dashboard: set default Prometheus job

Set default Prometheus job to "tarantool". After this patch, it will be
possible to import dashboard with community.ansible.ansible_dashboard,
if Prometheus is configured to collect Tarantool metrics with
"tarantool" job.

This patch also changes all examples job names to "tarantool".

Part of #138

Author: DifferentialOrange

CHANGELOG.md

@@ -9,6 +9,9 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ## Added
 - Options to build static dashboards
 
+## Changed
+- Set default Prometheus job to `tarantool`
+
 
 ## [1.3.0] - 2022-06-29
 Grafana revisions: [InfluxDB revision 13](https://grafana.com/api/dashboards/12567/revisions/13/download), [Prometheus revision 13](https://grafana.com/api/dashboards/13054/revisions/13/download), [InfluxDB TDG revision 2](https://grafana.com/api/dashboards/16405/revisions/2/download), [Prometheus TDG revision 2](https://grafana.com/api/dashboards/16406/revisions/2/download).

Makefile

@@ -1,3 +1,4 @@
+JOB ?= tarantool
 RATE_TIME_RANGE ?= 2m
 POLICY ?= autogen
 OUTPUT_STATIC_DASHBOARD ?= dashboard.json
@@ -14,10 +15,7 @@ ifndef DATASOURCE
 	@echo 1>&2 "DATASOURCE must be set"
 		false
 endif
-ifndef JOB
-	@echo 1>&2 "JOB must be set"
-		false
-endif
+	# JOB is optional, default is "tarantool"
 	# RATE_TIME_RANGE is optional, default is "2m"
 	jsonnet -J ./vendor -J . \
 		--ext-str DATASOURCE=${DATASOURCE} \

README.md

@@ -81,7 +81,7 @@ To set up an InfluxDB dashboard for monitoring example app, use the following va
 
 To set up an Prometheus dashboard for monitoring example app, use the following variables:
 
-- `Job`: `tarantool_app`;
+- `Job`: `tarantool`;
 - `Rate time range`: `2m`.
 
 ### Monitoring local app
@@ -116,7 +116,7 @@ to install build dependencies and dependencies that are required to run tests lo
 
 To build a static dashboard with no input and dynamic variables, run `make` commands.
 ```bash
-make DATASOURCE=MyPrometheus JOB=MyApp \
+make DATASOURCE=Prometheus JOB=tarantool \
      OUTPUT_STATIC_DASHBOARD=mydashboard.json build-static-prometheus
 ```
 Following targets are available:
@@ -127,7 +127,7 @@ Following targets are available:
 
 Variables for Prometheus targets:
 - `DATASOURCE`: name of a Prometheus data source;
-- `JOB`: name of a Prometheus job collecting your application metrics;
+- `JOB` (optional, default `tarantool`): name of a Prometheus job collecting your application metrics;
 - `RATE_TIME_RANGE` (optional, default `2m`): rps computation rate time range;
 - `OUTPUT_STATIC_DASHBOARD` (optional, default `dashboard.json`): compiled dashboard file.
 

dashboard/build/prometheus/dashboard.libsonnet

@@ -18,8 +18,7 @@ dashboard_raw(
   name='PROMETHEUS_JOB',
   label='Job',
   type='constant',
-  pluginId=null,
-  pluginName=null,
+  value='tarantool',
   description='Prometheus Tarantool metrics job'
 ).addInput(
   name='PROMETHEUS_RATE_TIME_RANGE',

dashboard/build/prometheus/tdg_dashboard.libsonnet

@@ -18,8 +18,7 @@ tdg_dashboard_raw(
   name='PROMETHEUS_JOB',
   label='Job',
   type='constant',
-  pluginId=null,
-  pluginName=null,
+  value='tarantool',
   description='Prometheus Tarantool metrics job'
 ).addInput(
   name='PROMETHEUS_RATE_TIME_RANGE',

doc/monitoring/grafana_dashboard.rst

@@ -79,7 +79,7 @@ metrics path as it was configured on Tarantool instances:
 ..  code-block:: yaml
 
     scrape_configs:
-      - job_name: "example_project"
+      - job_name: tarantool
         static_configs:
           - targets: 
             - "example_project:8081"

example_cluster/prometheus/alerts.yml

@@ -179,7 +179,7 @@ groups:
   rules:
   # Alert for CRUD module request errors.
   - alert: CRUDHighErrorRate
-    expr: rate(tnt_crud_stats_count{ job="tarantool_app", status="error" }[5m]) > 0.1
+    expr: rate(tnt_crud_stats_count{ job="tarantool", status="error" }[5m]) > 0.1
     for: 1m
     labels:
       severity: critical
@@ -190,7 +190,7 @@ groups:
 
   # Warning for CRUD module requests too long responses.
   - alert: CRUDHighLatency
-    expr: tnt_crud_stats{ job="tarantool_app", quantile="0.99" } > 0.1
+    expr: tnt_crud_stats{ job="tarantool", quantile="0.99" } > 0.1
     for: 1m
     labels:
       severity: warning
@@ -201,7 +201,7 @@ groups:
 
   # Warning for too many map reduce CRUD module requests.
   - alert: CRUDHighMapReduceRate
-    expr: rate(tnt_crud_map_reduces{ job="tarantool_app" }[5m]) > 0.1
+    expr: rate(tnt_crud_map_reduces{ job="tarantool" }[5m]) > 0.1
     for: 1m
     labels:
       severity: warning
@@ -214,12 +214,12 @@ groups:
 
 - name: tarantool-business
   rules:
-  # Warning for any endpoint of an instance in tarantool_app job that responds too long.
+  # Warning for any endpoint of an instance in tarantool job that responds too long.
   # Beware that metric name depends on name of the collector you use in HTTP metrics middleware
   # and request depends on type of this collector.
   # This example based on summary collector with default name.
   - alert: HTTPHighLatency
-    expr: http_server_request_latency{ job="tarantool_app", quantile="0.99" } > 0.1
+    expr: http_server_request_latency{ job="tarantool", quantile="0.99" } > 0.1
     for: 5m
     labels:
       severity: warning
@@ -228,12 +228,12 @@ groups:
       description: "Some {{ $labels.method }} requests to {{ $labels.path }} path with {{ $labels.status }} response status
         on '{{ $labels.alias }}' instance of job '{{ $labels.job }}' are processed too long."
 
