skipper

README

Skipper

Skipper is an HTTP router and reverse proxy for service composition. It's designed to handle >300k HTTP route definitions with detailed lookup conditions, and flexible augmentation of the request flow with filters. It can be used out of the box or extended with custom lookup, filter logic and configuration sources.

Main features:

An overview of deployments and data-clients shows some use cases to run skipper.

Skipper

• identifies routes based on the requests' properties, such as path, method, host and headers
• allows modification of the requests and responses with filters that are independently configured for each route
• simultaneously streams incoming requests and backend responses
• optionally acts as a final endpoint (shunt), e.g. as a static file server or a mock backend for diagnostics
• updates routing rules without downtime, while supporting multiple types of data sources — including etcd, Kubernetes Ingress, Innkeeper (deprecated), static files, route string and custom configuration sources
• can serve as a Kubernetes Ingress controller without reloads. You can use it in combination with a controller that will route public traffic to your skipper fleet; see AWS example
• shipped with eskip: a descriptive configuration language designed for routing rules

Skipper provides a default executable command with a few built-in filters. However, its primary use case is to be extended with custom filters, predicates or data sources. Go here for additional documentation.

A few examples for extending Skipper:

• Image server https://github.com/zalando-stups/skrop
• Plugins repository https://github.com/skipper-plugins/, plugin docs

Getting Started

Prerequisites/Requirements

In order to build and run Skipper, only the latest version of Go needs to be installed. Skipper can use Innkeeper or Etcd as data sources for routes, or for the simplest cases, a local configuration file. See more details in the documentation: https://godoc.org/github.com/zalando/skipper.

Installation

Skipper is 'go get' compatible. If needed, create a Go workspace first:

mkdir ws
cd ws
export GOPATH=$(pwd)
export PATH=$PATH:$GOPATH/bin

Get the Skipper packages:

GO111MODULE=on go get github.com/zalando/skipper/...

Running

Create a file with a route:

echo 'hello: Path("/hello") -> "https://www.example.org"' > example.eskip

Optionally, verify the file's syntax:

eskip check example.eskip

If no errors are detected nothing is logged, else a descriptive error is logged.

Start Skipper and make an HTTP request:

skipper -routes-file example.eskip &
curl localhost:9090/hello

Docker

To run the latest Docker container:

docker run registry.opensource.zalan.do/teapot/skipper:latest

To run eskip you first mount the .eskip file, into the container, and run the command

docker run \
  -v $(PWD)/doc-docker-intro.eskip:/doc-docker-intro.eskip \
  registry.opensource.zalan.do/teapot/skipper:latest eskip print doc-docker-intro.eskip

To run skipper you first mount the .eskip file, into the container, expose the ports and run the command

docker run -it \
    -v $(PWD)/doc-docker-intro.eskip:/doc-docker-intro.eskip \
    -p 9090:9090 \
    -p 9911:9911 \
    registry.opensource.zalan.do/teapot/skipper:latest skipper -routes-file doc-docker-intro.eskip

Skipper will then be available on http://localhost:9090

Authentication Proxy

Skipper can be used as an authentication proxy, to check incoming requests with Basic auth or an OAuth2 provider including audit logging. See the documentation at: https://godoc.org/github.com/zalando/skipper/filters/auth.

Working with the code

Working with the code requires Go1.11 or a higher version. Getting the code with the test dependencies (-t switch):

GO111MODULE=on go get -t github.com/zalando/skipper/...

Build and test all packages:

cd src/github.com/zalando/skipper
make deps
make install
make shortcheck

On Mac the tests may fail because of low max open file limit. Please make sure you have correct limits setup by following these instructions.

Working from IntelliJ / GoLand

To run or debug skipper from IntelliJ IDEA or GoLand, you need to create this configuration:

Parameter	Value
Template	Go Build
Run kind	Directory
Directory	skipper source dir + /cmd/skipper
Working directory	skipper source dir (usually the default)

Kubernetes Ingress

Skipper can be used to run as an Kubernetes Ingress controller. Details with examples of Skipper's capabilities and an overview you will can be found in our ingress-controller deployment docs.

For AWS integration, we provide an ingress controller https://github.com/zalando-incubator/kube-ingress-aws-controller, that manage ALBs in front of your skipper deployment. A production example, can be found in our Kubernetes configuration https://github.com/zalando-incubator/kubernetes-on-aws.

• Comparison with other Ingress controllers
• Comparison with service-mesh

Documentation

Skipper's Documentation and Godoc developer documentation, includes information about deployment use cases and detailed information on these topics:

• The Routing Mechanism
• Matching Requests
• Filters - Augmenting Requests and Responses
• Service Backends
• Route Definitions
• Data Sources: eskip file, etcd, Kubernetes, Route string
• Circuit Breakers
• Extending It with Customized Predicates, Filters, can be done by Plugins or Lua Scripts
• Predicates - additional predicates to match a route
• Proxy Packages
• Logging and Metrics
• Performance Considerations
• Rate Limiters
• Opentracing plugin or extend create your own
• WAF plugin - Web Application Firewall filter for Skipper routes

1 Minute Skipper introduction

The following example shows a skipper routes file in eskip format, that has 3 named routes: baidu, google and yandex.

% cat doc-1min-intro.eskip
baidu:
        Path("/baidu")
        -> setRequestHeader("Host", "www.baidu.com")
        -> setPath("/s")
        -> setQuery("wd", "godoc skipper")
        -> "http://www.baidu.com";
google:
        *
        -> setPath("/search")
        -> setQuery("q", "godoc skipper")
        -> "https://www.google.com";
yandex:
        * && Cookie("yandex", "true")
        -> setPath("/search/")
        -> setQuery("text", "godoc skipper")
        -> tee("http://127.0.0.1:12345/")
        -> "https://yandex.ru";

Matching the route:

• baidu is using Path() matching to differentiate the HTTP requests to select the route.
• google is the default matching with wildcard '*'
• yandex is the default matching with wildcard '*' if you have a cookie "yandex=true"

Request Filters:

• If baidu is selected, skipper sets the Host header, changes the path and sets a query string to the http request to the backend "http://www.baidu.com".
• If google is selected, skipper changes the path and sets a query string to the http request to the backend "https://www.google.com".
• If yandex is selected, skipper changes the path and sets a query string to the http request to the backend "https://yandex.ru". The modified request will be copied to "http://127.0.0.1:12345/"

Run skipper with the routes file doc-1min-intro.eskip shown above

% skipper -routes-file doc-1min-intro.eskip

To test each route you can use curl:

% curl -v localhost:9090/baidu
% curl -v localhost:9090/
% curl -v --cookie "yandex=true" localhost:9090/

To see the request that is made by the tee() filter you can use nc:

[terminal1]% nc -l 12345
[terminal2]% curl -v --cookie "yandex=true" localhost:9090/

3 Minutes Skipper in Kubernetes introduction

This introduction was moved to ingress controller documentation.

