role service-prometheus: update targets and alerts

This commit is contained in:
Julian Labus 2018-10-04 18:50:09 +02:00
parent a287a61d5b
commit 8bed1b7c5b
No known key found for this signature in database
GPG key ID: 8AF209F2C6B3572A
4 changed files with 12 additions and 33 deletions

View file

@ -26,6 +26,7 @@ prometheus_components:
- node_exporter
- blackbox_exporter
prometheus_opts: "--web.external-url=https://{{ http_prometheus_prefix }}.{{ http_domain_external }}"
alertmanager_opts: "--web.external-url=https://{{ http_prometheus_prefix }}.{{ http_domain_external }}/alertmanager/"
yanic_blacklist:

View file

@ -4,8 +4,6 @@ groups:
- alert: Icmp4Timeout
expr: probe_success{job="icmp4"} == 0
for: 5m
labels:
severity: page
annotations:
description: 'ICMP requests to the primary IPv4 address timed out'
summary: 'Instance {{ $labels.instance }} does not respond to ICMPv4 echo requests'
@ -13,8 +11,6 @@ groups:
- alert: Icmp6Timeout
expr: probe_success{job="icmp6"} == 0
for: 5m
labels:
severity: page
annotations:
description: 'ICMP requests to the primary IPv6 address timed out'
summary: 'Instance {{ $labels.instance }} does not respond to ICMPv6 echo requests'
@ -24,8 +20,6 @@ groups:
- alert: InstanceDown
expr: up{job="node"} == 0
for: 5m
labels:
severity: page
annotations:
description: 'The instance is down for more than 5 minutes'
summary: 'Instance {{ $labels.instance }} is down'
@ -48,8 +42,6 @@ groups:
- alert: InstanceHighCpuLong
expr: 100 - (avg(rate(node_cpu_seconds_total{mode="idle"}[5m])) BY (instance) * 100) > 90
for: 30m
labels:
severity: page
annotations:
description: 'CPU usage above 90% for more than 30m'
summary: 'Instance {{ $labels.instance }}: persistent cpu usage at {{ $value }}'
@ -58,8 +50,6 @@ groups:
- alert: InstanceLowMem
expr: node_memory_MemAvailable_bytes / 1024 / 1024 < node_memory_MemTotal_bytes / 1024 / 1024 / 10
for: 3m
labels:
severity: page
annotations:
description: 'Less than 10% of free memory'
summary: 'Instance {{ $labels.instance }}: {{ $value }}MB of free memory'
@ -68,8 +58,6 @@ groups:
- alert: InstanceLowDiskPrediction4Hours
expr: predict_linear(node_filesystem_free_bytes{device=~"/dev/.*",job="node"}[1h], 4 * 3600) < 0
for: 30m
labels:
severity: page
annotations:
description: 'Disk {{ $labels.mountpoint }} ({{ $labels.device }}) will be full in less than 4 hours'
summary: 'Instance {{ $labels.instance }}: Disk {{ $labels.mountpoint }} ({{ $labels.device }}) will be full in less than 4 hours'
@ -77,8 +65,6 @@ groups:
- alert: InstanceLowDiskPrediction12Hours
expr: predict_linear(node_filesystem_free_bytes{device=~"/dev/.*",job="node"}[3h], 12 * 3600) < 0
for: 60m
labels:
severity: page
annotations:
description: 'Disk {{ $labels.mountpoint }} ({{ $labels.device }}) will be full in less than 12 hours'
summary: 'Instance {{ $labels.instance }}: Disk {{ $labels.mountpoint }} ({{ $labels.device }}) will be full in less than 12 hours'
@ -86,8 +72,6 @@ groups:
- alert: InstanceLowDiskAbs
expr: node_filesystem_avail_bytes{mountpoint="/"} / 1024 / 1024 < 1024
for: 1m
labels:
severity: page
annotations:
description: 'Less than 1GB of free disk space left on the root filesystem'
summary: 'Instance {{ $labels.instance }}: {{ $value }}MB free disk space on {{ $labels.device }}'
@ -96,8 +80,6 @@ groups:
- alert: InstanceLowDiskPerc
expr: 100 * (node_filesystem_free_bytes / node_filesystem_size_bytes) < 10
for: 1m
labels:
severity: page
annotations:
description: 'Less than 10% of free disk space left on a device'
summary: 'Instance {{ $labels.instance }}: {{ $value }}% free disk space on {{ $labels.device }}'
@ -106,8 +88,6 @@ groups:
- alert: ServiceFailed
expr: node_systemd_unit_state{state="failed"} > 0
for: 1m
labels:
severity: page
annotations:
description: 'A systemd unit went into failed state'
summary: 'Instance {{ $labels.instance }}: Service {{ $labels.name }} failed'
@ -116,8 +96,6 @@ groups:
- alert: ServiceFlapping
expr: changes(node_systemd_unit_state{state="failed"}[5m]) > 5 or
(changes(node_systemd_unit_state{state="failed"}[1h]) > 15 unless changes(node_systemd_unit_state{state="failed"}[30m]) < 7)
labels:
severity: page
annotations:
description: 'A systemd service changed its state more than 5x/5min or 15x/1h'
summary: 'Instance {{ $labels.instance }}: Service {{ $labels.name }} is flapping'

View file

@ -3,11 +3,6 @@ global:
evaluation_interval: 15s # By default, scrape targets every 15 seconds.
# scrape_timeout is set to the global default (10s).
# The labels to add to any time series or alerts when communicating with
# external systems (federation, remote storage, Alertmanager).
external_labels:
monitor: 'master'
alerting:
alertmanagers:
- scheme: https
@ -51,23 +46,26 @@ scrape_configs:
- '{{ host }}:9281'
{% endfor %}
- job_name: "icmp6"
{% for job in ['icmp4','icmp6'] %}
- job_name: "{{ job }}"
metrics_path: /probe
params:
module: ["icmp6"]
module: ["{{ job }}"]
static_configs:
- targets:
{% for group in prometheus_groups %}
{% for host in groups[group] %}
{% if host != inventory_hostname %}
- "{{ host.rsplit('.freifunk-mwu.de')[0] }}.ffwi.org"
- "{{ host.rsplit('.freifunk-mwu.de')[0] }}.ffmz.org"
- "{{ host.rsplit('.')[0] }}.ffwi.org"
- "{{ host.rsplit('.')[0] }}.ffmz.org"
{% endif %}
{% endfor %}
{% endfor %}
{% for host, _ in bgp_mwu_servers.items() %}
{% if host not in ['extrasahne'] %}
- "{{ host }}.ffwi.org"
- "{{ host }}.ffmz.org"
{% endif %}
{% endfor %}
relabel_configs:
- source_labels: [__address__]
@ -76,3 +74,5 @@ scrape_configs:
target_label: instance
- target_label: __address__
replacement: 127.0.0.1:9115
{% endfor %}

View file

@ -9,7 +9,7 @@ modules:
tcp:
query_response:
- expect: "^SSH-2.0-"
icmp:
icmp4:
prober: icmp
icmp:
preferred_ip_protocol: "ip4"