-  # Warning for any endpoint of an instance in tarantool_app job that sends too much 4xx responses.
+  # Warning for any endpoint of an instance in tarantool job that sends too much 4xx responses.
   # Beware that metric name depends on name of the collector you use in HTTP metrics middleware
   # and request depends on type of this collector.
   # This example based on summary collector with default name.
   - alert: HTTPHighClientErrorRateInstance
-    expr: sum by (job, instance, method, path, alias) (rate(http_server_request_latency_count{ job="tarantool_app", status=~"^4\\d{2}$" }[5m])) > 10
+    expr: sum by (job, instance, method, path, alias) (rate(http_server_request_latency_count{ job="tarantool", status=~"^4\\d{2}$" }[5m])) > 10
     for: 1m
     labels:
       severity: warning
@@ -242,12 +242,12 @@ groups:
       description: "Too many {{ $labels.method }} requests to {{ $labels.path }} path 
         on '{{ $labels.alias }}' instance of job '{{ $labels.job }}' get client error (4xx) responses."
 
-  # Warning for any endpoint in tarantool_app job that sends too much 4xx responses (cluster overall).
+  # Warning for any endpoint in tarantool job that sends too much 4xx responses (cluster overall).
   # Beware that metric name depends on name of the collector you use in HTTP metrics middleware
   # and request depends on type of this collector.
   # This example based on summary collector with default name.
   - alert: HTTPHighClientErrorRate
-    expr: sum by (job, method, path) (rate(http_server_request_latency_count{ job="tarantool_app", status=~"^4\\d{2}$" }[5m])) > 20
+    expr: sum by (job, method, path) (rate(http_server_request_latency_count{ job="tarantool", status=~"^4\\d{2}$" }[5m])) > 20
     for: 1m
     labels:
       severity: warning
@@ -256,12 +256,12 @@ groups:
       description: "Too many {{ $labels.method }} requests to {{ $labels.path }} path
         on instances of job '{{ $labels.job }}' get client error (4xx) responses."
 
-  # Warning for any endpoint of an instance in tarantool_app job that sends 5xx responses.
+  # Warning for any endpoint of an instance in tarantool job that sends 5xx responses.
   # Beware that metric name depends on name of the collector you use in HTTP metrics middleware
   # and request depends on type of this collector.
   # This example based on summary collector with default name.
   - alert: HTTPServerErrors
-    expr: sum by (job, instance, method, path, alias) (rate(http_server_request_latency_count{ job="tarantool_app", status=~"^5\\d{2}$" }[5m])) > 0
+    expr: sum by (job, instance, method, path, alias) (rate(http_server_request_latency_count{ job="tarantool", status=~"^5\\d{2}$" }[5m])) > 0
     for: 1m
     labels:
       severity: warning
@@ -270,12 +270,12 @@ groups:
       description: "Some {{ $labels.method }} requests to {{ $labels.path }} path 
         on '{{ $labels.alias }}' instance of job '{{ $labels.job }}' get server error (5xx) responses."
 
-  # Warning for any endpoint of a router instance (with "router" in alias) in tarantool_app job that gets too little requests.
+  # Warning for any endpoint of a router instance (with "router" in alias) in tarantool job that gets too little requests.
   # Beware that metric name depends on name of the collector you use in HTTP metrics middleware
   # and request depends on type of this collector.
   # This example based on summary collector with default name.
   - alert: HTTPLowRequestRateRouter
-    expr: sum by (job, instance, alias) (rate(http_server_request_latency_count{ job="tarantool_app", alias=~"^.*router.*$" }[5m])) < 10
+    expr: sum by (job, instance, alias) (rate(http_server_request_latency_count{ job="tarantool", alias=~"^.*router.*$" }[5m])) < 10
     for: 5m
     labels:
       severity: warning

example_cluster/prometheus/prometheus.localapp.yml

@@ -27,7 +27,7 @@ scrape_configs:
     static_configs:
       - targets: ["localhost:9090"]
 
-  - job_name: "tarantool_app"
+  - job_name: "tarantool"
     static_configs:
       - targets: 
         - "host.docker.internal:8081"

example_cluster/prometheus/prometheus.tdg.yml

@@ -27,7 +27,7 @@ scrape_configs:
     static_configs:
       - targets: ["localhost:9090"]
 
-  - job_name: "tarantool_app"
+  - job_name: "tarantool"
     static_configs:
       - targets: 
         - "tdg:8080"

example_cluster/prometheus/prometheus.yml

@@ -27,7 +27,7 @@ scrape_configs:
     static_configs:
       - targets: ["localhost:9090"]
 
-  - job_name: "tarantool_app"
+  - job_name: "tarantool"
     static_configs:
       - targets: 
         - "app:8081"