ansible/roles/prometheus/templates/prometheus/alert.rules.yml.j2
2019-11-01 14:16:32 +01:00

62 lines
2 KiB
Django/Jinja

# {{ ansible_managed }}
{# As this is also Jinja2 it will conflict without a raw block #}
{# Depending of Prometheus Node exporter version, rules can change depending of version #}
{% raw %}
groups:
- name: alert.rules
rules:
# Alert for any instance that is unreachable for >3 minutes.
- alert: InstanceDown
expr: up == 0
for: 3m
labels:
severity: critical
annotations:
summary: "{{ $labels.instance }} est invisible depuis plus de 3 minutes !"
# Alert for out of memory
- alert: OutOfMemory
expr: (node_memory_MemFree_bytes + node_memory_Cached_bytes + node_memory_Buffers_bytes) / node_memory_MemTotal_bytes * 100 < 10
for: 5m
labels:
severity: warning
annotations:
summary: "Mémoire libre de {{ $labels.instance }} à {{ $value }}%."
# Alert for out of disk space
- alert: OutOfDiskSpace
expr: node_filesystem_free_bytes{fstype="ext4"} / node_filesystem_size_bytes{fstype="ext4"} * 100 < 10
for: 5m
labels:
severity: warning
annotations:
summary: "Espace libre de {{ $labels.mountpoint }} sur {{ $labels.instance }} à {{ $value }}%."
# Alert for out of inode space on disk
- alert: OutOfInodes
expr: node_filesystem_files_free{fstype="ext4"} / node_filesystem_files{fstype="ext4"} * 100 < 10
for: 5m
labels:
severity: warning
annotations:
summary: "Presque plus d'inodes disponibles ({{ $value }}% restant) dans {{ $labels.mountpoint }} sur {{ $labels.instance }}."
# Alert for high CPU usage
- alert: CpuUsage
expr: (100 - avg by (instance) (irate(node_cpu_seconds_total{mode="idle"}[5m])) * 100) > 75
for: 10m
labels:
severity: warning
annotations:
summary: "CPU sur {{ $labels.instance }} à {{ $value }}%."
# Check systemd unit (> buster)
- alert: SystemdServiceFailed
expr: node_systemd_unit_state{state="failed"} == 1
for: 10m
labels:
severity: warning
annotations:
summary: "{{ $labels.name }} a échoué sur {{ $labels.instance }}"
{% endraw %}