Wireless Access

Greetings,

 

We are installing our 1st AP275.  We have many AP175s installed useing Cisco 2960 POE switches using 2X power. These 175s are all working fine with POE+.  However, on the AP275, the radio keeps shutting down and the field tech says he gets a message (via console port) saying something like, "radio shutting down due to low power'.  I cant remember the exact message, but he is heading back out again today.  I overnighted him a mid-span external POE injecter (PowerDsine 9001GO) to try again today.  Does anyone have any other suggestions?

 

Thanks!

 

Mike

Did the installer use the ethernet cable glands?  From the install Guide:

 

 

Yes, The Glands were installed properly.  We even had him patch the 275 in right at the switch and got the same message.

What does the power output at the switch say for PoE details?

Seth,

 

I will post the statement.  Do you know what command to use to get this statement?

consumption g1/0/6
Interface  Consumption      Admin
           Configured    Consumption (Watts)
---------- -----------  -------------------

Gi1/0/6       YES                30.0

Mjotech,

 

Thank you for that information.  Please report back when the POE injector is installed.

OK,  we get the same alarm with the injector.  The tech consoles in, the gets the statement;

Shutting down Eth1 due to insufficient power".  Then says "shutting down Radio due to insufficient power". 

FYI.. We have a ticket open with TAC, but they dont have a clue either.

Incedentally, we have 5 AP275s at this site.  They all have the same error.

Mjotech,

 

Same switch platform?

 

Let's see what happens with the injector on the same cabling.

Same Cisco switch.  At the switch with two patch cables.  One going to data in of the POE injector and one in the data out of the POE injector.  Powering up the injector we get the same messages.

With the PowerDsine 9001GO, we still get;  Shutting down Eth1 due to insufficient power".  Then says "shutting down Radio due to insufficient power.  We have TAC on our conference bridge and he is as lost as we are right now.  Has anyone else seen this?

Mjotech,

 

I will let others answer about their experiences and let you continue to work with TAC.  Do you mind PMing me the ticket#?

TAC ticket 1541539.  Vasue (spell) is the tech.  Not sure we are making progress.  that is why Im reaching out to the community. Thanks to all for helping us through this as we are on day two.

What version of ArubaOS are you running, and what is the length of the cable plugged into the injector?  Can you get a reasonably short cable and try the same thing?

 

FYI - 

 

PoE+ is not 802.3at although people use them interchangeably.

 

PoE+ can mean 802.3af with higher power output.

 

That implies that you could have an output voltage as low as 44V.

 

If it is 44V and you attach an AP-27x to a long 24 gauge cable the voltage at the AP is out of range and will not operate.

 

Also note that our Aruba switch produces 56v on all ports

I undersatnd. However, we are working right at the switch.  We have hundreds of Cisco 2960 switches deployed.  I'm not certain what the output VDC is on the Cisco 2960, but working wight at the switch without a station cable should provide at least -48VDC to teh Access Point.

Right now we are working at the switch with just cat 6patch cords.

Mjotech:

 

The unit needs fully compliant 802.3at power 275 draws 23W .  The AP 175 only draws 16.

 

PoE+ is not necessarily 802.3at compliant.

 

What is the cable length?

Can you please elaborate on the statement, "PoE+ is not necessarily 802.3at compliant"?  I thought it was the same thing.

the two terms are used interchangably but they are not really equivalent.

 

Initially PoE+ was PoE with more output power.

 

PoE allows source ends (PSE) to operate with output voltages as low as 44V and clients on PoE(+) needed to accept voltages as low as 37V

 

There was a recognition in the standardization process that these were too low andlead to excess power losses in the cables.

 

802.3at specifies that the PSE must provide at least 50V at the source end and clients need to accept voltages as low as 42.5V

 

Thus a PoE+ device operating at the bottom of its voltage range (44V) has no chance of powering an 802.3at compliant device at the other end since the ohmic loss witll drop the voltage below tolerable limits.

 

802.3at also went on to indicate that the standard assumes 23 or 22 gauge wire and specs a maximum ohmic loss of 12.5 ohms.

 

Most standard cable (cat5 and cat 6) is still assembled with 24 guage wire. there are lots of good reasons for this. So to compensate you need to purchase device tht provide PSE voltages of 56V (typical with better quality sources) and you need to ensure that your cable runs are clean home runs with no degradation in the cable. patch panel and couplers incur additoinal power and voltage losses.

 

the indication that the switch is showing 30W drawn indicates that their is too much ohmic loss in the ethernet cable feeding the AP. this indiates 7W dropped across the cable  (which is a lot) and if the supply is under 54V at the source the voltage at the AP will be insufficient. (i.e. <42.5V). Bad connector crimping can impact contact losses at the two ends as well.

 

So, in short all 802.3at can be called PoE+ but not all PoE+ is 802.3at.

After reading through the Cisco literature, they state (below in bold)

 

Cisco Catalyst 2960-S Series Switches with LAN Base and LAN Lite Software

Cisco Catalyst 2960-S Series Switches with LAN Base Software:

● 10 and 1 Gigabit Ethernet uplink flexibility with Small Form-Factor Pluggable Plus (SFP+), providing business continuity and fast transition to 10 Gigabit Ethernet

● 24 or 48 ports of Gigabit Ethernet desktop connectivity

● Cisco FlexStack stacking module with 20 Gbps of throughput, allowing ease of operation with single configuration and simplified switch upgrade

● PoE+ with up to 30W per port that allows you to support the latest PoE+ capable devices

● Power supply options, with 740W or 370W fixed power supplies for PoE+ switches are available

 

Cisco Catalyst 2960 and 2960-S Series Switches Q and A

 

 

Q. What are the benefits of Power over Ethernet Plus?

 

A. Switches supporting Power over Ethernet Plus (PoE+) can supply power to any IEEE 802.3af or IEEE 802.3at compliant end device. PoE+ removes the need for wall power to each PoE+ enabled device and eliminates the cost for additional electrical cabling that would otherwise be necessary in IP phone and wireless LAN deployments. PoE switches also eliminate the need for power injectors and PoE midspans for powering IP devices. The list of devices that require more than the 15.4W provided by 802.3at is growing. Devices currently requiring more than 15.4W of PoE power include Cisco video IP phones and IP video cameras. Additional devices requiring more than 15.4W of power are coming, and the Cisco Catalyst 2960-S Series is leading the way to allow these devices to be deployed in the network.

Cisco Catalyst 2960-S Series Switches support IEEE 802.3at (PoE+) devices.

 

Q.How many end devices can the Cisco Catalyst 2960-S Series provide PoE and PoE+ to?

 

A.The Cisco Catalyst 2960S-48FPD-L and C2960S-48FPS-L can support 48 PoE ports with full Class 3 PoE power of 15.4W each, for a total output capacity of 740W. All Cisco Catalyst 2960-S Series Switches with PoE capability support PoE+. A complete list of Catalyst 2960 and 2960-S Series Switches, with their PoE capacities, is given in Table 1.

Table 1.PoE Capacity of the Cisco Catalyst 2960 and 2960-S Series Switches

Aruba TAC is still investigating and havent hearsd back since Friday.  We will escalate this today.  One of our engineers wanted to know if the AP275 can shut off LLDP??  Does anyone know if it could?

Yes, you can turn off (edited) the transmission of LLDP PDUs.

 

Go to AP Configuration, select the group that has your AP 275s, click AP, then ETH 0 then down to AP LLDP and uncheck PDU transmissions. Remember to hit apply!

We think this got it.  Not turning off LLDP on the controller (that did nothing), but turning LLDP on in the Cisco switch. After a couple minutes the AP275s came up.

---

@Mjoetech wrote:

We think this got it.  Not turning off LLDP on the controller (that did nothing), but turning LLDP on in the Cisco switch. After a couple minutes the AP275s came up.

