Just a minor issue, but when sorting workers by Origin (at least) its not sorting correctly when Origin name only contains int values (didn't test with string values yet) :-)

Be able to click on a worker or controller connection to get the extended information that rotom knows (eg source IP address of connection)

I can only select 5-6 devices for a job. if I select more then i got an error on the ui, but no error in the log.

To align with Dragonite/Golbat

Default.json should have memory monitor as false; this monitoring causes confusion in many cases and should likely be 'opt in'

The devices are not sorted correctly. So its difficult to select the right ones.

I think this is both an Issue/Feature Request. Currently, Rotom's dashboard and API are dynamic. Origin and Workers can drop off, so real-time monitoring scripts may not pick up if anything is offline. Kasmar is working on updating RDMMonitor to keep track of these changes, so this can provide a workaround. However, if both are restarted "like on a server restart", then everything is empty. Therefore, it would be nice to store Origin and Workers in a database as static placeholders so we can properly monitor workers and ensure uptime.

If 5/min is your target could you allow 5 at once and then cooldown for 60 seconds?

Following metrics would be nice to have via prometheus:

• Total workers
• Workers in Use
• Available Workers
• Device Metricen like ui (Origin, Load, Device id, is alive, MITM version, etc ...)
• Worker Metricen like ui (Origin, Worker id, is active, scanner worker name, last message)
• Job Queue count
• Job Queue status

Super minor quality of life feature to toggle Light/Dark mode.

Na-ji said he will accept the PR someone just need to add the UI element. I've volunteered myself but it isn't high on my priority list.

https://v1.nextui.org/docs/theme/dark-mode

Some way to show / record a rolling average connection duration or some other way to highlight devices that are connecting but continuously fail when a controller connects

Would it be possible to implement ad-hoc jobs in Rotom ?

What I mean is the ability to send any adb command to one or several device by simply inputing it in the web interface of Rotom, even if that command doesn't already exist as a Rotom Job.

This would be pretty useful, typically when needing to check if some app is running on a device, or if some parameter is correctly set.

I'm aware this could potentially represent a security issue if the rotom web app is left unprotected, but it's already the case for current jobs, even if more limited. Maybe Rotom could have one or 2 configuration settings to enable/disable jobs and/or ad-hoc jobs. It's then up to the user to secure its Rotom install if they have activated jobs.

Thanks

The information about a device, which is shown in the ui, is not available via prometheus metric.

This information/metric is essential for grafana dashboards and for monitoring the ATV devices.

Can you please add it?

Add a cooldown overwrite so the end user can calibrate according to their device needs. Would help people running a mix of android or iOS which have different cooldown needs.

Building Rotom with node 18.15.0 as per instructions result in the following build error.

https://pastebin.com/qLuqz4wr