[Ovmsdev] Can buses stop after some time

[Tom Parker]

On 08/07/18 04:53, Greg D. wrote:
>
> Is there a way to compare sent and received data, to see what was 
> lost, for example, were they 3 random single frames, or two in a row?  
> Might give an idea of the size of the distraction that's causing the 
> overflow.

This is important, checking for correctness during performance testing 
is often overlooked!

> Do we have any timing data from Tom's vehicle?  Just wondering what 
> the inter-frame gaps are, and such.  Since it locks up so quickly, a 
> Wireshark trace would be wonderful to pick through, or perhaps even 
> replay.

https://carrott.org/pcaps/leaf-2016-car-driving.pcap.bz2

I'm not sure of the time resolution of my can bus hardware (one of these 
https://www.mictronics.de/projects/usb-can-bus/ ), the timestamps look 
quite funky, it seems that they count up in 10uS increments and then 
jump 10ms. So I think maybe it only has 10mS resolution. If that is the 
case they appear to come pretty fast, maybe 22 frames per 10ms.

I haven't tried to play this recording back into the OVMS yet, but it is 
quite long. I didn't have the OVMS plugged in at the same time because I 
can't capture the car bus and feed it into the OVMS's can2 interface at 
the same time. Hence I don't know when or if can2 would stop working. I 
could also try recording on the OVMS itself but I haven't been able to 
make the sdcard work on my 3.0 unit, and recording there may well 
influence the problem without capturing it's cause.