---

I had the same exact issue with 2 AP 275 and Cisco 3750X switches.

 

I tried manually configuring the switch port to 30W and that did nothing to enable the radios on the AP.

 

The APs come up after I enabled LLDP on the switch.

Okay. What is the version of AruaOS?

ArubaOS 6.4.0.3

with an AP power draw of 23W and 100m of 24awg cable and temperature of 30C (86F)

Assumes home run with no breaks or couplers or patch panels in the run

    expect 18.5 ohms of resistance

 

Voltage at switch  Power Drop  Total Power Draw  Voltage at AP-270

50                                  6.8W                  29.8W                           39.2V Out of bounds for 802.3at

51                                  6.4W                  29.4W                           39.9V Out of bounds for 802.3at

52                                  5.9W                  28.9W                           41.3V Out of bounds for 802.3at

53                                  5.6W                  28.6W                           42.6V Marginal for 802.3at

54                                  5.3W                  28.3W                           44V     OK

 

If the voltage is low, or the run has extra ohmic losses (Patch panels, couplers, corroded wiring) then the AP-270 will not power up.

 

802.3at spec for full length was designed assuming 22 AWG cable which for 100m has a resistance of about 10.5 ohms

Mjoetech,

Thank you for sharing this with the community.

I dont get the Thanks.  Our engineers do!  But thanks just the same. :smileyhappy:

 

Mike J. Hall

General Growth Properties

You just have to enable LLDP on the switches!! :D

Thanks to YOU Mr DZ0.  You get the credit for figuring this out.  Thanks!!

Thanks, now Aruba needs to make the AP275 function if LLDP isn't available. When we forced the POE to static - it didn't help because they were relying on LLDP - EVEN if we disabled LLDP on the AP..

Message from Aruba engineering teams. I engaged them yesterday.

 

This is apparently driven from Cisco side

 

"Yes, cisco switch requires the PDs use CDP(for cisco PD) or LLDP (for non-cisco PD) to negotiate more than 15.4W power."

 

http://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst3750x_3560x/software/release/12-2_58_se/configuration/guide/3750xscg/swint.html#wp1874579

 

"The initial allocation for Class 0, Class 3, and Class 4 powered devices is 15.4 W. When a device starts up and uses CDP or LLDP to send a request for more than 15.4 W, it can be allocated up to the maximum of 30 W."

Thanks for engaging the engineering team.  We have a lot of these AP275s to put in and its nice to know we have the support.

I had a similar issue two weeks ago while performing a site survey with an AP-275. Many things gone wrong that day (had to hook the AP on customer's LAN since it was not an Instant AP, LLDP was missing on their switch template,...), and as usual, I did not have time to troubleshoot the whole thing onsite so I got the AP-275 powered with AC to get things done.

 

Now, reading this post got me thinking a bit more about the problems I encountered and the one described. 

 

I am fortunate enough to have built myself an 802.3at compliant power pack (more on that in a couple weeks) for my site surveys that lacks LLDP. How can we get around this? I now understand this might be mandatory, not only for Aruba Networks products, but for all 802.3at powered devices. 

 

Will it be possible for an Aruba Networks employee to fill a Request For Enhancement to circumvent the LLDP negociation? I don't think this should be restricted to AP-275 since the other APs uses the same LLDP negociation as well.

 

Please refer to PGallant as he has other similar issues over the last weeks as well.

 

Keep on your good work,

 

Hello,

 

Did you had any progress with the LLDP issue ? We are using Avaya PoE+ switch and I feel we have the same issue.

AP-275 shown as inactive in the controller and when the AP boots up, you will 1st get the output :

 

[   55.508012] Shutting down eth1 due to insufficient POE voltage [power profile 2]

 

And 1 minute after only :

[  113.660074] Enabling eth1 due to power change [power profile 1]

 

Really strange behavior...

 

Thanks

From my point of view, it seems that your issue is similar but different that what is reported in this tread.

 

Could this be a feature (!!!) of Avaya switches? Usually, Avaya switches powers Avaya phones (...) and these phones are power hungry.

 

I am unable to remember the manufacturer's name (might be Avaya or another company they bought) but they had a feature in their boot process to prevent PoE powered devices to request and obtained their power allocation simultanously. Some power supplies I saw requires to be warmed up before providing their maximum.

 

Have you tried powering an AP-275 on an already running switch? How fast it boots?

Thanks Simon for your reply. The Avaya switch is actually running (it is actually a stack of 4 switches) and we have AP93Hs and AP105s on those switches. I will try normally tomorrow to go there with a PoE injector 802.3at and see if it helps.

Let me know the outcome. I suspect it might be not sufficient since most (if not all) midspan injectors lacks LLDP support.

For the record, I was not able to power my AP-275 for the previously mentioned site survey using a midspan injector.

For those wanting an RFE filed as a work around regarding LLDP, it is best to work through your regional account reps and SEs so that the RFE is tied to specific accounts. RFEs filed without it are usually non-starters. 

 

Regarding any possibility of a workaround, note that EJohnson's covers in great detail the spec difference between POE+ and 802.3at. In many cases if you are hitting a suppled voltage issue, it's a PSE issue that cannot be solved by the AP since it's designed per the 803.2at spec regarding required voltage. It's not always just a wattage/power issue. The best thing would be to use 802.3at certified power injectors in cases where you hit limitations on older switch hardware or hardware not able to suport 802.3at. 

All

 

LLDP needs to be enabled on the switch side. The AP cannot do anything about that.

 

Cisco for sure, and probably others, restrict the amount of power available to 802.3af unless LLDP is enabled

 

Injectors do not operate the same way. 802.3at compliant injectors provide full power without needing LLDP. The 3at handshake simply applies in that case.

 

the AP-27x series draws 23W at the port on the device.

 

With 100m of 24 AWG cat5e cable the power draw at the source end is

 

So you need to insure that your midspan injector is actually 802.3at compliant and capable of providing 26W in this case.

 

 

Hi,

 

First, the switch is 802.3at capable. LLDP is enabled everywhere. After the AP boot and after 1 minute only you can see from the conroller :

 

(rcont1) #show ap debug system-status ap-name 6c:f3:7f:**** | begin "Power Status"
Power Status
------------
Operational State          POE-AT: No restrictions (Overridden by LLDP)
-----------------          --------------------------------------------
Current HW State           POE-AF: Ethernet port 1 disabled; All radios disabled;
LLDP Negotiated POE Power  25.5W

Hello again,

 

So, for information today I did the test with the PoE injector 802.3at capable. And the result was good as the AP worked just fine with it.

Now I can see one difference in the Power Status of the AP. Wihtout the PoE injector so I am getting that :

 

Operational State	POE-AT: No restrictions (Overridden by LLDP)
Current HW State	POE-AF: Ethernet port 1 disabled; All radios disabled;
LLDP Negotiated POE Power	25.5W

 

With the injector I am getting :

 

Operational State	POE-AT: No restrictions
Current HW State	POE-AT: No restrictions
LLDP Negotiated POE Power	25.5W

 

I am working with the Aruba support as well on that issue...

Hi,

 

For information, Aruba has found the problem and they are working on a fix.

Please be aware that this was addressed in 6.4.2 and beyond to ensure that the controller is updated with a change in powering status.

UPDATE:

 

After working for the past week with Aruba Tac we determined that this issues exists in 6.4.0 and 6.4.2, but does "NOT" exist in 6.4.1.  The issues is identified by Bug#106096 and will be address in the upcoming release.

This should be fixed in 6.4.2.1 which was released last Friday

Thanks Seth.  I just ran into this exact thing.

 

Its back!

I have a case where the customer is running the latest release AOS 6.4.3.7 and is having this issue

i also have a customer on AOS 6.4.3.4 with the issue

Both are frustrated, both say the AP's had been running for a year before they started acting up

on one the upgrade to aos 6.4.3.7 triggered it and the other they claim it just happened

 

Open a TAC case, see what logs can be pulled from the controller, etc. It's not common at all that APs working for a year with no change would just up and fall off again.

