TimeoutExceptions and stack configuration and/or loading?

[JavaLaurence]

Hi,

 

Is it possible that "fat" stacks with multiple masters and loads of bricklets driven by an aggressively *polling* program can be the cause of TimeoutExceptions?

 

I managed to get my Wifi signal strength to -60dB, and I still get way too many timeouts:

 

 

Starting Cellar Water Pump Control System @ Sun Mar 24 22:30:26 CET 2013

Running on Java 1.7.0_04 hosted by Mac OS X 10.7.5

Initializing TinkerForge stack..

Aha. Connected @ 21:30:26 because: CONNECT_REASON_REQUEST

Done.

[1] Ouch.. exception @ Sun Mar 24 22:30:46 CET 2013

[2] Ouch.. exception @ Sun Mar 24 22:31:10 CET 2013

[3] Ouch.. exception @ Sun Mar 24 22:31:39 CET 2013

[4] Ouch.. exception @ Sun Mar 24 22:32:08 CET 2013

[5] Ouch.. exception @ Sun Mar 24 22:32:21 CET 2013

[6] Ouch.. exception @ Sun Mar 24 22:32:40 CET 2013

[7] Ouch.. exception @ Sun Mar 24 22:33:37 CET 2013

[8] Ouch.. exception @ Sun Mar 24 22:33:43 CET 2013

[9] Ouch.. exception @ Sun Mar 24 22:34:29 CET 2013

[10] Ouch.. exception @ Sun Mar 24 22:34:46 CET 2013

[11] Ouch.. exception @ Sun Mar 24 22:36:03 CET 2013

[12] Ouch.. exception @ Sun Mar 24 22:36:21 CET 2013

[13] Ouch.. exception @ Sun Mar 24 22:36:29 CET 2013

[14] Ouch.. exception @ Sun Mar 24 22:36:41 CET 2013

[15] Ouch.. exception @ Sun Mar 24 22:36:45 CET 2013

[16] Ouch.. exception @ Sun Mar 24 22:36:57 CET 2013

[17] Ouch.. exception @ Sun Mar 24 22:37:02 CET 2013

[18] Ouch.. exception @ Sun Mar 24 22:37:13 CET 2013

[19] Ouch.. exception @ Sun Mar 24 22:37:50 CET 2013

[20] Ouch.. exception @ Sun Mar 24 22:38:11 CET 2013

Switching pump ON @ Sun Mar 24 22:38:43 CET 2013. Time since last: 22:38:43,959. Air: 1012.27,Temp: 12.81,Humid: 62.3,Light: 0.0,Heap: 2.93 MB,

[21] Ouch.. exception @ Sun Mar 24 22:39:14 CET 2013

[22] Ouch.. exception @ Sun Mar 24 22:39:21 CET 2013

[23] Ouch.. exception @ Sun Mar 24 22:39:45 CET 2013

[24] Ouch.. exception @ Sun Mar 24 22:39:48 CET 2013

[25] Ouch.. exception @ Sun Mar 24 22:39:53 CET 2013

[26] Ouch.. exception @ Sun Mar 24 22:40:14 CET 2013

[27] Ouch.. exception @ Sun Mar 24 22:40:49 CET 2013

[28] Ouch.. exception @ Sun Mar 24 22:40:59 CET 2013

[29] Ouch.. exception @ Sun Mar 24 22:41:26 CET 2013

[30] Ouch.. exception @ Sun Mar 24 22:41:30 CET 2013

[31] Ouch.. exception @ Sun Mar 24 22:41:32 CET 2013

[32] Ouch.. exception @ Sun Mar 24 22:41:47 CET 2013

[33] Ouch.. exception @ Sun Mar 24 22:41:51 CET 2013

 

The main new thing which I changed was hook up my LCD20x4.. and I'm updating the whole screen at 1Hz.. am I overloading the Master Brick with all the polling and LCD stuff?

 

[borg]

1hz should not be enough to overload anything, what do you mean with aggressively polling?

[JavaLaurence]

By "aggressively polling" I mean I'm polling all the bricklets in my stack every second, and updating the LCD to reflect the new values. I know 1Hz is peanuts, but I don;t know how you guys implemented your onboard software, so...

 

I've just written a much more sophisticated exception logging mechanism which will help me analyse the TimeoutException sources more easily. So far, it looks as if the Distance IR bricklet is the source of 98% of the exceptions.. and that just doesn;t make sense. What's so special about it? Maybe we're on to something 

[batti]

Mh, can you change the order of the Bricklets/Bricks? Does this change the behaviour?

[JavaLaurence]

I don't like plugging & unplugging bricklets too much, coz those connector pins look so incredibly fragile, but if you give me a hint at how to distribute my bricklets over my two Masters, I'll gladly give it a try.

 

My config is as shown in screenshot.