prysm-pulse/k8s/monitoring/alerts.yaml
2019-05-11 09:21:21 -07:00

31 lines
1000 B
YAML

apiVersion: v1
kind: ConfigMap
metadata:
name: prometheus-alerts
namespace: istio-system
data:
prometheus.rules.yml: |-
groups:
- name: prysm
rules:
- alert: stalled_chain
expr: delta(avg(state_last_slot > scalar(max(state_last_slot) - 100))[10m:30s]) < 50
for: 1m
annotations:
summary: No block slots advanced in 2 minutes
- alert: too_long_since_finality
expr: max(state_last_slot / 8) - floor(max(state_last_finalized_epoch)) > 10
for: 1m
annotations:
summary: No finality in 10 epochs
- alert: high_reorg_rate
expr: max(delta(reorg_counter[10m])) > 5
for: 1m
annotations:
summary: Some nodes are seeing more than 5 reorgs in 10 minutes
- alert: high_goroutines
expr: max_over_time(go_goroutines{component="beacon-chain"}[1m]) > 1000
for: 1m
annotations:
summary: Some nodes are experencing more than 1000 goroutines