Coredns: Metric coredns_cache_misses_total not showing up anymore in coredns 1.6.6

[tru-pro]

What happened:

Upgraded coredns from 1.6.5 to 1.6.6

What you expected to happen:

Cache plugin metrics should stay the same.

How to reproduce it (as minimally and precisely as possible):

1. run coredns 1.6.6 with cache and prometheus plugins enabled.
   docker run -d --name coredns --volume=/Users/u/coredns/:/root/ -p 53:53/udp -p 9153:9153/tcp coredns/coredns:1.6.6 -conf /root/Corefile
2. send one DNS request through coredns
   dig -t A google.com @127.0.0.1
3. get /metrics and look for "coredns_cache_misses_total"
   curl localhost:9153/metrics --raw -s | grep coredns_cache

Anything else we need to know?:

Last version that successfully increased this counter and emitted seems to be coredns 1.6.5.

Environment:

• the version of CoreDNS: 1.6.6
• Corefile:

.:53 {
    errors
    health
    prometheus :9153
    forward . /etc/resolv.conf
    cache 30
}

• logs, if applicable:
• metrics output after one DNS request (coredns 1.6.6):

curl 127.0.0.1:9153/metrics --raw -s | grep coredns_cache
# HELP coredns_cache_size The number of elements in the cache.
# TYPE coredns_cache_size gauge
coredns_cache_size{server="dns://:53",type="denial"} 0
coredns_cache_size{server="dns://:53",type="success"} 1

1. metrics output after one DNS request (coredns 1.6.5):

curl 127.0.0.1:9153/metrics --raw -s | grep coredns_cache
# HELP coredns_cache_misses_total The count of cache misses.
# TYPE coredns_cache_misses_total counter
coredns_cache_misses_total{server="dns://:53"} 1
# HELP coredns_cache_size The number of elements in the cache.
# TYPE coredns_cache_size gauge
coredns_cache_size{server="dns://:53",type="denial"} 0
coredns_cache_size{server="dns://:53",type="success"} 1

[zouyee]

/close