Temperature Reading not supported for Family

[brent.s]

I consoled into a switch that stopped working shortly after it was installed (within a month) and the following error keeps coming up on the console.

vtss_phy_chip_temp_get_private:7015 Temperature reading not supported for family: None

I don't know what this means and it fills the screen faster that I can read the results of an input command.

Any information would be much appreciated.

Brent

[sirhc]

Hi Brent,

Any information would be much appreciated.

Exactly, information is a good thing, speaking of needed information:
What model is it?

What firmware was it using?

What were the circumstances going on when it failed such as weather?

Do you have the unit back on your bench?

Have you tried holding in the reset button for 20 seconds while you apply power (it helps to be consoled into it to verify it is defaulting). After it starts the deafult process you MUST leave it set for 3-5 minutes as it needs to format a Flash partition.

Beyond that you can try a TFTP firmware receovery explained here: viewtopic.php?f=17&t=1073

If that fails your next step will be to RMA it and you can find out how to do this by reading posts #1 and #2 of this thread: viewtopic.php?f=6&t=1259

Thanks,
Chris

[brent.s]

The device failed when we attempted to plug in a second PoE device. The first device was 48V and the second device was 24V. When the second device was powered, the switch failed. It was running firmware wispswitch-1.3.9rc8, we had not updated to the newer 1.3.9 Stable.

There was not abnormal weather during this time. Just cool temperatures.

I do have the unit removed from the production network and at my desk where I can console in as needed.

I did reset the unit to factory default, but the error message continues to scroll on the screen.

I will work on the TFTP recovery of the unit, but I am not sure if it will work.

[sirhc]

Well try your best to do TFTP recovery, next step is read the RMA thread

[brent.s]

I got the following when I was going through the steps to recover the unit.

admin@Netonix_Switch:/tmp# for i in `cat /proc/switch/status | grep -v Down | cu
t -b 1-2`; do ./switch -t $i; done
cat: can't open '/proc/switch/status': No such file or directory
admin@Netonix_Switch:/tmp#


Please let me know if I entered the command wrong.

I don't see any switch folder in the /proc directory.

[brent.s]

The unit won't take the Firmware recovery. It drops during the XMODEM tranfer of the .bin firmware file.

I am going to be submitting an RMA form.

[jclendenan]

Did you get to the bottom of this? I just had a unit start to do it, and sadly don't have a spare to swap in.

We had a AF24, and a nbe-ac-19 on the same tower, grounding and transdectors on it. and the switch was power cycled, and the ethernet stopped working on the radios.

Then I noticed we were pulling 27w on the nanobeam cable, and 52 on the af24 cable.

Arrived onsite - restarted the switch, totally won't light up any phy interface on the switch,

Running on the console, I get the same issue. 3 days into the site, and it's already had a outage.

The nanobeam is shorting out the psu as well, so it's toast, but i wouldn't have expected it to take down a whole phy chip.

Thanks,

James

[sirhc]

The nanobeam is shorting out the psu as well, so it's toast, but i wouldn't have expected it to take down a whole phy chip.

Thanks,

James

This sounds like the switch was damaged due to a surge?
So what ever happened took out a radio and you do not understand how it could have taken out the switch as well?

How is your site grounded?

Here are some good posts on grounding:
viewtopic.php?f=30&t=188
viewtopic.php?f=30&t=1429

[jclendenan]

Thanks Chris.

I went through all the grounding info on the site, and the only thing that it might be related might be that the nano was directly into the switch, and everything else was bonded through a pair of transdectors and the RF Armour stand is also grounded down using #6 to the building ground.

The only place I can think of was that the building ground and the mast ground on the roof might be slightly different potentials as I can't see exactly where they are tied together.

I'll send the switch back, and you can give me an RCA for what went wrong depending on what part of the board got fried.

Full reset didn't do anything and the whole SOC PHY appears to have gone bad enough to no longer light network ports. Only serial console seems to be working now. (SFP's un-tested)

Thanks,

James