On Wed, Apr 26, 2017 at 8:21 PM, John Andersen wrote:

Maybe if you could include a link to the actual device it would help.

http://www.aaeon.com/en/p/bus-truck-box-computer-boxer-6301vs It is interesting how the PoE 802.3 devices are listed under IO and not Ethernet. But they are Ethernet. Or at least one of them is. The intel Ethernet devices are both working fine. As they usually do.

That is an unusual combination of features.

A usb connection to a 4 port PoE device ? I've Googled and the closest I could find was a two port USB32000SPT.

I agree. Something looks fishy. I think I should see 4 USB devices listed. Not just one. All I see is: Bus 001 Device 002: ID 8087:8008 Intel Corp. Bus 002 Device 002: ID 8087:8000 Intel Corp. Bus 003 Device 002: ID 0bda:8152 Realtek Semiconductor Corp. Bus 003 Device 003: ID 05f3:0405 PI Engineering, Inc. Bus 003 Device 004: ID 145f:01d3 Trust Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub

Are you sure that USB connection that you mention only in passing isn't the PoE power input port?

It is not a USB connection that we have. It is 4 RJ-45 connectors. The USB<->R8152 part is on the motherboard

Are you sure this doesn't contain a internal switch, in which case you wouldn't know about the Mac Addresses of the other ports unless you plugged into each one sequentially?

If there was only one RJ-45 maybe. But there are 4 physical connectors for these ports. And we are supposed to have 4 ports. I suspect what we really have is a 4-port PoE switch connected to a single 8152 chip. Not what we were expecting! Time to contact our friendly salesperson... -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Thu, Apr 27, 2017 at 10:27 AM, John Andersen wrote:

On April 26, 2017 11:52:01 PM PDT, Roger Oberholtzer wrote:

...

On Wed, Apr 26, 2017 at 8:21 PM, John Andersen wrote:

...

Maybe if you could include a link to the actual device it would help.

http://www.aaeon.com/en/p/bus-truck-box-computer-boxer-6301vs

It is interesting how the PoE 802.3 devices are listed under IO and not Ethernet. But they are Ethernet. Or at least one of them is. The intel Ethernet devices are both working fine. As they usually do.

...

That is an unusual combination of features.

A usb connection to a 4 port PoE device ? I've Googled and the closest I could find was a two port USB32000SPT.

I agree. Something looks fishy. I think I should see 4 USB devices listed. Not just one. All I see is:

Bus 001 Device 002: ID 8087:8008 Intel Corp. Bus 002 Device 002: ID 8087:8000 Intel Corp. Bus 003 Device 002: ID 0bda:8152 Realtek Semiconductor Corp. Bus 003 Device 003: ID 05f3:0405 PI Engineering, Inc. Bus 003 Device 004: ID 145f:01d3 Trust Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub

...

Are you sure that USB connection that you mention only in passing isn't the PoE power input port?

It is not a USB connection that we have. It is 4 RJ-45 connectors. The USB<->R8152 part is on the motherboard

...

Are you sure this doesn't contain a internal switch, in which case you wouldn't know about the Mac Addresses of the other ports unless you plugged into each one sequentially?

If there was only one RJ-45 maybe. But there are 4 physical connectors for these ports. And we are supposed to have 4 ports.

I suspect what we really have is a 4-port PoE switch connected to a single 8152 chip.

Not what we were expecting! Time to contact our friendly salesperson...

You're right about the Pie being listed as io ports. I suspect they are ports for transcievers or antennas or something.

The spec sheet says only two Ethernet ports.

The others are 802.11af, which is for some "white space" wifi sort of stuff.

https://en.m.wikipedia.org/wiki/IEEE_802.11af

Looks like stuff for cop cars, or roving instant wifi networks.

It is off that there are 4 RJ-45 connectors for these 802.11af ports. Why a connector if it is wireless. I am obviously missing some information... Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Sat, Apr 29, 2017 at 10:09 AM, Per Jessen wrote: We wanted 6 ports. Or at least 4 ports. Some of the devices we use are GigEVision cameras. Each requires a GBit Ethernet interface (with as big an MTU as the interface supports) as they are high resolution/frame rate. We have a number of systems with multiple PoE ports. Each has it's own interface. I don't see the description of this system saying that the 4 PoE is a switch connected to a third Ethernet interface (aside from the 2 intel ports). But that seems to be the case. Aargh! -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, May 2, 2017 at 10:28 AM, Per Jessen wrote:

No, I don't see that described either.

So, you have two Intel GigE interfaces, with separate RJ45 sockets. Is the Realtek R8152 interface for the POE switch then? That chip only does 10/100. It's also USB2, so max 480Mbit/s.

That was our initial concern. The purpose of this system is to try to eliminate the Electrolux TravelPower we use to generate A/C off the engine. We can do this now that we have removed the last bit of A/C equipment. And to try to eliminate as many cables and connectors as we can. We have already used PoE with our GigEVision cameras. So that is not new. Finding that the PoE is just one interface was not a happy surprise. PoE in a vehicle environment has it's gottchas. If the equipment that is powered has a long startup time (like a high-end GPS with an IMU), you must wait that additional time each time you restart the PC supplying the power. Despite our instructions, the vehicle users cannot shake their Windows bad habit of restarting the computer all the time... -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Tue, May 2, 2017 at 12:18 PM, Per Jessen wrote:

I guess a separate POE switch is not an option?

The goal here was to eliminate devices. And the dread cables that they bring with them. Cables and connectors account for the majority of failures. Vehicles are not the nicest environment for these things. In fact, that is one of the things about the computer: all components are soldered or otherwise permanently mounted. Our current need to occasionally open up a computer and jiggle the various bits so they are once again properly seated should be a thing of the past. The advent of intelligent systems in vehicles has led to systems that meet our needs becoming available. But wanting built-in swapable disk bays, ooddles of Ethernet ports, and a couple real RS-232 ports (if you expect to be a Stratum 1 time server) narrows the field. This is being tested in a new system configuration we are introducing. The downside to this approach is, of course, that this becomes a bigger point of failure. Taken as a whole, we expect it to fail less often. And since it is easier to replace this component when a failure occurs than locating what is amiss in a maze of separate wires and components, we expect easier field maintenance and less overall downtime). Or at least that is the theory... -- Roger Oberholtzer -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org