caddy/modules/caddyhttp/reverseproxy/upstreams.go
Kévin Dunglas f4bf4e0097
perf: use zap's Check() to prevent useless allocs (#6560)
* perf: use zap's Check() to prevent useless allocs

* fix

* fix

* fix

* fix

* restore previous replacer behavior

* fix linter
2024-09-13 11:16:37 -06:00

554 lines
16 KiB
Go

package reverseproxy
import (
"context"
"encoding/json"
"fmt"
weakrand "math/rand"
"net"
"net/http"
"strconv"
"sync"
"time"
"go.uber.org/zap"
"go.uber.org/zap/zapcore"
"github.com/caddyserver/caddy/v2"
)
func init() {
caddy.RegisterModule(SRVUpstreams{})
caddy.RegisterModule(AUpstreams{})
caddy.RegisterModule(MultiUpstreams{})
}
// SRVUpstreams provides upstreams from SRV lookups.
// The lookup DNS name can be configured either by
// its individual parts (that is, specifying the
// service, protocol, and name separately) to form
// the standard "_service._proto.name" domain, or
// the domain can be specified directly in name by
// leaving service and proto empty. See RFC 2782.
//
// Lookups are cached and refreshed at the configured
// refresh interval.
//
// Returned upstreams are sorted by priority and weight.
type SRVUpstreams struct {
// The service label.
Service string `json:"service,omitempty"`
// The protocol label; either tcp or udp.
Proto string `json:"proto,omitempty"`
// The name label; or, if service and proto are
// empty, the entire domain name to look up.
Name string `json:"name,omitempty"`
// The interval at which to refresh the SRV lookup.
// Results are cached between lookups. Default: 1m
Refresh caddy.Duration `json:"refresh,omitempty"`
// If > 0 and there is an error with the lookup,
// continue to use the cached results for up to
// this long before trying again, (even though they
// are stale) instead of returning an error to the
// client. Default: 0s.
GracePeriod caddy.Duration `json:"grace_period,omitempty"`
// Configures the DNS resolver used to resolve the
// SRV address to SRV records.
Resolver *UpstreamResolver `json:"resolver,omitempty"`
// If Resolver is configured, how long to wait before
// timing out trying to connect to the DNS server.
DialTimeout caddy.Duration `json:"dial_timeout,omitempty"`
// If Resolver is configured, how long to wait before
// spawning an RFC 6555 Fast Fallback connection.
// A negative value disables this.
FallbackDelay caddy.Duration `json:"dial_fallback_delay,omitempty"`
resolver *net.Resolver
logger *zap.Logger
}
// CaddyModule returns the Caddy module information.
func (SRVUpstreams) CaddyModule() caddy.ModuleInfo {
return caddy.ModuleInfo{
ID: "http.reverse_proxy.upstreams.srv",
New: func() caddy.Module { return new(SRVUpstreams) },
}
}
func (su *SRVUpstreams) Provision(ctx caddy.Context) error {
su.logger = ctx.Logger()
if su.Refresh == 0 {
su.Refresh = caddy.Duration(time.Minute)
}
if su.Resolver != nil {
err := su.Resolver.ParseAddresses()
if err != nil {
return err
}
d := &net.Dialer{
Timeout: time.Duration(su.DialTimeout),
FallbackDelay: time.Duration(su.FallbackDelay),
}
su.resolver = &net.Resolver{
PreferGo: true,
Dial: func(ctx context.Context, _, _ string) (net.Conn, error) {
//nolint:gosec
addr := su.Resolver.netAddrs[weakrand.Intn(len(su.Resolver.netAddrs))]
return d.DialContext(ctx, addr.Network, addr.JoinHostPort(0))
},
}
}
if su.resolver == nil {
su.resolver = net.DefaultResolver
}
return nil
}
func (su SRVUpstreams) GetUpstreams(r *http.Request) ([]*Upstream, error) {
suAddr, service, proto, name := su.expandedAddr(r)
// first, use a cheap read-lock to return a cached result quickly
srvsMu.RLock()
cached := srvs[suAddr]
srvsMu.RUnlock()
if cached.isFresh() {
return allNew(cached.upstreams), nil
}
// otherwise, obtain a write-lock to update the cached value
srvsMu.Lock()
defer srvsMu.Unlock()
// check to see if it's still stale, since we're now in a different
// lock from when we first checked freshness; another goroutine might
// have refreshed it in the meantime before we re-obtained our lock
cached = srvs[suAddr]
if cached.isFresh() {
return allNew(cached.upstreams), nil
