I am trying to create an intermediate layer between user and tcp, with Send and Receive functions. Currently, I am trying to integrate a context, so that the Send and Receive respects a context. However, I don't know how to make them respect the context's cancellation.
Until now, I got the following.
// c.underlying is a net.Conn
func (c *tcpConn) Receive(ctx context.Context) ([]byte, error) {
if deadline, ok := ctx.Deadline(); ok {
// Set the read deadline on the underlying connection according to the
// given context. This read deadline applies to the whole function, so
// we only set it once here. On the next read-call, it will be set
// again, or will be reset in the else block, to not keep an old
// deadline.
c.underlying.SetReadDeadline(deadline)
} else {
c.underlying.SetReadDeadline(time.Time{}) // remove the read deadline
}
// perform reads with
// c.underlying.Read(myBuffer)
return frameData, nil
}
However, as far as I understand that code, this only respects a context.WithTimeout or context.WithDeadline, and not a context.WithCancel.
If possible, I would like to pass that into the connection somehow, or actually abort the reading process.
How can I do that?
Note: If possible, I would like to avoid another function that reads in another goroutine and pushed a result back on a channel, because then, when calling cancel, and I am reading 2GB over the network, that doesn't actually cancel the read, and the resources are still used. If not possible in another way however, I would like to know if there is a better way of doing that than a function with two channels, one for a []byte result and one for an error.
EDIT:
With the following code, I can respect a cancel, but it doesn't abort the read.
// apply deadline ...
result := make(chan interface{})
defer close(result)
go c.receiveAsync(result)
select {
case res := <-result:
if err, ok := res.(error); ok {
return nil, err
}
return res.([]byte), nil
case <-ctx.Done():
return nil, ErrTimeout
}
}
func (c *tcpConn) receiveAsync(result chan interface{}) {
// perform the reads and push either an error or the
// read bytes to the result channel
If the connection can be closed on cancellation, you can setup a goroutine to shutdown the connection on cancellation within the Receive method. If the connection must be reused again later, then there is no way to cancel a Read in progress.
recvDone := make(chan struct{})
defer close(recvDone)
// setup the cancellation to abort reads in process
go func() {
select {
case <-ctx.Done():
c.underlying.CloseRead()
// Close() can be used if this isn't necessarily a TCP connection
case <-recvDone:
}
}()
It will be a little more work if you want to communicate the cancelation error back, but the CloseRead will provide a clean way to stop any pending TCP Read calls.
Related
This one is a tricky issue that bugs me quite a bit.
Essentially, I wrote an integration microservice that provides data streams from Binance crypto exchange using the Go client. A client sends a start messages, starts data stream for a symbol, and at some point, sends a close message to stop the stream. My implementation looks basically like this:
func (c BinanceClient) StartDataStream(clientType bn.ClientType, symbol, interval string) error {
switch clientType {
case bn.SPOT_LIVE:
wsKlineHandler := c.handlers.klineHandler.SpotKlineHandler
wsErrHandler := c.handlers.klineHandler.ErrHandler
_, stopC, err := binance.WsKlineServe(symbol, interval, wsKlineHandler, wsErrHandler)
if err != nil {
fmt.Println(err)
return err
} else {
c.state.clientSymChanMap[clientType][symbol] = stopC
return nil
}
...
}
The clientSymChanMap stores the stopChannel in a nested hashmap so that I can retrieve the stop channel later to stop the data feed. The stop function has been implemented accordingly:
func (c BinanceClient) StopDataStream(clientType bn.ClientType, symbol string) {
//mtd := "StopDataStream: "
stopC := c.state.clientSymChanMap[clientType][symbol]
if isClosed(stopC) {
DbgPrint(" Channel is already closed. Do nothing for: " + symbol)
} else {
close(stopC)
}
// Delete channel from the map otherwise the next StopAll throws a NPE due to closing a dead channel
delete(c.state.clientSymChanMap[clientType], symbol)
return
}
To prevent panics from already closed channels, I use a check function that returns true in case the channel is already close.
func isClosed(ch <-chan struct{}) bool {
select {
case <-ch:
return true
default:
}
return false
}
Looks nice, but has a catch. When I run the code with starting data for just one symbol, it starts and closes the datafeed exactly as expected.
However, when starting multiple data feeds, then the above code somehow never closes the websocket and just keeps streaming data forever. Without the isClosed check, I get panics of trying to close a closed channel, but with the check in place, well, nothing gets closed.
When looking at the implementation of the above binance.WsKlineServe function, it's quite obvious that it just wraps a new websocket with each invocation and then returns the done & stop channel.