For More details, please check out our Kubernetes ingress controller docs, our ingress usage and how to handle common backend problems in Kubernetes.

You should have a base understanding of Kubernetes and Ingress.

Packaging support

See https://github.com/zalando/skipper/blob/master/packaging/readme.md

Skipper uses Go modules, so you might need to add GO111MODULE=on in your custom build process.

In case you want to implement and link your own modules into your skipper, there is https://github.com/skipper-plugins organization to enable you to do so. In order to explain you the build process with custom Go modules there is https://github.com/skipper-plugins/skipper-tracing-build, that was used to build skipper's opentracing package. We moved the opentracing plugin source into the tracing package.

Community

User or developer questions can be asked in our public Google Group

We also have a slack channel #skipper in gophers.slack.com. Get an invite. If for some reason this link doesn't work, you can find more information about the gophers communities here.

Proposals

We do our proposals open in Skipper's Google drive. If you want to make a proposal feel free to create an issue and if it is a bigger change we will invite you to a document, such that we can work together.

Users

Zalando uses this project as shop frontend http router with 350000 routes, as Kubernetes ingress controller and runs several custom skipper instances that use skipper as library.

Sergio Ballesteros from spotahome

We also ran tests with several ingress controllers and skipper gave us the more reliable results. Currently we are running skipper since almost 2 years with like 20K Ingress rules. The fact that skipper is written in go let us understand the code, add features and fix bugs since all of our infra stack is golang.

In the media

Blog posts:

• Building our own open source http routing solution: Giving some context about why Skipper was created in the first place.
• Kubernetes in production @ ShopGun
• Hacker News Skipper – An HTTP router and reverse proxy for service composition

Conference/Meetups talks

• LISA 2018 - modern HTTP routing

Documentation

Overview

Package skipper provides an HTTP routing library with flexible configuration as well as a runtime update of the routing rules.

Skipper works as an HTTP reverse proxy that is responsible for mapping incoming requests to multiple HTTP backend services, based on routes that are selected by the request attributes. At the same time, both the requests and the responses can be augmented by a filter chain that is specifically defined for each route. Optionally, it can provide circuit breaker mechanism individually for each backend host.

Skipper can load and update the route definitions from multiple data sources without being restarted.

It provides a default executable command with a few built-in filters, however, its primary use case is to be extended with custom filters, predicates or data sources. For further information read 'Extending Skipper'.

Skipper took the core design and inspiration from Vulcand: https://github.com/mailgun/vulcand.

Quickstart

Skipper is 'go get' compatible. If needed, create a 'go workspace' first:

mkdir ws
cd ws
export GOPATH=$(pwd)
export PATH=$PATH:$GOPATH/bin

Get the Skipper packages:

go get github.com/zalando/skipper/...

Create a file with a route:

echo 'hello: Path("/hello") -> "https://www.example.org"' > example.eskip

Optionally, verify the syntax of the file:

eskip check example.eskip

Start Skipper and make an HTTP request:

skipper -routes-file example.eskip &
curl localhost:9090/hello

Routing Mechanism

The core of Skipper's request processing is implemented by a reverse proxy in the 'proxy' package. The proxy receives the incoming request, forwards it to the routing engine in order to receive the most specific matching route. When a route matches, the request is forwarded to all filters defined by it. The filters can modify the request or execute any kind of program logic. Once the request has been processed by all the filters, it is forwarded to the backend endpoint of the route. The response from the backend goes once again through all the filters in reverse order. Finally, it is mapped as the response of the original incoming request.

Besides the default proxying mechanism, it is possible to define routes without a real network backend endpoint. One of these cases is called a 'shunt' backend, in which case one of the filters needs to handle the request providing its own response (e.g. the 'static' filter). Actually, filters themselves can instruct the request flow to shunt by calling the Serve(*http.Response) method of the filter context.

Another case of a route without a network backend is the 'loopback'. A loopback route can be used to match a request, modified by filters, against the lookup tree with different conditions and then execute a different route. One example scenario can be to use a single route as an entry point to execute some calculation to get an A/B testing decision and then matching the updated request metadata for the actual destination route. This way the calculation can be executed for only those requests that don't contain information about a previously calculated decision.

For further details, see the 'proxy' and 'filters' package documentation.

Matching Requests

Finding a request's route happens by matching the request attributes to the conditions in the route's definitions. Such definitions may have the following conditions:

- method

- path (optionally with wildcards)

- path regular expressions

- host regular expressions

- headers

- header regular expressions

It is also possible to create custom predicates with any other matching criteria.

The relation between the conditions in a route definition is 'and', meaning, that a request must fulfill each condition to match a route.

For further details, see the 'routing' package documentation.

Filters - Augmenting Requests

Filters are applied in order of definition to the request and in reverse order to the response. They are used to modify request and response attributes, such as headers, or execute background tasks, like logging. Some filters may handle the requests without proxying them to service backends. Filters, depending on their implementation, may accept/require parameters, that are set specifically to the route.

For further details, see the 'filters' package documentation.

Service Backends

Each route has one of the following backends: HTTP endpoint, shunt, loopback or dynamic.

Backend endpoints can be any HTTP service. They are specified by their network address, including the protocol scheme, the domain name or the IP address, and optionally the port number: e.g. "https://www.example.org:4242". (The path and query are sent from the original request, or set by filters.)

A shunt route means that Skipper handles the request alone and doesn't make requests to a backend service. In this case, it is the responsibility of one of the filters to generate the response.

