Wired Intelligent Edge

Hi,

We are doing a big remote rollout of these switches. I now have had 3 of them where it appears POE is not working on. We only have 3 !AP 105s drawing power but the ports are dark. If I do a sh poe interface command the ports that the IAP's are plugged into show (code=24).

GE0/0/22
--------
GE0/0/22: Administratively Enable, Port status: Off
Maximum power: 30000 mW, Power consumption: 0 mW
Port voltage: 0 mV, Port current: 0 mA
PD class: Class-3, Priority: Low, PSE port status: off, (code=24)

Any ideas?

Wittwem,

Code 24 indicates that there is a "Voltage injection into the port". We're planning to replace the codes with more human understandable code in a upcoming maintenance release.

How are the IAPs cabled up? Are these IAPs already in the ceiling or just on a small patch cord connected to the S1500? If in the ceiling, any chance to bring them down and direct connect the switch on a short patch cord and see if the problem persists. It could indicate an issue with the longer run.

Best regards,

Madani

Hi Madani,

The AP's are already mounted in the ceiling, plus these have been deployed to remote locations. I don't have physical access to the equipment. I did have a tech connect the AP's directly to the switch with a small patch cord and we had the same result.

Thanks,

Mark

Mark,

I would recommend opening a TAC case.

Best regards,

Madani

Hi Madani,

Yeah I already have. The first in line tech told me to reset the switch to factory defaults. I can't do it at the moment because the switch is in production. I was just checking to see if anybody out here had run into this issue and had a solution instead of resetting the switch.

Thanks,

Mark

Mark,

Have you tried removing the poe-profile and re-applying to the ports in question. That should reset the PoE settings versus doing a factory default. Also, has rebooting had any effect?

Also, we have seen this once before:

http://community.arubanetworks.com/t5/Unified-Wired-Wireless-Access/S1524-not-power-the-115-APs-connected-to-it/m-p/130491/highlight/true#M27945

But I think the issue was caused by another device (not the APs) sending power down the line. What else might be connected to the switch?

Best regards,

Madani

Hi Madani,

The upstream port is connected to a Att router which does have POE enabled. Att manages the router so I cannot change the POE settings on it. What's wierd is we have deployed about 20 of these switches with the same hardware and the same configuration and they have all worked except these 3. So far.......

Thanks,

Mark

Mark,

It may not be the router upstream. Besides the APs, what else is attached, any little switches/hubs or other devices? Any possibility you can have everything disconnected but the IAPs and see if the issue disappears then slowly add devices?

Best regards,

Madani

Hi Madani,

I had a field tech on location who unplugged patch cords one at a time. POE did eventually start working. We couldn't determine exactly what device was causing this because we had a short time frame to work in. What's wierd is that we plugged everything back in and POE for the access points continued to work. I don't know what to make of it....

Thanks,

Mark

Mark,

Let me check with engineering if we need a port down state to reset that error code.

Best regards,

Madani

An update. We lost power to the switch this weekend. Upon power up we are in the same situation as before. The only ports that have POE enabled are the ports connected to the IAP 105s. If I do a sh poe interface I am receiving the same (code=24) on those ports.

GE0/0/21
--------
GE0/0/21: Administratively Enable, Port status: Off
Maximum power: 30000 mW, Power consumption: 0 mW
Port voltage: 0 mV, Port current: 0 mA
PD class: Class-3, Priority: Critical, PSE port status: off, (code=24)

I've tried downing all the ports except the uplink and the ports for the AP's but no luck.

One other thing I've notices the PSE port status is off. On a working switch the PSE port status is ON. I'm not sure what that signifies or how I would turn it "ON"

Thanks,

Mark

Mark,

Shutting down the interface will not clear the error code if the switch is still receiving the power load which could be coming from your PoE enabled AT&T Router. Talking with engineering, they asked if you could try a ethernet cross over cable between the router and the switch. We'll handle crossing the data pairs but it may stop the PoE mis-feed.

Best regards,

Madani

Hi Madani,

I find it hard to believe that would be the problem. We have 40 other locations that have the same network connections / equipment in them without any issues. I.E. AT&T router connected to an S1500 Aruba MAS with 3 access points drawing power. Is there a manual way to clear this error code? I'm not sure if I pointed this out or not but the ports with the error code are the ones that the IAP 105s are connected to.

I do have a TAC case open, maybe they can shed some light.

Thanks,

Mark

Mark,

As long as we are detecting voltage coming into the ethernet ports we cannot clear the error condition.

However now that I re-read your original post, you are shutting the ports down remotely and not disconnecting the cabling so it may not be the Router but something else at the site. In the lab we took three APs and attached them to an S1500 which powered up fine. We then purposely injected power into on the ether ports and the APs went down with the same error condition. Once we removed that voltage source, the issue when away. Unfortunately we need to isolate what port is getting that power from.

Best regards,

Madani

Is there a command that I can run to determine this or is this a manual process of unplugging cables one at a time?

Regards,

Mark

Mark,

Unfortunately it is a manual process of unplugging the cables as far as I know.

Just to make sure there isn't something else going on with the PoE PSU, can you post the "show inventory" output?