The documentation gives the following usage example:
wsKlineHandler := func(event *binance.WsKlineEvent) {
fmt.Println(event)
}
errHandler := func(err error) {
fmt.Println(err)
}
doneC, stopC, err := binance.WsKlineServe("LTCBTC", "1m", wsKlineHandler, errHandler)
if err != nil {
fmt.Println(err)
return
}
<-doneC
Because the doneC channel actually blocks, I removed it and thought that storing the stopC channel and then use it later to stop the datafeed would work. However, it only does so for one single instance. When multiple streams are open, this doesn't work anymore.
Any idea what that's the case and how to fix it?
Firstly, this is dangerous:
if isClosed(stopC) {
DbgPrint(" Channel is already closed. Do nothing for: " + symbol)
} else {
close(stopC) // <- can't be sure channel is still open
}
there is no guarantee that after your polling check of the channel state, that the channel will still be in that same state in the next line of code. So this code could in theory could panic if it's called concurrently.
If you want an asynchronous action to occur on the channel close - it's best to do this explicitly from its own goroutine. So you could try this:
go func() {
stopC := c.state.clientSymChanMap[clientType][symbol]
<-stopC
// stopC definitely closed now
delete(c.state.clientSymChanMap[clientType], symbol)
}()
P.S. you do need some sort of mutex on your map, since the delete is asynchronous - you need to ensure any adds to the map don't datarace with this.
P.P.S Channels are reclaimed by the GC when they go out of scope. If you are no longer reading from it - they do not need to be explicitly closed to be reclaimed by the GC.
Using channels for stopping a goroutine or closing something is very tricky. There are lots of things you can do wrong or forget to do.
context.WithCancel abstracts that complexity away, making the code more readable and maintainable.
Some code snippets:
ctx, cancel := context.WitchCancel(context.TODO())
TheThingToCancel(ctx, ...)
// Whenever you want to stop TheThingToCancel. Can be called multiple times.
cancel()
Then in a for loop you'd often have a select like this:
for {
select {
case <-ctx.Done():
return
default:
}
// do stuff
}
Here some code that is closer to your specific case of an open connection:
func TheThingToCancel(ctx context.Context) (context.CancelFunc, error) {
ctx, cancel := context.WithCancel(ctx)
conn, err := net.Dial("tcp", ":12345")
if err != nil {
cancel()
return nil, err
}
go func() {
<-ctx.Done()
_ = conn.Close()
}()
go func() {
defer func() {
_ = conn.Close()
// make sure context is always cancelled to avoid goroutine leak
cancel()
}()
var bts = make([]byte, 1024)
for {
n, err := conn.Read(bts)
if err != nil {
return
}
fmt.Println(bts[:n])
}
}()
return cancel, nil
}
It returns the cancel function to be able to close it from the outside.
Cancelling a context can be done many times over without a panic like would occur if a channel is closed multiple times. That is one advantage. Also you can derive contexts from other contexts and thereby close a lot of contexts that all stop different routines by closing a parent context. Carefully designed, this is very powerful for shutting down different routines belonging together that also need to be able to be shut down individually.
In a project the program receives data via websocket. This data needs to be processed by n algorithms. The amount of algorithms can change dynamically.
My attempt is to create some pub/sub pattern where subscriptions can be started and canceled on the fly. Turns out that this is a bit more challenging than expected.
Here's what I came up with (which is based on https://eli.thegreenplace.net/2020/pubsub-using-channels-in-go/):
package pubsub
import (
"context"
"sync"
"time"
)
type Pubsub struct {
sync.RWMutex
subs []*Subsciption
closed bool
}
func New() *Pubsub {
ps := &Pubsub{}
ps.subs = []*Subsciption{}
return ps
}
func (ps *Pubsub) Publish(msg interface{}) {
ps.RLock()
defer ps.RUnlock()
if ps.closed {
return
}
for _, sub := range ps.subs {
// ISSUE1: These goroutines apparently do not exit properly...