All symptoms with the issue where pointing toward the Cisco IOS version of the switch.

 

I have numerous AP-270 series running on 6.4.3.7 without any issues, in most of the worst scenarios. I have to admit they takes a longer time to boot and to have their config updated. Other than that, it is fairly stable.

 

Do not forget that you can bypass the PoE by using an AP-270 series AC patch cord.

 

Can you post the output of the following, if the switch is a Cisco of course?

 

sh version

sh lldp neighbors

sh power inline (port)

 

Regards,

Student_Apt_Bld3_4fl_18_245#sh version 

Cisco IOS Software, C2960X Software (C2960X-UNIVERSALK9-M), Version 15.0(2)EX5, RELEASE SOFTWARE (fc1)

Technical Support: http://www.cisco.com/techsupport

Copyright (c) 1986-2014 by Cisco Systems, Inc.

Compiled Fri 21-Feb-14 05:54 by prod_rel_team

 

ROM: Bootstrap program is C2960X boot loader

BOOTLDR: C2960X Boot Loader (C2960X-HBOOT-M) Version 15.2(2r)E1, RELEASE SOFTWARE (fc1)

 

Student_Apt_Bld3_4fl_18_245 uptime is 1 week, 4 days, 19 hours, 55 minutes

System returned to ROM by power-on

System restarted at 15:44:14 EDT Sun Apr 24 2016

System image file is "flash:/c2960x-universalk9-mz.150-2.EX5/c2960x-universalk9-mz.150-2.EX5.bin"

 

 

This product contains cryptographic features and is subject to United

States and local country laws governing import, export, transfer and

use. Delivery of Cisco cryptographic products does not imply

third-party authority to import, export, distribute or use encryption.

Importers, exporters, distributors and users are responsible for

compliance with U.S. and local country laws. By using this product you

agree to comply with applicable laws and regulations. If you are unable

to comply with U.S. and local laws, return this product immediately.

 

A summary of U.S. laws governing Cisco cryptographic products may be found at:

http://www.cisco.com/wwl/export/crypto/tool/stqrg.html

 

If you require further assistance please contact us by sending email to

export@cisco.com.

 

cisco WS-C2960X-48FPD-L (APM86XXX) processor (revision F0) with 524288K bytes of memory.

Processor board ID FCW1905B3S7

Last reset from power-on

2 Virtual Ethernet interfaces

1 FastEthernet interface

150 Gigabit Ethernet interfaces

6 Ten Gigabit Ethernet interfaces

The password-recovery mechanism is enabled.

 

512K bytes of flash-simulated non-volatile configuration memory.

Base ethernet MAC Address       : BC:F1:F2:C1:38:00

Motherboard assembly number     : 73-15964-02

Power supply part number        : 341-0527-01

Motherboard serial number       : FOC19052W21

Power supply serial number      : DCB1903610L

Model revision number           : F0

Motherboard revision number     : C0

Model number                    : WS-C2960X-48FPD-L

Daughterboard assembly number   : 73-14200-03

Daughterboard serial number     : FOC19053QP9

System serial number            : FCW1905B3S7

Top Assembly Part Number        : 800-41461-01

Top Assembly Revision Number    : C0

Version ID                      : V02

CLEI Code Number                : CMMLJ00ARB

Daughterboard revision number   : A0

Hardware Board Revision Number  : 0x12

 

 

Switch Ports Model              SW Version            SW Image                 

------ ----- -----              ----------            ----------               

*    1 52    WS-C2960X-48FPD-L  15.0(2)EX5            C2960X-UNIVERSALK9-M     

     2 52    WS-C2960X-48FPD-L  15.0(2)EX5            C2960X-UNIVERSALK9-M     

     3 52    WS-C2960X-48FPD-L  15.0(2)EX5            C2960X-UNIVERSALK9-M     

 

 

Switch 02

---------

Switch Uptime                   : 1 week, 4 days, 19 hours, 56 minutes 

Base ethernet MAC Address       : BC:F1:F2:85:65:80

Motherboard assembly number     : 73-15964-02

Power supply part number        : 341-0527-01

Motherboard serial number       : FOC19052WAL

Power supply serial number      : DCB19036263

Model revision number           : F0

Motherboard revision number     : C0

Model number                    : WS-C2960X-48FPD-L

Daughterboard assembly number   : 73-14200-03

Daughterboard serial number     : FOC190505K1

System serial number            : FCW1905B3NA

Top assembly part number        : 800-41461-01

Top assembly revision number    : C0

Version ID                      : V02

CLEI Code Number                : CMMLJ00ARB

Daughterboard revision number   : A0

 

Switch 03

---------

Switch Uptime                   : 1 week, 4 days, 19 hours, 56 minutes 

Base ethernet MAC Address       : BC:F1:F2:84:FB:00

Motherboard assembly number     : 73-15964-02

Power supply part number        : 341-0527-01

Motherboard serial number       : FOC19052X0C

Power supply serial number      : DCB190363JK

Model revision number           : F0

Motherboard revision number     : C0

Model number                    : WS-C2960X-48FPD-L

Daughterboard assembly number   : 73-14200-03

Daughterboard serial number     : FOC190503PW

System serial number            : FCW1905B3F5

Top assembly part number        : 800-41461-01

Top assembly revision number    : C0

Version ID                      : V02

CLEI Code Number                : CMMLJ00ARB

Daughterboard revision number   : A0

 

Configuration register is 0xF

 

Student_Apt_Bld3_4fl_18_245#

 

__________________________________________________________________________________________

 

Student_Apt_Bld3_4fl_18_245#sh lldp neighbors 

Capability codes:

    (R) Router, (B) Bridge, (T) Telephone, (C) DOCSIS Cable Device

    (W) WLAN Access Point, (P) Repeater, (S) Station, (O) Other

 

Device ID           Local Intf     Hold-time  Capability      Port ID

Stu                 Gi2/0/41       110        W               aca3.1ec0.6c5c

Stu                 Gi2/0/37       95         W               aca3.1ec0.6c78

Stu                 Gi2/0/40       103        W               aca3.1ec0.6c7e

 

Total entries displayed: 3

 

Student_Apt_Bld3_4fl_18_245#

 

___________________________________________________________________________________________

 

 

Student_Apt_Bld3_4fl_18_245#sh power inline gigabitEthernet 2/0/41

Interface Admin  Oper       Power   Device              Class Max

                            (Watts)                            

--------- ------ ---------- ------- ------------------- ----- ----

Gi2/0/41  auto   on         30.0    Ieee PD             4     30.0 

 

Interface  AdminPowerMax   AdminConsumption    

             (Watts)           (Watts)           

---------- --------------- --------------------  

 

Gi2/0/41              30.0                 30.0

Student_Apt_Bld3_4fl_18_245#sh power inline gigabitEthernet 2/0/37

Interface Admin  Oper       Power   Device              Class Max

                            (Watts)                            

--------- ------ ---------- ------- ------------------- ----- ----

Gi2/0/37  auto   on         30.0    Ieee PD             4     30.0 

 

Interface  AdminPowerMax   AdminConsumption    

             (Watts)           (Watts)           

---------- --------------- --------------------  

 

Gi2/0/37              30.0                 30.0

Student_Apt_Bld3_4fl_18_245#sh power inline gigabitEthernet 2/0/40

Interface Admin  Oper       Power   Device              Class Max

                            (Watts)                            

--------- ------ ---------- ------- ------------------- ----- ----

Gi2/0/40  auto   on         30.0    Ieee PD             4     30.0 

 

Interface  AdminPowerMax   AdminConsumption    

             (Watts)           (Watts)           

---------- --------------- --------------------  

 

Gi2/0/40              30.0                 30.0

Student_Apt_Bld3_4fl_18_245#

None of those are in reach of getting a laptop with a USB cable is there? Can you run a show log all | include the 275 AP-Name or mac (or both) to see if anything in the logs show a problem. Are they running as Inactive or do they not even show up in the controller?

