QuickApp crashed, Unknown error occurred:

[Nevicare 0]

Hi All,

 

In my own build QuickApp I get the message:

[18.04.2025] [09:46:01] [ERROR] [QUICKAPP1563]: QuickApp crashed

[18.04.2025] [09:46:01] [ERROR] [QUICKAPP1563]: Unknown error occurred:

 

This is happening very random, so the QA kan run for hours with no issues and it can crash minutes aparts.

The error code doesn't give anything hint on what the problem is or which line in the code the problem occurred.

 

Off course I build in my own debug lines, but that doesn't give me a hint.

There is a sleep op 60seconds in the code, It looks like het error occurs during that sleep..

 

Any hints how I can increase the debug level or see some more details in de console, that will be nice.

 

Thanks in advance.

 

[jgab 2,642]

Is it a lot of code? Can you post it here?
or can you make a minimal version of it that also crashes and post it?

People have been reporting this as long as the HC3 been out but there have never been any way to

"corner the bug", come up with code that always have this behavior....

[Nevicare 0]

The code is to big (700lines)..

 

I do some HTTP requests to an API like this.

function QuickApp:Get_WiFi_status(IPAddress)

net.HTTPClient():request('http://' .. IPAddress .. '/api/1/wifi_status', {

options = {

headers = {

Accept = "application/json"

},

method = "GET"

},

success = function(response)

if ((response.status == 200 or response.status == 201))

then

if ( DEBUG == "true" )

then

print ("API GET wifi_status: " .. response.data)

end

Table_WiFi_status = json.decode(response.data)

self:Get_lifetime(IPAddress)

else

error("Error: " .. response.status);

end

end,

error = function(message)

end

})

end

 

This is the SetTimeout use..

fibaro.setTimeout( Refresh_Rate*1000, function()

   self:CollectData()

end)

[Martin_N 154]

@Nevicare

Hi.

I have no idea what your Refresh_Rate parameter is. There will definitely be problems if you are querying a resource where you will process a more complex data structure and you will do it more often than you get responses.

Furthermore, I do not see a parameter with a suitable timeout for HTTP used elsewhere in the code. See:

net.HTTPClient({timeout=3000})

In practice, I have already encountered the need to extend this timeout up to 10 seconds.

 

 

[jgab 2,642]

No, I don't think that is the problem. One can have 0 in poll-time, the processing in-between will just push the timer into the future. 

Where in the code do you have the "sleep" ? is it a fibaro.sleep or were you referring to the setTimeout?

So, it seems to be impossible to come up with a limited example that exhibits these random crashes...

It makes me believe that the problem is elsewhere. I.e. the overall HC3 system runs low on some resources or have some other 

temporary issue and crashes QAs randomly that are in some specific state (being it a http requests, a sleep, or whatnots)

 

[Joep 200]

What @jgabsays, you have to narrow your code down to the minimal that triggers the crash. Or add a debug line with every action, so you can try to catch the last action before it crashes.