Richard, Marc, Jonathan, Erik, Dave:

On Monday 23rd the CHETA slow controls chassis was installed in the Vacuum Prep Lab (VPL). We connected its ECAT ethernet port to the MSR Beckhoff chassis AUX-1 port partially by copper (VPL to Rm163 into CUR) and partially by fibre (CUR CDS edge switch to MSR edge switch).

This appeared to work, but later we found out that the CDS core switch was complaining constantly of rapid movement of a MAC address on the ports used by CHETA. Further reading suggests ECAT does not strictly support unique MAC addresses per Beckhoff terminal, and it appears they may all have the same non-standard MAC address.

In addition, since Monday we have been seeing regular dropouts of the CUR network switch stacks, every 6 hours at 00:41, 06:41, 12:41 and 18:41 for about a minute apiece.

This afternoon we ran a long 148' cat5e ethernet cable from room 163 to the far end of the MSR. At 16:01 we routed the CHETA chassis connection to use this cable, we now have contiguous copper run between the two chassis.

This will fix the MAC movement issues on the CDS core, and time will tell if if fixes the CUR switch dropout.