Re: Ci automated deploy of seaside to Digital Ocean

On 10/05/2018 15:30, Esteban A. Maringolo wrote:
> How would you know when to kill the running vm+image? It might have
> active sessions.
> 
> If you don't care about that, what I used to manage a bunch of "worker
> images" (process groups) with supervisord [1], so in that case you'd
> stop all the related workers, copy the new image, and start the
> workers again.
My supervisord.conf entry for a pool of working images was:

[program:psworker]
command=/home/trentosur/perfectstore/pharo-vm/pharo --nodisplay
/home/trentosur/perfectstore/ps.image worker.st 818%(process_num)1d
process_name=%(program_name)s_%(process_num)02d ; process_name expr
(default %(program_name)s)
numprocs=2
directory=/home/trentosur/perfectstore
autostart=false
autorestart=true
user=trentosur
stopasgroup=true
killasgroup=true


Part of the worker.st file handling the port number was:

"Seaside server start"
Smalltalk isHeadless ifTrue: [
  Smalltalk commandLine arguments
    ifEmpty: [
      Transcript show: 'No port parameter was specified.'; cr.
      Smalltalk quitPrimitive. ]
    ifNotEmpty: [:args |
      | port |
      port := args first asNumber asInteger.
      Transcript show: 'Starting worker image at port ', port asString; cr.
      ZnZincServerAdaptor  startOn: port.
      ZnZincServerAdaptor default server debugMode: false.
    ]
  ]
  ifFalse: [
      | port |
      port := 8080.
      Transcript show: 'Starting worker image at port ', port asString; cr.
      ZnZincServerAdaptor  startOn: port.
      ZnZincServerAdaptor default server debugMode: true.
    ].


I hope it helps.

Best regards,



-- 
Esteban A. Maringolo
_______________________________________________
seaside mailing list
[email protected]
http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/seaside