}
if c := su.logger.Check(zapcore.DebugLevel, "refreshing SRV upstreams"); c != nil {
c.Write(
zap.String("service", service),
zap.String("proto", proto),
zap.String("name", name),
)
}
_, records, err := su.resolver.LookupSRV(r.Context(), service, proto, name)
if err != nil {
// From LookupSRV docs: "If the response contains invalid names, those records are filtered
// out and an error will be returned alongside the remaining results, if any." Thus, we
// only return an error if no records were also returned.
if len(records) == 0 {
if su.GracePeriod > 0 {
if c := su.logger.Check(zapcore.ErrorLevel, "SRV lookup failed; using previously cached"); c != nil {
c.Write(zap.Error(err))
}
cached.freshness = time.Now().Add(time.Duration(su.GracePeriod) - time.Duration(su.Refresh))
srvs[suAddr] = cached
return allNew(cached.upstreams), nil
}
return nil, err
}
if c := su.logger.Check(zapcore.WarnLevel, "SRV records filtered"); c != nil {
c.Write(zap.Error(err))
}
}
upstreams := make([]Upstream, len(records))
for i, rec := range records {
if c := su.logger.Check(zapcore.DebugLevel, "discovered SRV record"); c != nil {
c.Write(
zap.String("target", rec.Target),
zap.Uint16("port", rec.Port),
zap.Uint16("priority", rec.Priority),
zap.Uint16("weight", rec.Weight),
)
}
addr := net.JoinHostPort(rec.Target, strconv.Itoa(int(rec.Port)))
upstreams[i] = Upstream{Dial: addr}
}
// before adding a new one to the cache (as opposed to replacing stale one), make room if cache is full
if cached.freshness.IsZero() && len(srvs) >= 100 {
for randomKey := range srvs {
delete(srvs, randomKey)
break
}
}
srvs[suAddr] = srvLookup{
srvUpstreams: su,
freshness: time.Now(),
upstreams: upstreams,
}
return allNew(upstreams), nil
}
func (su SRVUpstreams) String() string {
if su.Service == "" && su.Proto == "" {
return su.Name
}
return su.formattedAddr(su.Service, su.Proto, su.Name)
}
// expandedAddr expands placeholders in the configured SRV domain labels.
// The return values are: addr, the RFC 2782 representation of the SRV domain;
// service, the service; proto, the protocol; and name, the name.
// If su.Service and su.Proto are empty, name will be returned as addr instead.
func (su SRVUpstreams) expandedAddr(r *http.Request) (addr, service, proto, name string) {
repl := r.Context().Value(caddy.ReplacerCtxKey).(*caddy.Replacer)
name = repl.ReplaceAll(su.Name, "")
if su.Service == "" && su.Proto == "" {
addr = name
return
}
service = repl.ReplaceAll(su.Service, "")
proto = repl.ReplaceAll(su.Proto, "")
addr = su.formattedAddr(service, proto, name)
return
}
// formattedAddr the RFC 2782 representation of the SRV domain, in
// the form "_service._proto.name".
func (SRVUpstreams) formattedAddr(service, proto, name string) string {
return fmt.Sprintf("_%s._%s.%s", service, proto, name)
}
type srvLookup struct {
srvUpstreams SRVUpstreams
freshness time.Time
upstreams []Upstream
}
func (sl srvLookup) isFresh() bool {
return time.Since(sl.freshness) < time.Duration(sl.srvUpstreams.Refresh)
}
type IPVersions struct {
IPv4 *bool `json:"ipv4,omitempty"`
IPv6 *bool `json:"ipv6,omitempty"`
}
func resolveIpVersion(versions *IPVersions) string {
resolveIpv4 := versions == nil || (versions.IPv4 == nil && versions.IPv6 == nil) || (versions.IPv4 != nil && *versions.IPv4)
resolveIpv6 := versions == nil || (versions.IPv6 == nil && versions.IPv4 == nil) || (versions.IPv6 != nil && *versions.IPv6)
switch {
case resolveIpv4 && !resolveIpv6:
return "ip4"
case !resolveIpv4 && resolveIpv6:
return "ip6"
default:
return "ip"
}
}
// AUpstreams provides upstreams from A/AAAA lookups.
// Results are cached and refreshed at the configured
// refresh interval.
type AUpstreams struct {
// The domain name to look up.
Name string `json:"name,omitempty"`
// The port to use with the upstreams. Default: 80
Port string `json:"port,omitempty"`
// The interval at which to refresh the A lookup.
// Results are cached between lookups. Default: 1m
Refresh caddy.Duration `json:"refresh,omitempty"`
