What is the failure handling of logic apps for failure of the logic app itself (i.e. all functions are ok)?
Consider for example a case where I have some trigger, then I call some REST API to get an authentication token, then I use that token to call some long process and I poll it until done.
Now the logic app crashes in the middle (e.g. because of a hardware failure on the azure side).
What would happen? would it simply fail? would it be restarted from the beginning (and possibly starting the long process multiple times)? would it be restarted from the state it was in (and risk the authentication token becoming stale?)
If it simply fails, is there a way for me to get a trigger on this happening and some input on where in the process it was (and what the trigger inputs were)?
