Spark Variables not returning

General
1

Not sure if this belongs in the Cloud section or not… Basically I have a fair amount of instability when polling a given core for it’s variables (not sure if I am actually polling the core directly or the cloud, so maybe it is the cloud?). I am basically just using curl via a shell script (just like the example) to grab the Spark Variable from one of my cores and post it to RRD.

I get a fair amount of not just dropping but long dropouts, sometimes just one but very often it drops out for 5, 10 and occasionally even 20 minutes. When I see the dropouts I jump to a shell and run the curl command directly and see the same general failure to poll the variable. I don’t have the error I get offhand but I will grab it on the next drop, in total it happens maybe 5-10 times a day.

If anyone has any input it would be greatly appreciated!

2

Any idea how many variables you are pulling down @ once? Are you by chance doing something else on your core which could tie up the wireless connection (transferring files/data or something)?

3

It’s 3 variables at a time, roughly every 20seconds or so, is that too much? I am worried that if I extend that out to 60 seconds I will just get longer timeouts…

J

4

Am I overrunning the number of times I can check this per minute or?

5

Could you try spreading them a bit more? Maybe try an interval of a second between polls. Although I don’t believe it’s currently limited, it definitely won’t hurt.

6

Hi @jboswell,

Hmm, I don’t think you’re being rate limited, but it’s possible the core is briefly offline when you’re checking?

Thanks,
David