mirrors/gotosocial
1
0
Fork 0
mirror of https://github.com/superseriousbusiness/gotosocial.git synced 2025-10-29 07:22:24 -05:00
Code Issues Projects Releases Packages Wiki Activity Actions

[performance] massively improved ActivityPub delivery worker efficiency (#2812)

Browse source 
* add delivery worker type that pulls from queue to httpclient package

* finish up some code commenting, bodge a vendored activity library change, integrate the deliverypool changes into transportcontroller

* hook up queue deletion logic

* support deleting queued http requests by target ID

* don't index APRequest by hostname in the queue

* use gorun

* use the original context's values when wrapping msg type as delivery{}

* actually log in the AP delivery worker ...

* add uncommitted changes

* use errors.AsV2()

* use errorsv2.AsV2()

* finish adding some code comments, add bad host handling to delivery workers

* slightly tweak deliveryworkerpool API, use advanced sender multiplier

* remove PopCtx() method, let others instead rely on Wait()

* shuffle things around to move delivery stuff into transport/ subpkg

* remove dead code

* formatting

* validate request before queueing for delivery

* finish adding code comments, fix up backoff code

* finish adding more code comments

* clamp minimum no. senders to 1

* add start/stop logging to delivery worker, some slight changes

* remove double logging

* use worker ptrs

* expose the embedded log fields in httpclient.Request{}

* ensure request context values are preserved when updating ctx

* add delivery worker tests

* fix linter issues

* ensure delivery worker gets inited in testrig

* fix tests to delivering messages to check worker delivery queue

* update error type to use ptr instead of value receiver

* fix test calling Workers{}.Start() instead of testrig.StartWorkers()

* update docs for advanced-sender-multiplier

* update to the latest activity library version

* add comment about not using httptest.Server{}
This commit is contained in:
kim 2024-04-11 10:45:35 +01:00 • committed by GitHub
commit a483bd9e38
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
32 changed files with 1285 additions and 374 deletions
Download patch file Download diff file Expand all files Collapse all files

16
internal/transport/controller.go
View file

@ -28,7 +28,6 @@ import (
"io"
"net/http"
"net/url"
"runtime"
"codeberg.org/gruf/go-byteutil"
"codeberg.org/gruf/go-cache/v3"
@ -56,24 +55,16 @@ type controller struct {
client pub.HttpClient
trspCache cache.TTLCache[string, *transport]
userAgent string
senders int // no. concurrent batch delivery routines.
}
// NewController returns an implementation of the Controller interface for creating new transports
func NewController(state *state.State, federatingDB federatingdb.DB, clock pub.Clock, client pub.HttpClient) Controller {
var (
host = config.GetHost()
proto = config.GetProtocol()
version = config.GetSoftwareVersion()
senderMultiplier = config.GetAdvancedSenderMultiplier()
host = config.GetHost()
proto = config.GetProtocol()
version = config.GetSoftwareVersion()
)
senders := senderMultiplier * runtime.GOMAXPROCS(0)
if senders < 1 {
// Clamp senders to 1.
senders = 1
}
c := &controller{
state: state,
fedDB: federatingDB,
@ -81,7 +72,6 @@ func NewController(state *state.State, federatingDB federatingdb.DB, clock pub.C
client: client,
trspCache: cache.NewTTL[string, *transport](0, 100, 0),
userAgent: fmt.Sprintf("gotosocial/%s (+%s://%s)", version, proto, host),
senders: senders,
}
return c