A loopback route executes the routing mechanism on current state of the request from the start, including the route lookup. This way it serves as a form of an internal redirect.

A dynamic route means that the final target will be defined in a filter. One of the filters in the chain must set the target backend url explicitly.

Route Definitions

Route definitions consist of the following:

- request matching conditions (predicates)

- filter chain (optional)

- backend

The eskip package implements the in-memory and text representations of route definitions, including a parser.

(Note to contributors: in order to stay compatible with 'go get', the generated part of the parser is stored in the repository. When changing the grammar, 'go generate' needs to be executed explicitly to update the parser.)

For further details, see the 'eskip' package documentation

Authentication and Authorization

Skipper has filter implementations of basic auth and OAuth2. It can be integrated with tokeninfo based OAuth2 providers. For details, see: https://godoc.org/github.com/zalando/skipper/filters/auth.

Data Sources

Skipper's route definitions of Skipper are loaded from one or more data sources. It can receive incremental updates from those data sources at runtime. It provides three different data clients:

- Kubernetes: Skipper can be used as part of a Kubernetes Ingress Controller implementation together with https://github.com/zalando-incubator/kube-ingress-aws-controller . In this scenario, Skipper uses the Kubernetes API's Ingress extensions as a source for routing. For a complete deployment example, see more details in: https://github.com/zalando-incubator/kubernetes-on-aws/ .

- Innkeeper: the Innkeeper service implements a storage for large sets of Skipper routes, with an HTTP+JSON API, OAuth2 authentication and role management. See the 'innkeeper' package and https://github.com/zalando/innkeeper.

- etcd: Skipper can load routes and receive updates from etcd clusters (https://github.com/coreos/etcd). See the 'etcd' package.

- static file: package eskipfile implements a simple data client, which can load route definitions from a static file in eskip format. Currently, it loads the routes on startup. It doesn't support runtime updates.

Skipper can use additional data sources, provided by extensions. Sources must implement the DataClient interface in the routing package.

Circuit Breaker

Skipper provides circuit breakers, configured either globally, based on backend hosts or based on individual routes. It supports two types of circuit breaker behavior: open on N consecutive failures, or open on N failures out of M requests. For details, see: https://godoc.org/github.com/zalando/skipper/circuit.

Running Skipper

Skipper can be started with the default executable command 'skipper', or as a library built into an application. The easiest way to start Skipper as a library is to execute the 'Run' function of the current, root package.

Each option accepted by the 'Run' function is wired in the default executable as well, as a command line flag. E.g. EtcdUrls becomes -etcd-urls as a comma separated list. For command line help, enter:

skipper -help

An additional utility, eskip, can be used to verify, print, update and delete routes from/to files or etcd (Innkeeper on the roadmap). See the cmd/eskip command package, and/or enter in the command line:

eskip -help

Extending Skipper

Skipper doesn't use dynamically loaded plugins, however, it can be used as a library, and it can be extended with custom predicates, filters and/or custom data sources.

Custom Predicates

To create a custom predicate, one needs to implement the PredicateSpec interface in the routing package. Instances of the PredicateSpec are used internally by the routing package to create the actual Predicate objects as referenced in eskip routes, with concrete arguments.

Example, randompredicate.go:

package main

import (
    "github.com/zalando/skipper/routing"
    "math/rand"
    "net/http"
)

type randomSpec struct {}

type randomPredicate struct {
    chance float64
}

func (s *randomSpec) Name() string { return "Random" }

func (s *randomSpec) Create(args []interface{}) (routing.Predicate, error) {
    p := &randomPredicate{.5}
    if len(args) > 0 {
        if c, ok := args[0].(float64); ok {
            p.chance = c
        }
    }

    return p, nil
}

func (p *randomPredicate) Match(_ *http.Request) bool {
    return rand.Float64() < p.chance
}

In the above example, a custom predicate is created, that can be referenced in eskip definitions with the name 'Random':

Random(.33) -> "https://test.example.org";
* -> "https://www.example.org"

Custom Filters

To create a custom filter we need to implement the Spec interface of the filters package. 'Spec' is the specification of a filter, and it is used to create concrete filter instances, while the raw route definitions are processed.

Example, hellofilter.go:

package main

import (
    "fmt"
    "github.com/zalando/skipper/filters"
)

type helloSpec struct {}

type helloFilter struct {
    who string
}

func (s *helloSpec) Name() string { return "hello" }

func (s *helloSpec) CreateFilter(config []interface{}) (filters.Filter, error) {
    if len(config) == 0 {
        return nil, filters.ErrInvalidFilterParameters
    }

    if who, ok := config[0].(string); ok {
        return &helloFilter{who}, nil
    } else {
        return nil, filters.ErrInvalidFilterParameters
    }
}

func (f *helloFilter) Request(ctx filters.FilterContext) {}

func (f *helloFilter) Response(ctx filters.FilterContext) {
    ctx.Response().Header.Set("X-Hello", fmt.Sprintf("Hello, %s!", f.who))
}

The above example creates a filter specification, and in the routes where they are included, the filter instances will set the 'X-Hello' header for each and every response. The name of the filter is 'hello', and in a route definition it is referenced as:

• -> hello("world") -> "https://www.example.org"

Custom Build

The easiest way to create a custom Skipper variant is to implement the required filters (as in the example above) by importing the Skipper package, and starting it with the 'Run' command.

Example, hello.go:

package main

import (
    "log"

    "github.com/zalando/skipper"
    "github.com/zalando/skipper/filters"
    "github.com/zalando/skipper/routing"
)

func main() {
    log.Fatal(skipper.Run(skipper.Options{
        Address: ":9090",
        RoutesFile: "routes.eskip",
        CustomPredicates: []routing.PredicateSpec{&randomSpec{}},
        CustomFilters: []filters.Spec{&helloSpec{}}}))
}

A file containing the routes, routes.eskip:

random:
    Random(.05) -> hello("fish?") -> "https://fish.example.org";
hello:
    * -> hello("world") -> "https://www.example.org"

Start the custom router:

go run hello.go

Proxy Package Used Individually

The 'Run' function in the root Skipper package starts its own listener but it doesn't provide the best composability. The proxy package, however, provides a standard http.Handler, so it is possible to use it in a more complex solution as a building block for routing.

