Remove systemd support. Add alertmanager. Remove old template

12 files changed, 170 insertions, 113 deletions

diff --git a/defaults/main.yml b/defaults/main.yml
index afe2627..7d67f8e 100644
--- a/defaults/main.yml
+++ b/defaults/main.yml
@@ -1,18 +1,15 @@
 ---
 
-# Use this to override the template for more complicated uses
-prometheus_template_source: prometheus/prometheus.yml
-
 prometheus_ui_listen: ':9090'
 prometheus_home_dir: /var/lib/prometheus
 prometheus_data_dir: '{{prometheus_home_dir}}/data'
 prometheus_retention_time: "730d"
 prometheus_log: '{{prometheus_home_dir}}/log'
 prometheus_user: prometheus
-prometheus_version: "2.15.1"
+prometheus_version: "2.22.1"
 # Should resolve to "linux"
 prometheus_os: "{{ansible_facts['system']|lower}}"
-prometheus_checksum: f8d2b9e8f9551a3dd6f476a43ea9f699d18d2cc6558baf0eebe0143d8f9c1d5a
+prometheus_checksum: 9001a9cb939e0a6d9f2b67d22506c620bc9457777272fced43274b032ba35f44
 prometheus_architecture: amd64
 prometheus_url: "https://github.com/prometheus/prometheus/releases/download/v{{prometheus_version}}/prometheus-{{prometheus_version}}.{{prometheus_os}}-{{prometheus_architecture}}.tar.gz"
 
@@ -26,10 +23,49 @@ prometheus_opts: |
        --web.enable-admin-api
 
 
-# The default template makes it easy to scrape Netdata from internal hosts
-prometheus_netdata_hosts:
-  - localhost:19999
+# Filtered through 'to_nice_yaml'
+# prometheus_config:
+
+alertmanager: true
+
+alertmanager_version: '0.21.0'
+alertmanager_checksum_alg: sha256
+alertmanager_checksum: 9ccd863937436fd6bfe650e22521a7f2e6a727540988eef515dde208f9aef232
+alertmanager_url: 'https://github.com/prometheus/alertmanager/releases/download/v{{alertmanager_version}}/alertmanager-{{alertmanager_version}}.{{prometheus_os}}-{{prometheus_architecture}}.tar.gz'
+alertmanager_port: '9093'
+# Note that this is also UDP
+alertmanager_cluster_port: '9094'
+alertmanager_data_dir: '{{prometheus_home_dir}}/alertmanager-data'
+alertmanager_log: '{{prometheus_home_dir}}/alertmanager.log'
+
+# Newlines are automatically replaced with spaces
+alertmanager_opts: |
+  --config.file "{{prometheus_home_dir}}/conf/alertmanager.yml"
+  --storage.path {{alertmanager_data_dir}}
+  --web.listen-address "0.0.0.0:{{alertmanager_port}}"
+  --cluster.listen-address "0.0.0.0:{{alertmanager_cluster_port}}"
+
+
+# You can either specify them by hand, or pull them out of the inventory with
+# a query()
+# alertmanager_nodes:
+#   - prometheus0.example.com
+#   - prometheus1.example.com
+# alertmanager_nodes: "{{query('inventory_hostnames', 'prometheus')}}"
+
+# Append the configured port number, used in the config in a few spots
+alertmanager_nodes_w_port: '{{alertmanager_nodes | forEachAppend(":" + alertmanager_cluster_port)}}'
 
