Merge branch 'thomasjpfan-prometheus2'

Author: vfarcic

Dockerfile

@@ -1,4 +1,4 @@
-FROM golang:1.8 AS build
+FROM golang:1.9.2 AS build
 ADD . /src
 WORKDIR /src
 RUN go get -t github.com/stretchr/testify/suite
@@ -8,7 +8,7 @@ RUN CGO_ENABLED=0 GOOS=linux go build -v -o docker-flow-monitor
 
 
 
-FROM prom/prometheus:v1.8.1
+FROM prom/prometheus:v2.0.0
 
 ENV GLOBAL_SCRAPE_INTERVAL=10s \
     ARG_CONFIG_FILE=/etc/prometheus/prometheus.yml \
@@ -24,7 +24,10 @@ ENTRYPOINT ["docker-flow-monitor"]
 HEALTHCHECK --interval=5s CMD /bin/check.sh
 
 COPY --from=build /src/docker-flow-monitor /bin/docker-flow-monitor
-RUN chmod +x /bin/docker-flow-monitor
 COPY check.sh /bin/check.sh
+
+USER root
 RUN chmod +x /bin/check.sh
+RUN chmod +x /bin/docker-flow-monitor
+USER nobody
 

Jenkinsfile

@@ -24,19 +24,15 @@ pipeline {
         branch "master"
       }
       steps {
-        withCredentials([usernamePassword(
-          credentialsId: "docker",
-          usernameVariable: "USER",
-          passwordVariable: "PASS"
-        )]) {
-          sh "docker login -u $USER -p $PASS"
-        }
-        sh "docker tag vfarcic/docker-flow-monitor vfarcic/docker-flow-monitor:${currentBuild.displayName}"
+        dockerLogin()
+        sh "docker tag vfarcic/docker-flow-monitor vfarcic/docker-flow-monitor:2-${currentBuild.displayName}"
         sh "docker image push vfarcic/docker-flow-monitor:latest"
-        sh "docker image push vfarcic/docker-flow-monitor:${currentBuild.displayName}"
-        sh "docker tag vfarcic/docker-flow-monitor-docs vfarcic/docker-flow-monitor-docs:${currentBuild.displayName}"
+        sh "docker image push vfarcic/docker-flow-monitor:2-${currentBuild.displayName}"
+        sh "docker tag vfarcic/docker-flow-monitor-docs vfarcic/docker-flow-monitor-docs:2-${currentBuild.displayName}"
         sh "docker image push vfarcic/docker-flow-monitor-docs:latest"
-        sh "docker image push vfarcic/docker-flow-monitor-docs:${currentBuild.displayName}"
+        sh "docker image push vfarcic/docker-flow-monitor-docs:2-${currentBuild.displayName}"
+        dockerLogout()
+        dfReleaseGithub("docker-flow-monitor")
       }
     }
     stage("deploy") {
@@ -47,8 +43,8 @@ pipeline {
         label "prod"
       }
       steps {
-        sh "docker service update --image vfarcic/docker-flow-monitor:${currentBuild.displayName} monitor_monitor"
-        sh "docker service update --image vfarcic/docker-flow-monitor-docs:${currentBuild.displayName} monitor_docs"
+        sh "docker service update --image vfarcic/docker-flow-monitor:2-${currentBuild.displayName} monitor_monitor"
+        sh "docker service update --image vfarcic/docker-flow-monitor-docs:2-${currentBuild.displayName} monitor_docs"
       }
     }
   }

docs/article.md

@@ -846,4 +846,4 @@ Before you leave, please remove the Docker machines we created and free the reso
 
 ```bash
 docker-machine rm -f swarm-1 swarm-2 swarm-3
-```
+```

docs/auto-scaling.md

@@ -321,19 +321,31 @@ You should see three alerts that correspond to the three labels define in the `m
 If you expand the *godemo_main_resp_time_above* alert, you'll see that DFM translated the service labels into the alert definition that follows.
 
 ```
-ALERT godemo_main_resp_tim_eabove
-  IF sum(rate(http_server_resp_time_bucket{job="go-demo_main",le="0.1"}[5m])) / sum(rate(http_server_resp_time_count{job="go-demo_main"}[5m])) < 0.99
-  LABELS {receiver="system", scale="up", service="go-demo_main"}
-  ANNOTATIONS {summary="Response time of the service go-demo_main is above 0.1"}
+alert: godemo_main_resptimeabove
+expr: sum(rate(http_server_resp_time_bucket{job="go-demo_main",le="0.1"}[5m]))
+  / sum(rate(http_server_resp_time_count{job="go-demo_main"}[5m])) < 0.99
+labels:
+  receiver: system
+  scale: up
+  service: go-demo_main
+  type: service
+annotations:
+  summary: Response time of the service go-demo_main is above 0.1
 ```
 
 Similarly, the *godemo_main_resp_time_below* alert is defined as follows.
 
