gcloud-lab/trading-platform/helm/trading-platform/templates/observability/solana-quant-bot-prometheus-rules.yaml
Hermes Agent da8fe22f6e feat(devops): add Solana quant bot K8s/Helm, CI/CD, observability, and funding
- Add solana-quant-bot deployment + ClusterIP service templates
- Add solana-rpc ExternalName service for Helius/QuickNode
- Add Alembic migration init job template
- Update values.yaml: solana-rpc, bot config, new Kafka topics, observability
- Update configmap with Solana RPC endpoint via internal DNS
- Update network policies: bot ingress/egress, postgres/redis/kafka access
- Update CI/CD workflow: build/push/deploy solana-quant-bot service
- Add Prometheus ServiceMonitor, PrometheusRules (alerts)
- Add Grafana dashboard ConfigMap (trades, latency, balance, P&L)
- Add initial funding script with dry-run and verification modes
2026-05-24 04:46:58 +00:00

70 lines
2.5 KiB
YAML

{{- if .Values.solanaQuantBot.enabled }}
{{- if .Values.alertmanager.enabled }}
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
name: {{ .Values.solanaQuantBot.name }}-alerts
namespace: {{ .Values.global.namespace }}
labels:
{{- include "trading-platform.labels" . | nindent 4 }}
app.kubernetes.io/part-of: solana-quant-bot
release: prometheus
spec:
groups:
- name: solana-quant-bot.rules
rules:
# Alert if bot pod is down
- alert: SolanaQuantBotDown
expr: up{job="{{ .Values.solanaQuantBot.name }}"} == 0
for: 5m
labels:
severity: critical
annotations:
summary: "Solana Quant Bot instance {{ '{{ $instance }}' }} is down"
description: "Bot has been unreachable for more than 5 minutes."
# High error rate on trades
- alert: SolanaQuantBotHighErrorRate
expr: |
sum(rate(solana_quant_bot_trade_errors_total[5m]))
/ sum(rate(solana_quant_bot_trades_total[5m])) > 0.05
for: 10m
labels:
severity: warning
annotations:
summary: "Solana Quant Bot trade error rate is >5%"
description: "Error rate: {{ '{{ $value }}' }} over 5m window."
# No trades executed in a long time (stale bot)
- alert: SolanaQuantBotNoTrades
expr: |
time() - increase(solana_quant_bot_last_trade_timestamp[1h]) > 3600
for: 15m
labels:
severity: warning
annotations:
summary: "Solana Quant Bot has not executed trades in 1 hour"
description: "Bot may be stuck or signals pipeline is blocked."
# High signal processing latency
- alert: SolanaQuantBotHighLatency
expr: |
histogram_quantile(0.95, sum(rate(solana_quant_bot_signal_latency_seconds_bucket[5m])) by (le)) > 2
for: 10m
labels:
severity: warning
annotations:
summary: "Solana Quant Bot p95 signal latency > 2s"
description: "p95 latency: {{ '{{ $value }}' }}s."
# Wallet balance below threshold
- alert: SolanaQuantBotLowBalance
expr: solana_quant_bot_sol_balance < 10
for: 5m
labels:
severity: critical
annotations:
summary: "Solana Quant Bot SOL balance is below 10 SOL"
description: "Current balance: {{ '{{ $value }}' }} SOL."
{{- end }}
{{- end }}