// Configures the DNS resolver used to resolve the
// domain name to A records.
Resolver *UpstreamResolver `json:"resolver,omitempty"`
// If Resolver is configured, how long to wait before
// timing out trying to connect to the DNS server.
DialTimeout caddy.Duration `json:"dial_timeout,omitempty"`
// If Resolver is configured, how long to wait before
// spawning an RFC 6555 Fast Fallback connection.
// A negative value disables this.
FallbackDelay caddy.Duration `json:"dial_fallback_delay,omitempty"`
// The IP versions to resolve for. By default, both
// "ipv4" and "ipv6" will be enabled, which
// correspond to A and AAAA records respectively.
Versions *IPVersions `json:"versions,omitempty"`
resolver *net.Resolver
logger *zap.Logger
}
// CaddyModule returns the Caddy module information.
func (AUpstreams) CaddyModule() caddy.ModuleInfo {
return caddy.ModuleInfo{
ID: "http.reverse_proxy.upstreams.a",
New: func() caddy.Module { return new(AUpstreams) },
}
}
func (au *AUpstreams) Provision(ctx caddy.Context) error {
au.logger = ctx.Logger()
if au.Refresh == 0 {
au.Refresh = caddy.Duration(time.Minute)
}
if au.Port == "" {
au.Port = "80"
}
if au.Resolver != nil {
err := au.Resolver.ParseAddresses()
if err != nil {
return err
}
d := &net.Dialer{
Timeout: time.Duration(au.DialTimeout),
FallbackDelay: time.Duration(au.FallbackDelay),
}
au.resolver = &net.Resolver{
PreferGo: true,
Dial: func(ctx context.Context, _, _ string) (net.Conn, error) {
//nolint:gosec
addr := au.Resolver.netAddrs[weakrand.Intn(len(au.Resolver.netAddrs))]
return d.DialContext(ctx, addr.Network, addr.JoinHostPort(0))
},
}
}
if au.resolver == nil {
au.resolver = net.DefaultResolver
}
return nil
}
func (au AUpstreams) GetUpstreams(r *http.Request) ([]*Upstream, error) {
repl := r.Context().Value(caddy.ReplacerCtxKey).(*caddy.Replacer)
// Map ipVersion early, so we can use it as part of the cache-key.
// This should be fairly inexpensive and comes and the upside of
// allowing the same dynamic upstream (name + port combination)
// to be used multiple times with different ip versions.
//
// It also forced a cache-miss if a previously cached dynamic
// upstream changes its ip version, e.g. after a config reload,
// while keeping the cache-invalidation as simple as it currently is.
ipVersion := resolveIpVersion(au.Versions)
auStr := repl.ReplaceAll(au.String()+ipVersion, "")
// first, use a cheap read-lock to return a cached result quickly
aAaaaMu.RLock()
cached := aAaaa[auStr]
aAaaaMu.RUnlock()
if cached.isFresh() {
return allNew(cached.upstreams), nil
}
// otherwise, obtain a write-lock to update the cached value
aAaaaMu.Lock()
defer aAaaaMu.Unlock()
// check to see if it's still stale, since we're now in a different
// lock from when we first checked freshness; another goroutine might
// have refreshed it in the meantime before we re-obtained our lock
cached = aAaaa[auStr]
if cached.isFresh() {
return allNew(cached.upstreams), nil
}
name := repl.ReplaceAll(au.Name, "")
port := repl.ReplaceAll(au.Port, "")
if c := au.logger.Check(zapcore.DebugLevel, "refreshing A upstreams"); c != nil {
c.Write(
zap.String("version", ipVersion),
zap.String("name", name),
zap.String("port", port),
)
}
ips, err := au.resolver.LookupIP(r.Context(), ipVersion, name)
if err != nil {
return nil, err
}
upstreams := make([]Upstream, len(ips))
for i, ip := range ips {
if c := au.logger.Check(zapcore.DebugLevel, "discovered A record"); c != nil {
c.Write(zap.String("ip", ip.String()))
}
upstreams[i] = Upstream{
Dial: net.JoinHostPort(ip.String(), port),
}
}
// before adding a new one to the cache (as opposed to replacing stale one), make room if cache is full
if cached.freshness.IsZero() && len(aAaaa) >= 100 {
for randomKey := range aAaaa {
delete(aAaaa, randomKey)
break
}
}
aAaaa[auStr] = aLookup{
aUpstreams: au,
freshness: time.Now(),
upstreams: upstreams,
}
return allNew(upstreams), nil
}
func (au AUpstreams) String() string { return net.JoinHostPort(au.Name, au.Port) }
type aLookup struct {
aUpstreams AUpstreams
