- Remove Cockpit remote_write from GitLab Prometheus (all metrics local) - Uninstall k8s-monitoring Helm (Alloy no longer pushes to Cockpit) - Deploy standalone node-exporter DaemonSet + kube-state-metrics - Add 10 extra Prometheus scrape configs: node-exporter, kube-state-metrics, kubelet, cadvisor, DCGM GPU, foxhunt services, Loki, Tempo, Promtail, prometheus.io annotated pods - Provision 9 Grafana dashboards via sidecar (ConfigMap-based): overview, services, GPU/training, infrastructure, logs, CI/CD pipelines, cluster overview, GPU overview, GitLab services - Expand Promtail to ingest kube-system + GitLab runner + node-level logs - Add dashboard.fxhnt.ai DNS record + nginx routing to Grafana - Update CI deploy job to apply new monitoring manifests Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
266 lines
8.4 KiB
YAML
266 lines
8.4 KiB
YAML
# Promtail — ships container logs to Loki (DaemonSet on all nodes)
|
|
apiVersion: v1
|
|
kind: ServiceAccount
|
|
metadata:
|
|
name: promtail
|
|
namespace: foxhunt
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
---
|
|
apiVersion: rbac.authorization.k8s.io/v1
|
|
kind: ClusterRole
|
|
metadata:
|
|
name: promtail
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
rules:
|
|
- apiGroups: [""]
|
|
resources: ["nodes", "nodes/proxy", "services", "endpoints", "pods"]
|
|
verbs: ["get", "watch", "list"]
|
|
---
|
|
apiVersion: rbac.authorization.k8s.io/v1
|
|
kind: ClusterRoleBinding
|
|
metadata:
|
|
name: promtail
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
subjects:
|
|
- kind: ServiceAccount
|
|
name: promtail
|
|
namespace: foxhunt
|
|
roleRef:
|
|
kind: ClusterRole
|
|
name: promtail
|
|
apiGroup: rbac.authorization.k8s.io
|
|
---
|
|
apiVersion: v1
|
|
kind: ConfigMap
|
|
metadata:
|
|
name: promtail-config
|
|
namespace: foxhunt
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
data:
|
|
promtail.yaml: |
|
|
server:
|
|
http_listen_port: 9080
|
|
positions:
|
|
filename: /run/promtail/positions.yaml
|
|
clients:
|
|
- url: http://loki.foxhunt.svc.cluster.local:3100/loki/api/v1/push
|
|
scrape_configs:
|
|
# Foxhunt application pods (structured Rust logs)
|
|
- job_name: foxhunt-pods
|
|
kubernetes_sd_configs:
|
|
- role: pod
|
|
relabel_configs:
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
action: keep
|
|
regex: foxhunt
|
|
# Drop noisy infra pods (gitlab internals, prometheus, grafana)
|
|
- source_labels: [__meta_kubernetes_pod_label_app]
|
|
action: drop
|
|
regex: (gitlab|prometheus|grafana|promtail)
|
|
- source_labels: [__meta_kubernetes_pod_label_app_kubernetes_io_name]
|
|
action: drop
|
|
regex: (promtail|node-exporter|kube-state-metrics)
|
|
# Map to containerd log path
|
|
- source_labels: [__meta_kubernetes_pod_uid, __meta_kubernetes_pod_container_name]
|
|
separator: /
|
|
target_label: __path__
|
|
replacement: /var/log/pods/*$1/*.log
|
|
# Standard K8s metadata labels
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
target_label: namespace
|
|
- source_labels: [__meta_kubernetes_pod_name]
|
|
target_label: pod
|
|
- source_labels: [__meta_kubernetes_pod_container_name]
|
|
target_label: container
|
|
- source_labels: [__meta_kubernetes_pod_label_app_kubernetes_io_name]
|
|
target_label: app
|
|
- source_labels: [__meta_kubernetes_pod_label_foxhunt_job__type]
|
|
target_label: job_type
|
|
- source_labels: [__meta_kubernetes_pod_label_foxhunt_model]
|
|
target_label: model
|
|
- source_labels: [__meta_kubernetes_pod_node_name]
|
|
target_label: node
|
|
pipeline_stages:
|
|
- json:
|
|
expressions:
|
|
level: level
|
|
message: fields.message
|
|
target: target
|
|
timestamp: timestamp
|
|
- labels:
|
|
level:
|
|
target:
|
|
- match:
|
|
selector: '{level=""}'
|
|
stages:
|
|
- regex:
|
|
expression: '(?P<level>TRACE|DEBUG|INFO|WARN|ERROR)'
|
|
- labels:
|
|
level:
|
|
|
|
# Kapsule system pods (kube-system, kube-node-lease, kube-public)
|
|
- job_name: kapsule-system
|
|
kubernetes_sd_configs:
|
|
- role: pod
|
|
relabel_configs:
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
action: keep
|
|
