I'm currently trying to establish a remote console connection to a Cisco 3750x switch via a tripp lite console server B096 doing RS232 over fiber conversion. I have several remote switches directly connected to this TRIPP Lite, and those all work fine using putty and reverse telnet or ssh to the tripp lite port that they are attached or via the tripp lite web interface.

My problem device is a 3750x switch in another building a few miles away, which I have attached a Serialcom Fiber to RS232 convertor (Single Mode).

There is one at each end of the fiber trunk between buildings with one end connected with straight thru serial connection to Tripp lite port and the other end connected with rollover to Cisco Switch. This configuration worked fine in my lab with just like a 10 meter fiber patch cord between the two converters and default Cisco and Tripp Lite console port settings.

But now that I have implemented it in the real scenario with the fiber run about 1 mile away; not working so well.
Even though Serialcom says convertor is good for distances of 12.4 miles.

I am able to make a connection to the switch after authenticating through the terminal server. However, it takes a very long time 3-5 minutes to return the switch login prompt. However, the switch login prompt just scrolls by and will not stop to allow entry of the password. It just repeatedly scrolls, as if someone is enter password but nothing being typed.

Both ends of the connection are set at 9600-8-N-1. I have experimented with the baud and stop bits settings to no avail, and either never returns switch login prompt or just does the same thing.

I not sure if the is a line/Fiber quality issue or Serial setting issue causing the problem; and not much of an expert when it comes to serial connections; and serialcom folks no help.

If anyone have any insight or ideas in terms of setting/configuration that could be causing this and or tweaks. I would be much appreciated.

 

I do wired serial but not FO and I'm not an IT guy so I don't even know what a "rollover to a Cisco" is. But by your own admission you have all the means of doing troubleshooting by substitution, to whit, "This configuration worked fine in my lab with just a 10 meter fiber patch cord between the two converters and the default Cisco and Tripp Lite console port settings."

So what's different from the lab to the installation?

- 1 mile of fiber run

- is the far end device a field device, different than the test unit that was tested in the lab?

- which other devices are field devices, different than those tested in the lab?

- How do the settings vary on the field devices from the lab test devices?

- are the test lab FO/232 converters the same ones used in field?

See where I'm going? The test lab devices worked OK. Why doesn't the same stuff (except the FO run) work in the field like it did in the lab?

Why don't you gather up all the gear used in the lab (and the near end field gear), including the "10 meter fiber patch cord" and carry it all to the far end. Then set up all the lab gear and prove that the lab gear all works at the far end with the 10 meter patch cord.

Start substituting field devices in the test setup until the system 'breaks' which indicates which device is causing the problem or test until you determine that the only thing you can't substitute is the 1 mile fiber run and then the FO is the problem.

One gotcha in comm devices is that many comm devices need to be power cycled when a setting is changed in order to recognize the changes. Caveat Emptor.

 

Hi,

Make sure that you 1 mile fiber connection is working OK. Need to do a loss test on each fiber of the cable. Hopefully you installed a multi-connductor fiber cable for such this situation. Also it should be Single Mode Fiber with the same Fiber Optic Gbics installed at each end.

Fiber connections are very reliable, but if there is an issue with one of the fibers in the bundle, it will not work at all.

Hope this helps,
Bob

 

Thanks for taking the time to respond.

So the short answer to all your questions/comments regarding troubleshooting...would be YES to all the above. everything is the same minus the extended fiber run.
In fact the same convertors were sent to the remote location for and used.

So, why can't I pick up the lab and move it to the field location. It's on a different continent a few thousand miles away.

Replacing "field devices" until something breaks --- Not an option "Mission Critical" live Network.

Was hoping that someone may have insight to RS232 comm settings over long haul that I may try. Been through dozen combination to no avail.

 

HI there

Check your sp modules if applicable are capable for shorthaul/longhaul, singlemode, multimode, etc. There a many different variants of modules dependent on the distance, throughput, conditions and application. You may be over/under shooting the target. You may also need to tweak your comms settings. The issues your experiencing with your terminal app sounds like a comms setting is not right, possibly baud rate.

If your sp modules are not suitable for the distance then this will effect your through put and drop packets and cause knock on effects.

My tuppence worth.

Happy fault finding.

Regards.
Gary

 

Are both fiber optic runs the same? 10 m and 1 mile both multi-mode or both single mode?
Do your FO drivers match the installed fiber?

Multi-mode still seems to be the default automation choice.

I would be surprised if the same single mode FO drivers worked with 10m and 1 mile runs. I have seen overpowered SM fiber circuits fail because of excessive reflections in the fiber.