Devices with only USB console port - Need a Console Server Solution

Miles Fidelman mfidelman at meetinghouse.net
Wed Feb 3 16:00:41 UTC 2016


Lantronix makes some serial_port-to-IP remote management devices, as 
well as KVM-over-IP.  Some of  them support USB interfaces.  Be warned - 
you probably want to run these through a VPN gatway (most of these 
things are built on old versions of embedded linux - there have been 
compromises - the same applies to IPMI boards - been bit by this 
personally).

Miles Fidelman

On 2/3/16 10:48 AM, Christopher Morrow wrote:
> On Wed, Feb 3, 2016 at 2:18 AM, Erik Sundberg <ESundberg at nitelusa.com> wrote:
>> Digi has something called USB Anywhere.  http://www.digi.com/products/usb-and-serial-connectivity/usb-over-ip-hubs/anywhereusb
>>
> #fail
> "COMING SOON: Security features, such as SSL and SNMPv3"
>
> :(
>
> "Creates systems redundancy and increases security"
>
> unless you consider ssl and snmpv3 security relevant I guess? Also of interest:
> "10/100 Mb switched Ethernet"  I hope your local-in-pop switch gear
> has 10/100/1000 and not just 1000 ports. This may be more problematic
> as the future progresses... (you can't get 100mbps ports on a qfx if I
> recall correctly, for example)
>
>> However I would like to limit the amount of equipment we deploy at a pop, the majority of our pop's don't have servers... Just Routers, Switches, Console Servers, and your other Network Hardware.
>>
> 'console server' is, in one view of the world, now 'usb console server' ...
>
>> The problem with USB is you can only wire a USB 2.0 Cable up to 15' (Per Google)....  And you have to purchase a cable premade.
>>
> this is a fairly salient point :( If I don't have a console server in
> each rack (or pair of racks) but as a row element, now I have
> significantly shorter row length before I can't console anymore.
>
>> Where as with a Serial Console you can go around 100', not to mention about just about everyone has a crimper, rj45 ends, and cat5 cable, to run and make cables as needed.
>>
> maybe the ubiquity of usb consoles will drive this i the right
> direction as well?
>
>> Assuming something is broke...With USB let's say you rely on remote hands to do a lot of work in the colo's. First they need to find a *Working Laptop*, then you have to walk the tech through downloading the drivers and installing them on there laptop. Hoping they have permissions to install software on there laptops. Plus if it's really broke and you get no output, you will never be sure if it's USB related or not. Where as serial it's just going to work, and it's easy to test to see if it's working on not by hooking up to another device.
>>
> my guess is that most / all tech's have a usb-serial dongle at this
> point, because who's laptop has serial ports anymore natively onboard?
>
> mostly you're outlining 'operational practices and norms are not
> accounted for yet in the usb-console design' right? which either is:
>    1) get out and write procedures/documentation for how this all should work
>    2) call back to 2005 and demand no usb in consoles on network equipment
>
> I don't think 2 is feasible :( but 1 sure is... Also, it's sort of
> funny to me that servers don't seem to be going this route?
>
>
>>
>>
>>
>>
>> -----Original Message-----
>> From: NANOG [mailto:nanog-bounces at nanog.org] On Behalf Of Dovid Bender
>> Sent: Tuesday, February 02, 2016 8:33 PM
>> To: Robert Drake <rdrake at direcpath.com>; NANOG <nanog-bounces at nanog.org>; nanog at nanog.org
>> Subject: Re: Devices with only USB console port - Need a Console Server Solution
>>
>> Why not use udev rules so the ports are persistent? I did that on a pi that I was using as an ice cast box. Based on the usb audio port on reboots I know which device is which stream.
>>
>>
>> Regards,
>>
>> Dovid
>>
>> -----Original Message-----
>> From: Robert Drake <rdrake at direcpath.com>
>> Sender: "NANOG" <nanog-bounces at nanog.org>Date: Tue, 2 Feb 2016 20:03:22
>> To: <nanog at nanog.org>
>> Subject: Re: Devices with only USB console port - Need a Console Server  Solution
>>
>>
>> On 2/2/2016 5:02 AM, Bjørn Mork wrote:
>>> No inside pictures :)
>>>
>>> Assuming that this is really an USB device, and that the console port is
>>> really an USB host port, it would be useful to know the USB decriptors
>>> of the device.  You wouldn't be willing to connect it to a Linux PC and
>>> run "lsusb -vd", would you?
>> I'm inconveniently consoled into one via a combination of remote desktop
>> into windows -- linux console on a virtual machine -- screen
>> /dev/ttyACM0.   Because of this posting lsusb -vd is taxing.
>>
>> Linux has full support for the device.  It sees it as cdc_acm.
>>
>> The vendor id is 0x04e2 (Exar Corp).  Product ID is 0x1410.   I've got
>> two connected right now.  This is in our lab and the windows box is
>> temporary.  Our intention is to use a raspberry pi for the terminal server.
>>
>> I'm obviously not in front of it, but I'm wondering if they can be
>> enumerated by something other than when they were plugged in. That's my
>> biggest hurdle for making a console server for them.. how to figure out
>> what router is connected to which USB port after a reboot, or someone
>> getting unpluggy with cables.
>>
>>> Bjørn
>>>
>> Robert
>>
>> ________________________________
>>
>> CONFIDENTIALITY NOTICE: This e-mail transmission, and any documents, files or previous e-mail messages attached to it may contain confidential information that is legally privileged. If you are not the intended recipient, or a person responsible for delivering it to the intended recipient, you are hereby notified that any disclosure, copying, distribution or use of any of the information contained in or attached to this transmission is STRICTLY PROHIBITED. If you have received this transmission in error please notify the sender immediately by replying to this e-mail. You must destroy the original transmission and its attachments without reading or saving in any manner. Thank you.

-- 
In theory, there is no difference between theory and practice.
In practice, there is.  .... Yogi Berra



More information about the NANOG mailing list