Logging and Metrics

Skipper provides detailed logging of failures, and access logs in Apache log format. Skipper also collects detailed performance metrics, and exposes them on a separate listener endpoint for pulling snapshots.

For details, see the 'logging' and 'metrics' packages documentation.

Performance Considerations

The router's performance depends on the environment and on the used filters. Under ideal circumstances, and without filters, the biggest time factor is the route lookup. Skipper is able to scale to thousands of routes with logarithmic performance degradation. However, this comes at the cost of increased memory consumption, due to storing the whole lookup tree in a single structure.

Benchmarks for the tree lookup can be run by:

go test github.com/zalando/skipper/routing -bench=Tree

In case more aggressive scale is needed, it is possible to setup Skipper in a cascade model, with multiple Skipper instances for specific route segments.

Example (RatelimitRegistryBinding)

s := &customRatelimitSpec{}

o := Options{
	Address:            ":9090",
	InlineRoutes:       `* -> customRatelimit() -> <shunt>`,
	EnableRatelimiters: true,
	EnableSwarm:        true,
	SwarmRedisURLs:     []string{":6379"},
	CustomFilters:      []filters.Spec{s},
	SwarmRegistry: func(registry *ratelimit.Registry) {
		s.registry = registry
	},
}

log.Fatal(Run(o))
// Example functions without output comments are compiled but not executed

Index

• Constants
• func Run(o Options) error
• type Options

Examples

• Package (RatelimitRegistryBinding)

Constants

const DefaultPluginDir = "./plugins"

Functions

func Run

func Run(o Options) error

Run skipper.

Types

type Options

type Options struct {
	// WaitForHealthcheckInterval sets the time that skipper waits
	// for the loadbalancer in front to become unhealthy. Defaults
	// to 0.
	WaitForHealthcheckInterval time.Duration

	// StatusChecks is an experimental feature. It defines a
	// comma separated list of HTTP URLs to do GET requests to,
	// that have to return 200 before skipper becomes ready
	StatusChecks []string

	// WhitelistedHealthcheckCIDR appends the whitelisted IP Range to the inernalIPS range for healthcheck purposes
	WhitelistedHealthCheckCIDR []string

	// Network address that skipper should listen on.
	Address string

	// EnableTCPQueue enables controlling the
	// concurrently processed requests at the TCP listener.
	EnableTCPQueue bool

	// ExpectedBytesPerRequest is used by the TCP LIFO listener.
	// It defines the expected average memory required to process an incoming
	// request. It is used only when MaxTCPListenerConcurrency is not defined.
	// It is used together with the memory limit defined in:
	// /sys/fs/cgroup/memory/memory.limit_in_bytes.
	//
	// See also: https://www.kernel.org/doc/Documentation/cgroup-v1/memory.txt
	ExpectedBytesPerRequest int

	// MaxTCPListenerConcurrency is used by the TCP LIFO listener.
	// It defines the max number of concurrently accepted connections, excluding
	// the pending ones in the queue.
	//
	// When undefined and the EnableTCPQueue is true,
	MaxTCPListenerConcurrency int

	// MaxTCPListenerQueue is used by the TCP LIFO listener.
	// If defines the maximum number of pending connection waiting in the queue.
	MaxTCPListenerQueue int

	// List of custom filter specifications.
	CustomFilters []filters.Spec

	// Urls of nodes in an etcd cluster, storing route definitions.
	EtcdUrls []string

	// Path prefix for skipper related data in the etcd storage.
	EtcdPrefix string

	// Timeout used for a single request when querying for updates
	// in etcd. This is independent of, and an addition to,
	// SourcePollTimeout. When not set, the internally defined 1s
	// is used.
	EtcdWaitTimeout time.Duration

	// Skip TLS certificate check for etcd connections.
	EtcdInsecure bool

	// If set this value is used as Bearer token for etcd OAuth authorization.
	EtcdOAuthToken string

	// If set this value is used as username for etcd basic authorization.
	EtcdUsername string

	// If set this value is used as password for etcd basic authorization.
	EtcdPassword string

	// If set enables skipper to generate based on ingress resources in kubernetes cluster
	Kubernetes bool

	// If set makes skipper authenticate with the kubernetes API server with service account assigned to the
	// skipper POD.
	// If omitted skipper will rely on kubectl proxy to authenticate with API server
	KubernetesInCluster bool

	// Kubernetes API base URL. Only makes sense if KubernetesInCluster is set to false. If omitted and
	// skipper is not running in-cluster, the default API URL will be used.
	KubernetesURL string

	// KubernetesHealthcheck, when Kubernetes ingress is set, indicates
	// whether an automatic healthcheck route should be generated. The
	// generated route will report healthyness when the Kubernetes API
	// calls are successful. The healthcheck endpoint is accessible from
	// internal IPs, with the path /kube-system/healthz.
	KubernetesHealthcheck bool

	// KubernetesHTTPSRedirect, when Kubernetes ingress is set, indicates
	// whether an automatic redirect route should be generated to redirect
	// HTTP requests to their HTTPS equivalent. The generated route will
	// match requests with the X-Forwarded-Proto and X-Forwarded-Port,
	// expected to be set by the load-balancer.
	KubernetesHTTPSRedirect bool

	// KubernetesHTTPSRedirectCode overrides the default redirect code (308)
	// when used together with -kubernetes-https-redirect.
	KubernetesHTTPSRedirectCode int

	// KubernetesIngressV1 will switch the dataclient to read ingress v1 resources, instead of v1beta1
	KubernetesIngressV1 bool

	// KubernetesIngressClass is a regular expression, that will make
	// skipper load only the ingress resources that have a matching
	// kubernetes.io/ingress.class annotation. For backwards compatibility,
	// the ingresses without an annotation, or an empty annotation, will
	// be loaded, too.
	KubernetesIngressClass string

	// KubernetesRouteGroupClass is a regular expression, that will make skipper
	// load only the RouteGroup resources that have a matching
	// zalando.org/routegroup.class annotation. Any RouteGroups without the
	// annotation, or which an empty annotation, will be loaded too.
	KubernetesRouteGroupClass string

