Hi guys, is there any plan to enable us to have (cloud) dashboards that contain statuses rather than just metrics? I have a few scenarios in mind:
“Service status” pages, so people can know if a service (in the generic sense) is online (maybe one could select/chain certain metrics or groups of metrics and their union would determine the up/down status of the service; or generic HTTP probes?
On/Off states: such as whether specific services are on or off.
Graphs with timelines are still desirable but a nice board with important services with a simple running/stopped etc would be great.
Numeric states: states where the number is important: for example, I’d really love to have a collector which checks apt/yum etc for the currently number of available packages to be installed; or a check for whether a system is in a pending reboot state or not.
Is this something that can be done now or is it in the plan?
With the release of the new Netdata charts we are also aiming to increase the summary dashboards available on the Overview or Single Node view. Example of what I mean by summary dashboards
“Service status” pages, so people can know if a service (in the generic sense) is online (maybe one could select/chain certain metrics or groups of metrics and their union would determine the up/down status of the service; or generic HTTP probes?
In this first iteration we will mostly use the existing charts, maybe introduce a new one, but there are plans to increase this chart options - maybe having one with a gauge representing a status (number 0/1 with red/green flag) could be feasible.
Also, not sure if you have seen this blogpost on How to monitor host reachability | Netdata Blog it could have some interesting insights into metrics and collectors, in case you don’t know about them.
On/Off states: such as whether specific services are on or off.
Graphs with timelines are still desirable but a nice board with important services with a simple running/stopped etc would be great.
Do you see these “running/stopped” flags to be sent to Netdata as some event or custom metric?
We have in our roadmap to deal with receiving Events from the Agent directly, so if an agent detected if some service went down or started it could be an event recorded and made available to Cloud.
Numeric states: states where the number is important: for example, I’d really love to have a collector which checks apt/yum etc for the currently number of available packages to be installed; or a check for whether a system is in a pending reboot state or not.
Here you mean something like the numeric summary chart we have for PostgreSQL DB Size or Uptime?
If yes, as we progress with the work on these summary dashboard, where we will also want to allow users to easily add them to custom dashboards, it makes total sense to allow users to display a given metric as a number or a table with numbers if user has multiple dataseries to be represented.
In terms of timelines and to summarize, what I can share is that we are working on the new charts and summary dashboards, we will plan to improve these and custom dashboard - these should happen in a near future.
For the agent Events this is up on our priorities but we don’t know yet when we will be able to pick this up.