-# Only inserted into the template if defined
-# prometheus_netdata_tls_hosts:
-#   - example.com:19443
+# alertmanager_conf:
+#   global:
+#     # The API URL to use for Slack notifications.
+#     slack_api_url: '{{slack_webhook_uri}}'
+#
+#   route:
+#     receiver: 'slack-notifications'
+#
+#   receivers:
+#     - name: 'slack-notifications'
+#       slack_configs:
+#         - channel: '#{{slack_alerts_channel}}'
+#           send_resolved: true
diff --git a/filter_plugins/__pycache__/forEachAppend.cpython-38.pyc b/filter_plugins/__pycache__/forEachAppend.cpython-38.pyc
new file mode 100644
index 0000000..124454c
--- /dev/null
+++ b/filter_plugins/__pycache__/forEachAppend.cpython-38.pyc
diff --git a/filter_plugins/forEachAppend.py b/filter_plugins/forEachAppend.py
new file mode 100644
index 0000000..b5d676d
--- /dev/null
+++ b/filter_plugins/forEachAppend.py
@@ -0,0 +1,10 @@
+#!/usr/bin/env python
+
+class FilterModule(object):
+    def filters(self):
+        return {
+            'forEachAppend': self.forEachAppend,
+        }
+
+    def forEachAppend(self, arr, _str):
+        return [ x +_str for x in arr ]
diff --git a/handlers/main.yml b/handlers/main.yml
new file mode 100644
index 0000000..abaa110
--- /dev/null
+++ b/handlers/main.yml
@@ -0,0 +1,8 @@
+- name: Restart prometheus
+  runit:
+    name: prometheus
+    state: restarted
+- name: Restart alertmanager
+  runit:
+    name: alertmanager
+    state: restarted
diff --git a/tasks/alertmanager.yml b/tasks/alertmanager.yml
new file mode 100644
index 0000000..4c9e4d3
--- /dev/null
+++ b/tasks/alertmanager.yml
@@ -0,0 +1,59 @@
+---
+- name: Download Alertmanager
+  get_url:
+    url: "{{alertmanager_url}}"
+    dest: "{{prometheus_home_dir}}/alertmanager-{{alertmanager_version}}.tgz"
+    mode: '0600'
+    owner: "{{prometheus_user}}"
+    checksum: '{{alertmanager_checksum_alg}}:{{alertmanager_checksum}}'
+- set_fact: "alertmanager_extract_dir=alertmanager-{{alertmanager_version}}.{{prometheus_os}}-{{prometheus_architecture}}"
+- name: Extract Alertmanager
+  shell: |
+    #!/bin/sh
+    set -e
+    if ! [ -e "{{prometheus_home_dir}}/{{alertmanager_extract_dir}}" ] ; then
+      su - '{{prometheus_user}}' -c 'tar xzf alertmanager-{{alertmanager_version}}.tgz'
+      exit 50
+    fi
+  register: res
+  changed_when: res is defined and res.rc == 50
+  ignore_errors: true
+  notify: Restart alertmanager
+- name: Link alertmanager directroy
+  file:
+    state: link
+    src: '{{alertmanager_extract_dir}}'
+    dest: '{{prometheus_home_dir}}/alertmanager'
+    force: yes
+- name: Create Data directory
+  file:
+    state: directory
+    dest: "{{alertmanager_data_dir}}"
+    mode: '0700'
+    owner: "{{prometheus_user}}"
+- name: Write alertmanager configuration file
+  copy:
+    content: '{{alertmanager_conf | to_nice_yaml}}'
+    dest: '{{prometheus_home_dir}}/conf/alertmanager.yml'
+    owner: '{{prometheus_user}}'
+    mode: '0600'
+  notify: Restart alertmanager
+  tags:
+    - configuration
+- name: Create Runit Directory
+  file:
+    state: directory
+    dest: /etc/sv/alertmanager
+    owner: root
+    mode: '0755'
+- name: Write runit service
+  template:
+    src: alertmanager.runit
+    dest: /etc/sv/alertmanager/run
+    owner: root
+    mode: '0755'
+- name: Enable Runit service
+  file:
+    state: link
+    src: /etc/sv/alertmanager
+    dest: /var/service/alertmanager
diff --git a/tasks/main.yml b/tasks/main.yml
index cb2645a..19ca131 100644
--- a/tasks/main.yml
+++ b/tasks/main.yml
@@ -22,15 +22,21 @@
     dest: "{{prometheus_home_dir}}/prometheus-{{prometheus_version}}.tgz"
     mode: '0600'
     owner: "{{prometheus_user}}"
-    checksum:  'sha256:{{prometheus_checksum}}'
+    checksum: 'sha256:{{prometheus_checksum}}'
 # There's a bug  in become_user that doesn't set permissions on the temp directory
 # which is asinine when my user has sudo persmissions.
 - name: Extract Prometheus
   shell: |
+    #!/bin/sh
+    set -e
     if ! [ -e "{{prometheus_home_dir}}/{{prometheus_extract_dir}}" ] ; then
       su - '{{prometheus_user}}' -c 'tar xzf prometheus-{{prometheus_version}}.tgz'
+      exit 50
     fi
-  changed_when: false
+  register: res
+  changed_when: res is defined and res.rc == 50
+  ignore_errors: true
+  notify: Restart Prometheus
 - name: Link prometheus directory
   file:
     state: link
@@ -45,15 +51,38 @@
     force: yes
   loop:
     - prometheus.yml
-- name: Write prometheus template
-  template:
-    src: '{{prometheus_template_source}}'
+- name: Write prometheus configuration file
+  copy:
+    content: '{{prometheus_config | to_nice_yaml}}'