 

Have you opened a TAC case yet? The IOS version and AOS versions are both new enough that it shouldn't be a firmware issue. your show LLDP neighbors LOOKS right, assuming those mac addresses and ports are from the three inactive 275s.

Sorry to have took so long to answer.

 

My customer is using Cisco 3750X switches, running 15.0(2) SE6 that was compiled on Apr 9, 2014. We had problems with earlier versions that were related to LLDP-MED negociation.

 

Compare the results of the command "show power inline (port)" of one AP-275 that works and from one that doesn't. You'll probably find it only draws 15.4W, which is insufficient. Per example, an AP-225 (which is a close parent of the AP-275) draws 22.4W. The AP-275 should draw as much if not a little more than an AP-225.

Running into similar issues with an AP-275 and a Cisco SG300-PP. 

Definitely an LLDP issue here as I get the following:

 

Power Status
------------
Operational State POE-AF: Ethernet port 1 disabled; All radios disabled;
----------------- ------------------------------------------------------
Current HW State POE-AF: Ethernet port 1 disabled; All radios disabled;
LLDP Negotiated POE Power 0.0W

 

The switch shows the following:

Port Powered Device State Status Priority Class
-------- -------------------- ---------------- ------------ -------- ---------
gi8 Auto On critical class0

Time range:
Power limit: 30.0 W
Inrush Test: Enable
Legacy Mode: Enable

Port Status: Port is on - valid capacitor/resistor detected
Overload Counter: 0
Short Counter: 0
Denied Counter: 0
Absent Counter: 0
Invalid Signature Counter: 0

 

LLDP is turned on the switch:

 

LLDP state: Enabled
Timer: 30 Seconds
Hold multiplier: 4
Reinit delay: 2 Seconds
Tx delay: 2 Seconds
Notifications Interval: 5 Seconds
LLDP packets handling: Filtering
Port State Optional TLVs Address Notifications
--------- ----------- -------------------- ----------------- ---------------
gi8 Rx and Tx SN, SC None Disabled

 

Anyone by any change using these switches? I only have one AP on this switch.

 

Thanks

I am having the same problem.  I tried to re-provision with no changes but that did not work for me.  My AP is actually an IAP-275.  my Switch is ExtremeNetworks Summit X440-48p-10G.  When I enable LLDP on the switch port the IAP-275 will power up the radios and Eth1 ports and broadcast the Instant SSID and I can connect to it with my wireless devices.  As soon as I convert it to a Campus AP to work with my Aruba 7210 Controller I start having issues with the Radios and Eth1 not powering up.  Show ap debug system-ststus ap-name WES-Football275 | begin "Power Status" shows the Eth1 and Radios as off and LLDP Negotiated POE Power  25.5W

 

Power Status
------------
Operational State POE-AT: No restrictions (Overridden by LLDP)
----------------- --------------------------------------------
Current HW State POE-AF: Ethernet port 1 disabled; All radios disabled;
LLDP Negotiated POE Power 25.5W

 

If I use a POE Injector it works fine or if I revert it back to an IAP it works fine.

#7210

From an Aruba SE.

 

It turned out to be an Extreme config issue.  Like Cisco, LLDP had to be configured:

 

Try configuring LLDP:

enable lldp ports <port #>

configure lldp port <port #> advertise vendor-specific dot3 power-via-mdi with-classification

Configure lldp port <port no> advertise vendor-specific med capabilities

disable inline-power port <port no>

enable inline-power port <port no>

I have already enabled and configured the LLDP on the Extreme switch.  You are right in the fact that it needs enabled and configured.  However it only works on my IAP-275 as an Instant AP.  Once I convert it to a Campus AP on my 7210 Controller it stops working.  Same switch, same configuration, same AP just only works as an IAP not a converted AP. It did not work at all untill I enabled and configured LLDP.  Why does the AP function differently with LLDP as an IAP vs AP?

#7210

there is an issue in 6.4.1 that if the turn up time is too long the AP does not update the controller when it switches.

 

So, upgrade to 6.4.2 latest will address that.

 

If the LED is green and you see

 

Operational State       POE-AT: No restrictions (Overridden by LLDP)

Current HW State      POE-AF: Ethernet port 1 disabled; All radios disabled

 

then you are are good to go. the operational state is the governing statement and you can ignore the HW state.

---

---

 

If the LED is green and I see

 

Operational State       POE-AT: No restrictions (Overridden by LLDP)

Current HW State      POE-AF: Ethernet port 1 disabled; All radios disabled

 

But when you look at the controller under configuration it shows an I flag for inactive and under monitoring it shows no band for either radio and it does not broadcast any SSIDs to be able to connect to.  So I am assuming that the radios are turned off otherwise I could see the SSIDs and the frequency/Band would show up and there would not be an I flag.  I am using code version 6.4.2.0 and going to upgrade to 6.4.2.1 in about a half hour.

one more update.

 

This is a known issue and has been targetted for correction in 6.4.2.2 and 6.4.3

 

The underlying cause is that if the LLDP power negotiation runs long the AP can sucessfully power up and turns the LED green but the controller is not made aware of the update in the power status.

 

A status push from the AP to the controller is the fix that is being implemented and this should address what you are seeing.

We run 6.4.2.2

and have the same problem with our 19 AP-275 APs and HP ProCurve switches.

 

My workaround...

1. Disable "PDU transmission" and "Reception of LLDP PDUs“ for eth0 of the AP group

2. Manually set the (reserved) PoE value for the switch port and disable LLDP

 

On a HP ProCurve switch ...

interface 17

   name "WLAN-AccessPoint"

   power-over-ethernet high

   poe-allocate-by value

   poe-value 33

 

All best

Martin

Time out !

 

"We run 6.4.2.2 and have the same problem with our 19 AP-275 APs and HP ProCurve switches."

 

Is this problem should have been fixed in 6.4.2.2? From what I understand, it is still present with the HP switches.

 

I thought the fix would cover other switches than the Cisco too.

 

Regards,

 

 

Simon

---

@weinema wrote:

We run 6.4.2.2 and have the same problem with our 19 AP-275 APs and HP ProCurve switches.

 

My workaround...

1. Disable "PDU transmission" and "Reception of LLDP PDUs“ for eth0 of the AP group

2. Manually set the (reserved) PoE value for the switch port and disable LLDP

 

On a HP ProCurve switch ...

interface 17

   name "WLAN-AccessPoint"

   power-over-ethernet high

   poe-allocate-by value

   poe-value 33

 

All best

Martin

---

 

I'm running into these issues as well with the 275.  I'm runnin 6.4.2.2.

Me also.. 

6.4.2.2 with Avaya 4850 GTS switches.

The reprovision fix does not help. 

 

#ODL1_01
Power Status
------------
Operational State POE-AT: No restrictions (Overridden by LLDP)
----------------- --------------------------------------------
Current HW State POE-AF: Ethernet port 1 disabled; All radios disabled
LLDP Negotiated POE Power 25.5W

There wasa bug fix in 6.4.2.3.

 

Bud ID: 106096

 

Symptom: The radios on an AP-270 Series access points were not enabled after receiving power

through a PoE+ source

 

The radio pushes an update to the controller.

 

Debugging comments:

 

LLDP needs to be on

 

With it on you need to ensure that the LED on the unit is NOT persistent RED.

 

If the LED goes off there is a bug that applies when the negotiation takes too long the radio may fail to notify the controller. This is fixed in 6.4.2.3

                A soft boot of the 275 may clear this.

