#
include_recipe "apache"
-include_recipe "apt"
+include_recipe "apt::grafana"
include_recipe "networking"
-include_recipe "timescaledb"
passwords = data_bag_item("prometheus", "passwords")
tokens = data_bag_item("prometheus", "tokens")
prometheus_exporter "fastly" do
port 8080
- listen_switch "endpoint"
- listen_type "url"
+ listen_switch "listen"
environment "FASTLY_API_TOKEN" => tokens["fastly"]
end
+prometheus_exporter "fastly_healthcheck" do
+ port 9696
+ scrape_interval "1m"
+ environment "FASTLY_API_TOKEN" => tokens["fastly"]
+end
+
+prometheus_exporter "statuscake" do
+ port 9595
+ scrape_interval "5m"
+ scrape_timeout "2m"
+ environment "STATUSCAKE_APIKEY" => tokens["statuscake"]
+end
+
cache_dir = Chef::Config[:file_cache_path]
-prometheus_version = "2.31.1"
-alertmanager_version = "0.23.0"
-karma_version = "0.93"
+prometheus_version = "2.45.0"
+alertmanager_version = "0.25.0"
+karma_version = "0.114"
directory "/opt/prometheus-server" do
owner "root"
strip_components 1
owner "root"
group "root"
- subscribes :extract, "remote_file[#{cache_dir}/prometheus.linux-amd64.tar.gz]"
+ subscribes :extract, "remote_file[#{cache_dir}/prometheus.linux-amd64.tar.gz]", :immediately
end
remote_file "#{cache_dir}/alertmanager.linux-amd64.tar.gz" do
strip_components 1
owner "root"
group "root"
- subscribes :extract, "remote_file[#{cache_dir}/alertmanager.linux-amd64.tar.gz]"
+ subscribes :extract, "remote_file[#{cache_dir}/alertmanager.linux-amd64.tar.gz]", :immediately
end
remote_file "#{cache_dir}/karma-linux-amd64.tar.gz" do
overwrite true
owner "root"
group "root"
- subscribes :extract, "remote_file[#{cache_dir}/karma-linux-amd64.tar.gz]"
-end
-
-package %w[
- prometheus
- prometheus-alertmanager
-]
-
-promscale_version = "0.10.0"
-promscale_extension_version = "0.3.2"
-
-database_version = node[:timescaledb][:database_version]
-database_cluster = "#{database_version}/main"
-
-postgresql_user "prometheus" do
- cluster database_cluster
- superuser true
-end
-
-postgresql_database "promscale" do
- cluster database_cluster
- owner "prometheus"
-end
-
-directory "/opt/promscale" do
- owner "root"
- group "root"
- mode "755"
-end
-
-remote_file "#{cache_dir}/promscale_extension.pg#{database_version}.x86_64.deb" do
- source "https://github.com/timescale/promscale_extension/releases/download/#{promscale_extension_version}/promscale_extension-#{promscale_extension_version}.pg#{database_version}.x86_64.deb"
- owner "root"
- group "root"
- mode "644"
- backup false
-end
-
-dpkg_package "promscale-extension-postgresql-#{database_version}" do
- source "#{cache_dir}/promscale_extension.pg#{database_version}.x86_64.deb"
- version "#{promscale_extension_version}-1"
-end
-
-directory "/opt/promscale/bin" do
- owner "root"
- group "root"
- mode "755"
-end
-
-remote_file "/opt/promscale/bin/promscale" do
- action :create
- source "https://github.com/timescale/promscale/releases/download/#{promscale_version}/promscale_#{promscale_version}_Linux_x86_64"
- owner "root"
- group "root"
- mode "755"
-end
-
-systemd_service "promscale" do
- description "Promscale Connector"
- type "simple"
- user "prometheus"
- exec_start "/opt/promscale/bin/promscale --db.uri postgresql:///promscale?host=/run/postgresql&port=5432 --db.connections-max 400"
- limit_nofile 16384
- private_tmp true
- protect_system "strict"
- protect_home true
- no_new_privileges true
-end
-
-systemd_service "promscale-maintenance" do
- description "Promscale Maintenance"
- type "simple"
- user "prometheus"
- exec_start "/usr/bin/psql --command='CALL prom_api.execute_maintenance()' promscale"
- private_tmp true
- protect_system "strict"
- protect_home true
- no_new_privileges true
-end
-
-systemd_timer "promscale-maintenance" do
- description "Promscale Maintenace"
- on_active_sec 1800
- on_unit_inactive_sec 1800
-end
-
-if node[:prometheus][:promscale]
- service "promscale" do
- action [:enable, :start]
- subscribes :restart, "remote_file[/opt/promscale/bin/promscale]"
- subscribes :restart, "systemd_service[promscale]"
- end
-
- service "promscale-maintenance.timer" do
- action [:enable, :start]
- end
-else
- service "promscale" do
- action [:disable, :stop]
- end
-
- service "promscale-maintenance.timer" do
- action [:disable, :stop]
- end
+ subscribes :extract, "remote_file[#{cache_dir}/karma-linux-amd64.tar.gz]", :immediately
end
search(:node, "roles:gateway") do |gateway|
- allowed_ips = gateway.interfaces(:role => :internal).map do |interface|
- "#{interface[:network]}/#{interface[:prefix]}"
- end
+ allowed_ips = gateway.ipaddresses(:role => :internal).map(&:subnet)
node.default[:networking][:wireguard][:peers] << {
:public_key => gateway[:networking][:wireguard][:public_key],
end
jobs = {}
+junos_targets = []
snmp_targets = []
