When specifying username and password, only the password is stored in teh username location when reloading

` "parameters": "--config Profile='abc'"``

results in "commandlineArguments":" --config"

C-troll doesn't always shut down OpenSpace on the nodes when using a multi node setup. That means you have to manually remote into them to shut down the process that runs OpenSpace. This happens both when you press x and when you click the Stop: ... button in C-troll.

Instead of creating a box widget + layout and adding the box, it should be possible to add the layout directly

Perhaps a machine in a cluster might wind up with a different drive letter or something. Right now i think they all need to be the same.

When creating an application in the editor it would be nice for users to have an picker instead of having to type in the path

Taking name and configuration of a program as commandline arguments in order to make Desktop shortcuts that start dome applications

Currently on my cluster I use the WWT controller to restart the nodes. If the WWT app can do it, I think we should be able to also.

1. Starting an application that terminates / or is terminated
2. The dropdown menu shows "Restart"
3. Restart the program (this time successfully)

Behavior: The dropdown menu still says "Restart"
Expected behavior: THe dropdown menu should say "Stop"

When already running a C-Troll instance, it should warn when trying to start another one

I tried to start notepad locally using given moc_core*.py scripts. Start and Exit works fine.
Kill command is not working. Any command given after Kill seems not working.

When pressing the "Kill all processes" button, the resulting processes are listed as "Crash Exit".

When using the "Kill" button, they show as "Normal Exit" instead

Effectively performing a Robocopy over the existing network connection to synchronize the working directory of a project (multiple projects?)

Moving the libraries into dynamic libraries and making the executables just link against it

Reason: Being able to mix and match them with other applications

This could be used to solve #43 by placing a robocopy synchronization script in the pre-run phase

Copy from OpenSpace/OpenSpace#2184

Right now when launching openspace with c-troll, if openspace exits too early, there is no way to see that logging that happens. So in the case of a bad configuration file, there is no way to find that out.

It would be nice to get access to this in the tray or maybe somewhere in the main C-Troll app

When creating / removing a large number of processes, the list gets very long, so it would be necessary to scroll that list

Sometimes when I use the editor, and save the application, if c-troll was running, then all the trays in the cluster would crash. Not sure exact repro steps.

When starting the REST API on port 5000 without the Tray running, but the C-Troll configured to look on localhost:5000, it will show localhost tray as being connected

When starting a program that fails to start, it presents the "Kill" button as an option, which should not be possible

right now they default to different ones....

I would be nice to have an option for applications to have the option to syncronize files before launching an application.

Currently there is no concept of a control/renderer connection in ctroll, this might have to be added, or some option to choose which node in the cluster has the files

Being able to select an option that will cause a Tray application to restart an application if it crashes or is closed (?)

Tray:

23:19:40.631  (Received [::1]): {"id":0,"secret":"","type":"ExitCommandMessage","version":1}
23:19:40.632  (Process Error): FailedToStart

When starting and stopping the same application over and over again, the process list is filled with remains of the previously exited processes, cluttering the list

When I startup the tray application, if it has an error binding to the socket, it logs it in the console, but gives no indication in the tray or alert message.

I think we should maybe provide a gray version of the icon for the tray and change it to gray when not connected, I have seen other applications do that.

When running powershell with arguments shutdown /s it doesn't work.
When checking output it looks like powershell was not able to understand the input (as in you've not given proper arguments).

I think something in C-Troll parses the /s incorrectly and thus the command does not work.

Suggestion: Let the user specify command line parameters in the application JSON configuration where specific cluster nodes can receive different parameters.

e.g. we have an executable that takes a different calibration file on each node as a command line parameter.
That means the command to be run on the nodes would be

Node 1:
<path/to/shared/drive>/program.exe  P1_A3.mpcdi

Node 2:
<path/to/shared/drive>/program.exe  P2_A3.mpcdi

I would like to have support for Zabbix i C-Troll.
To be able to monitor if the diffrent components (core, tray...) are running or not.
It's also possible to send data to Zabbix, such as if programs controlled by C-Troll is started (also which program, if you want to get detailed information).
This would be great for supoort staff, to see if the system is healthy and for managers to see usage of diffrent programs/shows. It's also possible in Zabbix to act on triggers, to send an e-mail or SMS if something is wrong.

OS specific monitoring, is best done with the Zabbix agents, which can gather information about CPU, Mem, Disk, Network usage to name a few items.