 ```
-ALERT godemo_main_resp_time_below
-  IF sum(rate(http_server_resp_time_bucket{job="go-demo_main",le="0.025"}[5m])) / sum(rate(http_server_resp_time_count{job="go-demo_main"}[5m])) > 0.75
-  LABELS {receiver="system", scale="down", service="go-demo_main"}
-  ANNOTATIONS {summary="Response time of the service go-demo_main is below 0.025"}
+alert: godemo_main_resptimebelow
+expr: sum(rate(http_server_resp_time_bucket{job="go-demo_main",le="0.025"}[5m]))
+  / sum(rate(http_server_resp_time_count{job="go-demo_main"}[5m])) > 0.75
+labels:
+  receiver: system
+  scale: down
+  service: go-demo_main
+  type: service
+annotations:
+  summary: Response time of the service go-demo_main is below 0.025
 ```
 
 Let's confirm that the `go-demo` stack is up-and-running.

docs/config.md

@@ -85,4 +85,4 @@ The formatting rules for the `REMOTE_WRITE` variables follow the same pattern as
 
 Additional scrapes can be added through files prefixed with `scrape_`. By default, all such files located in `/run/secrets` are automatically added to the `scrape_configs` section of the configuration. The directory can be changed by setting a different value to the environment variable `CONFIGS_DIR`.
 
-The simplest way to add scrape configs is to use Docker [secrets](https://docs.docker.com/engine/swarm/secrets/) or [configs](https://docs.docker.com/engine/swarm/configs/).
+The simplest way to add scrape configs is to use Docker [secrets](https://docs.docker.com/engine/swarm/secrets/) or [configs](https://docs.docker.com/engine/swarm/configs/).

docs/migration.md

@@ -0,0 +1,19 @@
+# Migration Guide
+
+*Docker Flow Monitor* (DFM) now supports Prometheus 2! Prometheus 2 includes a new storage subsystem that has reduce CPU usage and lower disk space usage compared to Prometheus 1.  This guide highlights issues that you may encounter when upgrading to Prometheus 2.
+
+## Database
+
+When upgrading from DFM backed by Prometheus 1 to Prometheus 2, DFM will create a new database supported by Prometheus 2. If you need to access the previous data scraped by Prometheus 1, downgrade your DFM instance using tag: `17.12.12-36`. DFM will launch with Prometheus 1, and continue using the previous database.
+
+## Backwards Compatibility
+
+The command line arguments for Prometheus 2 has changed. This is detailed in Prometheus's [Official Migration Guide](https://prometheus.io/docs/prometheus/latest/migration/). *DFM* will continue to support the previous command line arguments prefixed by `ARG_`:
+
+1. Setting `ARG_ALERTMANAGER_URL` configures the alertmanager correctly for Prometheus 2.
+2. `ARG_STORAGE_LOCAL_PATH` maps to `--storage.tsdb.path`.
+3. `ARG_STORAGE_LOCAL_RETENTION` maps to `--storage.tsdb.retention`.
+4. `ARG_QUERY_STALENESS-DELTA` maps to `--query.lookback-delta`.
+5. Setting `ARG_ENABLE-REMOTE-SHUTDOWN=true` sets flag `--web.enable-lifecycle`
+
+You can explore the new flags in Prometheus 2 by downloading the binary from their [Download Page](https://prometheus.io/download/) and running `./prometheus -h`.

mkdocs.yml

@@ -6,6 +6,7 @@ pages:
     - Auto-Scaling Services Using Instrumented Metrics: auto-scaling.md
   - Configuration: config.md
   - Usage: usage.md
+  - Migration Guide: migration.md
   - About:
     - Release Notes: release-notes.md
     - License: license.md

prometheus/alert.go

@@ -7,25 +7,28 @@ import (
 
 // GetAlertConfig returns Prometheus configuration snippet related to alerts.
 func GetAlertConfig(alerts map[string]Alert) string {