	// PathMode controls the default interpretation of ingress paths in cases
	// when the ingress doesn't specify it with an annotation.
	KubernetesPathMode kubernetes.PathMode

	// KubernetesNamespace is used to switch between monitoring ingresses in the cluster-scope or limit
	// the ingresses to only those in the specified namespace. Defaults to "" which means monitor ingresses
	// in the cluster-scope.
	KubernetesNamespace string

	// *DEPRECATED* KubernetesEnableEastWest enables cluster internal service to service communication, aka east-west traffic
	KubernetesEnableEastWest bool

	// *DEPRECATED* KubernetesEastWestDomain sets the cluster internal domain used to create additional routes in skipper, defaults to skipper.cluster.local
	KubernetesEastWestDomain string

	// KubernetesEastWestRangeDomains set the the cluster internal domains for
	// east west traffic. Identified routes to such domains will include
	// the KubernetesEastWestRangePredicates.
	KubernetesEastWestRangeDomains []string

	// KubernetesEastWestRangePredicates set the Predicates that will be
	// appended to routes identified as to KubernetesEastWestRangeDomains.
	KubernetesEastWestRangePredicates []*eskip.Predicate

	// KubernetesOnlyAllowedExternalNames will enable validation of ingress external names and route groups network
	// backend addresses, explicit LB endpoints validation agains the list of patterns in
	// AllowedExternalNames.
	KubernetesOnlyAllowedExternalNames bool

	// KubernetesAllowedExternalNames contains regexp patterns of those domain names that are allowed to be
	// used with external name services (type=ExternalName).
	KubernetesAllowedExternalNames []*regexp.Regexp

	// *DEPRECATED* API endpoint of the Innkeeper service, storing route definitions.
	InnkeeperUrl string

	// *DEPRECATED* Fixed token for innkeeper authentication. (Used mainly in
	// development environments.)
	InnkeeperAuthToken string

	// *DEPRECATED* Filters to be prepended to each route loaded from Innkeeper.
	InnkeeperPreRouteFilters string

	// *DEPRECATED* Filters to be appended to each route loaded from Innkeeper.
	InnkeeperPostRouteFilters string

	// *DEPRECATED* Skip TLS certificate check for Innkeeper connections.
	InnkeeperInsecure bool

	// *DEPRECATED* OAuth2 URL for Innkeeper authentication.
	OAuthUrl string

	// *DEPRECATED* Directory where oauth credentials are stored, with file names:
	// client.json and user.json.
	OAuthCredentialsDir string

	// *DEPRECATED* The whitespace separated list of OAuth2 scopes.
	OAuthScope string

	// File containing static route definitions. Multiple may be given comma separated.
	RoutesFile string

	// File containing route definitions with file watch enabled.
	// Multiple may be given comma separated. (For the skipper
	// command this option is used when starting it with the -routes-file flag.)
	WatchRoutesFile string

	// RouteURLs are URLs pointing to route definitions, in eskip format, with change watching enabled.
	RoutesURLs []string

	// InlineRoutes can define routes as eskip text.
	InlineRoutes string

	// Polling timeout of the routing data sources.
	SourcePollTimeout time.Duration

	// DefaultFilters will be applied to all routes automatically.
	DefaultFilters *eskip.DefaultFilters

	// CloneRoute is a PreProcessor, that will be applied to all routes automatically. It
	// will clone all matching routes and apply changes to the
	// cloned routes.
	CloneRoute *eskip.Clone

	// EditRoute will be applied to all routes automatically and
	// will apply changes to all matching routes.
	EditRoute *eskip.Editor

	// A list of custom routing pre-processor implementations that will
	// be applied to all routes.
	CustomRoutingPreProcessors []routing.PreProcessor

	// Deprecated. See ProxyFlags. When used together with ProxyFlags,
	// the values will be combined with |.
	ProxyOptions proxy.Options

	// Flags controlling the proxy behavior.
	ProxyFlags proxy.Flags

	// Tells the proxy maximum how many idle connections can it keep
	// alive.
	IdleConnectionsPerHost int

	// Defines the time period of how often the idle connections maintained
	// by the proxy are closed.
	CloseIdleConnsPeriod time.Duration

	// Defines ReadTimeoutServer for server http connections.
	ReadTimeoutServer time.Duration

	// Defines ReadHeaderTimeout for server http connections.
	ReadHeaderTimeoutServer time.Duration

	// Defines WriteTimeout for server http connections.
	WriteTimeoutServer time.Duration

	// Defines IdleTimeout for server http connections.
	IdleTimeoutServer time.Duration

	// Defines MaxHeaderBytes for server http connections.
	MaxHeaderBytes int

	// Enable connection state metrics for server http connections.
	EnableConnMetricsServer bool

	// TimeoutBackend sets the TCP client connection timeout for
	// proxy http connections to the backend.
	TimeoutBackend time.Duration

	// ResponseHeaderTimeout sets the HTTP response timeout for
	// proxy http connections to the backend.
	ResponseHeaderTimeoutBackend time.Duration

	// ExpectContinueTimeoutBackend sets the HTTP timeout to expect a
	// response for status Code 100 for proxy http connections to
	// the backend.
	ExpectContinueTimeoutBackend time.Duration

	// KeepAliveBackend sets the TCP keepalive for proxy http
	// connections to the backend.
	KeepAliveBackend time.Duration

	// DualStackBackend sets if the proxy TCP connections to the
	// backend should be dual stack.
	DualStackBackend bool

	// TLSHandshakeTimeoutBackend sets the TLS handshake timeout
	// for proxy connections to the backend.
	TLSHandshakeTimeoutBackend time.Duration

	// MaxIdleConnsBackend sets MaxIdleConns, which limits the
	// number of idle connections to all backends, 0 means no
	// limit.
	MaxIdleConnsBackend int

	// DisableHTTPKeepalives sets DisableKeepAlives, which forces
	// a backend to always create a new connection.
	DisableHTTPKeepalives bool

	// Flag indicating to ignore trailing slashes in paths during route
	// lookup.
	IgnoreTrailingSlash bool

	// Priority routes that are matched against the requests before
	// the standard routes from the data clients.
	PriorityRoutes []proxy.PriorityRoute

