[RFE] Handle SIGINT and interrupt command execution about kubectl-aks HOT 3 CLOSED

I did initial investigation regarding passing a cancelable context to BeginRunCommand and it seems it doesn't help with cancelling the actual run command and only effects the polling:

Error: failed to run command: error polling command response: context deadline exceeded

I will investigate if there is still an option to avoid conflicts when re-running the command.

from kubectl-aks.

I just realize that maybe we won't be able to do too much on this issue after re-reading the documentation:

You can't cancel a running script.

Maybe, what we can do is to handle this kind of signals and improve the UX:

$ kubectl az run-command "ip route" --node aks-agentpool-12345678-vmss000000
Running...
^C
Warn: The requested command hasn't finished yet...
Warn: Hit Ctrl+C again to exit, but please notice the command will continue running in the node anyway, and you will be unable to see the output or run another command until it finishes.
^C
Exiting...
$ 

from kubectl-aks.

I agree. I think having a warning like this can be useful so users are aware of the consequence of cancellation an unfinished command. Also, I am thinking of wrapping the run-command in timeout and set the default value to 300 seconds . Let me have a look into this.

from kubectl-aks.