regex: (kube-system|kube-node-lease|kube-public)
|
|
# Map to containerd log path
|
|
- source_labels: [__meta_kubernetes_pod_uid, __meta_kubernetes_pod_container_name]
|
|
separator: /
|
|
target_label: __path__
|
|
replacement: /var/log/pods/*$1/*.log
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
target_label: namespace
|
|
- source_labels: [__meta_kubernetes_pod_name]
|
|
target_label: pod
|
|
- source_labels: [__meta_kubernetes_pod_container_name]
|
|
target_label: container
|
|
- source_labels: [__meta_kubernetes_pod_label_app_kubernetes_io_name]
|
|
target_label: app
|
|
- source_labels: [__meta_kubernetes_pod_label_app]
|
|
target_label: kube_app
|
|
- source_labels: [__meta_kubernetes_pod_node_name]
|
|
target_label: node
|
|
pipeline_stages:
|
|
- match:
|
|
selector: '{namespace="kube-system"}'
|
|
stages:
|
|
- regex:
|
|
expression: '(?P<level>TRACE|DEBUG|INFO|WARN|ERROR|warning|error|fatal)'
|
|
- labels:
|
|
level:
|
|
|
|
# GitLab CI runner logs (foxhunt namespace, runner pods)
|
|
- job_name: gitlab-runners
|
|
kubernetes_sd_configs:
|
|
- role: pod
|
|
relabel_configs:
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
action: keep
|
|
regex: foxhunt
|
|
- source_labels: [__meta_kubernetes_pod_label_app]
|
|
action: keep
|
|
regex: gitlab-runner
|
|
- source_labels: [__meta_kubernetes_pod_uid, __meta_kubernetes_pod_container_name]
|
|
separator: /
|
|
target_label: __path__
|
|
replacement: /var/log/pods/*$1/*.log
|
|
- source_labels: [__meta_kubernetes_namespace]
|
|
target_label: namespace
|
|
- source_labels: [__meta_kubernetes_pod_name]
|
|
target_label: pod
|
|
- source_labels: [__meta_kubernetes_pod_container_name]
|
|
target_label: container
|
|
- replacement: gitlab-runner
|
|
target_label: app
|
|
- source_labels: [__meta_kubernetes_pod_node_name]
|
|
target_label: node
|
|
pipeline_stages:
|
|
- regex:
|
|
expression: '(?P<level>TRACE|DEBUG|INFO|WARN|ERROR|WARNING)'
|
|
- labels:
|
|
level:
|
|
|
|
# Kubernetes node-level logs (kubelet, containerd)
|
|
- job_name: kapsule-node-logs
|
|
static_configs:
|
|
- targets: [localhost]
|
|
labels:
|
|
job: kapsule-nodes
|
|
__path__: /var/log/syslog
|
|
- targets: [localhost]
|
|
labels:
|
|
job: kapsule-nodes
|
|
__path__: /var/log/kern.log
|
|
- targets: [localhost]
|
|
labels:
|
|
job: kapsule-nodes
|
|
__path__: /var/log/kubelet.log
|
|
relabel_configs:
|
|
- source_labels: [__path__]
|
|
regex: .*/(.+)\.log
|
|
target_label: log_type
|
|
pipeline_stages:
|
|
- regex:
|
|
expression: '(?P<level>TRACE|DEBUG|INFO|WARN|ERROR|warning|error|fatal)'
|
|
- labels:
|
|
level:
|
|
---
|
|
apiVersion: apps/v1
|
|
kind: DaemonSet
|
|
metadata:
|
|
name: promtail
|
|
namespace: foxhunt
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
spec:
|
|
selector:
|
|
matchLabels:
|
|
app.kubernetes.io/name: promtail
|
|
template:
|
|
metadata:
|
|
labels:
|
|
app.kubernetes.io/name: promtail
|
|
spec:
|
|
serviceAccountName: promtail
|
|
tolerations:
|
|
- effect: NoSchedule
|
|
operator: Exists
|
|
containers:
|
|
- name: promtail
|
|
image: grafana/promtail:3.4.2
|
|
args: ["-config.file=/etc/promtail/promtail.yaml"]
|
|
ports:
|
|
- containerPort: 9080
|
|
name: http
|
|
volumeMounts:
|
|
- name: config
|
|
mountPath: /etc/promtail
|
|
- name: varlog
|
|
mountPath: /var/log
|
|
readOnly: true
|
|
- name: containers
|
|
mountPath: /var/lib/docker/containers
|
|
readOnly: true
|
|
- name: pods
|
|
mountPath: /var/log/pods
|
|
readOnly: true
|
|
- name: run
|
|
mountPath: /run/promtail
|
|
env:
|
|
- name: HOSTNAME
|
|
valueFrom:
|
|
fieldRef:
|
|
fieldPath: spec.nodeName
|
|
resources:
|
|
requests:
|
|
cpu: 50m
|
|
memory: 64Mi
|
|
limits:
|
|
cpu: 200m
|
|
memory: 256Mi
|
|
securityContext:
|
|
readOnlyRootFilesystem: true
|
|
allowPrivilegeEscalation: false
|
|
volumes:
|
|
- name: config
|
|
configMap:
|
|
name: promtail-config
|
|
- name: varlog
|
|
hostPath:
|
|
path: /var/log
|
|
- name: containers
|
|
hostPath:
|
|
path: /var/lib/docker/containers
|
|
- name: pods
|
|
hostPath:
|
|
path: /var/log/pods
|
|
- name: run
|
|
emptyDir: {}
|