go func(ch chan interface{}) {
ch <- msg
}(sub.Data)
}
}
func (ps *Pubsub) Subscribe() (context.Context, *Subsciption, error) {
ps.Lock()
defer ps.Unlock()
// prep channel
ctx, cancel := context.WithCancel(context.Background())
sub := &Subsciption{
Data: make(chan interface{}, 1),
cancel: cancel,
ps: ps,
}
// prep subsciption
ps.subs = append(ps.subs, sub)
return ctx, sub, nil
}
func (ps *Pubsub) unsubscribe(s *Subsciption) bool {
ps.Lock()
defer ps.Unlock()
found := false
index := 0
for i, sub := range ps.subs {
if sub == s {
index = i
found = true
}
}
if found {
s.cancel()
ps.subs[index] = ps.subs[len(ps.subs)-1]
ps.subs = ps.subs[:len(ps.subs)-1]
// ISSUE2: close the channel async with a delay to ensure
// nothing will be written to the channel anymore
// via a pending goroutine from Publish()
go func(ch chan interface{}) {
time.Sleep(500 * time.Millisecond)
close(ch)
}(s.Data)
}
return found
}
func (ps *Pubsub) Close() {
ps.Lock()
defer ps.Unlock()
if !ps.closed {
ps.closed = true
for _, sub := range ps.subs {
sub.cancel()
// ISSUE2: close the channel async with a delay to ensure
// nothing will be written to the channel anymore
// via a pending goroutine from Publish()
go func(ch chan interface{}) {
time.Sleep(500 * time.Millisecond)
close(ch)
}(sub.Data)
}
}
}
type Subsciption struct {
Data chan interface{}
cancel func()
ps *Pubsub
}
func (s *Subsciption) Unsubscribe() {
s.ps.unsubscribe(s)
}
As mentioned in the comments there are (at least) two issues with this:
ISSUE1:
After a while of running in implementation of this I get a few errors of this kind:
goroutine 120624 [runnable]:
bm/internal/pubsub.(*Pubsub).Publish.func1(0x8586c0, 0xc00b44e880, 0xc008617740)
/home/X/Projects/bm/internal/pubsub/pubsub.go:30
created by bookmaker/internal/pubsub.(*Pubsub).Publish
/home/X/Projects/bm/internal/pubsub/pubsub.go:30 +0xbb
Without really understanding this it appears to me that the goroutines created in Publish() do accumulate/leak. Is this correct and what am I doing wrong here?
ISSUE2:
When I end a subscription via Unsubscribe() it occurs that Publish() tried to write to a closed channel and panics. To mitigate this I have created a goroutine to close the channel with a delay. This feel really off-best-practice but I was not able to find a proper solution to this. What would be a deterministic way to do this?
Heres a little playground for you to test with: https://play.golang.org/p/K-L8vLjt7_9
Before diving into your solution and its issues, let me recommend again another Broker approach presented in this answer: How to broadcast message using channel
Now on to your solution.
Whenever you launch a goroutine, always think of how it will end and make sure it does if the goroutine is not ought to run for the lifetime of your app.
// ISSUE1: These goroutines apparently do not exit properly...
go func(ch chan interface{}) {
ch <- msg
}(sub.Data)
This goroutine tries to send a value on ch. This may be a blocking operation: it will block if ch's buffer is full and there is no ready receiver on ch. This is out of the control of the launched goroutine, and also out of the control of the pubsub package. This may be fine in some cases, but this already places a burden on the users of the package. Try to avoid these. Try to create APIs that are easy to use and hard to misuse.
Also, launching a goroutine just to send a value on a channel is a waste of resources (goroutines are cheap and light, but you shouldn't spam them whenever you can).
You do it because you don't want to get blocked. To avoid blocking, you may use a buffered channel with a "reasonable" high buffer. Yes, this doesn't solve the blocking issue, in only helps with "slow" clients receiving from the channel.
To "truly" avoid blocking without launching a goroutine, you may use non-blocking send:
select {
case ch <- msg:
default:
// ch's buffer is full, we cannot deliver now
}
If send on ch can proceed, it will happen. If not, the default branch is chosen immediately. You have to decide what to do then. Is it acceptable to "lose" a message? Is it acceptable to wait for some time until "giving up"? Or is it acceptable to launch a goroutine to do this (but then you'll be back at what we're trying to fix here)? Or is it acceptable to get blocked until the client can receive from the channel...
Choosing a reasonable high buffer, if you encounter a situation when it still gets full, it may be acceptable to block until the client can advance and receive from the message. If it can't, then your whole app might be in an unacceptable state, and it might be acceptable to "hang" or "crash".
// ISSUE2: close the channel async with a delay to ensure
// nothing will be written to the channel anymore
// via a pending goroutine from Publish()
go func(ch chan interface{}) {
time.Sleep(500 * time.Millisecond)
close(ch)
}(s.Data)
Closing a channel is a signal to the receiver(s) that no more values will be sent on the channel. So always it should be the sender's job (and responsibility) to close the channel. Launching a goroutine to close the channel, you "hand" that job and responsibility to another "entity" (a goroutine) that will not be synchronized to the sender. This may easily end up in a panic (sending on a closed channel is a runtime panic, for other axioms see How does a non initialized channel behave?). Don't do that.
