoryMetrics,GetFileMetrics,DeviceInfoNode,FlowInformation
+
+## Metrics
+
+The following section defines the currently available metrics to be published 
by the MiNiFi C++ agent.
+
+NOTE: In Prometheus all metrics are extended with a `minifi_` prefix to mark 
the domain of the metric. For example the `connection_name` metric is published 
as `minifi_connection_name` in Prometheus.
+
+### QueueMetrics
+
+QueueMetrics is a system level metric that reports queue metrics for every 
connection in the flow.
+
+| Metric name          | Labels                                         | 
Description                                |
+|----------------------|------------------------------------------------|--------------------------------------------|
+| queue_data_size      | metric_class, connection_uuid, connection_name | Max 
queue size to apply back pressure      |
+| queue_data_size_max  | metric_class, connection_uuid, connection_name | Max 
queue data size to apply back pressure |
+| queue_size           | metric_class, connection_uuid, connection_name | 
Current queue size                         |
+| queue_size_max       | metric_class, connection_uuid, connection_name | 
Current queue data size                    |

Review Comment:
   What would you suggest chaning in the names? The 'minifi' prefix is applied 
in the Prometheus extension, I intended these names to be used in other metric 
collectors as well, so I only added a note above that in Prometheus these are 
extended with the `minifi` prefix. Do you think we should add a suffix here for 
the units?



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscr...@nifi.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org

Reply via email to