     dest: '{{prometheus_home_dir}}/conf/prometheus.yml'
     owner: '{{prometheus_user}}'
     mode: '0600'
-- name: Run Runit Specific tasks
-  include_tasks: systemd.yml
-  when: ansible_facts['distribution'].lower() == "ubuntu" or ansible_facts['distribution'].lower() == "centos"
-- name: Run Runit Specific tasks
-  include_tasks: runit.yml
-  when: ansible_facts['distribution'].lower() == "void" or ansible_facts['distribution'].lower() == "alpine"
+  notify: Restart prometheus
+  tags:
+    - configuration
+- name: Create Runit Directory
+  file:
+    state: directory
+    dest: /etc/sv/prometheus
+    owner: root
+    mode: '0755'
+- name: Write runit service
+  template:
+    src: prometheus.runit
+    dest: /etc/sv/prometheus/run
+    owner: root
+    mode: '0755'
+- name: Enable Runit service
+  file:
+    state: link
+    src: /etc/sv/prometheus
+    dest: /var/service/prometheus
+- name: Include Aertmanager tasks
+  include_tasks:
+    file: alertmanager.yml
+    apply:
+      tags:
+        - alertmanager
+  tags:
+    - alertmanager
+  when: alertmanager
diff --git a/tasks/runit.yml b/tasks/runit.yml
deleted file mode 100644
index d8bf8d4..0000000
--- a/tasks/runit.yml
+++ /dev/null
@@ -1,17 +0,0 @@
-- name: Create Runit Directory
-  file:
-    state: directory
-    dest: /etc/sv/prometheus
-    owner: root
-    mode: '0755'
-- name: Write runit service
-  template:
-    src: prometheus/runit/run
-    dest: /etc/sv/prometheus/run
-    owner: root
-    mode: '0755'
-- name: Enable Runit service
-  file:
-    state: link
-    src: /etc/sv/prometheus
-    dest: /var/service/prometheus
diff --git a/tasks/systemd.yml b/tasks/systemd.yml
deleted file mode 100644
index 031f036..0000000
--- a/tasks/systemd.yml
+++ /dev/null
@@ -1,9 +0,0 @@
-- name: Write Systemd Unit
-  template:
-    src: prometheus/systemd.unit
-    dest: /etc/systemd/system/prometheus.service
-- name: 'Enable Prometheus [Systemd]'
-  systemd:
-    enabled: yes
-    daemon_reload: yes
-    name: prometheus
diff --git a/templates/alertmanager.runit b/templates/alertmanager.runit
new file mode 100644
index 0000000..e300d47
--- /dev/null
+++ b/templates/alertmanager.runit
@@ -0,0 +1,6 @@
+#!/bin/sh
+exec chpst -u {{prometheus_user}} \
+	{{prometheus_home_dir}}/alertmanager/alertmanager \
+	{{alertmanager_opts | replace('\n', ' ')}} \
+	>> {{alertmanager_log}} 2>&1
+
diff --git a/templates/prometheus/runit/run b/templates/prometheus.runit
index b14e2d1..b14e2d1 100644
--- a/templates/prometheus/runit/run
+++ b/templates/prometheus.runit
diff --git a/templates/prometheus/prometheus.yml b/templates/prometheus/prometheus.yml
deleted file mode 100644
index 2099b8b..0000000
--- a/templates/prometheus/prometheus.yml
+++ /dev/null
@@ -1,53 +0,0 @@
-# my global config
-global:
-  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
-  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
-  # scrape_timeout is set to the global default (10s).
-
-# Alertmanager configuration
-alerting:
-  alertmanagers:
-  - static_configs:
-    - targets:
-      # - alertmanager:9093
-
-# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
-rule_files:
-  # - "first_rules.yml"
-  # - "second_rules.yml"
-
-# A scrape configuration containing exactly one endpoint to scrape:
-# Here it's Prometheus itself.
-scrape_configs:
-  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
-  - job_name: 'prometheus'
-
-    # metrics_path defaults to '/metrics'
-    # scheme defaults to 'http'.
-
-    static_configs:
-    - targets: ['localhost:9090']
-
-  - job_name: 'netdata'
-
-    metrics_path: /api/v1/allmetrics
-    params:
-      format: [ prometheus ]
-
-    static_configs:
-{% for h in prometheus_netdata_hosts %}      - targets: ['{{h}}']
-{%endfor%}
-
-{% if prometheus_netdata_tls_hosts is defined %}
-  - job_name: 'netdata_tls'
-
-    scheme: https
-
-    metrics_path: /api/v1/allmetrics
-    params:
-      format: [ prometheus ]
-
-    static_configs:
-{% for h in prometheus_netdata_tls_hosts %}      - targets: ['{{h}}']
-{%endfor%}
-{% endif %}
diff --git a/templates/prometheus/systemd.unit b/templates/prometheus/systemd.unit
deleted file mode 100644
index b1bd6cb..0000000
--- a/templates/prometheus/systemd.unit
+++ /dev/null
@@ -1,12 +0,0 @@
-# /etc/systemd/system/prometheus.service
-[Unit]
-Description=Prometheus Server
-Documentation=https://prometheus.io/docs/introduction/overview/
-After=network-online.target
-
-[Service]
-User=prometheus
-Restart=on-failure
-ExecStart={{prometheus_home_dir}}/prometheus/prometheus {{prometheus_opts | replace('\n', ' ')}}
-[Install]
-WantedBy=multi-user.target