Most of the time, netdata just shows overall system statistics, and the container statistics, I am unable to see the individual node's statistics.
1 in 10 times it shows after multiple hard refreshes.

Hello

I tried this script and it doesn't show up on Netdata. I've tried Debian, Ubuntu, and Xubuntu. None of them works. Please fix.

Thank you

I've recently set up a Netdata agent with Prometheus remote-write and wanted to set up a dashboard for system metrics overview, but found out that most of existing dashboards for Prometheus aren't maintained.
For example, there are some dashboards in Grafana listing:

• https://grafana.com/grafana/dashboards/7107-netdata/
• https://grafana.com/grafana/dashboards/12279-netdata/

Second dashboard still works, but requires some changes as metrics were changed since the last update.

My question is: will it be useful to maintain a dashboard for this use-case here?
It seems to me that hosting dashboard here will make it easily discoverable for community and will also allow maintaining dashboard in updated state by any community member.

If it sounds reasonable, I'd be happy to update dashboard above and send a PR/maintain it.

I'm looking into using Netdata for some other parts of my home lab setup, so I'd like to also build a dashboard for k8s metrics discovered by agent later on and some other dashboards which can be useful.

Can we have a script to uninstall plugins or at least some documentation? (as much fun as it would be to read through the scripts myself to see every single change they make...) :)

We should provide a tool (e.g a bash script) so that users can frictionlessly install community collectors and alerts.

Collectors

• python.d
• go.d
• charts.d

go.d is tricky, as the user will have to rebuild the go.d binary and replace the existing one. Moreover, we should somehow make netdata NOT to replace the edited go.d with every nightly release.

Alerts

Just copy the alert file in the proper directory. Since we will have multiple different alerts for the same collectors, we should have a structure where people can install specific alerts for a component.

e.g the goroutines alert for geth.

This could be done via a repo structure as: alerts/geth/goroutines.conf

UX

It should be as effortless as possible. For the users to use the community repo, we expect them to be more technical, since they are pushing the boundaries of what netdata can do.

There is a folder in this repo which is consumed in other repo's (for example, the go.d repo).
https://github.com/netdata/community/tree/main/devenv/.devcontainer
However, this appears to be kept up to date only manually.
Could we have some automation to copy, on change, from this repo to the consuming repos?
We already have config drift between the two repos mentioned.

Website: https://netdata-marketplace.pages.dev/
GitHub: https://github.com/OdysLam/netdata-

• Share MVP with the community
• Share MVP internally
• Fix absence of social image for sharing
• Refine about page (needs larger space between text)
• append installation instructions in every collector/alert, at the end of the page for ease of use
• Change community repo + readme to add mention for community plugins for statsd

• Define community collectors
• Define community alerts
• clone files to community repo
• remove files from main repo

repo structure

• apps/ --> applications building on or extending netdata (e.g netdata-api, configuration management, etc.)
• collectors/plugin/module/* --> source files for the particular collector as exists in the respective directory now
• collectors/.plugin./.module./module.yaml --> a yaml file describing the module. It will be used by the marketplace website
• alerts/module/alert.conf --> a conf file with a single alert definition. This will be appended into the alert file for the particular module. It enables us to have multiple alerts for the same module, without overloading a single configuration file.
  • example: /geth/goroutines.conf, /geth/chainhead.conf

Communicate this change/structure to users with a blog post

@shyamvalsan im getting this - any idea?

/usr/libexec/netdata/plugins.d/../charts.d/speedtest.chart.sh: line 47: 3.351*8/1000: syntax error: invalid arithmetic operator (error token is ".351*8/1000")