Yes, this was necessary because you launched goroutines to send. If you don't do that, then you may close "in-place", without launching a goroutine, because then the sender and closer will be the same entity: the Pubsub itself, whose sending and closing operations are protected by a mutex. So solving the first issue solves the second naturally.
In general if there are multiple senders for a channel, then closing the channel must be coordinated. There must be a single entity (often not any of the senders) that waits for all senders to finish, practically using a sync.WaitGroup, and then that single entity can close the channel, safely. See Closing channel of unknown length.
I am using the Sarama library written in Go to read from an error channel when I produce a message. The overall code looks like this which is enclosed within a function:
producer.AsyncProducer.Input() <- &sarama.ProducerMessage{Topic: topic, Key: nil, Value: sarama.ByteEncoder(message)}
go func() {
for err := range saramaProducer.Errors() {
if producer.callbacks.OnError != nil {
producer.callbacks.OnError(err)
}
}
}()
As my understanding of go routines goes, my go routine would keep iterating over the Errors() channel until it receives one. Is there a way to make it stop listening for errors once my function is done executing?
You can use another channel and a select to make the loop return.
var quit chan struct{}
go func() {
for {
select {
case err:=<-saramaProducer.Errors():
//handle errors
case <-quit:
return
}
}
}
defer func() { quit<-struct{}{} }()
The orignal for ... range loop does not iterate the channel until it gets one. Instead, it blocks until it gets an error, handle it, and waits for a new error again, until the channel close or the main returns.
There is a little problem about the above code, that owhen both quit and error channel is ready, the select picks one randomly, thus may cause a single error loss. If this is worth handling, just put another switch with default to get that error and then return.
is there a possible in go to defer a go routine, or a way to achieve the desired behaviour? The following background: I am pooling connections to a database in a channel. Basically in a handler I call
session, err := getSessionFromQueue()
// ...
// serving content to my client
// ...
go queueSession(session)
What I really would like to do is:
session, err := getSessionFromQueue()
defer go queueSession(session)
// ...
// serving content to my client
// ...
to avoid that my handler is hanging/crashing at some point and the session is not properly returned to the queue. The reason I want to run it as a go routine is that queueSession is potentially blocking for 1 second (in case the queue is full I wait for one second before I completely close the session).
Update
#abhink got me on the right track there. I solved the problem by putting the call to a goroutine in queueBackend.
func queueSession(mongoServer *Server) {
go func(mongoServer *Server) {
select {
case mongoQueue <- mongoServer:
// mongoServer stored in queue, done.
case <- time.After(1 * time.Second):
// cannot queue for whatever reason after 1 second
// abort
mongoServer.Close()
}
}(mongoServer)
}
Now I can simply call
defer queueSession(session)
and it is run as a goroutine.
There is no way to directly defer a goroutine. You can try something like this:
session, err := getSessionFromQueue()
defer func() {
go queueSession(session)
}()
It doesn't seem possible to have two way communication via channels with a goroutine which is performing file operations, unless you block the channel communication on the file operations. How can I work around the limits this imposes?
Another way to phrase this question...
If I have a loop similar to the following running in a goroutine, how can I tell it to close the connection and exit without blocking on the next Read?
func readLines(response *http.Response, outgoing chan string) error {
defer response.Body.Close()
reader := bufio.NewReader(response.Body)
for {
line, err := reader.ReadString('\n')
if err != nil {
return err
}
outgoing <- line
}
}
It's not possible for it to read from a channel that tells it when to close down because it's blocking on the network reads (in my case, that can take hours).
It doesn't appear to be safe to simply call Close() from outside the goroutine, since the Read/Close methods don't appear to be fully thread safe.
I could simply put a lock around references to response.Body that used inside/outside the routine, but would cause the external code to block until a pending read completes, and I specifically want to be able to interrupt an in-progress read.
To address this scenario, several io.ReadCloser implementations in the standard library support concurrent calls to Read and Close where Close interrupts an active Read.
The response body reader created by net/http Transport is one of those implementations. It is safe to concurrently call Read and Close on the response body.
You can also interrupt an active Read on the response body by calling the Transport CancelRequest method.
Here's how implement cancel using close on the body:
func readLines(response *http.Response, outgoing chan string, done chan struct{}) error {
cancel := make(chan struct{})
go func() {
select {
case <-done:
response.Body.Close()
case <-cancel:
return
}()
defer response.Body.Close()
defer close(cancel) // ensure that goroutine exits
reader := bufio.NewReader(response.Body)
for {
line, err := reader.ReadString('\n')
if err != nil {
return err
}
outgoing <- line
}
}
Calling close(done) from another goroutine will cancel reads on the body.