226
internal/transport/deliver.go
View file

@ -19,118 +19,188 @@ package transport
import (
"context"
"encoding/json"
"net/http"
"net/url"
"sync"
"codeberg.org/gruf/go-byteutil"
apiutil "github.com/superseriousbusiness/gotosocial/internal/api/util"
"github.com/superseriousbusiness/gotosocial/internal/config"
"github.com/superseriousbusiness/gotosocial/internal/gtscontext"
"github.com/superseriousbusiness/gotosocial/internal/gtserror"
"github.com/superseriousbusiness/gotosocial/internal/httpclient"
"github.com/superseriousbusiness/gotosocial/internal/transport/delivery"
)
func (t *transport) BatchDeliver(ctx context.Context, b []byte, recipients []*url.URL) error {
func (t *transport) BatchDeliver(ctx context.Context, obj map[string]interface{}, recipients []*url.URL) error {
var (
// errs accumulates errors received during
// attempted delivery by deliverer routines.
// accumulated delivery reqs.
reqs []*delivery.Delivery
// accumulated preparation errs.
errs gtserror.MultiError
// wait blocks until all sender
// routines have returned.
wait sync.WaitGroup
// mutex protects 'recipients' and
// 'errs' for concurrent access.
mutex sync.Mutex
// Get current instance host info.
domain = config.GetAccountDomain()
host = config.GetHost()
)
// Block on expect no. senders.
wait.Add(t.controller.senders)
for i := 0; i < t.controller.senders; i++ {
go func() {
// Mark returned.
defer wait.Done()
for {
// Acquire lock.
mutex.Lock()
if len(recipients) == 0 {
// Reached end.
mutex.Unlock()
return
}
// Pop next recipient.
i := len(recipients) - 1
to := recipients[i]
recipients = recipients[:i]
// Done with lock.
mutex.Unlock()
// Skip delivery to recipient if it is "us".
if to.Host == host || to.Host == domain {
continue
}
// Attempt to deliver data to recipient.
if err := t.deliver(ctx, b, to); err != nil {
mutex.Lock() // safely append err to accumulator.
errs.Appendf("error delivering to %s: %w", to, err)
mutex.Unlock()
}
}
}()
// Marshal object as JSON.
b, err := json.Marshal(obj)
if err != nil {
return gtserror.Newf("error marshaling json: %w", err)
}
// Wait for finish.
wait.Wait()
// Extract object IDs.
actID := getActorID(obj)
objID := getObjectID(obj)
tgtID := getTargetID(obj)
for _, to := range recipients {
// Skip delivery to recipient if it is "us".
if to.Host == host || to.Host == domain {
continue
}
// Prepare http client request.
req, err := t.prepare(ctx,
actID,
objID,
tgtID,
b,
to,
)
if err != nil {
errs.Append(err)
continue
}
// Append to request queue.
reqs = append(reqs, req)
}
// Push prepared request list to the delivery queue.
t.controller.state.Workers.Delivery.Queue.Push(reqs...)
// Return combined err.
return errs.Combine()
}
func (t *transport) Deliver(ctx context.Context, b []byte, to *url.URL) error {
func (t *transport) Deliver(ctx context.Context, obj map[string]interface{}, to *url.URL) error {
// if 'to' host is our own, skip as we don't need to deliver to ourselves...
if to.Host == config.GetHost() || to.Host == config.GetAccountDomain() {
return nil
}
// Deliver data to recipient.