	// Specifications of custom, user defined predicates.
	CustomPredicates []routing.PredicateSpec

	// Custom data clients to be used together with the default etcd and Innkeeper.
	CustomDataClients []routing.DataClient

	// CustomHttpHandlerWrap provides ability to wrap http.Handler created by skipper.
	// http.Handler is used for accepting incoming http requests.
	// It allows to add additional logic (for example tracing) by providing a wrapper function
	// which accepts original skipper handler as an argument and returns a wrapped handler
	CustomHttpHandlerWrap func(http.Handler) http.Handler

	// CustomHttpRoundTripperWrap provides ability to wrap http.RoundTripper created by skipper.
	// http.RoundTripper is used for making outgoing requests (backends)
	// It allows to add additional logic (for example tracing) by providing a wrapper function
	// which accepts original skipper http.RoundTripper as an argument and returns a wrapped roundtripper
	CustomHttpRoundTripperWrap func(http.RoundTripper) http.RoundTripper

	// WaitFirstRouteLoad prevents starting the listener before the first batch
	// of routes were applied.
	WaitFirstRouteLoad bool

	// SuppressRouteUpdateLogs indicates to log only summaries of the routing updates
	// instead of full details of the updated/deleted routes.
	SuppressRouteUpdateLogs bool

	// Dev mode. Currently this flag disables prioritization of the
	// consumer side over the feeding side during the routing updates to
	// populate the updated routes faster.
	DevMode bool

	// Network address for the support endpoints
	SupportListener string

	// Deprecated: Network address for the /metrics endpoint
	MetricsListener string

	// Skipper provides a set of metrics with different keys which are exposed via HTTP in JSON
	// You can customize those key names with your own prefix
	MetricsPrefix string

	// EnableProfile exposes profiling information on /profile of the
	// metrics listener.
	EnableProfile bool

	// BlockProfileRate calls runtime.SetBlockProfileRate(BlockProfileRate) if non zero value, deactivate with <0
	BlockProfileRate int

	// MutexProfileFraction calls runtime.SetMutexProfileFraction(MutexProfileFraction) if non zero value, deactivate with <0
	MutexProfileFraction int

	// MemProfileRate calls runtime.SetMemProfileRate(MemProfileRate) if non zero value, deactivate with <0
	MemProfileRate int

	// Flag that enables reporting of the Go garbage collector statistics exported in debug.GCStats
	EnableDebugGcMetrics bool

	// Flag that enables reporting of the Go runtime statistics exported in runtime and specifically runtime.MemStats
	EnableRuntimeMetrics bool

	// If set, detailed response time metrics will be collected
	// for each route, additionally grouped by status and method.
	EnableServeRouteMetrics bool

	// If set, a counter for each route is generated, additionally
	// grouped by status and method. It differs from the automatically
	// generated counter from `EnableServeRouteMetrics` because it will
	// always contain the status and method labels, independently of the
	// `EnableServeMethodMetric` and `EnableServeStatusCodeMetric` flags.
	EnableServeRouteCounter bool

	// If set, detailed response time metrics will be collected
	// for each host, additionally grouped by status and method.
	EnableServeHostMetrics bool

	// If set, a counter for each host is generated, additionally
	// grouped by status and method. It differs from the automatically
	// generated counter from `EnableServeHostMetrics` because it will
	// always contain the status and method labels, independently of the
	// `EnableServeMethodMetric` and `EnableServeStatusCodeMetric` flags.
	EnableServeHostCounter bool

	// If set, the detailed total response time metrics will contain the
	// HTTP method as a domain of the metric. It affects both route and
	// host splitted metrics.
	EnableServeMethodMetric bool

	// If set, the detailed total response time metrics will contain the
	// HTTP Response status code as a domain of the metric. It affects
	// both route and host splitted metrics.
	EnableServeStatusCodeMetric bool

	// If set, detailed response time metrics will be collected
	// for each backend host
	EnableBackendHostMetrics bool

	// EnableAllFiltersMetrics enables collecting combined filter
	// metrics per each route. Without the DisableMetricsCompatibilityDefaults,
	// it is enabled by default.
	EnableAllFiltersMetrics bool

	// EnableCombinedResponseMetrics enables collecting response time
	// metrics combined for every route.
	EnableCombinedResponseMetrics bool

	// EnableRouteResponseMetrics enables collecting response time
	// metrics per each route. Without the DisableMetricsCompatibilityDefaults,
	// it is enabled by default.
	EnableRouteResponseMetrics bool

	// EnableRouteBackendErrorsCounters enables counters for backend
	// errors per each route. Without the DisableMetricsCompatibilityDefaults,
	// it is enabled by default.
	EnableRouteBackendErrorsCounters bool

	// EnableRouteStreamingErrorsCounters enables counters for streaming
	// errors per each route. Without the DisableMetricsCompatibilityDefaults,
	// it is enabled by default.
	EnableRouteStreamingErrorsCounters bool

	// EnableRouteBackendMetrics enables backend response time metrics
	// per each route. Without the DisableMetricsCompatibilityDefaults, it is
	// enabled by default.
	EnableRouteBackendMetrics bool

	// EnableRouteCreationMetrics enables the OriginMarker to track route creation times. Disabled by default
	EnableRouteCreationMetrics bool

	// When set, makes the histograms use an exponentially decaying sample
	// instead of the default uniform one.
	MetricsUseExpDecaySample bool

	// Use custom buckets for prometheus histograms.
	HistogramMetricBuckets []float64

	// The following options, for backwards compatibility, are true
	// by default: EnableAllFiltersMetrics, EnableRouteResponseMetrics,
	// EnableRouteBackendErrorsCounters, EnableRouteStreamingErrorsCounters,
	// EnableRouteBackendMetrics. With this compatibility flag, the default
	// for these options can be set to false.
	DisableMetricsCompatibilityDefaults bool

	// Implementation of a Metrics handler. If provided this is going to be used
	// instead of creating a new one based on the Kind of metrics wanted. This
	// is useful in case you want to report metrics to a custom aggregator.
	MetricsBackend metrics.Metrics

