diff options
Diffstat (limited to 'f3s/argocd-apps')
| -rw-r--r-- | f3s/argocd-apps/monitoring/prometheus.yaml | 32 |
1 files changed, 32 insertions, 0 deletions
diff --git a/f3s/argocd-apps/monitoring/prometheus.yaml b/f3s/argocd-apps/monitoring/prometheus.yaml index 3d88e39..f59be60 100644 --- a/f3s/argocd-apps/monitoring/prometheus.yaml +++ b/f3s/argocd-apps/monitoring/prometheus.yaml @@ -58,6 +58,25 @@ spec: kubeSchedulerRecording: false kubeScheduler: false + # Enable the textfile collector in the node_exporter DaemonSet so + # check-nfs-mount.sh can expose nfs_mount_monitor_consecutive_failures + # as a Prometheus metric without needing a separate exporter. + # The hostPath mount maps /var/lib/node_exporter/textfile_collector + # on the node into the container; the extra arg tells node_exporter + # to scan that directory for .prom files. + prometheus-node-exporter: + extraArgs: + - --collector.textfile.directory=/host/textfile_collector + extraVolumes: + - name: textfile-collector + hostPath: + path: /var/lib/node_exporter/textfile_collector + type: DirectoryOrCreate + extraVolumeMounts: + - name: textfile-collector + mountPath: /host/textfile_collector + readOnly: true + prometheus: prometheusSpec: additionalArgs: @@ -110,6 +129,16 @@ spec: group_by: ['alertname', 'name', 'severity'] group_wait: 10s repeat_interval: 6h + # NFS auto-repair alerts from r0/r1/r2 — short group_wait so + # operators are notified quickly when the mount breaks. + # repeat_interval is short: NFS outages are urgent and + # the auto-reboot cycle takes only ~30 s per node. + - matchers: + - component = "nfs" + receiver: 'nfs-alerts' + group_by: ['alertname', 'host', 'severity'] + group_wait: 10s + repeat_interval: 30m # Container image CVEs from Trivy Operator (see trivy-operator ArgoCD app) - matchers: - component = "trivy" @@ -124,6 +153,9 @@ spec: - name: 'argocd-alerts' # ArgoCD-specific receiver - alerts visible in UI only # Future: add email/slack/webhook configuration here + - name: 'nfs-alerts' + # NFS auto-repair alerts — visible in Alertmanager UI + # Future: add webhook/email for on-call paging here - name: 'trivy-alerts' # Trivy Operator CVE alerts - visible in Alertmanager UI; add webhook/email when desired inhibit_rules: |