Closed rstoermer closed 2 years ago
Alright, found the error. The memory limit we were setting in Kubernetes was too low. It would be better if nodemon would be a bit more expressive than just saying: "Killed"
The killed
message came from your code, not from nodemon. You can see this from your logged output:
[data-broker] [nodemon] spawning
[data-broker] [nodemon] child pid: 31
[data-broker] [nodemon] watching 115 files
[data-broker] Killed
[data-broker] [nodemon] app crashed - waiting for file changes before starting...
Everything from nodemon is prefixed with [nodemon]
.
nodemon -v
: 2.0.16Expected behaviour
Nodemon starts process
Actual behaviour
Code that worked yesterday with nodemon isn't working anymore and just reporting "Killed" in terminal. The container is not crashing instantly, but after 10 - 15 seconds. However the first line in my entry file is never reached, so I assume an error in nodemon. The same code is running without nodemon when doing "node ./dist/app.js" after building the js files.
I am starting via skaffold with a Helm deployment in a Kubernetes cluster...
nodemon.json
Waiting for deployments to stabilize...
rs
[data-broker] [nodemon] or send SIGUSR2 to 18 to restart [data-broker] [nodemon] ignoring: *.json temp//* logs ./node_modules/*/ node_modules_production types/hl7// [data-broker] [nodemon] watching path(s): .* [data-broker] [nodemon] watching extensions: ts,json [data-broker] [nodemon] startingts-node ./src/app.ts
[data-broker] [nodemon] spawning [data-broker] [nodemon] child pid: 31 [data-broker] [nodemon] watching 115 files [data-broker] Killed [data-broker] [nodemon] app crashed - waiting for file changes before starting...Steps to reproduce
If applicable, please append the
--dump
flag on your command and include the output here ensuring to remove any sensitive/personal details or tokens.