Merge branch 'tezos_signer_forwarder' into midl4

charts/tezos-signer-forwarder/Chart.yaml

@@ -0,0 +1,6 @@
+apiVersion: v2
+name: tezos-signer-forwarder
+description: A chart for tezos-signer-forwarder
+type: application
+version: 0.0.0
+appVersion: "10.0"

charts/tezos-signer-forwarder/scripts/entrypoint.sh

@@ -0,0 +1,3 @@
+#!/bin/sh
+
+/usr/sbin/sshd -D -e -p ${TUNNEL_ENDPOINT_PORT}

charts/tezos-signer-forwarder/scripts/signer_exporter.py

@@ -0,0 +1,53 @@
+#!/usr/bin/env python
+import os
+from flask import Flask, request, jsonify
+import requests
+
+import logging
+log = logging.getLogger('werkzeug')
+log.setLevel(logging.ERROR)
+
+application = Flask(__name__)
+
+readiness_probe_path = os.getenv("READINESS_PROBE_PATH")
+signer_port = os.getenv("SIGNER_PORT")
+signer_metrics = os.getenv("SIGNER_METRICS") == "true"
+
+# https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/
+# Configured readiness probe timeoutSeconds is 5s, timeout sync request before that.
+SIGNER_CONNECT_TIMEOUT = 4.5
+
+@application.route('/metrics', methods=['GET'])
+def prometheus_metrics():
+    '''
+    Prometheus endpoint
+    This combines:
+    * the metrics from the signer, which themselves are a combination of the
+      prometheus node-exporter and custom probes (power status, etc)
+    * the `unhealthy_signers_total` metric exported by this script, verifying
+      whether the signer URL configured upstream returns a 200 OK
+    '''
+
+    try:
+        probe = requests.get(f"http://localhost:{signer_port}{readiness_probe_path}", timeout=SIGNER_CONNECT_TIMEOUT)
+    except requests.exceptions.ConnectTimeout:
+        #Timeout connect to node
+        probe = None
+    except requests.exceptions.ReadTimeout:
+        #Timeout read from node
+        probe = None
+    except requests.exceptions.RequestException:
+        probe = None
+    if probe and signer_metrics:
+        try:
+            healthz = requests.get(f"http://localhost:{signer_port}/healthz").text
+        except requests.exceptions.RequestException:
+            healthz = None
+    else:
+        healthz = None
+    return '''# number of unhealthy signers - should be 0 or 1
+unhealthy_signers_total %s
+%s''' % (0 if probe else 1, healthz or "")
+
+if __name__ == "__main__":
+   application.run(host = "0.0.0.0", port = 31732, debug = False)

charts/tezos-signer-forwarder/templates/_helpers.tpl

@@ -0,0 +1,62 @@
+{{/*
+Expand the name of the chart.
+*/}}
+{{- define "tezos-signer-forwarder.name" -}}
+{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Create a default fully qualified app name.
+We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec).
+If release name contains chart name it will be used as a full name.
+*/}}
+{{- define "tezos-signer-forwarder.fullname" -}}
+{{- if .Values.fullnameOverride }}
+{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- $name := default .Chart.Name .Values.nameOverride }}
+{{- if contains $name $.Release.Name }}
+{{- .Release.Name | trunc 63 | trimSuffix "-" }}
+{{- else }}
+{{- printf "%s-%s" $.Release.Name $name | trunc 63 | trimSuffix "-" }}
+{{- end }}
+{{- end }}
+{{- end }}
+
+{{/*
+Create chart name and version as used by the chart label.
+*/}}
+{{- define "tezos-signer-forwarder.chart" -}}
+{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }}
+{{- end }}
+
+{{/*
+Common labels
+*/}}
+{{- define "tezos-signer-forwarder.labels" -}}
+helm.sh/chart: {{ include "tezos-signer-forwarder.chart" . }}
+{{ include "tezos-signer-forwarder.selectorLabels" . }}
+{{- if .Chart.AppVersion }}
+app.kubernetes.io/version: {{ .Chart.AppVersion | quote }}
+{{- end }}
+app.kubernetes.io/managed-by: {{ .Release.Service }}
+{{- end }}
+
+{{/*
+Selector labels
+*/}}
+{{- define "tezos-signer-forwarder.selectorLabels" -}}
+app.kubernetes.io/name: {{ include "tezos-signer-forwarder.name" . }}
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
+{{/*
+Create the name of the service account to use
+*/}}
+{{- define "tezos-signer-forwarder.serviceAccountName" -}}
+{{- if .Values.serviceAccount.create }}
+{{- default (include "tezos-signer-forwarder.fullname" .) .Values.serviceAccount.name }}
+{{- else }}
+{{- default "default" .Values.serviceAccount.name }}
+{{- end }}
+{{- end }}

