We are using a CCR1016-12G (version 6.18) as our main router for one of our remote sites. It is connected on ether2 to a GE cisco switch. At times, the router would show the port as “running”, but we would have no communication with any devices (including the switch) behind it. As this is a remote site, our only solution is to reboot the entire CCR router. This situation happened 3-4 times and would happen about once every 2-3 weeks. Checking our log files, we found that there were TCS mismatches being logged. We then removed auto-negotiation on both the router and the switch and set both to full-duplex, 100 Mbps. This seemed to solve the issue until this morning. Our Solarwinds server caught a message saying that the speed had changed on the CCR router back to 1GBps and we dropped connection to all devices behind it again. We checked the ethernet settings and it was still set to no auto-negotiation and still at 100 Mbps. We rebooted the router to regain access to our switch, and it also showed no auto-negotation and 100 Mbps. Any ideas as to what could be causing this issue? We’ve upgraded to 6.25 in an attempt to solve this issue, however after checking the change-logs, nothing shows any issues with layer 1 connections.