We have a similar situation.

 

Upgraded our 7220 to AOS 6.4.2.3, however we're still not able to bring up AP-275 radios.  

 

We're getting PoE+ power from a stack of Cisco 3750x's.  

 

We do not have LLDP enabled, however we have taken PoE limitations out of the picture.  

 

Does anyone know if there is a PoE+ doc or a specific Cisco Doc that explictly states LLDP needs to be enabled for devices requiring 802.3at power?  

 

I've heard of a few folks that have had success by using LLDP, just wondering if that is an absolute requirement, & how that comes into play when using midspan PoE+ injectors. 

 

I've opended a TAC case, but we'll see how far I get before the holidays.  

 

It would also be of great value if the AP-275, or any 802.3at AP, could report how much power it is seeing from its PSE. 

 

Thanks, 

 

#7220

We had the same problem but with Extreme Network switches.  Once we turned on and configured LLDP the AP would work as an IAP but not when converted to a campus AP.  Once we upgraded the Controller code to 6.4.2.3 the AP works in both IAP and Campus AP mode from the controller as long as LLDP is configured and enabled on the switch.  I have seen other post saying Cisco also has to have LLDP turned on.  PoE+ injectors do not need nor do they have LLDP.  They just work. (Most of them anyway)

LLDP is explicitly required in order for a cisco device to. Provide more than 15W

Upgrading tonight, will see how it goes.

 

Edit: all good, stable APs for 24 hours. So 6.4.2.3 solves the issues for Avaya 4500 switches with LLDP enabled.

---

@weinema wrote:

We run 6.4.2.2

and have the same problem with our 19 AP-275 APs and HP ProCurve switches.

 

My workaround...

1. Disable "PDU transmission" and "Reception of LLDP PDUs“ for eth0 of the AP group

2. Manually set the (reserved) PoE value for the switch port and disable LLDP

 

On a HP ProCurve switch ...

interface 17

   name "WLAN-AccessPoint"

   power-over-ethernet high

   poe-allocate-by value

   poe-value 33

 

All best

Martin

---

Hi!

 

This fixed it for me too on an HP 2910al-24G PoE+ Switch with an AP325.

I did not change the eth0 configuration on the Mobility Controller (7005).

The cold boot of the AP325 works. It shows up 'Power: 802.3af POE during the initial boot, but after 22s it detects the power availability: 'Enabling eth1 due to power change [power profile 1]'.

(I'm running ArubaOS 6.4.4.9). Also good to know the meaning of the color of the power led (yellow: insufficient power).

 

Switch Port Power Status:

 

HP-E2910al-24G-PoE(config)# show power 3

 Status and Counters - Port Power Status for port 3

  Power Enable      : Yes
                                       LLDP Detect       : enabled
  Priority          : high              Configured Type   :
  AllocateBy        : value             Value             : 33 W
  Power Class       : 4                 Detection  Status : Delivering

  Over Current Cnt  : 0                 MPS Absent Cnt    : 3
  Power Denied Cnt  : 0                 Short Cnt         : 0

  Voltage           : 55.0 V            Current           : 120 mA
  Power             : 25.0 W    Pre-std Detect : off

HP-E2910al-24G-PoE(config)#

 

Thank you weinema!

 

Manfred

 

 

#AP325

Hello Everyone,

 

I need some help here. I am having this issue with AP225, 7005 controller and a Cisco 3850 715 Watt switch. I tried everything in this thread and I cannot get these CAP's to come up with the full power mode. Has there been any more info on this? And why doesn't the AP come up when you dedicate the power to 30w? Seems that the voltage is there so why not reponse from the AP? 

 

Aruba Controller

Name:	Aruba Operating System Software.
Model:	Aruba7005-US
Version:	6.5.0.1

 

 

Cisco 3850:

 

Chassis id: 04bd.88cc.e122
Port id: 04bd.88cc.e122
Port Description: eth0
System Name: 04:bd:88:cc:e1:22

System Description:
ArubaOS (MODEL: 225), Version 6.5.0.1 (56440)

Time remaining: 118 seconds
System Capabilities: B,W
Enabled Capabilities: W
Management Addresses:
IP: 10.10.111.100
Auto Negotiation - supported, enabled
Physical media capabilities:
1000baseT(FD)
100base-TX(FD)
100base-TX(HD)
10base-T(FD)
10base-T(HD)
Media Attachment Unit type: 30
Vlan ID: - not advertised
PoE+ Power-via-MDI TLV:
Power Pair: Signal
Power Class: Class 4
Power Device Type: Type 2 PD
Power Source: PSE
Power Priority: unknown
Power Requested: 20200 mW
Power Allocated: 0 mW

 

Global LLDP Information:
Status: ACTIVE
LLDP advertisements are sent every 30 seconds
LLDP hold time advertised is 120 seconds
LLDP interface reinitialisation delay is 2 seconds

 

 

Cisco IOS Software, IOS-XE Software, Catalyst L3 Switch Software (CAT3K_CAA-UNIVERSALK9-M), Version 03.06.00E RELEASE SOFTWARE (fc3)

 

I read that someone had to upgrade the swith to 3.06.03 to get it to work. Has anyone had this issue on a Cisco 3850? 

Here >>> 

 

https://www.reddit.com/r/networking/comments/402ont/cisco_switch_poe_config_with_aruba_access_point/ 

 

It works like a charm on my old HP v1910 24G PoE switch. never thought that adding 300 watts would not be enough LOL

 

Thanks

Joseph

 

 

Joseph,

 

Do you have any other PoE devices (cameras, phones...) connected to the same switch ?

 

(..)

PoE+ Power-via-MDI TLV:
Power Pair: Signal
Power Class: Class 4
Power Device Type: Type 2 PD
Power Source: PSE
Power Priority: unknown
Power Requested: 20200 mW
Power Allocated: 0 mW

 

I'm wondering if the version of IOS-XE that you are running have the same LLDP features bugs as I ran into. It may sound a bit silly but would it be possible to upgrade the switch/stack to newest IOS at least to clear out that possibility?

Hi Simon

Thanks for the reply. No I do not just two 225's. I plan on upgrading this switch tomorrow. But what really bothers me is that I can set the switch to power inline max 30000 and Aruba doesn't work. That sucks. The problem is it doesn't have enough power. OK no problem,I will give you 30w and you still won't accept it? Why?

 

Thanks again. 

Joseph

LLDP-MED is essential for PoE+ power negotiation. If LLDP is broken in your version of IOS-XE, you won't get any power out of that switch.

 

While testing some of the first AP-275s, we could not figured out what was wrong. We ended up looking into and matching the debug logs of the AP-275 and the Cisco 3750X we had to find the issue. Anyway, one of our LAN admins pointed out the issue on a Cisco support forum while I ran into another forum with a post related to a similar PoE+ issues but with security cameras. It did not took me a long time to add up the numbers...

 

Let us know the outcome.

Hi Simon,

 

That is interesting so no LLDP-MED no power? What about setting the max manually as I mentioned? You say that there will not be 30w on the cable if I do this? Has anyone tested that with a tool?

 

Also further reading on Cisco software release a guy said that he up graded to the version I have and it broke his Avaya phones. He has to go back to 3.3. I am upgrading tonight to 3.6.5E I think this is the fix too. 

 

I will test the cable before I do this though I want to see it with a meter.

Thanks I will update you as soon as I know.

Joseph

Joseph,

 

Please accept my apologies. LLDP-MED is essential for PoE+ (802.3at). Regular PoE (802.3af) uses a resistance based mechanism to detect if power over Ethernet is required, so LLDP is not used for PoE first gen devices.

 

You don't want to provide PoE to all ports all the time. Specifiying the PoE maximum power only applies to what amount of power a device will receive if it request PoE. Per example, you might want to limit all ports to 15.4W to equalize the load on each port, whatever the powered device requested. 

