* Laying down kubernetes middleware foundation

* Duplicated a bunch of code form etcd middleware
* Duplicated code hacked to compile and load as a separate middleware

* Adding verbose build option to Makefile

* Removing stubzone and tls support

tls and stubzone support was carried over from base etcd middleware code.
Removing to simplify the kube middleware implementation. (For now.)

* Adding conf directory for sample conf files

* Removing stubzone support from query handler

* Remove upstream and proxy from k8s corefile.

Not sure that upstream or proxy makes sense for a k8s backed zone.

* Comment out use of singleflight serialization

* Removing parsing support for "upstream" directive from k8s

* Removing upstream directive parsing code

* Removing CNAME and TXT lookup implementation

* Create README.md

Brain-dump of DNS record name assembly and open work items.

* Adding notes about wildcard handling

* Adding basic k8s API client

* Fleshing out methods on k8s connector

* Remove PathPrefix from middleware init

* Removing incorrect plural

* Adding brute-force k8s service lookup functions

* Initializing k8s API connector during startup

* Hacking around to call k8s connector

* Parsing incoming domain name into serviceName and namespace

* Improving and simplifying k8s zone matching and label segmentation

* Removing unused functions carried over from etcd middleware

* Adding basic return of k8s data to DNS client

* updated debugging println statements to flag with "[debug]"
* removed code in kubernetes.go::Records that was a hold-over from etcd middleware.
* Removed some random exploratory hacking.

* Minior README.md updates

* Updating with demo instructions

* Updating README.md with CoreFile and removing completed TODO items

* Updating conf file and README to reflect DNS response cache works

* Disabling DNS response caching

* Adding debug statement on entry to Records()

* Changing port number in exampes to port 53.

* Misc style and clarity changes

* Removing empty function definitions

* Adding comment to track future cleanup

* Refactoring README to follow style of other middleware

* Exposing dataobject field (typo)

Always set the auth and rd bits in the reply. And reverse the
ownername of debug queries so that manual matching is much easier.

With this you can retreive the raw data that the etcd middleware
used to create the reply. The debug data is put in TXT records
that are stuffed in the CH classs. This is only enabled if you
specify `debug` in the etcd stanza.

You can retrieve it by prefixing your query with 'o-o.debug.'
For instance:

; <<>> DiG 9.10.3-P4-Ubuntu <<>> @localhost -p 1053 SRV o-o.debug.production.*.skydns.local
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 47798
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 3

;; OPT PSEUDOSECTION:
; EDNS: version: 0, flags:; udp: 4096
;; QUESTION SECTION:
;o-o.debug.production.*.skydns.local. IN	SRV

;; ANSWER SECTION:
production.*.skydns.local. 154	IN	SRV	10 50 8080 service1.example.com.
production.*.skydns.local. 154	IN	SRV	10 50 8080 service2.example.com.

;; ADDITIONAL SECTION:
skydns.local.skydns.east.production.rails.1. 154 CH TXT	"service1.example.com:8080(10,0,,false)[0,]"
skydns.local.skydns.west.production.rails.2. 154 CH TXT	"service2.example.com:8080(10,0,,false)[0,]"

When setting the TTL on all RR in the message we would also do this
for the OPT RR. This is wrong as the OPT RR does *not* have a TTL.

Define locations for keys and secondary zones, 'n stuff.

Add a bunch of tests as well.

Add port 53 in the proxy host if not specified.
Check if the host is actually an IP address (v4 or v6)
Remove the http headers and other TODOs

Make middleware that sets up a (http) handler survive a graceful
restart. We calls the middleware's Shutdown function(s). If restart
fails the Start function is called again.

* middleware/health: OK
* middleware/pprof: OK
* middleware/metrics: OK

All restart OK.

Fix CoreDNS graceful reloading. This uses the same stuff as Caddy
(obviously), but extends it for UDP listeners as well. Also add to the
README that we *will* call Shutdown for middleware.

Fixes #4

* Changed reference to Caddy over to CoreDNS

* Removing references to caddy

* Fixed misleading error message to reference coredns

* Cleaning up references to caddy

* Adding clean and deps targets

Muscle memory is resulting in "make clean" commands.

* Adding test target to makefile

* More "Caddy" cleanup

Add pprof middleware, enabled by pprof directive.

Don't silently hide failures, barf on startup. Also add more integration
tests that should catch some of these things.

This adds an online dnssec middleware. The middleware will sign
responses on the fly. Negative responses are signed with NSEC black
lies.

Local variable shadowing in global const. Now subsystem is back to
'dns'.

Add prometheus metrics to the cache handler. This just used prometheus,
if the metrics middleware does not setup the handler, there is nobody
reading these metrics, but they are still reported. Seems the simplest
solution while keeping the whole middleware separation in tact.

Also fix failing test in setup/error.

Put error back in the correct place in the directives.go. Also don't
make it a pointer. If it *is* a pointer the buildstack function does
not correctly set the Next Handler. Don't understand *why* this is
different from Caddy. Anyway this fixes it, with the caveat that
the error log file is now openend earlier in the startup.

Fixes #127

The endpoint for etcd as wrongly set if a proxy was configured.
Clean up some documentation in the process as well.

* Add middleware/cache

Add a caching middleware that caches nxdomain, nodata and successful
responses. It differentiates between DNSSEC on normal DNS replies.

Each reply is compress and scrubbed so it will fit the specific client
asking for it.

* first simple test, less exporting of stuff

* more

* Add middleware/cache

Add a caching middleware that caches nxdomain, nodata and successful
responses. It differentiates between DNSSEC on normal DNS replies.

Each reply is compressed and scrubbed so it will fit the specific client
asking for it. The TTL is decremented with the time spend in the cache.
There is syntax that allows you to cap the TTL for all records, no
matter what. This allows for a shortlived cache, just to absorb query
peaks.

+Tests

* cache test infrastructure

* Testing

A target of a cname must be put in the answer section. Fix this and add
a AAAA test for quad-AAAA CNAME responses.

Fixes #125

Return a delegation when seeing one while traversing the tree in
search of an answer.

Put the SOA and NS record in the zone.Apex as these are to be handled
somewhat special.

Lowercase record on insert to make compares easier. This lowercases
all RR that have domain names in their rdata as well.

Instead of watching a single file watch the entire directory and
catch the .Name from the event. On any event, just reload the damn
thing. This also fixes the problem of loosing events when the inode
changes.

Zone reloading