	// Output file for the application log. Default value: /dev/stderr.
	//
	// When /dev/stderr or /dev/stdout is passed in, it will be resolved
	// to os.Stderr or os.Stdout.
	//
	// Warning: passing an arbitrary file will try to open it append
	// on start and use it, or fail on start, but the current
	// implementation doesn't support any more proper handling
	// of temporary failures or log-rolling.
	ApplicationLogOutput string

	// Application log prefix. Default value: "[APP]".
	ApplicationLogPrefix string

	// Enables logs in JSON format
	ApplicationLogJSONEnabled bool

	// ApplicationLogJsonFormatter, when set and JSON logging is enabled, is passed along to to the underlying
	// Logrus logger for application logs. To enable structured logging, use ApplicationLogJSONEnabled.
	ApplicationLogJsonFormatter *log.JSONFormatter

	// Output file for the access log. Default value: /dev/stderr.
	//
	// When /dev/stderr or /dev/stdout is passed in, it will be resolved
	// to os.Stderr or os.Stdout.
	//
	// Warning: passing an arbitrary file will try to open for append
	// it on start and use it, or fail on start, but the current
	// implementation doesn't support any more proper handling
	// of temporary failures or log-rolling.
	AccessLogOutput string

	// Disables the access log.
	AccessLogDisabled bool

	// Enables logs in JSON format
	AccessLogJSONEnabled bool

	// AccessLogStripQuery, when set, causes the query strings stripped
	// from the request URI in the access logs.
	AccessLogStripQuery bool

	// AccessLogJsonFormatter, when set and JSON logging is enabled, is passed along to to the underlying
	// Logrus logger for access logs. To enable structured logging, use AccessLogJSONEnabled.
	AccessLogJsonFormatter *log.JSONFormatter

	DebugListener string

	// Path of certificate(s) when using TLS, mutiple may be given comma separated
	CertPathTLS string
	// Path of key(s) when using TLS, multiple may be given comma separated. For
	// multiple keys, the order must match the one given in CertPathTLS
	KeyPathTLS string

	// TLS Settings for Proxy Server
	ProxyTLS *tls.Config

	// Client TLS to connect to Backends
	ClientTLS *tls.Config

	// TLSMinVersion to set the minimal TLS version for all TLS configurations
	TLSMinVersion uint16

	// Flush interval for upgraded Proxy connections
	BackendFlushInterval time.Duration

	// Experimental feature to handle protocol Upgrades for Websockets, SPDY, etc.
	ExperimentalUpgrade bool

	// ExperimentalUpgradeAudit enables audit log of both the request line
	// and the response messages during web socket upgrades.
	ExperimentalUpgradeAudit bool

	// MaxLoopbacks defines the maximum number of loops that the proxy can execute when the routing table
	// contains loop backends (<loopback>).
	MaxLoopbacks int

	// EnableBreakers enables the usage of the breakers in the route definitions without initializing any
	// by default. It is a shortcut for setting the BreakerSettings to:
	//
	// 	[]circuit.BreakerSettings{{Type: BreakerDisabled}}
	//
	EnableBreakers bool

	// BreakerSettings contain global and host specific settings for the circuit breakers.
	BreakerSettings []circuit.BreakerSettings

	// EnableRatelimiters enables the usage of the ratelimiter in the route definitions without initializing any
	// by default. It is a shortcut for setting the RatelimitSettings to:
	//
	// 	[]ratelimit.Settings{{Type: DisableRatelimit}}
	//
	EnableRatelimiters bool

	// RatelimitSettings contain global and host specific settings for the ratelimiters.
	RatelimitSettings []ratelimit.Settings

	// EnableRouteLIFOMetrics enables metrics for the individual route LIFO queues, if any.
	EnableRouteLIFOMetrics bool

	// OpenTracing enables opentracing
	OpenTracing []string

	// OpenTracingInitialSpan can override the default initial, pre-routing, span name.
	// Default: "ingress".
	OpenTracingInitialSpan string

	// OpenTracingExcludedProxyTags can disable a tag so that it is not recorded. By default every tag is included.
	OpenTracingExcludedProxyTags []string

	// OpenTracingDisableFilterSpans flag is used to disable creation of spans representing request and response filters.
	OpenTracingDisableFilterSpans bool

	// OpenTracingLogFilterLifecycleEvents flag is used to enable/disable the logs for events marking request and
	// response filters' start & end times.
	OpenTracingLogFilterLifecycleEvents bool

	// OpenTracingLogStreamEvents flag is used to enable/disable the logs that marks the
	// times when response headers & payload are streamed to the client
	OpenTracingLogStreamEvents bool

	// OpenTracingBackendNameTag enables an additional tracing tag containing a backend name
	// for a route when it's available (e.g. for RouteGroups)
	OpenTracingBackendNameTag bool

	// PluginDir defines the directory to load plugins from, DEPRECATED, use PluginDirs
	PluginDir string
	// PluginDirs defines the directories to load plugins from
	PluginDirs []string

	// FilterPlugins loads additional filters from modules. The first value in each []string
	// needs to be the plugin name (as on disk, without path, without ".so" suffix). The
	// following values are passed as arguments to the plugin while loading, see also
	// https://opensource.zalando.com/skipper/reference/plugins/
	FilterPlugins [][]string

	// PredicatePlugins loads additional predicates from modules. See above for FilterPlugins
	// what the []string should contain.
	PredicatePlugins [][]string

	// DataClientPlugins loads additional data clients from modules. See above for FilterPlugins
	// what the []string should contain.
	DataClientPlugins [][]string

	// Plugins combine multiple types of the above plugin types in one plugin (where
	// necessary because of shared data between e.g. a filter and a data client).
	Plugins [][]string

	// DefaultHTTPStatus is the HTTP status used when no routes are found
	// for a request.
	DefaultHTTPStatus int

	// EnablePrometheusMetrics enables Prometheus format metrics.
	//
	// This option is *deprecated*. The recommended way to enable prometheus metrics is to
	// use the MetricsFlavours option.
	EnablePrometheusMetrics bool

	// An instance of a Prometheus registry. It allows registering and serving custom metrics when skipper is used as a
	// library.
	// A new registry is created if this option is nil.
	PrometheusRegistry *prometheus.Registry

