Wireless Access

Customer just installed 6 new AP105's and 4 of the 6 will not come up on the 3400 controller rinning 6.1.2.1. One of the six came up immediately and then a day later another came up and I was able to provision both. The remaining for....no dice! When I do a Show log system I see the AP's attempt but then they reboot with a message of:

Reboot Reason: AP rebooted Sun Dec 2 15:40:47 cst 2012; SAPD: Reboot after image upgrade failed: 255

Had the cutomer send me the ap boot log off the AP and the following is what is there:

APBoot 1.2.4.4 (build 26618)
Built: 2011-01-07 at 13:42:04

Model: AP-10x
CPU: AR7161 revision: A2
Clock: 680 MHz, DDR clock: 340 MHz, Bus clock: 170 MHz
DRAM: . 128 MB
POST1: mem 0 mem 1 mem 2 mem 3 passed
Copy: ....    done
Flash: 16 MB
PCI: scanning bus 0 ...
dev fn venID devID class rev MBAR0 MBAR1 MBAR2 MBAR3
00 00 168c 0029 00002 01 10000000 00000000 00000000 00000000
01 00 168c 0029 00002 01 10010000 00000000 00000000 00000000
Net: eth0
Radio: ar922x#0, ar922x#1

Hit <Enter> to stop autoboot: 2  1  0
Checking image @ 0xbf100000
Invalid image format version: 0xffffffff
Checking image @ 0xbf800000
Invalid image format version: 0xffffffff
eth0 up: 100 Mb/s full duplex
DHCP broadcast 1
DHCP broadcast 2
DHCP broadcast 3
DHCP broadcast 4
DHCP broadcast 5

Retry count exceeded; starting again
eth0 up: 100 Mb/s full duplex
DHCP broadcast 1
DHCP broadcast 2
DHCP broadcast 3
DHCP broadcast 4
DHCP broadcast 5

Just repeats over and over...

any suggestions?

I have asked that the port be full duplex and all that and it appears it is. The switch is a Cisco 2960 POE.

I just requested that he make sure the port is set to portfast...havent heard back

He claims if he plugs his laptop into the same port on the switch he is able to get an address

Thanks

David

It looks like it is not getting an ip address....  You can type "DHCP" at the apboot> prompt to confirm this.

He needs to make sure that he has not run out of addresses on his DHCP server.

Its not, otherwise why would his PC get an IP when plugged into the same port?

His laptop may get an address b/c maybe it already had an active lease from a prior connection.   The symptom is consistent with IP lease issues, especially when you said initially only 1 came up; then the 2nd a day later (likely when a lease freed up).   

Have him stop the boot process on one of the APs.   At the prompt, type dhcp.  This will have the AP attempt to obtain an address.   If it does not get one, have him check the DHCP server pool.   

well doesnt this is the ap boot capture pretty state that its not getting an IP?

DHCP broadcast 1
DHCP broadcast 2
DHCP broadcast 3
DHCP broadcast 4
DHCP broadcast 5

Retry count exceeded; starting again
eth0 up: 100 Mb/s full duplex
DHCP broadcast 1
DHCP broadcast 2
DHCP broadcast 3
DHCP broadcast 4
DHCP broadcast 5

Also why would they show up in monitoring as down and have an IP associated with them?

Yes, it is pretty clear that it is not getting an IP based on that capture.   Why they show up as down with an IP, I can't say.....but it definitely appears to be DHCP at this point.   Have him check the DHCP server pool, or give the AP a static IP on that network:

apboot> setenv ipaddr x.x.x.x
apboot> setenv netmask 255.255.255.0
apboot> setenv gatewayip x.x.x.y

apboot> setenv dnsip y.y.y.y   (if controller lookup uses DNS)

apboot> save

now I am getting this:

Image is signed; verifying checksum... failed! 0xe3c7bec8
tftpboot failed: retrying in one second
eth0 up: 100 Mb/s full duplex
Using eth0 device
TFTP from server 10.200.11.10; our IP address is 10.200.0.177
Filename 'mips32.ari'.
Load address: 0x2000000
Loading: *#########################################################
done
Bytes transferred = 3694028 (385dcc hex)

Image is signed; verifying checksum... failed! 0xe3c7bec8
tftpboot failed: retrying in one second
eth0 up: 100 Mb/s full duplex
Using eth0 device
TFTP from server 10.200.11.10; our IP address is 10.200.0.177
Filename 'mips32.ari'.
Load address: 0x2000000
Loading: *#########################################################
done
Bytes transferred = 3694028 (385dcc hex)

he checked DHCP and it is not full

He is using a /16 mask so no issue there

With AP image issues, I have had mixed results with wiping the AP using the "clear os" command at the apboot> prompt.    Worth a shot.

Otherwise, I recommend you call a case in with support and show them that output to see if they can determine why the AP is not liking the software image it is getting from the controller.

I am rebooting the controller tonight and see if that shakes it loose. I will be onsite in the morning to try a few things like Clear OS and manually pointing the AP to a different controller and see if it pulls down the image correctly. If those don't work then TAC can figure it out! I think I have exhausted all options. Thanks for the suggestions!

What brand of switch you got?

At least on alcatel switch i get that problem in which the AP wont get the IP address if the AP is on another vlan than the DHCP server is... even if i configured the ip helper...

I need to set the ip helper forwarding  time on 0 to make it work....

That migh be your issue...

If it not alcaltel stilll try to put the forwarding time on 0 on the ip helper to see if that solve your problem....

You can put the static ip address but that would be like a work around i guess...

Cheers

Carlos

Crisco 2960 POE

I rebooted the controller lastnight to no avail

The customer rebooted it again this morning and the APs came up! Go figure

I was at the point of upgrading the code on the controller thinking the AP FW in it was corrupted. May still do that.

Four more APs are being installed tonight, so we will see