[etherlab-users] Reasons for Loss of All Slaves with no Erros

[Henry Bausley]

I had this type of  problem and in my case it turned out the NIC adapter
was buggy and switching to a new NIC resolved the issue.
https://www.mail-archive.com/etherlab-users@etherlab.org/msg02722.html


The way we found the problem was

1.) Log the time when ecrt_master_recieve and ecrt_master_send . This
will indicate perhaps an error in your app. (In our case  the time
ecrt_master_receive/ecrt_master_send were called were perfect with very
low jitter.)

2.) Monitor with wireshark to guarantee data is being sent every cycle
(in my case data was not on the wire even though ecrt_master_send was
called).  This pointed to hardware

I also received suggestions from this forum to check Lost Link Counter
Register (0x0310:0x0313) and  RX Error Counter Register (0x0300:0x0307)
but that did not help in my case.



On Fri, 2016-07-29 at 15:43 +0930, Mike O'Connor wrote:

> Hi All
> 
> I'm having a real problem where the machine loses all slaves modules, 
> but I can not find any reason for this. The are no cable issues or 
> packet error that I can see.
> 
> We are using the generic interface and our scan rate is 1000/sec, the 
> machines runs happy for sometimes hours/a full day then suddenly it will 
> say 0 slaves responding. Our code will then stop scanning the bus as a 
> safety step.
> 
> Jul 29 09:55:27 SixHeadGrab kernel: [176187.764409] EtherCAT WARNING 0: 
> 4 datagrams UNMATCHED!
> Jul 29 09:55:27 SixHeadGrab kernel: [176187.796046] EtherCAT WARNING: 
> Datagram ffff880137e42a98 (domain0-0-main) was SKIPPED 4 times.
> Jul 29 09:55:27 SixHeadGrab kernel: [176188.201052] EtherCAT 0: Domain 
> 0: 7 working counter changes - now 25/25.
> Jul 29 09:59:57 SixHeadGrab kernel: [176457.840059] EtherCAT 0: 0 
> slave(s) responding on main device.
> 
> We get very few UNMATCHED datagrames but do get the on occasion. As you 
> can see they do not match up as there is almost 5 minutes between the 
> last UNMATCHED and the lost of the IO.
> 
> I'm looking for ideas on what can cause this or what steps I might take 
> to identify the issue.
> 
> Thanks
> Mike
> _______________________________________________
> etherlab-users mailing list
> etherlab-users at etherlab.org
> http://lists.etherlab.org/mailman/listinfo/etherlab-users




Outbound scan for Spam or Virus by Barracuda at Delta Tau

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20160729/d078dd52/attachment-0004.htm>