return t.deliver(ctx, b, to)
}
// Marshal object as JSON.
b, err := json.Marshal(obj)
if err != nil {
return gtserror.Newf("error marshaling json: %w", err)
}
func (t *transport) deliver(ctx context.Context, b []byte, to *url.URL) error {
url := to.String()
// Use rewindable bytes reader for body.
var body byteutil.ReadNopCloser
body.Reset(b)
req, err := http.NewRequestWithContext(ctx, "POST", url, &body)
// Prepare http client request.
req, err := t.prepare(ctx,
getActorID(obj),
getObjectID(obj),
getTargetID(obj),
b,
to,
)
if err != nil {
return err
}
req.Header.Add("Content-Type", string(apiutil.AppActivityLDJSON))
req.Header.Add("Accept-Charset", "utf-8")
rsp, err := t.POST(req, b)
if err != nil {
return err
}
defer rsp.Body.Close()
if code := rsp.StatusCode; code != http.StatusOK &&
code != http.StatusCreated && code != http.StatusAccepted {
return gtserror.NewFromResponse(rsp)
}
// Push prepared request to the delivery queue.
t.controller.state.Workers.Delivery.Queue.Push(req)
return nil
}
// prepare will prepare a POST http.Request{}
// to recipient at 'to', wrapping in a queued
// request object with signing function.
func (t *transport) prepare(
ctx context.Context,
actorID string,
objectID string,
targetID string,
data []byte,
to *url.URL,
) (
*delivery.Delivery,
error,
) {
url := to.String()
// Use rewindable reader for body.
var body byteutil.ReadNopCloser
body.Reset(data)
// Prepare POST signer.
sign := t.signPOST(data)
// Update to-be-used request context with signing details.
ctx = gtscontext.SetOutgoingPublicKeyID(ctx, t.pubKeyID)
ctx = gtscontext.SetHTTPClientSignFunc(ctx, sign)
// Prepare a new request with data body directed at URL.
r, err := http.NewRequestWithContext(ctx, "POST", url, &body)
if err != nil {
return nil, gtserror.Newf("error preparing request: %w", err)
}
// Set the standard ActivityPub content-type + charset headers.
r.Header.Add("Content-Type", string(apiutil.AppActivityLDJSON))
r.Header.Add("Accept-Charset", "utf-8")
// Validate the request before queueing for delivery.
if err := httpclient.ValidateRequest(r); err != nil {
return nil, err
}
return &delivery.Delivery{
ActorID: actorID,
ObjectID: objectID,
TargetID: targetID,
Request: httpclient.WrapRequest(r),
}, nil
}
// getObjectID extracts an object ID from 'serialized' ActivityPub object map.
func getObjectID(obj map[string]interface{}) string {
switch t := obj["object"].(type) {
case string:
return t
case map[string]interface{}:
id, _ := t["id"].(string)
return id
default:
return ""
}
}
// getActorID extracts an actor ID from 'serialized' ActivityPub object map.
func getActorID(obj map[string]interface{}) string {
switch t := obj["actor"].(type) {
case string:
return t
case map[string]interface{}:
id, _ := t["id"].(string)
return id
default:
return ""
}
}
// getTargetID extracts a target ID from 'serialized' ActivityPub object map.
func getTargetID(obj map[string]interface{}) string {
switch t := obj["target"].(type) {
case string:
return t
case map[string]interface{}:
id, _ := t["id"].(string)
return id
default:
return ""
}
}