charts/tezos-signer-forwarder/templates/alertmanagerconfig.yaml

@@ -0,0 +1,63 @@
+{{- if .Values.alertmanagerConfig.enabled }}
+{{- range .Values.signers }}
+{{- if .monitoring_email }}
+{{ $signer := . }}
+{{- range .endpoints }}
+{{- if .alert_when_down }}
+apiVersion: monitoring.coreos.com/v1alpha1
+kind: AlertmanagerConfig
+metadata:
+  name: tezos-signer-{{ $signer.name }}-{{ .alias }}-email
+  labels:
+{{- toYaml $.Values.alertmanagerConfig.labels | nindent 4 }}
+spec:
+  route:
+    groupBy: ['job']
+    groupWait: 30s
+    groupInterval: 5m
+    repeatInterval: 12h
+    receiver: 'email_{{ $signer.name }}'
+    matchers:
+    - name: service
+      value: tezos-remote-signer-{{ $signer.name }}
+      regex: false
+    - name: alertType
+      value: tezos-remote-signer-alert
+      regex: false
+    - name: tezos_endpoint_name
+      value: {{ .alias }}
+      regex: false
+    continue: false
+
+  receivers:
+  - name: 'email_{{ $signer.name }}'
+    emailConfigs:
+    - to: "{{ $signer.monitoring_email }}"
+      sendResolved: true
+      headers:
+      - key: subject
+        value: '{{`[{{ .Status | toUpper }}{{ if eq .Status "firing" }}:{{ .Alerts.Firing | len }}{{ end }}] {{ .CommonLabels.alertname }}`}}'
+      html: >-
+        {{`{{ if eq .Status "firing" }}
+        Attention Required for Tezos Remote Signer:
+        {{ else }}
+        Resolved Alert for Tezos Remote Signer:
+        {{ end }}
+        {{ range .Alerts -}}
+        {{ .Annotations.summary }}
+        {{ end }}`}}
+      text: >-
+        {{`{{ if eq .Status "firing" }}
+        Attention Required for Tezos Remote Signer:
+        {{ else }}
+        Resolved Alert for Tezos Remote Signer:
+        {{ end }}
+        {{ range .Alerts -}}
+        {{ .Annotations.summary }}
+        {{ end }}`}}
+---
+{{- end }}
+{{- end }}
+{{- end }}
+{{- end }}
+{{- end }}

charts/tezos-signer-forwarder/templates/config.yaml

@@ -0,0 +1,12 @@
+{{- range .Values.signers }}
+{{- $name := .name }}
+{{- range $i, $endpoint := .endpoints }}
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: tezos-signer-forwarder-config-{{ $name }}-{{ $i }}
+data:
+  authorized_keys: "{{ $endpoint.ssh_pubkey }} signer"
+---
+{{- end }}
+{{- end }}

