I have a simple server running in node.js using connect:
var server = require('connect').createServer();
//actions...
server.listen(3000);
In my code I have actual handlers, but thats the basic idea. The problem I keep getting is
EADDRINUSE, Address already in use
I receive this error when running my application again after it previously crashed or errors. Since I am not opening a new instance of terminal I close out the process with ctr + z.
I am fairly certain all I have to do is close out the server or connection. I tried calling server.close() in process.on('exit', ...); with no luck.
First, you would want to know which process is using port 3000
sudo lsof -i :3000
this will list all PID listening on this port, once you have the PID you can terminate it with the following:
kill -9 <PID>
where you replace <PID> by the process ID, or the list of process IDs, the previous command output.
You can also go the command line route:
ps aux | grep node
to get the process ids.
Then:
kill -9 PID
Doing the -9 on kill sends a SIGKILL (instead of a SIGTERM).
SIGTERM has been ignored by node for me sometimes.
I hit this on my laptop running win8. this worked.
Run cmd.exe as 'Administrator':
C:\Windows\System32>taskkill /F /IM node.exe
SUCCESS: The process "node.exe" with PID 11008 has been terminated.
process.on('exit', ..) isn't called if the process crashes or is killed. It is only called when the event loop ends, and since server.close() sort of ends the event loop (it still has to wait for currently running stacks here and there) it makes no sense to put that inside the exit event...
On crash, do process.on('uncaughtException', ..) and on kill do process.on('SIGTERM', ..)
That being said, SIGTERM (default kill signal) lets the app clean up, while SIGKILL (immediate termination) won't let the app do anything.
Check the PID i.e. id of process running on port 3000 with below command :
lsof -i tcp:3000
It would output something like following:
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
node 5805 xyz 12u IPv6 63135 0t0 TCP *:3000 (LISTEN)
Now kill the process using :
kill -9 5805
For macOS Monterey(12.0):
Apple introduced some changes for AirPlay on macOS Monterey. Now, it uses 5000 and 7000 ports. If you are using these ports in your project, you need to disable this feature.
System Preferences > Sharing > untick AirPlay Receiver
For macOS Ventura(13.0) and above users:
System Settings > General > disable AirPlay Receiver
I found this solution, try it
Give permission use sudo
sudo pkill node
I usually use
npx kill-port 3000
or on my mac.
killall node
Rewriting #Gerard 's comment in my answer:
Try pkill nodejs or pkill node if on UNIX-like OS.
This will kill the process running the node server running on any port.
Worked for me.
Linux
Run ps and determine the PID of your node process.
Then, run sudo kill PID
Windows
Use tasklist to display the list of running processes:
tasklist /O
Then, kill the node process like so (using the PID obtained from the tasklist command):
taskkill /pid PID
Here is a one liner (replace 3000 with a port or a config variable):
kill $(lsof -t -i:3000)
For windows open Task Manager and find node.exe processes. Kill all of them with End Task.
I was getting this error once and took many of the approaches here.
My issues was that I had two app.listen(3000); calls in the same app.js script. The first app.listen() succeeded where the second threw the error.
Another useful command I came across that helped me debug was sudo fuser -k 3000/tcp which will kill any rogue processes you might have started (some processes may restart, e.g. if run with forever.js, but it was useful for me).
For Visual Studio Noobs like me
You may be running the process in other terminals!
After closing the terminal in Visual Studio, the terminal just disappears.
I manually created a new one thinking that the previous one was destroyed. In reality, every time I was clicking on New Terminal I was actually creating a new one on top of the previous ones.
So I located the first terminal and... Voila, I was running the server there.
Windows by Cmd
1/2. search => write cmd => open node.js command prompt
2/2. Run windows command: taskkill
Ends one or more tasks or processes.
taskkill /f /im node.exe
/f - force ended
/im - Specifies the image name of the process to be terminated.
node.exe - executable file
Windows - Mannualy by Task Manager
This command is the same as going to Task Manager under the details tab & select node tasks (Tidy in my opinion).
And end task
Visual studio
Sometimes there is more than one terminal/task (client/server and so on).
Select and close by ctrl + c.
You may run into scenarios where even killing the thread or process won't actually terminate the app (this happens for me on Linux and Windows every once in a while). Sometimes you might already have an instance running that you didn't close.
As a result of those kinds of circumstances, I prefer to add to my package.json:
"scripts": {
"stop-win": "Taskkill /IM node.exe /F",
"stop-linux": "killall node"
},
I can then call them using:
npm run stop-win
npm run stop-Linux
You can get fancier and make those BIN commands with an argument flag if you want. You can also add those as commands to be executed within a try-catch clause.
