Spark Cloud Timeout / Core Request Limit?

There have been several times today where I received a 408 timeout from the cloud API when making calls to retrieve a variable value. I was just toying around with pulling the analog value from a thermistor. It seemed to happen more often when I would poll every 2 seconds, but it would also happen if I increased the polling interval to 5 or 10 seconds. Even with the sketch whittled down to the basics (no sampling/averaging), I still get a 408 timeout from the cloud API.

I know my Internet connection is good (fiber, 100Mbps up/down, “Gig City” FTW) as is my wifi. I monitored both when I was getting the timeouts. Is this a problem with the API or the Spark or maybe both? I’ve included the sample sketch below.


int temp_raw = 0;

void setup() {
pinMode(A0, INPUT);
Spark.variable(“temp_raw”, &temp_raw, INT);

void loop() {
temp_raw = analogRead(A0);

A 408 when reading a variable means the API server didn’t hear back from the Core within a reasonable time. This is often the result of the Core reconnecting to the server or Wi-Fi or otherwise being momentarily unreachable.