Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 3 years ago.
Improve this question
Many languages have their own high-level non-blocking HTTP client, for example, python's aiohttp. Namely, they send out HTTP requests; do not wait for response; When response arrives they make some kind of callbacks.
My questions are
is there a Go package for that?
or we just create a goroutine in which we use normal HTTP clients?
which way is better?
Other languages have such features because when they block waiting for request they block the thread they are using. This is the case for Java, Python or NodeJS. Therefore to make them useful, the developers needed to implement such long-standing blocking operations with callbacks. The root cause of that is the usage of the C library beneath that blocks threads on input-output operations.
Go does not use C library (only in some cases, but it can be turned off) and makes system calls by itself. While doing this the thread that executes current goroutine parks it and executes another goroutine. Therefore you can have enormous number of blocked goroutines without running out of threads. Goroutines are cheap with regard to memory, threads are operating system entities.
In Go using goroutines is better. There is no need for creating asynchronous client because of the above.
For comparison in Java you would quickly end up with multiple threads. The next step would be pooling them as they are costly. Pooling means limiting the concurrency.
As others have stated, goroutines are the way to go (pun intended).
Minimal Example:
type nonBlocking struct {
Response *http.Response
Error error
}
const numRequests = 2
func main() {
nb := make(chan nonBlocking, numRequests)
wg := &sync.WaitGroup{}
for i := 0; i < numRequests; i++ {
wg.Add(1)
go Request(nb)
}
go HandleResponse(nb, wg)
wg.Wait()
}
func Request(nb chan nonBlocking) {
resp, err := http.Get("http://example.com")
nb <- nonBlocking{
Response: resp,
Error: err,
}
}
func HandleResponse(nb chan nonBlocking, wg *sync.WaitGroup) {
for get := range nb {
if get.Error != nil {
log.Println(get.Error)
} else {
log.Println(get.Response.Status)
}
wg.Done()
}
}
Yip, built into the standard library, just not usable by a simple function call out of the box.
Take this example
package main
import (
"flag"
"log"
"net/http"
"sync"
"time"
)
var url string
var timeout time.Duration
func init() {
flag.StringVar(&url, "url", "http://www.stackoverflow.com", "url to GET")
flag.DurationVar(&timeout, "timeout", 5*time.Second, "timeout for the GET operation")
}
func main() {
flag.Parse()
// We use the channel as our means to
// hand the response over
rc := make(chan *http.Response)
// We need a waitgroup because all goroutines exit when main exits
var wg sync.WaitGroup
// We are spinning up an async request
// Increment the counter for our WaitGroup.
// What we are basically doing here is to tell the WaitGroup
// "Hey, there is one more task you have to wait for!"
wg.Add(1)
go func() {
// Notify the WaitGroup that one task is done as soon
// as we exit the goroutine.
defer wg.Done()
log.Printf("Doing GET request on \"%s\"", url)
resp, err := http.Get(url)
if err != nil {
log.Printf("GET for %s: %s", url, err)
}
// We send the reponse downstream
rc <- resp
// Now, the goroutine exits, the defered call to wg.Done()
// is executed.
}()
// And here we do our async processing.
// Note that you could have done the processing in the first goroutine
// as well, since http.Get would be a blocking operation and any subsequent
// code in the goroutine would have been excuted only after the Get returned.
// However, I put te processing into its own goroutine for demonstration purposes.
wg.Add(1)
go func() {
// As above
defer wg.Done()
log.Println("Doing something else")
// Setting up a timer for a timeout.
// Note that this could be done using a request with a context, as well.
to := time.NewTimer(timeout).C
select {
case <-to:
log.Println("Timeout reached")
// Exiting the goroutine, the deferred call to wg.Done is executed
return
case r := <-rc:
if r == nil {
log.Printf("Got no useful response from GETting \"%s\"", url)
// Exiting the goroutine, the deferred call to wg.Done is executed
return
}
log.Printf("Got response with status code %d (%s)", r.StatusCode, r.Status)
log.Printf("Now I can do something useful with the response")
}
}()
// Now we have set up all of our tasks,
// we are waiting until all of them are done...
wg.Wait()
log.Println("All tasks done, exiting")
}
If you look at this closely, we have all building blocks to make GETting an URL and processing the response async. We can start to abstract this a bit:
package main
import (
"flag"
"log"
"net/http"
"time"
)
var url string
var timeout time.Duration
func init() {
flag.StringVar(&url, "url", "http://www.stackoverflow.com", "url to GET")
flag.DurationVar(&timeout, "timeout", 5*time.Second, "timeout for the GET operation")
}
type callbackFunc func(*http.Response, error) error
func getWithCallBack(u string, callback callbackFunc) chan error {
// We create a channel which we can use to notify the caller of the
// result of the callback.
c := make(chan error)
go func() {
c <- callback(http.Get(u))
}()
return c
}
func main() {
flag.Parse()
c := getWithCallBack(url, func(resp *http.Response, err error) error {
if err != nil {
// Doing something useful with the err.