 

The very first AP220 series firmwares required full PoE+. Now, you can run these AP220 from normal PoE, with limited functionnalities. (Which makes me think about which firmware version are you using for these AP225?)

 

Never the lest, I would not run any AP220 and newer out of regular PoE. It just does not makes any sense to not upgrade if the switches cannot meet this requirement.

Hi Simon

 

Yes they work they just have the amber power ligh on indication limited functionality. 

 

I agree on not wantting to put full power on all ports but for the situation of just two AP 225 or just to test that the AP is gettting enough juice isn't that why you have the commands to set priority and max power?

 

I also agree I never intended to use af PoE that is why I chose this switch and the previous one too. 

 

Thanks for the help and the information. Definitley learned something which is good.

 

The 802.3af and 802.3at protocols requires discovery and negociation. No  network manufacturers will go against implementing such fuctionnality that will break its protocol compliancy.

 

None of us can claim it never misconnect a network cable. Per such, the discovery and negotiation parts of these protocols are essentials to protect not only the switch ports but the connected equipments as well.

Hello Simon

Just finished the upgrade to cat3k_caa-universalk9.SPA.03.06.05.E.152-2.E5.bin and it fixed it! Makes more sense now that you explained why but I hope this helps someone else!

 

Joseph

Hi Joseph,

 

I'm glad to read this. The release of the IOS-XE version you were running matched in time the one we were using at the time. I could not confirm that version of IOS-XE had the bug but the dates of various IOS (-XE) versions that I knew about were matching with the ones I was sure about.

 

Lastly, please give me kudos if you think I deserve them.

 

Keep on your good work.

Hi Simon

 

Yes this is good news. here is the switch and IOS XE

 

Switch Ports Model SW Version SW Image Mode
------ ----- ----- ---------- ---------- ----
* 1 32 WS-C3850-24P 03.06.05E cat3k_caa-universalk9 INSTALL

 

• 03.06.00E  will break the lldp
• versions before will not 
• 03.06.03E is good too.

Joseph,

 

Thanks for the info.

 

Let"s hope that other members of this Community will share the version numbers that are known to be working as well as the ones that does not.

 

I'd like to go a bit further by adding this list should not be specific to Cisco as it should include all other manufacturers as well.

Hello everyone,

 

Sorry for bumping the thread, I didn't find another thread related to my issue.

 

I replaced a HP ProCurve 2520 by an Aruba J9776A 2530-24G, which connects 4 IAP-228.

 

My IAP-228 on this switch are connected through POE+ Power Injector, and I am facing the same issue :

 

"Power Supply                : POE-AF
LLDP Power                  : 0.0W
Current Operational State   : Ethernet port 1 disabled; All radios disabled;"

LLDP is enabled for sure ; show lldp info remote 1 on my switch :

 

"
 LLDP Remote Device Information Detail

  Local Port   : 1
  ChassisType  : mac-address
  ChassisId    : 40 e3 d6 c9 xx xx
  PortType     : mac-...
  PortId       : 40 e3 d6 c9 xx xx
  SysName      : Borne_5
  System Descr : ArubaOS (MODEL: 228), Version Aruba AP
  PortDescr    : eth0

  System Capabilities Supported  : bridge, wlan-access-point
  System Capabilities Enabled    :

  Remote Management Address
     Type    : ipv4
     Address : 10.200.0.9
"

 

Switch and IAPs are running last firmware :

 

"

Aruba Operating System Software.
ArubaOS (MODEL: 228), Version 6.5.3.0
Website: http://www.arubanetworks.com
(c) Copyright 2017 Hewlett Packard Enterprise Development LP.
Compiled on 2017-05-24 at 22:08:47 AST (build 59775) by p4build
FIPS Mode :disabled

"

 

 

Is there any command to disable LLDP check while APs are booting on Instant OS?

Does anybody found a reliable workaround?

 

Thanks

Jeremy

 

 

 

One important thing to be aware of, is that power injectors are by nature passive. So if there is an 802.3at injector between the switch and AP, but the switch is enabled for 802.3af, the injector will not enable AT power and will instead pass on the AF power (which will not be enough to power the 228). If you have an AT enabled POE switch, you would need to change the default POE settings to enable LLDP and allow AT power levels. 

 

Hi Jerrod

 

Thanks for your feedback but I have the same issue on Aruba 2530 NOT POE...

According to what you said, my POE injectors are inevitably 802.3at because some APs have booted correctly

 

I don't know what to conclude...

If the POE injector fires up and powers the AP before the switch port and POE fire up (or if the switch is rebooted while the POE injector is connected), then the POE injector will power the AP first. 

 

Are the HP switches you are using 802.3at capable or no? That model has both 802.3at and non-802.3at models, so if you have the J-numbers that will help.

We just purchased an AP275 and behold we had the same issue with Cisco switches.  The AP gets classified as a class 4, but only negotiates 15.4W. Even after enabling LLDP on the switch, the issue was still present.  To get it to pull the proper power, we had to set the power to static on the port.

 

Switch# config ter

Switch (config)# lldp run

Switch (config)# interface gi6/5

Switch (config-if)# power inline static max 30000

 

After doing this we verified on the controller that the AP was negotiating according to LLDP 25.5W.

 

 

****UPDATE****

After our group looking at the issue, one engineer noticed that even though the AP was receiving a Class 4, it was not being marked with the Capability flag on the switch. After looking around the LLDP settings on port 0 on the controller, we enabled the "Capabilities" option under the default LLDP profile for port 0.  We removed the static power config from above and it all worked.  The port was flagging the AP port with a W and it was getting the proper power automatically.  You do still need LLDP enabled of course on the switch.

 

 

Thanks Seth.

 

Unfortunately, in the release note of the 6.4.2.1 I don't see anything related at the LLDP issues with the AP-275 and the dev team working on that case hasn't came back to us yet.

But anyway, in the time being we are using PoE Injectors as temporary solution.

 

JB

I have the exact same problem with HPE Aruba 2920-48G-PoE+ 740Watts switch. AP275 goes into Monitor mode. Logged a TAC case also. WLC is a Aruba7030 running on 6.4.2.3.

 

Tried to power up the AP with an external PoE injector (PowerDsine 9001GR PN: PD-9001GR/AT/AC) but no success. Below is the config in the switch ports. There are other Aruba AP's (2 x 225's & 3 x 205's) connected on this switch, they work just fine.  

 

Will continue in finding a solution and keep you all posted

 

interface 2     

power-over-ethernet critical     

poe-value 33     

poe-lldp-detect enabled

Hi Ishan,

 

The first known ArubaOS version to work well in that series is 6.4.2.5 but you'll be better served to run the latest (6.4.2.12 I believe). If I remember correctly, 6.4.2.5 or 6.4.2.7-8 does contain a bug fix for AP-275.

 

Anyway, you'll be better served by running a newer version.

 

Keep us posted.

Hello Sforget,

 

An hour ago i have updated to 6.4.3.6 (latest). Issue still remains the same. I also had 277 AP's which were rebooting on the old code but after upgrading they are stable now. Biggest worry is both 275 & 277 goes into monitor mode as before. Additionally tried to power up these AP's with Various switches like Cisco & HPE Comware switches. Issue remains the same :(

Ishan,

 

Sorry to read this. I didn't had the chance to test any HPE/Aruba switches yet. However, I can confirm that Cisco IOS 15.0(1)SE2 have problems with PoE+ LLDP negociation. We got it resolved by upgrading the IOS to 15.0(2)SE6 (neither about this version as well). You might have to check this out as well.

 

I only have a single AP-277 that is still awaiting a controller upgrade to 6.4.3.x to be usuable. Our two AP-275s on the same controller (6.4.2.12) and switches (Cisco 3750X with IOS 15.0(2)SE6) runs perfectly.

 

Regards.

