Externel NFS Storage Health Check

[Deprecated]

Please see new project - nfs-exporter

Check Externel NFS

NFS Storage, Prometheus and k8s cluster

architecture

How to check external NFS server?

Using Command showmount -e ${NFS_SERVER_IP} to check NFS server's health.

If the showmount command gets normally pod nfs-health-check are always Running, it will be 'Error' or 'CrashLoopBackup' status to the contrary.

(showmount will return exit status when get error.)

Last, we can use prometheus rules to monitor nfs-health-check pod to get external NFS server status.

Prometheus Rules

- alert: NFSServerDown
annotations:
    summary: 'External NFS server is down!'
expr: |
    # get pod restart rate from 5m to check nfs server status
    rate(kube_pod_container_status_restarts_total{pod=~"(nfs-health-check).*"}[5m]) > 0
for: 5m
labels:
    severity: critical

make images

make docker

Compile

make build

ENV

Using environment variables NFS_SERVER, NFS_PATH to point the external NFS server.

Like this in deploy.yaml

containers:
- name: nfs-health-check
    image: aixeshunter/nfs-health-check:v1.1
    env:
    - name: NFS_SERVER
        value: 127.0.0.1
    - name: NFS_PATH
        value: /opt/nfs