// Add additional cases as needed.
switch err {
case http.ErrNotSupported:
log.Printf("GET not supported for \"%s\"", url)
}
return err
}
log.Printf("GETting \"%s\": Got response with status code %d (%s)", url, resp.StatusCode, resp.Status)
return nil
})
if err := <-c; err != nil {
log.Printf("Error GETting \"%s\": %s", url, err)
}
log.Println("All tasks done, exiting")
}
And there you Go (pun intended): Async processing of GET requests.
Related
I have written an API that makes DB calls and does some business logic. I am invoking a goroutine that must perform some operation in the background.
Since the API call should not wait for this background task to finish, I am returning 200 OK immediately after calling the goroutine (let us assume the background task will never give any error.)
I read that goroutine will be terminated once the goroutine has completed its task.
Is this fire and forget way safe from a goroutine leak?
Are goroutines terminated and cleaned up once they perform the job?
func DefaultHandler(w http.ResponseWriter, r *http.Request) {
// Some DB calls
// Some business logics
go func() {
// some Task taking 5 sec
}()
w.WriteHeader(http.StatusOK)
}
I would recommend always having your goroutines under control to avoid memory and system exhaustion.
If you are receiving a spike of requests and you start spawning goroutines without control, probably the system will go down soon or later.
In those cases where you need to return an immediate 200Ok the best approach is to create a message queue, so the server only needs to create a job in the queue and return the ok and forget. The rest will be handled by a consumer asynchronously.
Producer (HTTP server) >>> Queue >>> Consumer
Normally, the queue is an external resource (RabbitMQ, AWS SQS...) but for teaching purposes, you can achieve the same effect using a channel as a message queue.
In the example you'll see how we create a channel to communicate 2 processes.
Then we start the worker process that will read from the channel and later the server with a handler that will write to the channel.
Try to play with the buffer size and job time while sending curl requests.
package main
import (
"fmt"
"log"
"net/http"
"time"
)
/*
$ go run .
curl "http://localhost:8080?user_id=1"
curl "http://localhost:8080?user_id=2"
curl "http://localhost:8080?user_id=3"
curl "http://localhost:8080?user_id=....."
*/
func main() {
queueSize := 10
// This is our queue, a channel to communicate processes. Queue size is the number of items that can be stored in the channel
myJobQueue := make(chan string, queueSize) // Search for 'buffered channels'
// Starts a worker that will read continuously from our queue
go myBackgroundWorker(myJobQueue)
// We start our server with a handler that is receiving the queue to write to it
if err := http.ListenAndServe("localhost:8080", myAsyncHandler(myJobQueue)); err != nil {
panic(err)
}
}
func myAsyncHandler(myJobQueue chan<- string) http.HandlerFunc {
return func(rw http.ResponseWriter, r *http.Request) {
// We check that in the query string we have a 'user_id' query param
if userID := r.URL.Query().Get("user_id"); userID != "" {
select {
case myJobQueue <- userID: // We try to put the item into the queue ...
rw.WriteHeader(http.StatusOK)
rw.Write([]byte(fmt.Sprintf("queuing user process: %s", userID)))
default: // If we cannot write to the queue it's because is full!
rw.WriteHeader(http.StatusInternalServerError)
rw.Write([]byte(`our internal queue is full, try it later`))
}
return
}
rw.WriteHeader(http.StatusBadRequest)
rw.Write([]byte(`missing 'user_id' in query params`))
}
}
func myBackgroundWorker(myJobQueue <-chan string) {
const (
jobDuration = 10 * time.Second // simulation of a heavy background process
)
// We continuosly read from our queue and process the queue 1 by 1.
// In this loop we could spawn more goroutines in a controlled way to paralelize work and increase the read throughput, but i don't want to overcomplicate the example.
for userID := range myJobQueue {
// rate limiter here ...
