2026-10-07 21:20:01 +02:00
|
|
|
groups:
|
|
|
|
|
- name: syllabus
|
|
|
|
|
rules:
|
|
|
|
|
- alert: DisqueSature
|
|
|
|
|
expr: 100 - (node_filesystem_avail_bytes{mountpoint="/",fstype!="tmpfs"} / node_filesystem_size_bytes{mountpoint="/",fstype!="tmpfs"} * 100) > 80
|
|
|
|
|
for: 5m
|
|
|
|
|
labels:
|
|
|
|
|
severite: avertissement
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "Disque / rempli a plus de 80%"
|
|
|
|
|
description: "Occupation du disque racine superieure a 80% depuis plus de 5 minutes."
|
|
|
|
|
|
|
|
|
|
- alert: MemoireSaturee
|
|
|
|
|
expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes)) * 100 > 90
|
|
|
|
|
for: 5m
|
|
|
|
|
labels:
|
|
|
|
|
severite: avertissement
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "RAM utilisee a plus de 90%"
|
|
|
|
|
description: "Memoire disponible inferieure a 10% depuis plus de 5 minutes."
|
|
|
|
|
|
|
|
|
|
- alert: ServiceOpenStackArrete
|
|
|
|
|
expr: openstack_nova_agent_state == 0 or openstack_neutron_agent_state == 0
|
|
|
|
|
for: 2m
|
|
|
|
|
labels:
|
|
|
|
|
severite: critique
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "Un agent OpenStack est arrete"
|
|
|
|
|
description: "Un agent Nova ou Neutron signale un etat different de UP depuis plus de 2 minutes."
|
|
|
|
|
|
2026-10-09 10:39:50 +02:00
|
|
|
- alert: CertificatExpireBientot
|
|
|
|
|
expr: (syllabus_certificat_expiration_timestamp - time()) < 14 * 86400
|
|
|
|
|
for: 1h
|
|
|
|
|
labels:
|
|
|
|
|
severite: critique
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "Le certificat HTTPS de {{ $labels.domaine }} expire dans moins de 14 jours"
|
|
|
|
|
description: "Le renouvellement automatique (certbot.timer) a echoue : voir docs/runbooks/https.md."
|
|
|
|
|
|
|
|
|
|
- alert: CertificatNonMesure
|
|
|
|
|
expr: absent(syllabus_certificat_expiration_timestamp)
|
|
|
|
|
for: 2d
|
|
|
|
|
labels:
|
|
|
|
|
severite: avertissement
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "L'expiration du certificat HTTPS n'est plus mesuree"
|
|
|
|
|
description: "Le fichier certificat.prom du collecteur textfile n'est plus produit : verifier le cron syllabus-expiration-certificat."
|
|
|
|
|
|
2026-10-09 17:28:50 +02:00
|
|
|
- alert: RestaurationNonTestee
|
|
|
|
|
expr: (time() - syllabus_derniere_restauration_reussie_timestamp) > 8 * 86400
|
|
|
|
|
for: 1h
|
|
|
|
|
labels:
|
|
|
|
|
severite: avertissement
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "Aucun test de restauration du portail reussi depuis plus de 8 jours"
|
|
|
|
|
description: "Voir journalctl -u syllabus-test-restauration et docs/runbooks/sauvegardes.md."
|
|
|
|
|
|
2026-10-07 21:20:01 +02:00
|
|
|
- alert: SauvegardeAncienne
|
|
|
|
|
expr: (time() - syllabus_derniere_sauvegarde_reussie_timestamp) > 26 * 3600
|
|
|
|
|
for: 10m
|
|
|
|
|
labels:
|
|
|
|
|
severite: critique
|
|
|
|
|
annotations:
|
|
|
|
|
summary: "Derniere sauvegarde reussie il y a plus de 26h"
|
|
|
|
|
description: "Le timer de sauvegarde Dedibackup n'a pas reussi depuis plus de 26 heures : verifier /root/.dedibackup et le journal du service."
|