Handling 1006 errors and unexepected disconnections

Pusher version: 2.2.4

My organization has been using Pusher for awhile and we’ve noticed that occasionally it’ll just stop working. In most cases, it logs a 1066 error like so:

Pusher : Error : {"type":"WebSocketError","error":{"type":"PusherError","data":{"code":1006}}}

and will not send/receive any more events. Another error I’ve seen with less frequency is:

Pusher : Error : {"type":"WebSocketError","error":{"type":"PusherError","data":{"code":4201,"message":"Pong reply not received"}}}

Usually this event occurs if the user’s computer has been put to sleep and then re-awoken – but not always. I’m only occasionally able to reproduce it on my machine.

Any ideas on what I could do to ensure that Pusher connections stay alive or refresh them when they die? Also any recommendations on testing error/disconnected/fail scenarios with Pusher?

Sorry for the really vague issue but it’s all I have to work with.

[amrendraw3]

The code 1006 error occurs mostly because of missing cluster key while initializing pusher client. It should something like:

var client = new Pusher('YOUR_APP_KEY', {cluster: 'CLUSTER_VALUE', encrypted: true});

[aeolusheath]

@tquiroga pusher has auto-reconenct func