Nodemon conflicts with zero-downtime

ghost commented 6 years ago

Steps to reproduce:

Install nodemon npm install -g nodemon
change pump.io settings to children: 1
run pump with nodemon nodemon bin/pump -c pump-config.json
change a file: example lib/app.js and saves.
see the logs and the console shows: restarting due to changes...

the restart is unsuccessful and the log shows:

Received SIGUSR2 but ignoring because there aren't enough workers for zero-downtime restart"

With children 2 and databank{redis/disk}:

"Zero-downtime restarts are only supported on MongoDB for the time being"

strugee commented 6 years ago

Can you jack your logLevel up to trace and then post the full logs? At least from when it says "starting zero-downtime restart" or whatever

ghost commented 6 years ago

sure:

log level: default

with children 2:

[nodemon] 1.17.3
[nodemon] to restart at any time, enter `rs`
[nodemon] watching: *.*
[nodemon] starting `node bin/pump -c pump.io.json`
{"name":"pump.io","hostname":"parabola","pid":30477,"level":40,"msg":"`config.secret` is either unset or set to the sample value; this is very insecure and should be changed ASAP","time":"2018-04-14T18:20:03.304Z","v":0}
{"name":"pump.io","hostname":"parabola","pid":30476,"level":40,"msg":"`config.secret` is either unset or set to the sample value; this is very insecure and should be changed ASAP","time":"2018-04-14T18:20:03.309Z","v":0}
{"name":"pump.io","hostname":"parabola","pid":30477,"level":30,"msg":"Listening on 80 for host 127.0.1.2","time":"2018-04-14T18:20:03.335Z","v":0}
{"name":"pump.io","hostname":"parabola","pid":30476,"level":30,"msg":"Listening on 80 for host 127.0.1.2","time":"2018-04-14T18:20:03.340Z","v":0}
[nodemon] restarting due to changes...
{"name":"pump.io","hostname":"parabola","pid":30470,"component":"cluster","level":40,"msg":"Zero-downtime restarts are only supported on MongoDB for the time being","time":"2018-04-14T18:20:06.453Z","v":0}

with children 1:

[nodemon] 1.17.3
[nodemon] to restart at any time, enter `rs`
[nodemon] watching: *.*
[nodemon] starting `node bin/pump -c pump.io.json`
{"name":"pump.io","hostname":"parabola","pid":2432,"level":40,"msg":"`config.secret` is either unset or set to the sample value; this is very insecure and should be changed ASAP","time":"2018-04-14T19:29:45.172Z","v":0}
{"name":"pump.io","hostname":"parabola","pid":2432,"level":30,"msg":"Listening on 80 for host 127.0.1.2","time":"2018-04-14T19:29:45.202Z","v":0}
[nodemon] restarting due to changes...
{"name":"pump.io","hostname":"parabola","pid":2426,"component":"cluster","level":40,"msg":"Received SIGUSR2 but ignoring because there aren't enough workers for zero-downtime restart","time":"2018-04-14T19:29:48.178Z","v":0}

config:

  {
   "driver":  "disk",
  "params":  {"dir": "/tmp/pumpio3/"},
  "secret":  "my dog has fleas",
  "noweb":  false,
  "site":  "Test pump",
  "owner":  "Pump",
  "ownerURL":  "http://example.com/",
  "port":  80,
  "hostname":  "local.pump.io",
  "address":  "127.0.1.2",
  "nologger":  false,
  "children": 1 // or 2
 }

pump-io / pump.io

Nodemon conflicts with zero-downtime #1577