-	templateString := `{{range .}}
-ALERT {{.AlertNameFormatted}}
-  IF {{.AlertIf}}{{if .AlertFor}}
-  FOR {{.AlertFor}}{{end}}
-  {{- if .AlertLabels}}
-  LABELS {
-    {{- range $key, $value := .AlertLabels}}
-    {{$key}} = "{{$value}}",
+	templateString := `groups:
+- name: alert.rules
+  rules:
+  {{- range . }}
+  - alert: {{ .AlertNameFormatted }}
+    expr: {{ .AlertIf }}
+    {{- if .AlertFor }}
+    for: {{ .AlertFor }}
+    {{- end }}
+    {{- if .AlertLabels }}
+    labels:
+    {{- range $key, $value := .AlertLabels }}
+      {{ $key }}: {{ $value }}
     {{- end}}
-  }
-  {{- end}}
-  {{- if .AlertAnnotations}}
-  ANNOTATIONS {
-    {{- range $key, $value := .AlertAnnotations}}
-    {{$key}} = "{{$value}}",
+    {{- end }}
+    {{- if .AlertAnnotations }}
+    annotations:
+    {{- range $key, $value := .AlertAnnotations }}
+      {{ $key }}: "{{ $value }}"
     {{- end}}
-  }
-  {{- end}}
-{{end}}`
+    {{- end }}
+  {{- end }}`
 	tmpl, _ := template.New("").Parse(templateString)
 	var b bytes.Buffer
 	tmpl.Execute(&b, alerts)

prometheus/alert_test.go

@@ -2,8 +2,9 @@ package prometheus
 
 import (
 	"fmt"
-	"github.com/stretchr/testify/suite"
 	"testing"
+
+	"github.com/stretchr/testify/suite"
 )
 
 type AlertTestSuite struct {
@@ -24,14 +25,15 @@ func TestAlertUnitTestSuite(t *testing.T) {
 // GetAlertConfig
 
 func (s *AlertTestSuite) Test_GetAlertConfig_ReturnsConfigWithData() {
-	expected := ""
+	expected := `groups:
+- name: alert.rules
+  rules:`
 	alerts := s.getTestAlerts()
 	for _, i := range []int{1, 2} {
 		expected += fmt.Sprintf(`
-ALERT alertNameFormatted%d
-  IF alert-if-%d
-  FOR alert-for-%d
-`, i, i, i)
+  - alert: alertNameFormatted%d
+    expr: alert-if-%d
+    for: alert-for-%d`, i, i, i)
 	}
 
 	actual := GetAlertConfig(alerts)
@@ -40,18 +42,19 @@ ALERT alertNameFormatted%d
 }
 
 func (s *AlertTestSuite) Test_GetAlertConfig_ReturnsConfigWithLabels_WhenPresent() {
-	expected := ""
+	expected := `groups:
+- name: alert.rules
+  rules:`
 	alerts := s.getTestAlerts()
 	for _, i := range []int{1, 2} {
 		expected += fmt.Sprintf(`
-ALERT alertNameFormatted%d
-  IF alert-if-%d
-  FOR alert-for-%d
-  LABELS {
-    alert-label-%d-1 = "alert-label-value-%d-1",
-    alert-label-%d-2 = "alert-label-value-%d-2",
-  }
-`, i, i, i, i, i, i, i)
+  - alert: alertNameFormatted%d
+    expr: alert-if-%d
+    for: alert-for-%d
+    labels:
+      alert-label-%d-1: alert-label-value-%d-1
+      alert-label-%d-2: alert-label-value-%d-2`,
+			i, i, i, i, i, i, i)
 		key := fmt.Sprintf("alert-name-%d", i)
 		alert := alerts[key]
 		alert.AlertLabels = map[string]string{
@@ -67,18 +70,19 @@ ALERT alertNameFormatted%d
 }
 
 func (s *AlertTestSuite) Test_GetAlertConfig_ReturnsConfigWithAnnotations_WhenPresent() {
-	expected := ""
+	expected := `groups:
+- name: alert.rules
+  rules:`
 	alerts := s.getTestAlerts()
 	for _, i := range []int{1, 2} {
 		expected += fmt.Sprintf(`
-ALERT alertNameFormatted%d
-  IF alert-if-%d
-  FOR alert-for-%d
-  ANNOTATIONS {
-    alert-annotation-%d-1 = "alert-annotation-value-%d-1",
-    alert-annotation-%d-2 = "alert-annotation-value-%d-2",
-  }
-`, i, i, i, i, i, i, i)
+  - alert: alertNameFormatted%d
+    expr: alert-if-%d
+    for: alert-for-%d
+    annotations:
+      alert-annotation-%d-1: "alert-annotation-value-%d-1"
+      alert-annotation-%d-2: "alert-annotation-value-%d-2"`,
+			i, i, i, i, i, i, i)
 		key := fmt.Sprintf("alert-name-%d", i)
 		alert := alerts[key]
 		alert.AlertAnnotations = map[string]string{