tbideas / pi-steroid Goto Github PK

When the parent process dies (for example because of the network error below) - the child process created stays alive and keeps hogging resources (a serial port for example).

We need to make sure that:

• The parent process does not die on network errors
• If it dies anyway, the parent process kills any child process with him

/usr/local/lib/node_modules/pi-steroid/node_modules/ddp/node_modules/ws/lib/WebSocket.js:175
    else throw new Error('not opened');
               ^
Error: not opened
    at WebSocket.send (/usr/local/lib/node_modules/pi-steroid/node_modules/ddp/node_modules/ws/lib/WebSocket.js:175:16)
    at DDPClient._send (/usr/local/lib/node_modules/pi-steroid/node_modules/ddp/lib/ddp-client.js:77:15)
    at DDPClient.call (/usr/local/lib/node_modules/pi-steroid/node_modules/ddp/lib/ddp-client.js:222:8)
    at AsteroidClient._writeConsole (/usr/local/lib/node_modules/pi-steroid/lib/asteroid_client.js:145:19)
    at null.<anonymous> (/usr/local/lib/node_modules/pi-steroid/lib/asteroid_client.js:28:10)
    at EventEmitter.emit (events.js:98:17)
    at Socket.<anonymous> (/usr/local/lib/node_modules/pi-steroid/lib/asteroid_vm.js:75:10)
    at Socket.EventEmitter.emit (events.js:95:17)
    at Socket.<anonymous> (_stream_readable.js:699:14)
    at Socket.EventEmitter.emit (events.js:92:17)
Forever detected script exited with code: 8

I'm getting a 400 error from the pi-steroid client and it's not reporting back and updating pijs.io. It seems I was able to add my pi, but since then it hasn't updated logs or been able to successfully connect at all. What would be the cause?

EDIT: at some point the error switched from a 400 to a 502.

If the program was not updated, there is no need to stop and relaunch the program after a reconnect to the server

Forever seems to start with the system-installed nodejs (if it exists) instead of the one we ship: probably because /usr/bin is before /usr/local/bin in the initd scripts.

Fix: force the path in the startup scripts.

Initial report: https://pijs.uservoice.com/admin/tickets/2

when running the installation script and at the end of the process I
see the following error:

piJS.io: Launching pi-steroid
info: Forever processing file: /usr/local/bin/pi-steroid

/usr/local/lib/node_modules/forever/lib/forever.js:404
monitor.send(JSON.stringify(options));
^
TypeError: Object # has no method 'send'
at Object.startDaemon
(/usr/local/lib/node_modules/forever/lib/forever.js:404:11)
at /usr/local/lib/node_modules/forever/lib/forever/cli.js:229:13
at /usr/local/lib/node_modules/forever/lib/forever/cli.js:134:5
at Object.oncomplete
(/usr/local/lib/node_modules/forever/lib/forever.js:343:11)

It looks like all the connect() callbacks (at the node-ddp level - or at the asteroid-client level) are all stacked up and run at once when the connection is re-established.

For more security and also for users who are behing http proxy who do not re-cognize the Upgrade command.