Home/ Questions/Q 9205109
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-17T23:52:30+00:00

This is executed for each cluster process using https://github.com/fent/socket.io-clusterhub : var store = new

  • 0

This is executed for each cluster process using https://github.com/fent/socket.io-clusterhub:

var store = new (require('socket.io-clusterhub'));
var io = require('socket.io').listen(config.ports.socket);
io.configure(function(){
    io.set('store',store);
});
console.log('Socket service listening on port '+config.ports.socket);
io.sockets.on('connection',function(socket){
    if(typeof fn === 'function'){fn(socket);}
});

But it looks like all 4 threads are attempting to register the same user with different socket id’s:

debug: client authorized
info: handshake authorized XEGx6zAP7rSWGQcaxmUB
debug: client authorized
info: handshake authorized r0Vcr7Z-r_9MnylyxmUC
debug: client authorized
info: handshake authorized 7RWUgoCuZPIu0IyQxmUj
debug: client authorized
info: handshake authorized mq0zKCgeDkisr-rbxmUk

Then it disconnects outright:

debug: setting request GET /socket.io/1/websocket/XEGx6zAP7rSWGQcaxmUB
debug: set heartbeat interval for client XEGx6zAP7rSWGQcaxmUB
debug: websocket writing 7:::1+0
warn: client not handshaken client should reconnect
info: transport end (error)
debug: set close timeout for client XEGx6zAP7rSWGQcaxmUB
debug: cleared close timeout for client XEGx6zAP7rSWGQcaxmUB
debug: cleared heartbeat interval for client XEGx6zAP7rSWGQcaxmUB
debug: discarding transport
debug: setting request GET /socket.io/1/websocket/r0Vcr7Z-r_9MnylyxmUC
debug: set heartbeat interval for client r0Vcr7Z-r_9MnylyxmUC
debug: setting request GET /socket.io/1/websocket/7RWUgoCuZPIu0IyQxmUj
debug: set heartbeat interval for client 7RWUgoCuZPIu0IyQxmUj
debug: websocket writing 7:::1+0
warn: client not handshaken client should reconnect
info: transport end (error)
debug: set close timeout for client 7RWUgoCuZPIu0IyQxmUj
debug: cleared close timeout for client 7RWUgoCuZPIu0IyQxmUj
debug: cleared heartbeat interval for client 7RWUgoCuZPIu0IyQxmUj
debug: discarding transport

And it fails to run the callback routine at all, even though it continues to act like r0Vcr7Z-r_9MnylyxmUC
is still active:

debug: cleared heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: set heartbeat interval for client r0Vcr7Z-r_9MnylyxmUC
debug: emitting heartbeat for client r0Vcr7Z-r_9MnylyxmUC
debug: websocket writing 2::
debug: set heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: got heartbeat packet
debug: cleared heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: set heartbeat interval for client r0Vcr7Z-r_9MnylyxmUC
debug: emitting heartbeat for client r0Vcr7Z-r_9MnylyxmUC
debug: websocket writing 2::
debug: set heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: got heartbeat packet
debug: cleared heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: set heartbeat interval for client r0Vcr7Z-r_9MnylyxmUC
debug: emitting heartbeat for client r0Vcr7Z-r_9MnylyxmUC
debug: websocket writing 2::
debug: set heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: got heartbeat packet
debug: cleared heartbeat timeout for client r0Vcr7Z-r_9MnylyxmUC
debug: set heartbeat interval for client r0Vcr7Z-r_9MnylyxmUC
debug: emitting heartbeat for client r0Vcr7Z-r_9MnylyxmUC

Any ideas? Thanks!

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    Since socket.io usually uses long-polling to talk to the server, each new request can be handled by any node in the cluster. Thus each node needs to have server state of the client to be able to repond to the request. The main issue is that during initial handshake socket.io needs to make few requests before creating a valid session, and they can be distributed across the cluster which adds complexity to the code. Clusterhub approach to sync everything using cross-process messaging doesn’t looks efficient to me. I would suggest looking at sticky session approach, where each client is guaranteed to work with the same node, always. You can configure load-balancer in front of your nodejs server, or try something like https://github.com/indutny/sticky-session

    • 0