WIP Add default alertmanager route and receiver

This commit is contained in:
2026-07-06 11:35:32 +02:00
parent 5e7a0c07b0
commit 65c1933a99
3 changed files with 57 additions and 7 deletions
@@ -9,6 +9,27 @@ node.default["kosmos_prometheus"]["alertmanager"]["checksum"] = "8ce11c42e8a6dfb
node.default["kosmos_prometheus"]["node_exporter"]["version"] = "1.11.1"
node.default["kosmos_prometheus"]["node_exporter"]["checksum"] = "9f5ea48e5bc7b656f8a91a32e7d7deb89f70f73dabd0d974418aca15f37d6810"
node.default["kosmos_prometheus"]["alertmanager"]["route"] = {
"receiver" => "xmpp",
"group_by" => ["alertname"],
"group_wait" => "30s",
"group_interval" => "5m",
"repeat_interval" => "3h",
"routes" => [],
}
node.default["kosmos_prometheus"]["alertmanager"]["receivers"] = [
{
"name" => "xmpp",
"webhook_configs" => [
{
"send_resolved" => true,
"room" => "info@kosmos.chat",
}
]
}
]
node.default["kosmos_prometheus"]["global"] = {
"scrape_interval" => "30s",
"evaluation_interval" => "30s",
@@ -26,4 +47,6 @@ node.default["kosmos_prometheus"]["jobs"] = {
"prometheus" => { "targets" => [{ "target" => "localhost:#{node["kosmos_prometheus"]["web"]["port"]}", "instance" => "localhost" }] },
}
node.default["kosmos_prometheus"]["rule_files"] = []
node.default["kosmos_prometheus"]["rule_files"] = [
"/etc/prometheus/rules/*.yml"
]
@@ -55,12 +55,20 @@ file "/usr/local/bin/alertmanager" do
notifies :restart, "service[alertmanager]", :delayed
end
# TODO use Auth header once supported by hubot-incoming-webhook
credentials = data_bag_item("credentials", "hal8000_xmpp")
webhook_url = "https://hal8000.kosmos.chat/incoming/#{credentials['webhook_token']}"
template "/etc/prometheus/alertmanager.yml" do
source "alertmanager.yml.erb"
owner "root"
group "alertmanager"
mode "0644"
variables route: node["kosmos_prometheus"]["alertmanager"]["route"],
receivers: node["kosmos_prometheus"]["alertmanager"]["receivers"],
webhook_url: webhook_url
notifies :restart, "service[alertmanager]", :delayed
sensitive true
end
systemd_unit "alertmanager.service" do
@@ -2,11 +2,30 @@ global:
resolve_timeout: 5m
route:
receiver: default
group_by: ['alertname']
group_wait: 30s
group_interval: 5m
repeat_interval: 3h
<% @route.each do |k, v| %>
<% if k == "routes" && v.is_a?(Array) && !v.empty? %>
routes:
<% v.each do |child| %>
- receiver: <%= child["receiver"] %>
<% child.each do |ck, cv| %>
<% next if ck == "receiver" %>
<%= ck %>: <%= cv.is_a?(Array) ? cv.inspect : cv %>
<% end %>
<% end %>
<% else %>
<%= k %>: <%= v.is_a?(Array) ? v.inspect : v %>
<% end %>
<% end %>
receivers:
- name: default
<% @receivers.each do |receiver| %>
- name: <%= receiver["name"] %>
<% receiver["webhook_configs"].each do |wc| %>
webhook_configs:
- url: "<%= @webhook_url %>"
send_resolved: <%= wc["send_resolved"] %>
payload:
room: "<%= wc["room"] %>"
message: '{{ .Status | toUpper }}: {{ .CommonLabels.alertname }}{{ if .CommonLabels.instance }} on {{ .CommonLabels.instance }}{{ end }}{{ if .CommonLabels.vm_host }} (host: {{ .CommonLabels.vm_host }}){{ end }} — {{ .CommonAnnotations.summary }}'
<% end %>
<% end %>