freshness time.Time
upstreams []Upstream
}
func (al aLookup) isFresh() bool {
return time.Since(al.freshness) < time.Duration(al.aUpstreams.Refresh)
}
// MultiUpstreams is a single dynamic upstream source that
// aggregates the results of multiple dynamic upstream sources.
// All configured sources will be queried in order, with their
// results appended to the end of the list. Errors returned
// from individual sources will be logged and the next source
// will continue to be invoked.
//
// This module makes it easy to implement redundant cluster
// failovers, especially in conjunction with the `first` load
// balancing policy: if the first source returns an error or
// no upstreams, the second source's upstreams will be used
// naturally.
type MultiUpstreams struct {
// The list of upstream source modules to get upstreams from.
// They will be queried in order, with their results appended
// in the order they are returned.
SourcesRaw []json.RawMessage `json:"sources,omitempty" caddy:"namespace=http.reverse_proxy.upstreams inline_key=source"`
sources []UpstreamSource
logger *zap.Logger
}
// CaddyModule returns the Caddy module information.
func (MultiUpstreams) CaddyModule() caddy.ModuleInfo {
return caddy.ModuleInfo{
ID: "http.reverse_proxy.upstreams.multi",
New: func() caddy.Module { return new(MultiUpstreams) },
}
}
func (mu *MultiUpstreams) Provision(ctx caddy.Context) error {
mu.logger = ctx.Logger()
if mu.SourcesRaw != nil {
mod, err := ctx.LoadModule(mu, "SourcesRaw")
if err != nil {
return fmt.Errorf("loading upstream source modules: %v", err)
}
for _, src := range mod.([]any) {
mu.sources = append(mu.sources, src.(UpstreamSource))
}
}
return nil
}
func (mu MultiUpstreams) GetUpstreams(r *http.Request) ([]*Upstream, error) {
var upstreams []*Upstream
for i, src := range mu.sources {
select {
case <-r.Context().Done():
return upstreams, context.Canceled
default:
}
up, err := src.GetUpstreams(r)
if err != nil {
if c := mu.logger.Check(zapcore.ErrorLevel, "upstream source returned error"); c != nil {
c.Write(
zap.Int("source_idx", i),
zap.Error(err),
)
}
} else if len(up) == 0 {
if c := mu.logger.Check(zapcore.WarnLevel, "upstream source returned 0 upstreams"); c != nil {
c.Write(zap.Int("source_idx", i))
}
} else {
upstreams = append(upstreams, up...)
}
}
return upstreams, nil
}
// UpstreamResolver holds the set of addresses of DNS resolvers of
// upstream addresses
type UpstreamResolver struct {
// The addresses of DNS resolvers to use when looking up the addresses of proxy upstreams.
// It accepts [network addresses](/docs/conventions#network-addresses)
// with port range of only 1. If the host is an IP address, it will be dialed directly to resolve the upstream server.
// If the host is not an IP address, the addresses are resolved using the [name resolution convention](https://golang.org/pkg/net/#hdr-Name_Resolution) of the Go standard library.
// If the array contains more than 1 resolver address, one is chosen at random.
Addresses []string `json:"addresses,omitempty"`
netAddrs []caddy.NetworkAddress
}
// ParseAddresses parses all the configured network addresses
// and ensures they're ready to be used.
func (u *UpstreamResolver) ParseAddresses() error {
for _, v := range u.Addresses {
addr, err := caddy.ParseNetworkAddressWithDefaults(v, "udp", 53)
if err != nil {
return err
}
if addr.PortRangeSize() != 1 {
return fmt.Errorf("resolver address must have exactly one address; cannot call %v", addr)
}
u.netAddrs = append(u.netAddrs, addr)
}
return nil
}
func allNew(upstreams []Upstream) []*Upstream {
results := make([]*Upstream, len(upstreams))
for i := range upstreams {
results[i] = &Upstream{Dial: upstreams[i].Dial}
}
return results
}
var (
srvs = make(map[string]srvLookup)
srvsMu sync.RWMutex
aAaaa = make(map[string]aLookup)
aAaaaMu sync.RWMutex
)
// Interface guards
var (
_ caddy.Provisioner = (*SRVUpstreams)(nil)
_ UpstreamSource = (*SRVUpstreams)(nil)
_ caddy.Provisioner = (*AUpstreams)(nil)
_ UpstreamSource = (*AUpstreams)(nil)
)