323
internal/transport/delivery/delivery.go Normal file
View file

@ -0,0 +1,323 @@
// GoToSocial
// Copyright (C) GoToSocial Authors admin@gotosocial.org
// SPDX-License-Identifier: AGPL-3.0-or-later
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package delivery
import (
"context"
"slices"
"time"
"codeberg.org/gruf/go-runners"
"codeberg.org/gruf/go-structr"
"github.com/superseriousbusiness/gotosocial/internal/gtscontext"
"github.com/superseriousbusiness/gotosocial/internal/httpclient"
"github.com/superseriousbusiness/gotosocial/internal/log"
"github.com/superseriousbusiness/gotosocial/internal/queue"
)
// Delivery wraps an httpclient.Request{}
// to add ActivityPub ID IRI fields of the
// outgoing activity, so that deliveries may
// be indexed (and so, dropped from queue)
// by any of these possible ID IRIs.
type Delivery struct {
// ActorID contains the ActivityPub
// actor ID IRI (if any) of the activity
// being sent out by this request.
ActorID string
// ObjectID contains the ActivityPub
// object ID IRI (if any) of the activity
// being sent out by this request.
ObjectID string
// TargetID contains the ActivityPub
// target ID IRI (if any) of the activity
// being sent out by this request.
TargetID string
// Request is the prepared (+ wrapped)
// httpclient.Client{} request that
// constitutes this ActivtyPub delivery.
Request httpclient.Request
// internal fields.
next time.Time
}
func (dlv *Delivery) backoff() time.Duration {
if dlv.next.IsZero() {
return 0
}
return time.Until(dlv.next)
}
// WorkerPool wraps multiple Worker{}s in
// a singular struct for easy multi start/stop.
type WorkerPool struct {
// Client defines httpclient.Client{}
// passed to each of delivery pool Worker{}s.
Client *httpclient.Client
// Queue is the embedded queue.StructQueue{}
// passed to each of delivery pool Worker{}s.
Queue queue.StructQueue[*Delivery]
// internal fields.
workers []*Worker
}
// Init will initialize the Worker{} pool
// with given http client, request queue to pull
// from and number of delivery workers to spawn.
func (p *WorkerPool) Init(client *httpclient.Client) {
p.Client = client
p.Queue.Init(structr.QueueConfig[*Delivery]{
Indices: []structr.IndexConfig{
{Fields: "ActorID", Multiple: true},
{Fields: "ObjectID", Multiple: true},
{Fields: "TargetID", Multiple: true},
},
})
}
// Start will attempt to start 'n' Worker{}s.
func (p *WorkerPool) Start(n int) (ok bool) {
if ok = (len(p.workers) == 0); ok {
p.workers = make([]*Worker, n)
for i := range p.workers {
p.workers[i] = new(Worker)
p.workers[i].Client = p.Client
p.workers[i].Queue = &p.Queue
ok = p.workers[i].Start() && ok
}
}
return
}
// Stop will attempt to stop contained Worker{}s.
func (p *WorkerPool) Stop() (ok bool) {
if ok = (len(p.workers) > 0); ok {
for i := range p.workers {
ok = p.workers[i].Stop() && ok
p.workers[i] = nil
}
p.workers = p.workers[:0]
}
return
}
// Worker wraps an httpclient.Client{} to feed
// from queue.StructQueue{} for ActivityPub reqs
// to deliver. It does so while prioritizing new
// queued requests over backlogged retries.
type Worker struct {
// Client is the httpclient.Client{} that
// delivery worker will use for requests.
Client *httpclient.Client
// Queue is the Delivery{} message queue
// that delivery worker will feed from.
Queue *queue.StructQueue[*Delivery]
// internal fields.
backlog []*Delivery
