mirror of
https://github.com/octoleo/syncthing.git
synced 2024-12-23 11:28:59 +00:00
916ec63af6
This is a new revision of the discovery server. Relevant changes and non-changes: - Protocol towards clients is unchanged. - Recommended large scale design is still to be deployed nehind nginx (I tested, and it's still a lot faster at terminating TLS). - Database backend is leveldb again, only. It scales enough, is easy to setup, and we don't need any backend to take care of. - Server supports replication. This is a simple TCP channel - protect it with a firewall when deploying over the internet. (We deploy this within the same datacenter, and with firewall.) Any incoming client announces are sent over the replication channel(s) to other peer discosrvs. Incoming replication changes are applied to the database as if they came from clients, but without the TLS/certificate overhead. - Metrics are exposed using the prometheus library, when enabled. - The database values and replication protocol is protobuf, because JSON was quite CPU intensive when I tried that and benchmarked it. - The "Retry-After" value for failed lookups gets slowly increased from a default of 120 seconds, by 5 seconds for each failed lookup, independently by each discosrv. This lowers the query load over time for clients that are never seen. The Retry-After maxes out at 3600 after a couple of weeks of this increase. The number of failed lookups is stored in the database, now and then (avoiding making each lookup a database put). All in all this means clients can be pointed towards a cluster using just multiple A / AAAA records to gain both load sharing and redundancy (if one is down, clients will talk to the remaining ones). GitHub-Pull-Request: https://github.com/syncthing/syncthing/pull/4648
167 lines
3.6 KiB
Go
167 lines
3.6 KiB
Go
package mg
|
|
|
|
import (
|
|
"context"
|
|
"fmt"
|
|
"reflect"
|
|
"runtime"
|
|
"strings"
|
|
"sync"
|
|
|
|
"github.com/magefile/mage/types"
|
|
)
|
|
|
|
type onceMap struct {
|
|
mu *sync.Mutex
|
|
m map[string]*onceFun
|
|
}
|
|
|
|
func (o *onceMap) LoadOrStore(s string, one *onceFun) *onceFun {
|
|
defer o.mu.Unlock()
|
|
o.mu.Lock()
|
|
|
|
existing, ok := o.m[s]
|
|
if ok {
|
|
return existing
|
|
}
|
|
o.m[s] = one
|
|
return one
|
|
}
|
|
|
|
var onces = &onceMap{
|
|
mu: &sync.Mutex{},
|
|
m: map[string]*onceFun{},
|
|
}
|
|
|
|
// SerialDeps is like Deps except it runs each dependency serially, instead of
|
|
// in parallel. This can be useful for resource intensive dependencies that
|
|
// shouldn't be run at the same time.
|
|
func SerialDeps(fns ...interface{}) {
|
|
checkFns(fns)
|
|
ctx := context.Background()
|
|
for _, f := range fns {
|
|
runDeps(ctx, f)
|
|
}
|
|
}
|
|
|
|
// SerialCtxDeps is like CtxDeps except it runs each dependency serially,
|
|
// instead of in parallel. This can be useful for resource intensive
|
|
// dependencies that shouldn't be run at the same time.
|
|
func SerialCtxDeps(ctx context.Context, fns ...interface{}) {
|
|
checkFns(fns)
|
|
for _, f := range fns {
|
|
runDeps(ctx, f)
|
|
}
|
|
}
|
|
|
|
// CtxDeps runs the given functions as dependencies of the calling function.
|
|
// Dependencies must only be of type: github.com/magefile/mage/types.FuncType.
|
|
// The function calling Deps is guaranteed that all dependent functions will be
|
|
// run exactly once when Deps returns. Dependent functions may in turn declare
|
|
// their own dependencies using Deps. Each dependency is run in their own
|
|
// goroutines. Each function is given the context provided if the function
|
|
// prototype allows for it.
|
|
func CtxDeps(ctx context.Context, fns ...interface{}) {
|
|
checkFns(fns)
|
|
runDeps(ctx, fns...)
|
|
}
|
|
|
|
// runDeps assumes you've already called checkFns.
|
|
func runDeps(ctx context.Context, fns ...interface{}) {
|
|
mu := &sync.Mutex{}
|
|
var errs []string
|
|
var exit int
|
|
wg := &sync.WaitGroup{}
|
|
for _, f := range fns {
|
|
fn := addDep(ctx, f)
|
|
wg.Add(1)
|
|
go func() {
|
|
defer func() {
|
|
if v := recover(); v != nil {
|
|
mu.Lock()
|
|
if err, ok := v.(error); ok {
|
|
exit = changeExit(exit, ExitStatus(err))
|
|
} else {
|
|
exit = changeExit(exit, 1)
|
|
}
|
|
errs = append(errs, fmt.Sprint(v))
|
|
mu.Unlock()
|
|
}
|
|
wg.Done()
|
|
}()
|
|
if err := fn.run(); err != nil {
|
|
mu.Lock()
|
|
errs = append(errs, fmt.Sprint(err))
|
|
exit = changeExit(exit, ExitStatus(err))
|
|
mu.Unlock()
|
|
}
|
|
}()
|
|
}
|
|
|
|
wg.Wait()
|
|
if len(errs) > 0 {
|
|
panic(Fatal(exit, strings.Join(errs, "\n")))
|
|
}
|
|
}
|
|
|
|
func checkFns(fns []interface{}) {
|
|
for _, f := range fns {
|
|
if err := types.FuncCheck(f); err != nil {
|
|
panic(err)
|
|
}
|
|
}
|
|
}
|
|
|
|
// Deps runs the given functions with the default runtime context
|
|
func Deps(fns ...interface{}) {
|
|
CtxDeps(context.Background(), fns...)
|
|
}
|
|
|
|
func changeExit(old, new int) int {
|
|
if new == 0 {
|
|
return old
|
|
}
|
|
if old == 0 {
|
|
return new
|
|
}
|
|
if old == new {
|
|
return old
|
|
}
|
|
// both different and both non-zero, just set
|
|
// exit to 1. Nothing more we can do.
|
|
return 1
|
|
}
|
|
|
|
func addDep(ctx context.Context, f interface{}) *onceFun {
|
|
var fn func(context.Context) error
|
|
if fn = types.FuncTypeWrap(f); fn == nil {
|
|
// should be impossible, since we already checked this
|
|
panic("attempted to add a dep that did not match required type")
|
|
}
|
|
|
|
n := name(f)
|
|
of := onces.LoadOrStore(n, &onceFun{
|
|
fn: fn,
|
|
ctx: ctx,
|
|
})
|
|
return of
|
|
}
|
|
|
|
func name(i interface{}) string {
|
|
return runtime.FuncForPC(reflect.ValueOf(i).Pointer()).Name()
|
|
}
|
|
|
|
type onceFun struct {
|
|
once sync.Once
|
|
fn func(context.Context) error
|
|
ctx context.Context
|
|
}
|
|
|
|
func (o *onceFun) run() error {
|
|
var err error
|
|
o.once.Do(func() {
|
|
err = o.fn(o.ctx)
|
|
})
|
|
return err
|
|
}
|