[etherlab-users] master fsm datagram

[Peter P]

Thank you very much for the explanation. That helped me a lot to know where to look. I changed the frequency, which did not directly helped, but rather showed me where the error had to be due to its frequency change. The program tried to sync slave clocks even though I only use one slave.
Now there is only 3 unmatched datagrams at start-up but they don't seem to make problems at the moment.

Thank you again.

> From: gavinl at compacsort.com
> To: peter.ploetner at mytum.de; etherlab-users at etherlab.org
> Date: Fri, 12 Dec 2014 20:55:27 +1300
> Subject: Re: [etherlab-users] master fsm datagram
> 
> On 12 December 2014 19:01, quoth Peter P:
> > does anybody know how the master-fsm exactly works? The master 
> > fsm datagram in my program times out/gets skipped even before I 
> > activate the slave. I assume it has something to do with the 
> > ec_fsm_master in the fsm_masater.h, is that correct?
> > 
> > How can it be that there is a problem with a datagram that 
> > should be internal to the ethercat master?
> 
> All datagrams are external -- they're actual network packets (or technically
> fragments of packets).  It's called the "master fsm" datagram simply because
> it's the one owned and sent out by the master FSM.  As the master FSM is
> responsible for determining the number and identity of the slaves on the
> network, this is in fact the most likely one to fail if you have
> communication problems.
> 
> The most likely issue (other than hardware or wiring failure) is that you
> are trying to run faster than your network drivers/OS can handle.
> 
> 
> _______________________________________________
> etherlab-users mailing list
> etherlab-users at etherlab.org
> http://lists.etherlab.org/mailman/listinfo/etherlab-users
 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20141212/7ef5690e/attachment-0004.htm>