search(:node, "recipes:prometheus\\:\\:default").sort_by(&:name).each do |client|
if exporter.is_a?(Hash)
name = exporter[:name]
address = exporter[:address]
+ sni = exporter[:sni]
+ labels = Array(exporter[:labels])
+ scrape_interval = exporter[:scrape_interval]
+ scrape_timeout = exporter[:scrape_timeout]
metric_relabel = exporter[:metric_relabel] || []
else
name = key
address = exporter
+ sni = nil
+ labels = []
+ scrape_interval = nil
+ scrape_timeout = nil
metric_relabel = []
end
jobs[name] ||= []
jobs[name] << {
:address => address,
+ :sni => sni,
:instance => client.name.split(".").first,
+ :labels => labels,
+ :scrape_interval => scrape_interval,
+ :scrape_timeout => scrape_timeout,
:metric_relabel => metric_relabel
}
end
+ Hash(client[:prometheus][:junos]).each do |instance, details|
+ junos_targets << {
+ :instance => instance,
+ :target => details[:address],
+ :address => client[:prometheus][:addresses]["junos"],
+ :labels => Array(details[:labels])
+ }
+ end
+
Hash(client[:prometheus][:snmp]).each do |instance, details|
snmp_targets << {
:instance => instance,
:target => details[:address],
- :module => details[:module],
+ :modules => details[:modules],
:address => client[:prometheus][:addresses]["snmp"],
:labels => Array(details[:labels])
}
register_target false
end
+package "prometheus"
+
systemd_service "prometheus-executable" do
service "prometheus"
dropin "executable"
- exec_start "/opt/prometheus-server/prometheus/prometheus --config.file=/etc/prometheus/prometheus.yml --web.external-url=https://prometheus.openstreetmap.org/prometheus --storage.tsdb.path=/var/lib/prometheus/metrics2"
+ exec_start "/opt/prometheus-server/prometheus/prometheus --config.file=/etc/prometheus/prometheus.yml --web.external-url=https://prometheus.openstreetmap.org/prometheus --storage.tsdb.path=/var/lib/prometheus/metrics2 --storage.tsdb.retention.time=540d"
+ timeout_stop_sec 300
notifies :restart, "service[prometheus]"
end
owner "root"
group "root"
mode "644"
- variables :jobs => jobs, :snmp_targets => snmp_targets, :certificates => certificates
+ variables :jobs => jobs, :junos_targets => junos_targets, :snmp_targets => snmp_targets, :certificates => certificates
end
template "/etc/prometheus/alert_rules.yml" do
subscribes :restart, "archive_file[#{cache_dir}/prometheus.linux-amd64.tar.gz]"
end
-systemd_service "prometheus-alertmanager-executable" do
- service "prometheus-alertmanager"
- dropin "executable"
+systemd_service "prometheus-alertmanager" do
+ description "Prometheus alert manager"
+ type "simple"
+ user "prometheus"
exec_start "/opt/prometheus-server/alertmanager/alertmanager --config.file=/etc/prometheus/alertmanager.yml --storage.path=/var/lib/prometheus/alertmanager --web.external-url=https://prometheus.openstreetmap.org/alertmanager"
+ exec_reload "/bin/kill -HUP $MAINPID"
+ timeout_stop_sec 20
+ restart "on-failure"
notifies :restart, "service[prometheus-alertmanager]"
end
+link "/usr/local/bin/promtool" do
+ to "/opt/prometheus-server/prometheus/promtool"
+end
+
template "/etc/prometheus/alertmanager.yml" do
source "alertmanager.yml.erb"
owner "root"
mode "644"
end
+directory "/var/lib/prometheus/alertmanager" do
+ owner "prometheus"
+ group "prometheus"
+ mode "755"
+end
+
service "prometheus-alertmanager" do
action [:enable, :start]
subscribes :reload, "template[/etc/prometheus/alertmanager.yml]"
+ subscribes :restart, "systemd_service[prometheus-alertmanager]"
subscribes :restart, "archive_file[#{cache_dir}/alertmanager.linux-amd64.tar.gz]"
end
-template "/etc/prometheus/amtool.yml" do
+directory "/etc/amtool" do
+ owner "root"
+ group "root"
+ mode "755"
+end
+
+template "/etc/amtool/config.yml" do
source "amtool.yml.erb"
owner "root"
group "root"
mode "644"
end
+link "/usr/local/bin/amtool" do
+ to "/opt/prometheus-server/alertmanager/amtool"
+end
+
template "/etc/prometheus/karma.yml" do
source "karma.yml.erb"
owner "root"
description "Alert dashboard for Prometheus Alertmanager"
user "prometheus"
exec_start "/opt/prometheus-server/karma/karma-linux-amd64 --config.file=/etc/prometheus/karma.yml"
- private_tmp true
- private_devices true
- protect_system "full"
- protect_home true
- no_new_privileges true
+ sandbox :enable_network => true
restart "on-failure"
end
service "prometheus-karma" do
action [:enable, :start]
- subscribes :reload, "template[/etc/prometheus/karma.yml]"
+ subscribes :restart, "template[/etc/prometheus/karma.yml]"
subscribes :restart, "archive_file[#{cache_dir}/karma-linux-amd64.tar.gz]"
+ subscribes :restart, "systemd_service[prometheus-karma]"
end
package "grafana-enterprise"
apache_module "alias"
apache_module "proxy_http"
+apache_module "proxy_wstunnel"
ssl_certificate "prometheus.openstreetmap.org" do
domains ["prometheus.openstreetmap.org", "prometheus.osm.org"]
projects / chef.git / blobdiff