charts/tezos-signer-forwarder/templates/prometheusrule.yaml

@@ -0,0 +1,51 @@
+{{- if .Values.prometheusRule.enabled }}
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  labels:
+{{- toYaml .Values.prometheusRule.labels | nindent 4 }}
+  name: tezos-remote-signer-rules
+spec:
+  groups:
+  - name: tezos-remote-signer.rules
+    rules:
+    - alert: SignerPowerLoss
+      annotations:
+        description: 'Remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" has lost power'
+        summary: 'Remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" has lost power'
+      expr: power{namespace="{{ .Release.Namespace }}"} != 0
+      for: 1m
+      labels:
+        severity: critical
+        alertType: tezos-remote-signer-alert
+    - alert: SignerWiredNetworkLoss
+      annotations:
+        description: 'Remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" has lost wired internet connection'
+        summary: 'Tezos remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" has lost wired internet connection'
+      expr: wired_network{namespace="{{ .Release.Namespace }}"} != 0
+      for: 1m
+      labels:
+        severity: critical
+        alertType: tezos-remote-signer-alert
+---
+apiVersion: monitoring.coreos.com/v1
+kind: PrometheusRule
+metadata:
+  labels:
+{{- toYaml .Values.prometheusRule.labels | nindent 4 }}
+  name: tezos-remote-signer-reachability-rules
+spec:
+  groups:
+  - name: tezos-remote-signer.rules
+    rules:
+    - alert: NoRemoteSigner
+      annotations:
+        description: 'Remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" is down'
+        summary: 'Remote signer "{{`{{ $labels.tezos_endpoint_name }}`}}" for baker "{{`{{ $labels.tezos_baker_name }}`}}" is down or unable to sign.'
+      expr: unhealthy_signers_total{namespace="{{ .Release.Namespace }}"} != 0
+      for: 1m
+      labels:
+        severity: critical
+        alertType: tezos-remote-signer-alert
+---
+{{- end }}

charts/tezos-signer-forwarder/templates/secret.yaml

@@ -0,0 +1,7 @@
+apiVersion: v1
+kind: Secret
+metadata:
+  name: tezos-signer-forwarder-secret-{{ .Values.name }}
+data:
+  ssh_host_ecdsa_key: |
+{{ println .Values.secrets.signer_target_host_key | b64enc | indent 4 -}}

charts/tezos-signer-forwarder/templates/service.yaml

@@ -0,0 +1,50 @@
+apiVersion: v1
+kind: Service
+metadata:
+  name: tezos-remote-signer-ssh-ingress-{{ .Values.name }}
+  annotations:
+{{ toYaml .Values.service_annotations | indent 4 }}
+spec:
+  type: LoadBalancer
+  selector:
+    app.kubernetes.io/name: tezos-signer-forwarder
+  ports:
+{{- range .Values.signers }}
+{{- $name := .name }}
+  # undocumented k8s feature to make a service route to different pods
+  # based on the port - allows to reuse the same public ip in all cloud
+  # providers. For it to work, ports need to have names.
+  # https://github.com/kubernetes/kubernetes/issues/24875#issuecomment-794596576
+{{- range $i, $endpoint := .endpoints }}
+  - port: {{ $endpoint.tunnel_endpoint_port }}
+    name: ssh-{{ trunc 9 $name }}-{{ $i }}
+    targetPort: ssh-{{ trunc 9 $name }}-{{ $i }}
+{{- end }}
+{{- end }}
+  # ensures that remote signers can always ssh
+  publishNotReadyAddresses: true
+{{ if .Values.load_balancer_ip }}
+  loadBalancerIP: {{ .Values.load_balancer_ip }}
+{{ end }}
+---
+{{- range .Values.signers }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: tezos-remote-signer-{{ .name }}
+  labels:
+    app.kubernetes.io/name: tezos-signer-forwarder
+    tezos_baker_name: {{ .name }}
+spec:
+  selector:
+    app.kubernetes.io/name: tezos-signer-forwarder
+    tezos_baker_name: {{ .name }}
+  ports:
+  - port: {{ .signer_port }}
+    name: signer
+  - port: 31732
+    name: metrics
+  # make sure that the service always targets the same signer, when HA is in use.
+  sessionAffinity: ClientIP
+---
+{{- end }}

charts/tezos-signer-forwarder/templates/servicemonitor.yaml

@@ -0,0 +1,25 @@
+{{- if .Values.serviceMonitor.enabled }}
+{{- range .Values.signers }}
+apiVersion: monitoring.coreos.com/v1
+kind: ServiceMonitor
+metadata:
+  labels:
+    app.kubernetes.io/name: tezos-signer-forwarder
+  name: tezos-remote-signer-monitoring-{{ .name }}
+spec:
+  endpoints:
+  - port: metrics
+    path: /metrics
+    # default scrape timeout of 10 can be too small for remote raspberry pis
+    scrapeTimeout: "20s"
+  selector:
+    matchLabels:
+      app.kubernetes.io/name: tezos-signer-forwarder
+      tezos_baker_name: {{ .name }}
+  targetLabels:
+  - tezos_baker_name
+  podTargetLabels:
+  - tezos_endpoint_name
+---
+{{- end }}
+{{- end }}