service runners.Service
}
// Start will attempt to start the Worker{}.
func (w *Worker) Start() bool {
return w.service.GoRun(w.run)
}
// Stop will attempt to stop the Worker{}.
func (w *Worker) Stop() bool {
return w.service.Stop()
}
// run wraps process to restart on any panic.
func (w *Worker) run(ctx context.Context) {
if w.Client == nil || w.Queue == nil {
panic("not yet initialized")
}
log.Infof(ctx, "%p: started delivery worker", w)
defer log.Infof(ctx, "%p: stopped delivery worker", w)
for returned := false; !returned; {
func() {
defer func() {
if r := recover(); r != nil {
log.Errorf(ctx, "recovered panic: %v", r)
}
}()
w.process(ctx)
returned = true
}()
}
}
// process is the main delivery worker processing routine.
func (w *Worker) process(ctx context.Context) {
if w.Client == nil || w.Queue == nil {
// we perform this check here just
// to ensure the compiler knows these
// variables aren't nil in the loop,
// even if already checked by caller.
panic("not yet initialized")
}
loop:
for {
// Get next delivery.
dlv, ok := w.next(ctx)
if !ok {
return
}
// Check whether backoff required.
const min = 100 * time.Millisecond
if d := dlv.backoff(); d > min {
// Start backoff sleep timer.
backoff := time.NewTimer(d)
select {
case <-ctx.Done():
// Main ctx
// cancelled.
backoff.Stop()
return
case <-w.Queue.Wait():
// A new message was
// queued, re-add this
// to backlog + retry.
w.pushBacklog(dlv)
backoff.Stop()
continue loop
case <-backoff.C:
// success!
}
}
// Attempt delivery of AP request.
rsp, retry, err := w.Client.DoOnce(
&dlv.Request,
)
if err == nil {
// Ensure body closed.
_ = rsp.Body.Close()
continue loop
}
if !retry {
// Drop deliveries when no
// retry requested, or they
// reached max (either).
continue loop
}
// Determine next delivery attempt.
backoff := dlv.Request.BackOff()
dlv.next = time.Now().Add(backoff)
// Push to backlog.
w.pushBacklog(dlv)
}
}
// next gets the next available delivery, blocking until available if necessary.
func (w *Worker) next(ctx context.Context) (*Delivery, bool) {
loop:
for {
// Try pop next queued.
dlv, ok := w.Queue.Pop()
if !ok {
// Check the backlog.
if len(w.backlog) > 0 {
// Sort by 'next' time.
sortDeliveries(w.backlog)
// Pop next delivery.
dlv := w.popBacklog()
return dlv, true
}
select {
// Backlog is empty, we MUST
// block until next enqueued.
case <-w.Queue.Wait():
continue loop
// Worker was stopped.
case <-ctx.Done():
return nil, false
}
}
// Replace request context for worker state canceling.
ctx := gtscontext.WithValues(ctx, dlv.Request.Context())
dlv.Request.Request = dlv.Request.Request.WithContext(ctx)
return dlv, true
}
}
// popBacklog pops next available from the backlog.
func (w *Worker) popBacklog() *Delivery {
if len(w.backlog) == 0 {
return nil
}
// Pop from backlog.
dlv := w.backlog[0]
// Shift backlog down by one.
copy(w.backlog, w.backlog[1:])
w.backlog = w.backlog[:len(w.backlog)-1]
return dlv
}
// pushBacklog pushes the given delivery to backlog.
func (w *Worker) pushBacklog(dlv *Delivery) {
w.backlog = append(w.backlog, dlv)
}
// sortDeliveries sorts deliveries according
// to when is the first requiring re-attempt.
func sortDeliveries(d []*Delivery) {
slices.SortFunc(d, func(a, b *Delivery) int {
const k = +1
switch {
case a.next.Before(b.next):
return +k
case b.next.Before(a.next):
return -k
default:
return 0
}
})
}

