kylegato / kube-dns-monitoring

Basic docker image that runs an nslookup every 30 seconds to ensure DNS is functioning correctly, otherwise outputs an error to stdout.

kube-dns-monitoring

Basic docker image that runs an host lookup against the kube-dns service running in the kube-system namespace every 30 seconds to ensure DNS is functioning correctly, otherwise outputs an error to stdout.

Usage

Run it as a daemonset in the kube-system namespace.

kubectl create -f ./kube-dns-monitoring-ds.yaml

Monitoring

1. Check the pod for # of restarts (EX: kubectl -n kube-system get pods | grep kube-dns-monitoring)
2. Use your favorite log aggregator (Splunk/LogDNA/Papertrail) to setup alerts on the output from the script

Sample outputs:

DNS Okay:

1518716342 - Thu Feb 15 17:39:02 UTC 2018 - OK: DNS working as expected.

DNS Not Okay:

1518716173 - Thu Feb 15 17:36:13 UTC 2018 - ERROR: DNS Lookup for kube-dns.kube-system.svc.cluster.local failed.