// go func(u string){
log.Printf("processing user: %s, started", userID)
time.Sleep(jobDuration)
log.Printf("processing user: %s, finisehd", userID)
// }(userID)
}
}
There is no "goroutine cleaning" you have to handle, you just launch goroutines and they'll be cleaned when the function launched as a goroutine returns. Quoting from Spec: Go statements:
When the function terminates, its goroutine also terminates. If the function has any return values, they are discarded when the function completes.
So what you do is fine. Note however that your launched goroutine cannot use or assume anything about the request (r) and response writer (w), you may only use them before you return from the handler.
Also note that you don't have to write http.StatusOK, if you return from the handler without writing anything, that's assumed to be a success and HTTP 200 OK will be sent back automatically.
See related / possible duplicate: Webhook process run on another goroutine
#icza is absolutely right there is no "goroutine cleaning" you can use a webhook or a background job like gocraft. The only way I can think of using your solution is to use the sync package for learning purposes.
func DefaultHandler(w http.ResponseWriter, r *http.Request) {
// Some DB calls
// Some business logics
var wg sync.WaitGroup
wg.Add(1)
go func() {
defer wg.Done()
// some Task taking 5 sec
}()
w.WriteHeader(http.StatusOK)
wg.wait()
}
you can wait for a goroutine to finish using &sync.WaitGroup:
// BusyTask
func BusyTask(t interface{}) error {
var wg = &sync.WaitGroup{}
wg.Add(1)
go func() {
// busy doing stuff
time.Sleep(5 * time.Second)
wg.Done()
}()
wg.Wait() // wait for goroutine
return nil
}
// this will wait 5 second till goroutune finish
func main() {
fmt.Println("hello")
BusyTask("some task...")
fmt.Println("done")
}
Other way is to attach a context.Context to goroutine and time it out.
//
func BusyTaskContext(ctx context.Context, t string) error {
done := make(chan struct{}, 1)
//
go func() {
// time sleep 5 second
time.Sleep(5 * time.Second)
// do tasks and signle done
done <- struct{}{}
close(done)
}()
//
select {
case <-ctx.Done():
return errors.New("timeout")
case <-done:
return nil
}
}
//
func main() {
fmt.Println("hello")
ctx, cancel := context.WithTimeout(context.TODO(), 2*time.Second)
defer cancel()
if err := BusyTaskContext(ctx, "some task..."); err != nil {
fmt.Println(err)
return
}
fmt.Println("done")
}
I have the following code in Go using the semaphore library just as an example:
package main
import (
"fmt"
"context"
"time"
"golang.org/x/sync/semaphore"
)
// This protects the lockedVar variable
var lock *semaphore.Weighted
// Only one go routine should be able to access this at once
var lockedVar string
func acquireLock() {
err := lock.Acquire(context.TODO(), 1)
if err != nil {
panic(err)
}
}
func releaseLock() {
lock.Release(1)
}
func useLockedVar() {
acquireLock()
fmt.Printf("lockedVar used: %s\n", lockedVar)
releaseLock()
}
func causeDeadLock() {
acquireLock()
// calling this from a function that's already
// locked the lockedVar should cause a deadlock.
useLockedVar()
releaseLock()
}
func main() {
lock = semaphore.NewWeighted(1)
lockedVar = "this is the locked var"
// this is only on a separate goroutine so that the standard
// go "deadlock" message doesn't print out.
go causeDeadLock()
// Keep the primary goroutine active.
for true {
time.Sleep(time.Second)
}
}
Is there a way to get the acquireLock() function call to print a message after a timeout indicating that there is a potential deadlock but without unblocking the call? I would want the deadlock to persist, but a log message to be written in the event that a timeout is reached. So a TryAcquire isn't exactly what I want.
An example of what I want in psuedo code:
afterFiveSeconds := func() {
fmt.Printf("there is a potential deadlock\n")
}
lock.Acquire(context.TODO(), 1, afterFiveSeconds)
The lock.Acquire call in this example would call the afterFiveSeconds callback if the Acquire call blocked for more than 5 seconds, but it would not unblock the caller. It would continue to block.
