Node.js Monitoring

[tomfree]

Hi,
I am using a shared hosting that uses cpanel and I happily run a prototypical node.js app in there. Now I wanna do this for real and need some monitoring for it and wonder what's the best way to do this.

Seems there are no out of the box apps to do this and the regular monitoring is regarding websites but not apps (i.e. i would expect for each route of my rest apis to get a view on response times and http 4xx and 5xx stattus codes over time).
I then thought about installing a Grafana agent to push metrics to a cloud instance of Grafana but cant install it due to rights issues.

So I wonder if you can run node.js via cpanel there must be a way to monitor this. What am I missing?

Would really appreciate a hint. Have been fighting on this for too long already.

Cheers
Tom

 

[cPRex]

Hey there! I don't personally have any tools on my end that would help with this. We do have an older feature request here that hasn't gotten much traction:

https://features.cpanel.net/topic/node-js-logs-monitoring-and-watching

but I'll bring that up with our team during next week's features meeting to see if that's something they'd consider adding now that Node.JS is more widely used.

 

[tomfree]

@cPRex - that'd be great. Maybe a good point to push it is also that offering node.js support essentially requires an infrastructure to ensure it runs reliably and otherwise it is more for prototypical use.

I would argue that in regards to priorities from the ticket that:
1.) Monitoring really needs to be covered since there is hardly an alternative. E.g. express-status-monitor that one could use just keeps data in memory for some short time and thus u will usually miss the times that in the end you wanna see and thus to me is not suitable for professional use. To keep efforts low on your end it may be an option to refer to things like Grafana-Cloud and essentially enable running a Grafana/Prometheus Agent as an installable app and require from app that it exposes metric infirmation in a standard way (E.g. Prometheus style metrics or InfluxDB Line format...)
2.) Restart in case of case of changes would be nice from my pov, especially if one could specifiy via regex or so which files to monitor. E.g. config directory if one increased log level or so...
3.) Logging I would personally argue that node frameworks like nest.js offer goods ways to handle exactly that, so I wouldn't see it as a top listed item

Would absolutely love to get an update on this.

 

[cPRex]

I'll bring up the monitoring portion of the issue with the team next week, and then it's up to the product owners to decide how that gets handled.

For your second point, you could submit a feature request using the link in my signature. I'll see that, get it approved, and then talk to the team about it.