	// MetricsFlavours sets the metrics storage and exposed format
	// of metrics endpoints.
	MetricsFlavours []string

	// LoadBalancerHealthCheckInterval enables and sets the
	// interval when to schedule health checks for dead or
	// unhealthy routes
	LoadBalancerHealthCheckInterval time.Duration

	// ReverseSourcePredicate enables the automatic use of IP
	// whitelisting in different places to use the reversed way of
	// identifying a client IP within the X-Forwarded-For
	// header. Amazon's ALB for example writes the client IP to
	// the last item of the string list of the X-Forwarded-For
	// header, in this case you want to set this to true.
	ReverseSourcePredicate bool

	// EnableOAuth2GrantFlow, enables OAuth2 Grant Flow filter
	EnableOAuth2GrantFlow bool

	// OAuth2AuthURL, the url to redirect the requests to when login is required.
	OAuth2AuthURL string

	// OAuth2TokenURL, the url where the access code should be exchanged for the
	// access token.
	OAuth2TokenURL string

	// OAuth2RevokeTokenURL, the url where the access and refresh tokens can be
	// revoked during a logout.
	OAuth2RevokeTokenURL string

	// OAuthTokeninfoURL sets the the URL to be queried for
	// information for all auth.NewOAuthTokeninfo*() filters.
	OAuthTokeninfoURL string

	// OAuthTokeninfoTimeout sets timeout duration while calling oauth token service
	OAuthTokeninfoTimeout time.Duration

	// OAuth2SecretFile contains the filename with the encryption key for the
	// authentication cookie and grant flow state stored in Secrets.
	OAuth2SecretFile string

	// OAuth2ClientID, the OAuth2 client id of the current service, used to exchange
	// the access code.
	OAuth2ClientID string

	// OAuth2ClientSecret, the secret associated with the ClientID, used to exchange
	// the access code.
	OAuth2ClientSecret string

	// OAuth2ClientIDFile, the path of the file containing the OAuth2 client id of
	// the current service, used to exchange the access code.
	OAuth2ClientIDFile string

	// OAuth2ClientSecretFile, the path of the file containing the secret associated
	// with the ClientID, used to exchange the access code.
	OAuth2ClientSecretFile string

	// OAuth2CallbackPath contains the path where the OAuth2 callback requests with the
	// authorization code should be redirected to. Defaults to /.well-known/oauth2-callback
	OAuth2CallbackPath string

	// OAuthTokenintrospectionTimeout sets timeout duration while calling oauth tokenintrospection service
	OAuthTokenintrospectionTimeout time.Duration

	// OAuth2AuthURLParameters the additional parameters to send to OAuth2 authorize and token endpoints.
	OAuth2AuthURLParameters map[string]string

	// OAuth2AccessTokenHeaderName the name of the header to which the access token
	// should be assigned after the oauthGrant filter.
	OAuth2AccessTokenHeaderName string

	// OAuth2TokeninfoSubjectKey the key of the subject ID attribute in the
	// tokeninfo map. Used for downstream oidcClaimsQuery compatibility.
	OAuth2TokeninfoSubjectKey string

	// OAuth2TokenCookieName the name of the cookie that Skipper sets after a
	// successful OAuth2 token exchange. Stores the encrypted access token.
	OAuth2TokenCookieName string

	// CompressEncodings, if not empty replace default compression encodings
	CompressEncodings []string

	// OIDCSecretsFile path to the file containing key to encrypt OpenID token
	OIDCSecretsFile string

	// OIDCDistributedClaimsTimeout sets timeout duration while calling Distributed Claims endpoint.
	OIDCDistributedClaimsTimeout time.Duration

	// SecretsRegistry to store and load secretsencrypt
	SecretsRegistry *secrets.Registry

	// CredentialsPaths directories or files where credentials are stored one secret per file
	CredentialsPaths []string

	// CredentialsUpdateInterval sets the interval to update secrets
	CredentialsUpdateInterval time.Duration

	// API Monitoring feature is active (feature toggle)
	ApiUsageMonitoringEnable                bool
	ApiUsageMonitoringRealmKeys             string
	ApiUsageMonitoringClientKeys            string
	ApiUsageMonitoringRealmsTrackingPattern string
	// *DEPRECATED* ApiUsageMonitoringDefaultClientTrackingPattern
	ApiUsageMonitoringDefaultClientTrackingPattern string

	// Default filters directory enables default filters mechanism and sets the directory where the filters are located
	DefaultFiltersDir string

	// WebhookTimeout sets timeout duration while calling a custom webhook auth service
	WebhookTimeout time.Duration

	// MaxAuditBody sets the maximum read size of the body read by the audit log filter
	MaxAuditBody int

	// EnableSwarm enables skipper fleet communication, required by e.g.
	// the cluster ratelimiter
	EnableSwarm bool
	// redis based swarm
	SwarmRedisURLs          []string
	SwarmRedisPassword      string
	SwarmRedisHashAlgorithm string
	SwarmRedisDialTimeout   time.Duration
	SwarmRedisReadTimeout   time.Duration
	SwarmRedisWriteTimeout  time.Duration
	SwarmRedisPoolTimeout   time.Duration
	SwarmRedisMinIdleConns  int
	SwarmRedisMaxIdleConns  int
	// swim based swarm
	SwarmKubernetesNamespace          string
	SwarmKubernetesLabelSelectorKey   string
	SwarmKubernetesLabelSelectorValue string
	SwarmPort                         int
	SwarmMaxMessageBuffer             int
	SwarmLeaveTimeout                 time.Duration
	// swim based swarm for local testing
	SwarmStaticSelf  string // 127.0.0.1:9001
	SwarmStaticOther string // 127.0.0.1:9002,127.0.0.1:9003

	// SwarmRegistry specifies an optional callback function that is
	// called after ratelimit registry is initialized
	SwarmRegistry func(*ratelimit.Registry)

	// ClusterRatelimitMaxGroupShards specifies the maximum number of group shards for the clusterRatelimit filter
	ClusterRatelimitMaxGroupShards int

	// KubernetesEnableTLS enables kubernetes to use resources to terminate tls
	KubernetesEnableTLS bool
	// contains filtered or unexported fields
}

Options to start skipper.