mirror of
https://github.com/luxfi/monitoring.git
synced 2026-10-30 15:24:48 +00:00
91 lines
2.8 KiB
YAML
91 lines
2.8 KiB
YAML
groups:
|
|
- name: lux_chain_alerts
|
|
interval: 30s
|
|
rules:
|
|
# Node availability
|
|
- alert: LuxNodeDown
|
|
expr: up{job="luxd"} == 0
|
|
for: 5m
|
|
labels:
|
|
severity: critical
|
|
annotations:
|
|
summary: "Lux node is down"
|
|
description: "The Lux node has been down for more than 5 minutes."
|
|
|
|
# C-Chain alerts
|
|
- alert: CChainNotSyncing
|
|
expr: rate(avalanche_C_last_accepted_height[5m]) == 0
|
|
for: 10m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "C-Chain is not syncing"
|
|
description: "C-Chain height has not increased in the last 10 minutes."
|
|
|
|
- alert: CChainHighRejectionRate
|
|
expr: rate(avalanche_C_blks_rejected_count[5m]) > rate(avalanche_C_blks_accepted_count[5m]) * 0.1
|
|
for: 5m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "C-Chain high block rejection rate"
|
|
description: "C-Chain is rejecting more than 10% of blocks."
|
|
|
|
# P-Chain alerts
|
|
- alert: PChainNotSyncing
|
|
expr: rate(avalanche_P_last_accepted_height[5m]) == 0
|
|
for: 10m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "P-Chain is not syncing"
|
|
description: "P-Chain height has not increased in the last 10 minutes."
|
|
|
|
# X-Chain alerts
|
|
- alert: XChainNotSyncing
|
|
expr: rate(avalanche_X_last_accepted_height[5m]) == 0
|
|
for: 10m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "X-Chain is not syncing"
|
|
description: "X-Chain height has not increased in the last 10 minutes."
|
|
|
|
# Network connectivity
|
|
- alert: LowPeerCount
|
|
expr: avalanche_network_peers < 5
|
|
for: 10m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "Low peer count"
|
|
description: "Node has fewer than 5 peers connected."
|
|
|
|
# Resource alerts
|
|
- alert: HighCPUUsage
|
|
expr: rate(process_cpu_seconds_total{job="luxd"}[5m]) * 100 > 80
|
|
for: 10m
|
|
labels:
|
|
severity: warning
|
|
annotations:
|
|
summary: "High CPU usage on Lux node"
|
|
description: "CPU usage is above 80% for more than 10 minutes."
|
|
|
|
- alert: HighMemoryUsage
|
|
expr: (1 - (node_memory_MemAvailable_bytes / node_memory_MemTotal_bytes)) * 100 > 90
|
|
for: 10m
|
|
labels:
|
|
severity: critical
|
|
annotations:
|
|
summary: "High memory usage"
|
|
description: "Memory usage is above 90% for more than 10 minutes."
|
|
|
|
# Database alerts
|
|
- alert: PostgresDown
|
|
expr: up{job="postgres"} == 0
|
|
for: 5m
|
|
labels:
|
|
severity: critical
|
|
annotations:
|
|
summary: "PostgreSQL is down"
|
|
description: "PostgreSQL database has been down for more than 5 minutes." |