FYI, you can kill the process in one command sudo fuser -k 3000/tcp. This can be done for all other ports like 8000, 8080 or 9000 which are commonly used for development.
ps aux | grep node
kill -9 [PID] (provided by above command)
Description:
ps will give the process status, aux provide the list of a: all users processes, u: user own processes, x: all other processes not attached to terminal.
pipe symbol: | will pass the result of ps aux to manipulate further.
grep will search the string provided(node in our case) from the list provided by ps aux.
First find out what is running using:
sudo lsof -nP -i4TCP:3000 | grep LISTEN
You will get something like:
php-fpm 110 root 6u IPv4 0x110e2ba1cc64b26d 0t0 TCP 127.0.0.1:3000 (LISTEN)
php-fpm 274 _www 0u IPv4 0x110e2ba1cc64b26d 0t0 TCP 127.0.0.1:3000 (LISTEN)
php-fpm 275 _www 0u IPv4 0x110e2ba1cc64b26d 0t0 TCP 127.0.0.1:3000 (LISTEN)
Then you can kill the process as followed:
sudo kill 110
Then you will be able to run without getting the listen EADDRINUSE :::3000 errors
Really simply for all OS's ..
npx kill-port 3000
Although your problem is as mentioned above you need to catch the different ways node can exit for example
process.on('uncaughtException', (err, origin) => {
console.log(err);
});
// insert other handlers.
bash$ sudo netstat -ltnp | grep -w ':3000'
- tcp6 0 0 :::4000 :::* LISTEN 31157/node
bash$ kill 31157
PowerShell users:
Taskkill /IM node.exe /F
UI solution For Windows users: I found that the top answers did not work for me, they seemed to be commands for Mac or Linux users. I found a simple solution that didn't require any commands to remember: open Task Manager (ctrl+shift+esc). Look at background processes running. Find anything Node.js and end the task.
After I did this the issue went away for me. As stated in other answers it's background processes that are still running because an error was previously encountered and the regular exit/clean up functions didn't get called, so one way to kill them is to find the process in Task Manager and kill it there. If you ran the process from a terminal/powerShell you can usually use ctrl+c to kill it.
Task Manager (ctrl+alt+del) ->
Processes tab ->
select the "node.exe" process and hit "End Process"
Just in case check if you have added this line multiple times by mistake
app.listen(3000, function() {
console.log('listening on 3000')
});
The above code is for express but just check if you are trying to use the same port twice in your code.
In windows users: open task manager and end task the nodejs.exe file, It works fine.
On Windows, I was getting the following error:
EADDRINUSE: address already in use :::8081.
Followed these steps:
Opened CMD as Admin
Ran the folowing
command netstat -ano|findstr "PID :8081"
got the following processes:
killed it via:
taskkill /pid 43144 /f
On MAC you can do like this:
raghavkhunger#MacBook-Air ~ % lsof -i tcp:8081
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
node 23722 username 24u IPv6 0xeed16d7ccfdd347 0t0 TCP *:sunproxyadmin (LISTEN)
username#MacBook-Air ~ % kill -9 23722
With due respect to all the answers in the form, I would like to add a point.
I found that when I terminate a node app on error using Ctrl + Z, the very next time when I try to open it got the same error EADDRINUSE.
When I use Ctrl + C to terminate a node app, the next time I opened it, it did without a hitch.
Changing the port number to something other than the one in error solved the issue.
using netstat to get all node processes with the port they are using and then kill the only one you want by PID
netstat -lntp | grep node
you will get all node processes
tcp6 0 0 :::5744 :::* LISTEN 3864/node
and then when you get the PID (3864) just kill the processes by PID
kill -HUP PID
You may use hot-node to prevent your server from crashing/ run-time-errors. Hot-node automatically restarts the nodejs application for you whenever there is a change in the node program[source] / process[running node program].
Install hot-node using npm using the global option:
npm install -g hotnode
I am aware that adding a '&' in the end makes it run as a background but does it also mean that it runs as a daemon?
Like:
celery -A project worker -l info &
celery -A project worker -l info --detach
I am sure that the first one runs in a background however the second as stated in the document runs in the background as a daemon.
I would love to know the main difference of the commands above
They are different!