Hello Sforget,

 

Even i manage to get 275 to work on the Cisco switch most of the IDF's have HPE/Aruba switches so it is not going to solve my issue. Aruba TAC says that this problem is a known issue and they have fixed in an earlier software update, if the problem is the same they will collect logs and packet captures to escalate it to R&D. Lets hope for the best fingers crossed.....

Adding to my previous post:

 

• Can you capture the boot sequence's console output?  I'd like to know if you have any error message showing there. I tend to suspect PoE+ negotiation.
• Also, have you tried booting them under AC power? This would confirm or eliminate the PoE as a source of the problems. 

Sforget,

 

I do not have an external AC power adapter. I only have a PoE+ injector. Below boot sequence is by plugging into the HP switch. Will post the output when it is connected with the Injector as well.

 

APBoot 1.5.3.14 (build 46206)
Built: 2014-09-26 at 18:50:05

Model: AP-27x
CPU0: P1020E, Version: 1.1, (0x80ec0011)
Core: E500, Version: 5.1, (0x80212051)
Clock:
CPU0: 800 MHz
CPU1: 800 MHz
CCB: 400 MHz
DDR: 333.333 MHz (666.667 MT/s data rate) (Asynchronous)
LBC: 25 MHz
L1: D-cache 32KB enabled
I-cache 32KB enabled
I2C: ready
DRAM: Configuring DDR for 666.667 MT/s data rate
DDR: 512 MB (DDR3, 32-bit, CL=5, ECC off)
POST1: memory passed
Flash: 32 MB
L2: 256 KB enabled
Power: 802.3af POE
PCIe1: RC, link up, x1
dev fn venID devID class rev MBAR0 MBAR1 MBAR2 MBAR3
00 00 14e4 4360 00002 03 80000004 00000000 00000000 00000000
PCIe2: RC, link up, x1
dev fn venID devID class rev MBAR0 MBAR1 MBAR2 MBAR3
00 00 14e4 4360 00002 03 a0000004 00000000 00000000 00000000
Net: eth0, eth1
Radio: bcm43460#0, bcm43460#1

Hit <Enter> to stop autoboot: 0
Checking image @ 0xee000000
Copying image from 0xee000000

Image is signed; verifying checksum... passed
Signer Cert OK
Policy Cert OK
RSA signature verified.
[ 0.000000] Flash variant: default
[ 0.000000]
[ 0.000000] Aruba Networks
[ 0.000000] ArubaOS Version 6.4.3.6 (build 52927 / label #52927)
[ 0.000000] Built by p4build@lesbos on 2015-12-11 at 15:25:39 PST (gcc version 4.5.1)
[ 0.000000]
[ 0.000000] Memory: 509532k/524288k available (9180k kernel code, 14756k reserved, 516k data, 360k bss, 5736k init)
[ 0.002902] Processor 1 found.
[ 0.156826] Brought up 2 CPUs
[ 0.233337] PCI: Probing PCI hardware
[ 0.276388] pci 0000:00:00.0: PCI bridge to [bus 01-ff]
[ 0.339389] pci 0001:02:00.0: PCI bridge to [bus 03-ff]
[ 0.401613] pci 0000:00:00.0: enabling device (0106 -> 0107)
[ 0.468805] pci 0001:02:00.0: enabling device (0106 -> 0107)
[ 4.581557] Kprobe smoke test started
[ 4.657746] Kprobe smoke test passed successfully
[ 4.829659] ee000000.nor: Found 1 x16 devices at 0x0 in 16-bit bank
[ 4.904704] Amd/Fujitsu Extended Query Table at 0x0040
[ 4.966221] number of CFI chips: 1
[ 5.018906] fsl-gianfar ethernet.2: enabled errata workarounds, flags: 0x4
[ 5.102336] /proc/gfar_eth0_stats created
[ 5.152628] fsl-gianfar ethernet.3: enabled errata workarounds, flags: 0x4
[ 5.236060] /proc/gfar_eth1_stats created
[ 5.359060] ad7418 1-0048: cannot read configuration register
[ 5.428196] ad7418 1-0049: cannot read configuration register
[ 5.497537] Enabling watchdog on CPU 0
[ 5.497546] Enabling watchdog on CPU 1
[ 5.587400] Masking ICV Error interrupt
[ 5.888223]
[ 5.888228] Starting Kernel SHA1 KAT ...Completed Kernel SHA1 KAT
[ 5.952659] Starting Kernel HMAC SHA1 FIPS KAT ...Completed Kernel HMAC SHA1 FIPS KAT
[ 5.960271] Starting Kernel HMAC SHA256 FIPS KAT ...Completed Kernel HMAC SHA256 FIPS KAT
[ 5.960956] Starting Kernel HMAC SHA384 FIPS KAT ...Completed Kernel HMAC SHA384 FIPS KAT
[ 5.961378] Starting Kernel HMAC SHA512 FIPS KAT ...Completed Kernel HMAC SHA512 FIPS KAT
[ 5.961824] Starting Kernel AES-CBC FIPS KAT ...Completed Kernel AES-CBC FIPS KAT
[ 5.983843] Starting Kernel AES-CCM FIPS KAT ...Completed Kernel AES-CCM FIPS KAT
[ 5.984313] Starting Kernel 3DES-CBC FIPS KAT ...Completed Kernel 3DES-CBC FIPS KAT
[ 6.645665] Starting Kernel HMAC-SHA1 KAT ...Completed Kernel HMAC-SHA1 KAT
[ 6.749803] Starting Kernel DES KAT ...Completed Kernel DES KAT
[ 6.821755] Starting Kernel AES KAT ...Completed Kernel AES KAT
[ 6.865557]
AP-type has_ble_support: .
Domain Name: arubanetworks.com
No panic info available
[ 11.090455] p1020_soc: P1020 Misc Utils Driver version 1
Ethernet port 1 mode: active-standby
[ 26.562709] ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 26.636387] ADDRCONF(NETDEV_UP): bond0: link is not ready
[ 26.726037] bonding: bond0: enslaving eth0 as a backup interface with a down link.
[ 26.874322] bonding: bond0: enslaving eth1 as a backup interface with a down link.
[ 27.171872] AP xml model 75, num_radios 2 (jiffies 13369)
[ 27.236494] init_asap_mod: installation:0
[ 27.284439] radio 0: band 1 ant 0 max_ssid 16
[ 27.336540] radio 1: band 0 ant 0 max_ssid 16
[ 27.388655] ethernet_device_event: dev eth0 is up
Starting watchdog process...
Getting an IP address...
[ 31.162833] PHY: eth0 (mdio@ffe24000:00) - Link is Up - 1000/Full
[ 31.443261] bonding: bond0: link status definitely up for interface eth0.
[ 31.524545] bonding: bond0: making interface eth0 the new active one.
[ 31.601701] bonding: bond0: first active interface up!
[ 31.663303] ADDRCONF(NETDEV_CHANGE): bond0: link becomes ready
[ 31.956145] Enabling eth1 due to power change [power profile 1]
172.16.70.43 255.255.255.128 172.16.70.1
Running ADP...Done. Master is 172.16.70.102
[ 38.302069] wifi0: AP type AP-275, radio 0, max_bssids 16
[ 38.704683] wifi1: AP type AP-275, radio 1, max_bssids 16
Clearing P1020 PCIe Error Status
shutting down watchdog process (nanny will restart it)...

<<<<< Welcome to the Access Point >>>>>

Ishan,

 

I notice from your posted Procurve config, you don't have the "poe-allocate-by value" statement

applied.  Have you tried that?  Also have you looked at the overall switch poe power budget?

 

I'm on 6.4.3.6 wth AP-275s on HP Procurve 2920s and they are fine.  There should not be much

difference with the HPE/Aruba branded product.  I have two 2920s attached to 275s one at an old

software versions and one at WB.15.18.0007 working fine.