I think I've found a solution to my problem.
func acquireLock() {
timeoutChan := make(chan bool)
go func() {
select {
case <-time.After(time.Second * time.Duration(5)):
fmt.Printf("potential deadlock while acquiring semaphore\n")
case <-timeoutChan:
break
}
}()
err := lock.Acquire(context.TODO(), 1)
close(timeoutChan)
if err != nil {
panic(err)
}
}
I am fairly new to golang and its concurrency principles. My use-case involves performing multiple http requests(for a single entity), on batch of entities. If any of the http request fails for an entity, I need to stop all parallel http requests for it. Also, I have to manage counts of entities failed with errors. I am trying to implement errorgroup inside entities goroutines, such that if any http request fails for a single entity the errorgroup terminates and return error to its parent goroutine. But I am not sure how to maintain count of errors.
func main(entity[] string) {
errorC := make(chan string) // channel to insert failed entity
var wg sync.WaitGroup
for _, link := range entity {
wg.Add(1)
// Spawn errorgroup here. errorgroup_spawn
}
go func() {
wg.Wait()
close(errorC)
}()
for msg := range errorC {
// here storing error entityIds somewhere.
}
}
and errorgroup like this
func errorgroup_spawn(ctx context.Context, errorC chan string, wg *sync.WaitGroup) { // and other params
defer (*wg).Done()
goRoutineCollection, ctxx := errgroup.WithContext(ctx)
results := make(chan *result)
goRoutineCollection.Go(func() error {
// http calls for single entity
// if error occurs, push it in errorC, and return Error.
return nil
})
go func() {
goRoutineCollection.Wait()
close(result)
}()
return goRoutineCollection.Wait()
}
PS: I was also thinking to apply nested errorgroups, but can't think to maintain error counts, while running other errorgroups
Can anyone guide me, is this a correct approach to handle such real world scenarios?
One way to keep track of errors is to use a status struct to keep track of which error came from where:
type Status struct {
Entity string
Err error
}
...
errorC := make(chan Status)
// Spawn error groups with name of the entity, and when error happens, push Status{Entity:entityName,Err:err} to the chanel
You can then read all errors from the error channel and figure out what failed why.
Another option is not to use errorgroups at all. This makes things more explicit, but whether it is better or not is debatable:
// Keep entity statuses
statuses:=make([]Status,len(entity))
for i, link := range entity {
statuses[i].Entity=link
wg.Add(1)
go func(i index) {
defer wg.Done()
ctx, cancel:=context.WithCancel(context.Background())
defer cancel()
// Error collector
status:=make(chan error)
defer close(status)
go func() {
for st:=range status {
if st!=nil {
cancel() // Stop all calls
// store first error
if statuses[i].Err==nil {
statuses[i].Err=st
}
}
}
}()
innerWg:=sync.WaitGroup{}
innerWg.Add(1)
go func() {
defer innerWg.Done()
status<- makeHttpCall(ctx)
}()
innerWg.Add(1)
go func() {
defer innerWg.Done()
status<- makeHttpCall(ctx)
}()
...
innerWg.Wait()
}(i)
}
When everything is done, statuses will contain all entities and corresponding statuses.
I am attempting to create a poller in Go that spins up and every 24 hours executes a function.
I want to also be able to stop the polling, I'm attempting to do this by having a done channel and passing down an empty struct to stop the for loop.
In my tests, the for just loops infinitely and I can't seem to stop it, am I using the done channel incorrectly? The ticker case works as expected.
Poller struct {
HandlerFunc HandlerFunc
interval *time.Ticker
done chan struct{}
}
func (p *Poller) Start() error {
for {
select {
case <-p.interval.C:
err := p.HandlerFunc()
if err != nil {
return err
}
case <-p.done:
return nil
}
}
}
func (p *Poller) Stop() {
p.done <- struct{}{}
}
Here is the test that's exeuting the code and causing the infinite loop.
poller := poller.NewPoller(
testHandlerFunc,
time.NewTicker(1*time.Millisecond),
)
err := poller.Start()
assert.Error(t, err)
poller.Stop()
Seems like problem is in your use case, you calling poller.Start() in blocking maner, so poller.Stop() is never called. It's common, in go projects to call goroutine inside of Start/Run methods, so, in poller.Start(), i would do something like that:
func (p *Poller) Start() <-chan error {
errc := make(chan error, 1 )
go func() {
defer close(errc)
for {
select {
case <-p.interval.C:
err := p.HandlerFunc()
if err != nil {
errc <- err
return
}
case <-p.done:
return
}
}
}
return errc
}
Also, there's no need to send empty struct to done channel. Closing channel like close(p.done) is more idiomatic for go.
