cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2726
Views
10
Helpful
12
Replies
Rps-Cheers
Rising star

AP3802I power on by AIR-PWR-50=,but the wired which connect to a switch is down

we buy 4 APs,the PID is AIR-AP3802I-H-K9,we use AIR-PWR-50=,and AP3802 can power up,but when we connect ap to switch,we find
ap wired0 always down,the status as below:
Interface IP-Address Method Status Protocol Speed
wired0 unassigned unset down up n/a
wired1 unassigned unset down down n/a
wifi0 unassigned unset administatively down down n/a
wifi1 unassigned unset administatively down down n/a

This problem is so strange, please help us to see,thx

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !
12 REPLIES 12
Leo Laohoo
VIP Community Legend


@Rps-Cheers wrote:
Jan 24 22:49:22 kernel: [*01/24/2018 22:49:22.4143] Discovery Request sent to 192.168.155.148, discovery type STATIC_CONFIG(1)

I can see the AP booting up to a static IP address of 192.168.155.148.  

So what, exactly, is the problem?

What kind of switch is the AP connected to?

What happens if the AP is connected to a computer?

The problem is described as follows:
The AP3802 can be up and running normally with AIR-PWR-50 =, but when the mG interface (that is, the POE interface) is connected to the switch, enter show ip interface bri under the console and see the following result:

Interface IP-Address Method Status Protocol Speed
wired0 unassigned unset down up n / a

Switch can be ruled out the problem, because we have other 3802I AP connected 2960,3750,3650 other switches can be normal.

Is this a problem with version? But 8.2.151 is a factory-shipped version, I know other companies are this version of you, but wired0 can be up.

In addition, I saw some people say that the fault version does not match, I do not know if this is not the problem? Will the AP's firmware version should be how to look?

 

Currently tested 8.5.110.0 version can do the request,However, I think this is not the root cause.

 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !
Leo Laohoo
VIP Community Legend

I think I have a very good idea what's going on. Just to make sure, the symptoms are the following:
1. AP is showing as up/up when connected to the switch interface;
2. When issued with the switch command "sh cdp neighbor" the output is a blank;
3. The AP is not getting an IP address.
Did I sum it up completely?

No, it is this:
1, the AP can be started, but when the POE interface connected to the switch, the AP console "show ip inter bri" shows the down / up
2, AP's wired0 interface is down, then certainly can not get the IP address
3, because the wired0 interface is down, so even if the AP is configured with a static IP address, and the controller will not work.
4, very strange, four new AP3802I are such a problem.

Therefore, the most fundamental reason is that the AP can be powered on (using AIR-PWR-50 =, which is recommended by the official website). However, when the POE interface is connected to the switch, the AP's wired0 interface is down through the CLI

The following is the AP connection switch interface state:

AP # show ip inter bri
Interface IP-Address Method Status Protocol Speed
wired0 unassigned unset down up n / a

Thanks ~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !
Leo Laohoo
VIP Community Legend

If the APs are connected to a Cisco switch, do the following:
1. Command: test cable tdr interface <BLAH>;
2. If 3650/3850 switch, wait for approximately 7 to 10 seconds while the process runs;
3. Command: sh cable tdr interface <BLAH>; and
4. Post the complete output to #3.

OK ,

But,what meaning about "<BLAH>"?these commands enter in AP and SW?

AP:test cable tdr interface <BLAH>;

SW:sh cable tdr interface <BLAH>;

 

 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !
Leo Laohoo
VIP Community Legend

The commands are for the switch.
<BLAH> = port(s) where the AP(s) is/are connected to.

OK,thanks,i will test~

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !

i have resloved the issue,it's a bug of cisco....
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !

I am experiencing the same issue. Can you tell us what was the bug or explain how you got this issue resolved?

 

Thanks in advance for your help

sure,i had opened a cisco tac case.this is a internal bug of cisco,if you just have one or two ap,i think you can rma,because of some ap3800 that made at Jan or Feb just have the issue.if you want to fix it,you can refer to below:
NOTE: It is imperative this procedure be done ONLY on an AP affected by this specific issue - Verify against affected list or check the AP console output as detailed below.

DO NOT perform this procedure if in doubt. AP may become un-usable if modified un-necessarily.

=============================
IDENTIFY an AFFECTED UNIT :
=============================

Check the Serial # identified if affected.

On an affected AP, you should see the following when AP is in bad state

AP PHY ports don't come up, so AP does not get IP and does not join WLC.

during Boot check the board type :

Board: Barbados-3K

and in Uboot: (hit esc during bootup)
u-boot>> phy_type
Phytype: Aquantia

and in devshell

AP:/# ls -lrt /firmware
total 204
-rwxr-xr-x 1 root root 256 Jul 24 2017 fw_update.tgz.sig
-rw-r--r-- 1 root root 203132 Jul 24 2017 fw_update.tgz

(Firmware file exists which it should not for this Model of Board)


========================
Recovery Procedure :
========================

Follow the procedure as below:

1) Connect to the serial console with a terminal application

2) Boot the AP, break to U-BOOT by hitting 'ESC'

3) 1st boot: check board type and remove firmware

u-boot>> dump_board_env
Confirm again: Board ID: 3K

u-boot>> ubi part fs

Creating 1 MTD partitions on "nand0":
0x000000200000-0x000010000000 : "mtd=2"
UBI: attaching mtd1 to ubi0

u-boot>> ubi remove firmware

Remove UBI volume firmware (id 4)

u-boot>> ubi create firmware 0x800000

Creating dynamic volume firmware of size 8388608

u-boot>> reset

4) 2nd boot: Boot in COS AP image and check wired port status

AP#show controller wired 0
wired0 Link encap:Ethernet HWaddr 58:AC:78:DE:55:9A eMac Status: UP --> Port is UP
UP BROADCAST RUNNING PROMISC MULTICAST MTU:2400 Metric:1

(Please make sure you complete the entire process even if your AP gets an IP address)

5) 3rd boot: check UBI firmware info and set FACTORY_RESET

u-boot>> ubi part fs

u-boot>> ubi info

Look: UBI: number of user volumes: 8

u-boot>> ubi info layout

Look: UBI: name firmware

u-boot>>setenv FACTORY_RESET 2
u-boot>>saveenv

u-boot>> reset

6) After boot, AP will get the PHY ports to be up . AP will reboot and should work on any version of code.

7) An additional check can also be done to make sure the firmware is deleted

(from devshell)

AP:/# ls -lrt /firmware
total 0
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Rps-Cheers | If it solves your problem, please mark as answer. Thanks !
Leo Laohoo
VIP Community Legend

Thanks for this.  I'm going to have a play with this procedure Monday.

Create
Recognize Your Peers
Content for Community-Ad