managarten/docker/promtail/config.yaml
Till JS 7908995a29 feat(monitoring): structured logging, Promtail alignment, GlitchTip config, status page
- Upgrade shared-logger to dual-mode: JSON lines in production, console
  in dev. Adds configureLogger() for service name + request ID.
- Add requestLogger middleware to shared-hono with request ID generation
  and structured request/response logging.
- Align Promtail config with new JSON field names (requestId, ts, service).
- Add PUBLIC_GLITCHTIP_DSN + PUBLIC_UMAMI_WEBSITE_ID to mana-web docker config.
- Add /status page that polls all backend /health endpoints server-side.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-02 17:23:52 +02:00

116 lines
3.5 KiB
YAML

server:
http_listen_port: 9080
grpc_listen_port: 0
positions:
filename: /tmp/positions.yaml
clients:
- url: http://loki:3100/loki/api/v1/push
batchwait: 3s
batchsize: 1048576 # 1 MB
scrape_configs:
- job_name: docker
docker_sd_configs:
- host: unix:///var/run/docker.sock
refresh_interval: 30s
filters:
# Only collect from our compose project
- name: label
values: ["com.docker.compose.project"]
relabel_configs:
# Extract compose service name → label "service"
- source_labels: ["__meta_docker_container_label_com_docker_compose_service"]
target_label: "service"
# Extract container name → label "container"
- source_labels: ["__meta_docker_container_name"]
regex: "/(.*)"
target_label: "container"
# Extract compose project → label "project"
- source_labels: ["__meta_docker_container_label_com_docker_compose_project"]
target_label: "project"
# Tier labels based on container name prefix for easy filtering
# mana-infra-* → tier=infra
- source_labels: ["container"]
regex: "mana-infra-.*"
target_label: "tier"
replacement: "infra"
# mana-core-* → tier=core
- source_labels: ["container"]
regex: "mana-core-.*"
target_label: "tier"
replacement: "core"
# mana-auth/credits/user/subscriptions/analytics → tier=auth
- source_labels: ["container"]
regex: "mana-(auth|credits|user|subscriptions|analytics)"
target_label: "tier"
replacement: "auth"
# mana-app-* → tier=app
- source_labels: ["container"]
regex: "mana-app-.*"
target_label: "tier"
replacement: "app"
# mana-mon-* → tier=monitoring
- source_labels: ["container"]
regex: "mana-mon-.*"
target_label: "tier"
replacement: "monitoring"
# mana-matrix-* → tier=matrix
- source_labels: ["container"]
regex: "mana-matrix-.*"
target_label: "tier"
replacement: "matrix"
# mana-game-* → tier=games
- source_labels: ["container"]
regex: "mana-game-.*"
target_label: "tier"
replacement: "games"
# mana-service-* → tier=service
- source_labels: ["container"]
regex: "mana-service-.*"
target_label: "tier"
replacement: "service"
# Default tier for anything unmatched
- source_labels: ["tier"]
regex: "^$"
target_label: "tier"
replacement: "other"
pipeline_stages:
# Drop monitoring container logs to save space (they're noisy)
- match:
selector: '{tier="monitoring"}'
action: drop
# Try to parse JSON logs (Go services, Hono services)
- json:
expressions:
level: level
msg: msg
error: error
status: status
method: method
path: path
duration: duration
request_id: requestId
service_name: service
# Fall back: extract level from common log patterns
- regex:
expression: '(?i)(?P<level>error|warn|info|debug|fatal|panic)'
# Normalize level label
- labels:
level:
service_name:
# Add timestamp from log if available
- timestamp:
source: ts
format: RFC3339Nano
fallback_formats:
- "2006-01-02T15:04:05.000Z"
- "2006-01-02 15:04:05"
action_on_failure: fudge