There is no explicit way in Go to broadcast an event to go routines for something like cancellation. Instead its idiomatic to create a channel that when closed signifies a message such as cancelling any work it has to do. Something like this is a viable pattern:
var done = make(chan struct{})
func cancelled() bool {
select {
case <-done:
return true
default:
return false
}
}
Go-routines can call cancelled to poll for a cancellation.
Then your main loop can respond to such an event but make sure you drain any channels that might cause go-routines to block.
for {
select {
case <-done:
// Drain whatever channels you need to.
for range someChannel { }
return
//.. Other cases
}
}
I'm trying to stop all clients connected to a stream server from server side.
Actually I'm using GracefulStop method to handle it gracefully.
I am waiting for os.Interrupt signal on a channel to perform a graceful stop for gRPC. but it gets stuck on server.GracefulStop() when the client is connected.
func (s *Service) Subscribe(_ *empty.Empty, srv clientapi.ClientApi_SubscribeServer) error {
ctx := srv.Context()
updateCh := make(chan *clientapi.Update, 100)
stopCh := make(chan bool)
defer func() {
stopCh<-true
close(updateCh)
}
go func() {
ticker := time.NewTicker(1 * time.Second)
defer func() {
ticker.Stop()
close(stopCh)
}
for {
select {
case <-stopCh:
return
case <-ticker.C:
updateCh<- &clientapi.Update{Name: "notification": Payload: "sample notification every 1 second"}
}
}
}()
for {
select {
case <-ctx.Done():
return ctx.Err()
case notif := <-updateCh:
err := srv.Send(notif)
if err == io.EOF {
return nil
}
if err != nil {
s.logger.Named("Subscribe").Error("error", zap.Error(err))
continue
}
}
}
}
I expected the context in method ctx.Done() could handle it and break the for loop.
How to close all response streams like this one?
Create a global context for your gRPC service. So walking through the various pieces:
Each gRPC service request would use this context (along with the client context) to fulfill that request
os.Interrupt handler would cancel the global context; thus canceling any currently running requests
finally issue server.GracefulStop() - which should wait for all the active gRPC calls to finish up (if they haven't see the cancelation immediately)
So for example, when setting up the gRPC service:
pctx := context.Background()
globalCtx, globalCancel := context.WithCancel(pctx)
mysrv := MyService{
gctx: globalCtx
}
s := grpc.NewServer()
pb.RegisterMyService(s, mysrv)
os.Interrupt handler initiates and waits for shutdown:
globalCancel()
server.GracefulStop()
gRPC methods:
func(s *MyService) SomeRpcMethod(ctx context.Context, req *pb.Request) error {
// merge client and server contexts into one `mctx`
// (client context will cancel if client disconnects)
// (server context will cancel if service Ctrl-C'ed)
mctx, mcancel := mergeContext(ctx, s.gctx)
defer mcancel() // so we don't leak, if neither client or server context cancels
// RPC WORK GOES HERE
// RPC WORK GOES HERE
// RPC WORK GOES HERE
// pass mctx to any blocking calls:
// - http REST calls
// - SQL queries etc.
// - or if running a long loop; status check the context occasionally like so:
// Example long request (10s)
for i:=0; i<10*1000; i++ {
time.Sleep(1*time.Milliscond)
// poll merged context
select {
case <-mctx.Done():
return fmt.Errorf("request canceled: %s", mctx.Err())
default:
}
}
}
And:
func mergeContext(a, b context.Context) (context.Context, context.CancelFunc) {
mctx, mcancel := context.WithCancel(a) // will cancel if `a` cancels
go func() {
select {
case <-mctx.Done(): // don't leak go-routine on clean gRPC run
case <-b.Done():
mcancel() // b canceled, so cancel mctx
}
}()
return mctx, mcancel
}
Typically clients need to assume that RPCs can terminate (e.g. due to connection errors or server power failure) at any moment. So what we do is GracefulStop, sleep for a short time period to allow in-flight RPCs an opportunity to complete naturally, then hard-Stop the server. If you do need to use this termination signal to end your RPCs, then the answer by #colminator is probably the best choice. But this situation should be unusual, and you may want to spend some time analyzing your design if you do find it is necessary to manually end streaming RPCs at server shutdown.