 

I actually have not needed to put special lldp statements on these since 6.4.3.2 or so, but I do

still have these statements on one of the interfaces:

 

interface 3
ip source-lockdown
poe-allocate-by value
poe-value 26
untagged vlan XXX
lldp config ipAddrEnable XX.XX.XX.XX
...bunch of AAA statements...
spanning-tree admin-edge-port
spanning-tree bpdu-protection
exit

Hello All,

 

Thank you so much for helping me. At the end issue was a completely different one. One of the great TAC guys from Aruba namely Ramesh helped me to figure this out. It turned out that the AP275 & 277 are not supported as per my countries regulations which is Qatar in this case. As a result AP's are going to monitoring mode rather than AP mode. So we have created a different regulatory domain profile with a country of Saudi Arabia (Almost similar to my countries regulatory domain setting) and applied it to the group of outdoor AP's i have. Voila!!! it worked like a charm. 

 

So hope this will be helpful to all.

 

Thanks again for all the efforts put up by the community.

Hello Simon,

 

The below output while connected to the PoE+ Injector

 

APBoot 1.5.3.14 (build 46206)
Built: 2014-09-26 at 18:50:05

Model: AP-27x
CPU0: P1020E, Version: 1.1, (0x80ec0011)
Core: E500, Version: 5.1, (0x80212051)
Clock:
CPU0: 800 MHz
CPU1: 800 MHz
CCB: 400 MHz
DDR: 333.333 MHz (666.667 MT/s data rate) (Asynchronous)
LBC: 25 MHz
L1: D-cache 32KB enabled
I-cache 32KB enabled
I2C: ready
DRAM: Configuring DDR for 666.667 MT/s data rate
DDR: 512 MB (DDR3, 32-bit, CL=5, ECC off)
POST1: memory passed
Flash: 32 MB
L2: 256 KB enabled
Power: 802.3at POE
PCIe1: RC, link up, x1
dev fn venID devID class rev MBAR0 MBAR1 MBAR2 MBAR3
00 00 14e4 4360 00002 03 80000004 00000000 00000000 00000000
PCIe2: RC, link up, x1
dev fn venID devID class rev MBAR0 MBAR1 MBAR2 MBAR3
00 00 14e4 4360 00002 03 a0000004 00000000 00000000 00000000
Net: eth0, eth1
Radio: bcm43460#0, bcm43460#1

Hit <Enter> to stop autoboot: 0
Checking image @ 0xee000000
Copying image from 0xee000000

Image is signed; verifying checksum... passed
Signer Cert OK
Policy Cert OK
RSA signature verified.
[ 0.000000] Flash variant: default
[ 0.000000]
[ 0.000000] Aruba Networks
[ 0.000000] ArubaOS Version 6.4.3.6 (build 52927 / label #52927)
[ 0.000000] Built by p4build@lesbos on 2015-12-11 at 15:25:39 PST (gcc version 4.5.1)
[ 0.000000]
[ 0.000000] Memory: 509532k/524288k available (9180k kernel code, 14756k reserved, 516k data, 360k bss, 5736k init)
[ 0.002902] Processor 1 found.
[ 0.156821] Brought up 2 CPUs
[ 0.233332] PCI: Probing PCI hardware
[ 0.276383] pci 0000:00:00.0: PCI bridge to [bus 01-ff]
[ 0.339385] pci 0001:02:00.0: PCI bridge to [bus 03-ff]
[ 0.401610] pci 0000:00:00.0: enabling device (0106 -> 0107)
[ 0.468800] pci 0001:02:00.0: enabling device (0106 -> 0107)
[ 4.580736] Kprobe smoke test started
[ 4.657749] Kprobe smoke test passed successfully
[ 4.829785] ee000000.nor: Found 1 x16 devices at 0x0 in 16-bit bank
[ 4.904821] Amd/Fujitsu Extended Query Table at 0x0040
[ 4.966341] number of CFI chips: 1
[ 5.019016] fsl-gianfar ethernet.2: enabled errata workarounds, flags: 0x4
[ 5.102457] /proc/gfar_eth0_stats created
[ 5.152747] fsl-gianfar ethernet.3: enabled errata workarounds, flags: 0x4
[ 5.236177] /proc/gfar_eth1_stats created
[ 5.359190] ad7418 1-0048: cannot read configuration register
[ 5.428324] ad7418 1-0049: cannot read configuration register
[ 5.497673] Enabling watchdog on CPU 0
[ 5.497682] Enabling watchdog on CPU 1
[ 5.587526] Masking ICV Error interrupt
[ 5.888355]
[ 5.888360] Starting Kernel SHA1 KAT ...Completed Kernel SHA1 KAT
[ 5.952663] Starting Kernel HMAC SHA1 FIPS KAT ...Completed Kernel HMAC SHA1 FIPS KAT
[ 5.960272] Starting Kernel HMAC SHA256 FIPS KAT ...Completed Kernel HMAC SHA256 FIPS KAT
[ 5.960957] Starting Kernel HMAC SHA384 FIPS KAT ...Completed Kernel HMAC SHA384 FIPS KAT
[ 5.961379] Starting Kernel HMAC SHA512 FIPS KAT ...Completed Kernel HMAC SHA512 FIPS KAT
[ 5.961823] Starting Kernel AES-CBC FIPS KAT ...Completed Kernel AES-CBC FIPS KAT
[ 5.983802] Starting Kernel AES-CCM FIPS KAT ...Completed Kernel AES-CCM FIPS KAT
[ 5.984281] Starting Kernel 3DES-CBC FIPS KAT ...Completed Kernel 3DES-CBC FIPS KAT
[ 6.645801] Starting Kernel HMAC-SHA1 KAT ...Completed Kernel HMAC-SHA1 KAT
[ 6.749905] Starting Kernel DES KAT ...Completed Kernel DES KAT
[ 6.821857] Starting Kernel AES KAT ...Completed Kernel AES KAT
[ 6.865660]
AP-type has_ble_support: .
Domain Name: arubanetworks.com
No panic info available
[ 11.084803] p1020_soc: P1020 Misc Utils Driver version 1
Ethernet port 1 mode: active-standby
[ 26.555468] ADDRCONF(NETDEV_UP): eth0: link is not ready
[ 26.629206] ADDRCONF(NETDEV_UP): bond0: link is not ready
[ 26.718525] bonding: bond0: enslaving eth0 as a backup interface with a down link.
[ 26.866968] bonding: bond0: enslaving eth1 as a backup interface with a down link.
[ 27.165039] AP xml model 75, num_radios 2 (jiffies 13365)
[ 27.229673] init_asap_mod: installation:0
[ 27.277617] radio 0: band 1 ant 0 max_ssid 16
[ 27.329720] radio 1: band 0 ant 0 max_ssid 16
[ 27.381835] ethernet_device_event: dev eth0 is up
Starting watchdog process...
[ 30.470437] PHY: eth0 (mdio@ffe24000:00) - Link is Up - 1000/Full
Getting an IP ad[ 30.639315] bonding: bond0: link status definitely up for interface eth0.
dress...
[ 30.736522] bonding: bond0: making interface eth0 the new active one.
[ 30.824066] bonding: bond0: first active interface up!
[ 30.885795] ADDRCONF(NETDEV_CHANGE): bond0: link becomes ready
172.16.70.43 255.255.255.128 172.16.70.1
Running ADP...Done. Master is 172.16.70.102
[ 36.395277] wifi0: AP type AP-275, radio 0, max_bssids 16
[ 36.775839] Enabling eth1 due to power change [power profile 1]
[ 36.868969] wifi1: AP type AP-275, radio 1, max_bssids 16
Clearing P1020 PCIe Error Status
shutting down watchdog process (nanny will restart it)...

<<<<< Welcome to the Access Point >>>>>