RPDO message timestamp issue

[gunstr]

Hi,

I'm using the canopen library in an application configured as a local node. To detect loss of communication to the master I check the time delta between consecutive RPDO,s that are being sent from the master as discussed in #286 . It works well, but as the system has no RTC I have seen issues when the clock is updated by the NTP client which in most cases happens shortly after boot but in some cases can happen also at a later stage of operation.

Do you have any suggestion how to handle this in a better way than permanently disabling NTP (that is what I have done now). One option would be to configure the timestamp to use time.monotonic instead but what I can find out it's not suported by the underlying python-can library.

I would appreciate to hear your thoughts on this!