Best regards,

Madani

Madani,

As requested:

Show Inventory
--------------


System Card Slot        : 0
SC Serial #             : CE0002037 (Date: 10/14/13)
SC Model Name           : ArubaS1500-24P
HW MAC Addr             : 00:0b:86:90:84:77 to 00:0b:86:90:84:b6
CPLD Version            : Rev: 1.12 (0x1.c)
PoE Firmware Version    : 2.4.8 (Build: 48)
CPU Assembly #          : 2010179B (Rev: 08.00)
CPU Serial #            : AD31014142 (Date: 09/28/13)
Power Supply            : Present (580W)
                        : 12V System Voltage Ok
                        : 56V PoE Voltage Ok
System Temperature      : CPU Temp                : 54  C
                        : Mid Temp                : 53  C
                        : DPE Temp                : 59  C
                        : Ambient Temp            : 30  C
System Voltages         : 12000mV Sense           : 12062.5 mV
                        : 5000mV 5VDD             : 4974.0 mV
                        : 3300mV Sense            : 3334.4 mV
                        : 900mV CPU               : 935.5 mV
System Fan Tachometers  : Power Supply Fan 0      : 4054  RPM (Low Speed)
                        : Power Supply Fan 1      : 4041  RPM (Low Speed)
                        : FAN 0                   : 7520  RPM (Normal Speed)
                        : FAN 1                   : 7692  RPM (Normal Speed)

Okay, not a PSU issue. I was interested in the 56v line:

Power Supply            : Present (580W)
                        : 12V System Voltage Ok
                        : 56V PoE Voltage Ok

Hi Madani,

I worked with some staff at the location and we unplugged ehternet cords one by one and we were able to get the AP's to power up. The device that was plugged in was an HP Thin Client.

Please note these devices do not supply nor use POE.

Thanks,

Mark

Mark,

It doesn't have to be a PoE device providing power toward us. PoE PSE devices are typically the culprit as they are capable of sending voltage down a link. Do all the sites have these thin clients? Anything different about this particular one, a unique model perhaps? Can the patch cord to this one be replaced? Perhaps its faulty and picking up some voltage from somewhere?

Best regards,

Madani

Hi Madni,

Sorry for the late follow up. Yes all the locations that we are deploying these switches / access points to have these HP thin clients. Out of the 44 sites that we have deployed to we have had 3 of them with these issues. Our solution was to move that particular unit to the AT&T multifunction router. ( It has switchports as well). It works fine when it's connected to that piece of hardware.

Thanks,

Mark

Mark,

Can you let me know what model thin client?

Best regards,

Madani

Madani,

It is a HP t5630 thin client.

Regards,

Mark

Thanks Mark.

I have a similar issue with a MAS 1500.

Each interface has the following error/code:

GE0/0/16
--------
GE0/0/16: Administratively Enable, Port status: Off
Maximum power: 30000 mW, Power consumption: 0 mW
Port voltage: 0 mV, Port current: 0 mA
PD class: Class-0, Priority: High, PSE port status: off, (code=37)

I have opened a cas with TAC but we have been unable to resolve it as of yet.

There are (5) AP 105's connected. The other (3) connections are to the firewall, controller, and the customers LAN (Which is an Allied Telesis switch).

The AP's do not power up. What does "code=37" mean?

Thanks!

-Herman

Herman,

Code=37 means "Port is off, Unknown Status (indicating Software issue)". In 7.4.0.0 we updated the outputs to show text instead of codes so I assume you are running a 7.3.X.X varient? We'll be including the same improvement in 7.3.2.4 releasing later this month.

We also updated the PoE code on the S1500s in ArubaOS 7.3.2.1 so what version of software are you currently running?

Best regards,

Madani

Thanks for the response. I'm running 7.3.2.2 on the S1500. I connected the AP's to a S2500 and I did not have an issue. I will verify the code on the S2500 and upgrade the code on the S1500 to the latest version and see what the results are.

Thanks

The version of code on the S2500 wouldn't really matter since the two platforms actually have a different version of their PoE chipset.

Upgrading to 7.4.0.1 on the S1500 would at least replace the "Code" numbers with more understandable text but since you're already running 7.3.2.2, you alrady have the PoE chipset update.

What's your TAC case number?

Best regards,

Madani

Ok, thanks. I won't waste my time looking at the S2500.

 The Case Number is: 1621590

At this point Aruba is processing an RMA. I actually requested the RMA while the engineer looked into what the code meant. It was explained that I can not have both. I can either request an RMA or wait 2-3 days for an engineer to respond to what the code meant...I decided to go the RMA route.

Herman,

I looked over the case but don't have enough information on whether this is indeed a hardware or software issue. Since you've already requested the RMA, let me know how things go with the replacement and if you're still having trouble respond on this thread and we'll grab some more info to troubleshoot.

Best regards,

Madani

Thank you.

I am troubleshooting the affected switch remotely. I did upgrade that switch to the latest version of code (7.4.0.1). I still am getting the "PSE port status: off, unknown port status".

I received the new switch and PoE does work on that device. I checked the PoE status before and after I upgraded the code version.

Herman,

I'm not sure if it is possible or not from your location but can you download the tech-support logs from the switch. You can do this either through the GUI on the maintenance tab or in the CLI you can generate them using "tar logs" and then transfer them out using TFTP/SCP/FTP.

Best regards,

Madani

Hellos Madjali

do you know what code 26 means?

we got a switch which suddenly the poe desont work

Administratively Enable, Port status: Off, Power consumption: 0 mW
PSE port status: off, (code=26)

Cheers

Carlos