"&" version is background , but not run as daemon, daemon process will detach with terminal.
in C language ,daemon can write in code :
fork()
setsid()
close(0) /* and /dev/null as fd 0, 1 and 2 */
close(1)
close(2)
fork()
This ensures that the process is no longer in the same process group as the terminal and thus won't be killed together with it. The IO redirection is to make output not appear on the terminal.(see:https://unix.stackexchange.com/questions/56495/whats-the-difference-between-running-a-program-as-a-daemon-and-forking-it-into)
a daemon make it to be in its own session, not be attached to a terminal, not have any file descriptor inherited from the parent open to anything, not have a parent caring for you (other than init) have the current directory in / so as not to prevent a umount... while "&" version do not
Yes the process will be ran as a daemon, or background process; they both do the same thing.
You can verify this by looking at the opt parser in the source code (if you really want to verify this):
. cmdoption:: --detach
Detach and run in the background as a daemon.
https://github.com/celery/celery/blob/d59518f5fb68957b2d179aa572af6f58cd02de40/celery/bin/beat.py#L12
https://github.com/celery/celery/blob/d59518f5fb68957b2d179aa572af6f58cd02de40/celery/platforms.py#L365
Ultimately, the code below is what detaches it in the DaemonContext. Notice the fork and exit calls:
def _detach(self):
if os.fork() == 0: # first child
os.setsid() # create new session
if os.fork() > 0: # pragma: no cover
# second child
os._exit(0)
else:
os._exit(0)
return self
Not really. The process started with & runs in the background, but is attached to the shell that started it, and the process output goes to the terminal.
Meaning, if the shell dies or is killed (or the terminal is closed), that process will be sent a HUG signal and will die as well (if it doesn't catch it, or if its output goes to the terminal).
The command nohup detaches a process (command) from the shell and redirects its I/O, and prevents it from dying when the parent process (shell) dies.
Example:
You can see that by opening two terminals. In one run
sleep 500 &
in the other one run ps -ef to see the list of processes, and near the bottom something like
me 1234 1201 ... sleep 500
^ ^
process id parent process (shell)
close the terminal in which sleep sleeps in the background, and then do a ps -ef again, the sleep process is gone.
A daemon job is usually started by the system (its owner may be changed to a regular user) by upstart or init.
I have a nodejs server which will be started using nohup. So basically it will not have stdin.
I have created daemon for my nodejs server which does all start stop restart status functions.
But I dont want to stop the nodejs server immediately. I want to stop it after certain tasks. So I have to send command to the server's stdin. So that shutdown will be controlled by the server itself. Yes I can do that by creating a API. But its my last option.
Yes my server takes input from stdin and its working well. But using nohup I will not have stdin therefore no way to enter command in that case.
So my question is, if there is a way so that I can change the stdin to other file and my server will take input from it in the same way it does in case of stdin.
If you are in Linux or Mac you can send signals to your server process. In Windows you can too but I don't know the command.
For example if you want to reload your server you can do
kill -s SIGUSR2 processId
And in your server
process.on('SIGUSR2', function() {
console.log('Got SIGUSR2, reloading server...');
});
I'm running a node server on port 5100. It executes a shell script as child process which in turn starts another node server.
var execFile = require('child_process').execFile;
execFile("./testscript", args, {cwd: cwd}, function (error, stdout, stderr) {
console.log("error.." + error);
console.log("stdout.." + stdout);
console.log("stderr.." + stderr);
})
In script i'm running a node process by nohup as :
nohup node server.js PORT=5200 2>/dev/null 1>/dev/null &
Now node#5200 has parentid of testscript which in turn has parentid of node#5100.
When i kill PID of node#5100 everything stops.
I want node#5200 to still run when testscript and node#5100 have exited.
You need to set the detached option when you create the child process. This is sort of like nohup. See the documentation here: http://nodejs.org/api/child_process.html#child_process_options_detached
What this does is to make the child process the leader of a new "process group," so when the parent process group dies, the child can continue running.
Edit: the documentation says:
When using the detached option to start a long-running process, the process will not stay running in the background unless it is provided with a stdio configuration that is not connected to the parent. If the parent's stdio is inherited, the child will remain attached to the controlling terminal.
So you must write code like this:
var child = spawn(cmd, args, {detached: true, stdio: ['ignore', 'ignore', 'ignore']});
child.unref();
Or you can use their example to have stdout redirect to a file, etc.
When i spawned a process from a file that ran once, my child process had perfect behavior as parent file exited normally.
Issue was raised as i was spawning process by a server file that was killed by Ctrl+c so it gave an error code on exit and as a result whole process group got SIGTERM which lead to killing of child process.
To successfully implement spawning i used double spawn i.e spawned a child process from my server file that further spawned child_process that ran my other node server. So i was able to avoid SIGTERM to reach grandchild and fulfill my requirement.