Pre-requisite n00b disclaimer: I am new to Cisco managed switches and the CLI, but I have a basic understanding of what I am doing (I think).
We have a few older CNC machines that support FTP for machine file transfers. For the longest time these machines have used FTP from a server running on a local off-network PC. The two are connected via an Ethernet cable with static IPs and FTP works all the time, every time. That PC is a large time bomb, and doesn't adhere to policy, so I'm trying to get rid of it (them).
The machine has settings for IP, subnet mask, but NO DEFAULT GATEWAY.
I started the process by adding a second NIC to our Network FTP server to get it on the same VLAN and IP range so it doesn't have to traverse a different VLAN. Tried to pull a file and it worked once and never again. Unable to repeat the results I setup RSPAN to monitor the traffic with wireshark. I also attempted to setup a static route in the switch to direct the machine IP to the FTP server IP. With RSPAN I discovered that the FTP request doesn't ever come accross the port, but the machine IP is pingable on that VLAN from the ftp server. I also tested the line to the switch, and it is good. I plugged it back into the local computer, and it will transfer all day with request after request.
Before I left today, I noticed that the switch port was negotiating 10mbs (expected) with half-duplex (unexpected) to the machine. I'll verify what it is negotiating to the local computer tomorrow. The machine is old enough that the Ethernet adapter could very well be half-duplex.
I also tested the FTP between the machine and the local computer with an unmanaged switch between, and it works just fine as well.
My two problem indicators are potentially the half-duplex negotiation, and an "ethernet check failure" that shows up in WireShark during FTP/ping on the local computer, and when pinging the machine over the network.
I'm running out of ideas to troubleshoot the issue. What would you recommend as next steps?
No comments:
Post a Comment