Merge pull request #115 from tarantool/114-iproto-overall

Add overall net requests panel for Prometheus

Author: DifferentialOrange

CHANGELOG.md

@@ -5,6 +5,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## Unreleased
+### Added
+- Overall net requests panel for Prometheus dashboard
+
 ### Changed
 - Use green color for low values in overall panels
 - Configure default collector in example cluster

dashboard/panels/cluster.libsonnet

@@ -162,7 +162,7 @@ local prometheus = grafana.prometheus;
     decimals=0,
     unit='none',
     expr=std.format('sum(up{job=~"%s"})', job),
-  ) { gridPos: { w: 8, h: 3 } },
+  ) { gridPos: { w: 6, h: 3 } },
 
   memory_used_stat(
     title='',
@@ -193,7 +193,7 @@ local prometheus = grafana.prometheus;
     decimals=2,
     unit='bytes',
     expr=std.format('sum(tnt_slab_arena_used{job=~"%s"})', job),
-  ) { gridPos: { w: 4, h: 5 } },
+  ) { gridPos: { w: 3, h: 3 } },
 
   memory_reserved_stat(
     title='',
@@ -224,7 +224,7 @@ local prometheus = grafana.prometheus;
     decimals=2,
     unit='bytes',
     expr=std.format('sum(tnt_slab_quota_size{job=~"%s"})', job),
-  ) { gridPos: { w: 4, h: 5 } },
+  ) { gridPos: { w: 3, h: 3 } },
 
   space_ops_stat(
     title='',
@@ -257,7 +257,7 @@ local prometheus = grafana.prometheus;
     decimals=3,
     unit='ops',
     expr=std.format('sum(rate(tnt_stats_op_total{job=~"%s"}[%s]))', [job, rate_time_range]),
-  ) { gridPos: { w: 4, h: 4 } },
+  ) { gridPos: { w: 4, h: 5 } },
 
   http_rps_stat(
     title='',
@@ -275,7 +275,7 @@ local prometheus = grafana.prometheus;
       else (
         if datasource == '${DS_PROMETHEUS}' then
           |||
-            Overall rate of requests processed on Tarantool instances (all methods and response codes).
+            Overall rate of HTTP requests processed on Tarantool instances (all methods and response codes).
             If Tarantool instance is not available for Prometheus metrics extraction now, its contribution is not counted.
             If `No data` displayed, check up your 'rate_time_range' variable.
           |||
@@ -290,7 +290,40 @@ local prometheus = grafana.prometheus;
     decimals=3,
     unit='reqps',
     expr=std.format('sum(rate(http_server_request_latency_count{job=~"%s"}[%s]))', [job, rate_time_range]),
-  ) { gridPos: { w: 4, h: 4 } },
+  ) { gridPos: { w: 4, h: 5 } },
+
+  net_rps_stat(
+    title='',
+    description=null,
+
+    datasource=null,
+    measurement=null,
+    job=null,
+    rate_time_range=null,
+  ):: overview_stat(
+    title=title,
+    description=(
+      if description != null then
+        description
+      else (
+        if datasource == '${DS_PROMETHEUS}' then
+          |||
+            Overall rate of network requests processed on Tarantool instances.
+            If Tarantool instance is not available for Prometheus metrics extraction now, its contribution is not counted.
+            If `No data` displayed, check up your 'rate_time_range' variable.
+          |||
+        else
+          null
+      )
+    ),
+    datasource=datasource,
+    measurement=measurement,
+    job=job,
+    stat_title='Overall net load:',
+    decimals=3,
+    unit='reqps',
+    expr=std.format('sum(rate(tnt_net_requests_total{job=~"%s"}[%s]))', [job, rate_time_range]),
+  ) { gridPos: { w: 4, h: 5 } },
 
   local cartridge_issues(
     title,

dashboard/section.libsonnet

@@ -42,29 +42,35 @@ local vinyl = import 'panels/vinyl.libsonnet';
     cluster.health_overview_stat(
       datasource=datasource,
       job=job,
-    ) { gridPos: { w: 8, h: 3, x: 12, y: 1 } },
+    ) { gridPos: { w: 6, h: 3, x: 12, y: 1 } },
 
     cluster.memory_used_stat(
       datasource=datasource,
       job=job,
-    ) { gridPos: { w: 4, h: 5, x: 12, y: 4 } },
+    ) { gridPos: { w: 3, h: 3, x: 18, y: 1 } },
 
     cluster.memory_reserved_stat(
       datasource=datasource,
       job=job,
-    ) { gridPos: { w: 4, h: 5, x: 16, y: 4 } },
+    ) { gridPos: { w: 3, h: 3, x: 21, y: 1 } },
 
-    cluster.space_ops_stat(
+    cluster.http_rps_stat(
       datasource=datasource,
       job=job,
       rate_time_range=rate_time_range,
-    ) { gridPos: { w: 4, h: 4, x: 20, y: 1 } },
+    ) { gridPos: { w: 4, h: 5, x: 12, y: 4 } },
 
-    cluster.http_rps_stat(
+    cluster.net_rps_stat(
+      datasource=datasource,
+      job=job,
+      rate_time_range=rate_time_range,
+    ) { gridPos: { w: 4, h: 5, x: 16, y: 4 } },
+
+    cluster.space_ops_stat(
       datasource=datasource,
       job=job,
       rate_time_range=rate_time_range,
-    ) { gridPos: { w: 4, h: 4, x: 20, y: 5 } },
+    ) { gridPos: { w: 4, h: 5, x: 20, y: 4 } },
 
     cluster.cartridge_warning_issues(
       datasource=datasource,