205
internal/transport/delivery/delivery_test.go Normal file
View file

@ -0,0 +1,205 @@
package delivery_test
import (
"fmt"
"io"
"math/rand"
"net"
"net/http"
"strconv"
"strings"
"testing"
"codeberg.org/gruf/go-byteutil"
"github.com/superseriousbusiness/gotosocial/internal/config"
"github.com/superseriousbusiness/gotosocial/internal/httpclient"
"github.com/superseriousbusiness/gotosocial/internal/queue"
"github.com/superseriousbusiness/gotosocial/internal/transport/delivery"
)
func TestDeliveryWorkerPool(t *testing.T) {
for _, i := range []int{1, 2, 4, 8, 16, 32} {
t.Run("size="+strconv.Itoa(i), func(t *testing.T) {
testDeliveryWorkerPool(t, i, generateInput(100*i))
})
}
}
func testDeliveryWorkerPool(t *testing.T, sz int, input []*testrequest) {
wp := new(delivery.WorkerPool)
wp.Init(httpclient.New(httpclient.Config{
AllowRanges: config.MustParseIPPrefixes([]string{
"127.0.0.0/8",
}),
}))
if !wp.Start(sz) {
t.Fatal("failed starting pool")
}
defer wp.Stop()
test(t, &wp.Queue, input)
}
func test(
t *testing.T,
queue *queue.StructQueue[*delivery.Delivery],
input []*testrequest,
) {
expect := make(chan *testrequest)
errors := make(chan error)
// Prepare an HTTP test handler that ensures expected delivery is received.
handler := http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
errors <- (<-expect).Equal(r)
})
// Start new HTTP test server listener.
l, err := net.Listen("tcp", "127.0.0.1:0")
if err != nil {
t.Fatal(err)
}
defer l.Close()
// Start the HTTP server.
//
// specifically not using httptest.Server{} here as httptest
// links that server with its own http.Client{}, whereas we're
// using an httpclient.Client{} (well, delivery routine is).
srv := new(http.Server)
srv.Addr = "http://" + l.Addr().String()
srv.Handler = handler
go srv.Serve(l)
defer srv.Close()
// Range over test input.
for _, test := range input {
// Generate req for input.
req := test.Generate(srv.Addr)
r := httpclient.WrapRequest(req)
// Wrap the request in delivery.
dlv := new(delivery.Delivery)
dlv.Request = r
// Enqueue delivery!
queue.Push(dlv)
expect <- test
// Wait for errors from handler.
if err := <-errors; err != nil {
t.Error(err)
}
}
}
type testrequest struct {
method string
uri string
body []byte
}
// generateInput generates 'n' many testrequest cases.
func generateInput(n int) []*testrequest {
tests := make([]*testrequest, n)
for i := range tests {
tests[i] = new(testrequest)
tests[i].method = randomMethod()
tests[i].uri = randomURI()
tests[i].body = randomBody(tests[i].method)
}
return tests
}
var methods = []string{
http.MethodConnect,
http.MethodDelete,
http.MethodGet,
http.MethodHead,
http.MethodOptions,
http.MethodPatch,
http.MethodPost,
http.MethodPut,
http.MethodTrace,
}
// randomMethod generates a random http method.
func randomMethod() string {
return methods[rand.Intn(len(methods))]
}
// randomURI generates a random http uri.
func randomURI() string {
n := rand.Intn(5)
p := make([]string, n)
for i := range p {
p[i] = strconv.Itoa(rand.Int())
}
return "/" + strings.Join(p, "/")
}
// randomBody generates a random http body DEPENDING on method.
func randomBody(method string) []byte {
if requiresBody(method) {
return []byte(method + " " + randomURI())
}
return nil
}
// requiresBody returns whether method requires body.
func requiresBody(method string) bool {
switch method {
case http.MethodPatch,
http.MethodPost,
http.MethodPut:
return true
default:
return false
}
}
// Generate will generate a real http.Request{} from test data.
func (t *testrequest) Generate(addr string) *http.Request {
var body io.ReadCloser
if t.body != nil {
var b byteutil.ReadNopCloser
b.Reset(t.body)
body = &b
}
req, err := http.NewRequest(t.method, addr+t.uri, body)
if err != nil {
panic(err)
}
return req
}
// Equal checks if request matches receiving test request.
func (t *testrequest) Equal(r *http.Request) error {
// Ensure methods match.
if t.method != r.Method {
return fmt.Errorf("differing request methods: t=%q r=%q", t.method, r.Method)
}
// Ensure request URIs match.
if t.uri != r.URL.RequestURI() {
return fmt.Errorf("differing request urls: t=%q r=%q", t.uri, r.URL.RequestURI())
}
// Ensure body cases match.
if requiresBody(t.method) {
// Read request into memory.
b, err := io.ReadAll(r.Body)
if err != nil {
return fmt.Errorf("error reading request body: %v", err)
}
// Compare the request bodies.
st := strings.TrimSpace(string(t.body))
sr := strings.TrimSpace(string(b))
if st != sr {
return fmt.Errorf("differing request bodies: t=%q r=%q", st, sr)
}
}
return nil
}

7
internal/transport/transport.go
View file

@ -51,10 +51,10 @@ type Transport interface {
POST(*http.Request, []byte) (*http.Response, error)
// Deliver sends an ActivityStreams object.
Deliver(ctx context.Context, b []byte, to *url.URL) error
Deliver(ctx context.Context, obj map[string]interface{}, to *url.URL) error
// BatchDeliver sends an ActivityStreams object to multiple recipients.
BatchDeliver(ctx context.Context, b []byte, recipients []*url.URL) error
BatchDeliver(ctx context.Context, obj map[string]interface{}, recipients []*url.URL) error
/*
GET functions
@ -77,7 +77,8 @@ type Transport interface {
Finger(ctx context.Context, targetUsername string, targetDomain string) ([]byte, error)
}
// transport implements the Transport interface.
// transport implements
// the Transport interface.
type transport struct {
controller *controller
pubKeyID string