Wireless Access

Hi guys.  I have a wierd issue, and I hope someone can help.  I bought a few used AP-125's to add to our network.  I have always had AP-65's and have just started adding the AP-125's recently.  I have read three different things on resetting an AP-125, and I am not sure which one is correct.  I have done the purge, save, reset method.  I have done the purge, clear os method.  And I have done the clear os, purgeenv, saveenv, reset method.  The standard purge doesn't seem to do anything, while the clear os one seems to kill the image on the AP.  I now have a Invalid Image Format version: 0xffffffff error that I can't seem to get around.  Any ideas on how to fix the error, and what the proper reset method is?  Thanks.

What exactly you are trying to purge or in others words what exactly is not working after you purge AP ?

Can you share the console output ?

I am actually just trying to factory reset them so I can add them to my controller. 

Purge command from apboot  would be the proper command to erase the config on the AP and this will revert back the AP to factory default settings. Find below.

 

apboot> help
? - alias for 'help'
boot - boot the OS image
clear - clear the OS image or other information
dhcp - invoke DHCP client to obtain IP/boot params
factory_reset - emulate factory_reset functionality
help - print online help
mfginfo - show manufacturing info
ping - send ICMP ECHO_REQUEST to network host
printenv - print environment variables
purgeenv - restore default environment variables
reset - Perform RESET of the CPU
saveenv - save environment variables to persistent storage
setenv - set environment variables
tftpboot - boot image via network using TFTP protocol
upgrade - upgrade the APBoot or OS image
version - print monitor version

 

Are we able to do DHCP and then try to ping the controller ip address and default-gateway of the AP itself. If the AP is unable to reach master controller and if it unable to down load the image then we may be running into this issue.

 

Please verify the master controller ip address and gateway is pingable from apboot mode.

 

Thanks!

 

<<<<<<<<<<<Kudos please if my suggestions helped you>>>>>>>>>>>>>>>>>>>>>>>>

To confirm that the purged worked you can do a printenv and that will show you the variables or config on the AP like the name , ap-group .

After you purge it is the AP not able to reach the controller ?

I am beginning to think maybe they are just bad.  I can't get an IP even with the DHCP command.  One of them has both ethernet 1 and 2 lights lit up even though I only have one cable plugged in, and they other one only gets a light on ethernet port 2 when the cable is in it, but ethernet 1 will not light up.  I have one though that lights up ethernet 1 like it should, but it will not get a dhcp address.

Do you have any other working APs on the same segment you are connecting that AP 125 ?

Yes.  And I have used the same cables and setup for clearing and provisioning the AP-125's, so like I said, I am beginning to think I have some bad ones that I need to send back, since they were used in the first place.

Actually, I just looked and it does show down on the en0 and en1 ports even if I have anything plugged into them.

You might got the bad one; please raise the case with TAC to RMA.

 

Thanks

I do have another one that I have reset to factory using purge, save, reset, and it shows up in the installation list on the controller like it should.  The problem it has though is that it says upgrading, then it reboots, then it upgrades, reboots.  It is a non-stop cycle.  Any ideas? 

 

It does show an error of:

 

MTD do_erase_oneblock():  software timeout   erase err    

What AOS are you using ?

On the controller do a show log system all | include <device mac>

Can you please share the console output ?

Make there's enough IPs available to give on your dhcp server ?

Also check that Poe is properly provided ?

How many of these 125s are not able to come up ?

6.1.3.7

---

@gbguy wrote:
6.1.3.7

---

GBguy,

 

There is an issue where APs on 3.x cannot be upgraded to 6.x directly without a stop at the latest 5.x.  you could be running into that issue.  The only solution is to safe your config, downgrade the controller to the latest 5.x and see if that specific access point will come up.  If it does, upgrade back to 6.1.3.7 and then boot from your saved running config.

 

At issue is the the 3.x AP had a hard limit on the size of the image it would accept.  In 6.x the images exceed that, so direct upgrades from 3.x to 6.x would have reboots.  The 5.x code relaxes the limit so that access points can accept upgrades that were previously restricted by 3.x.

 

If the access point you obtained was on 3.x, it could not be directly upgraded to 6.x.  Does that make sense?

Yes that makes sense. I have an old 2400 that has version 5 on it. I'll see if it'll upgrade it correctly on Friday. Thanks!

Well, that didn't help.  Tried 5.0.4.12 and even went back to 3.4.5.1 and it has the same erase err.  Wonder if this one is bad too?

Try the following at the apboot prompt:

 

flash erase bfc80000 bfc81000
flash erase oxbfc60000 0xbfc6ffff
upgrade
tftp boot
dhcp

flash isn't a vaild command.

What is Aruba's policy on returning an AP that I didn't buy from them?  I know they have a lifetime warranty, but does it matter who owns it or if it was bought from Ebay?

Hi cappalli

 

sorry for digging this older thread out , would like to reset an AP125 which was a RemoteAP , i tried "purge, save, boot" as i used those commands since years , but this AP125 is really weird, i tried with purgeenv, purve, clear os , all the usual commands.

 

after boot, it still has the RemoteAP env infos , .

 

tried with your "flash" commands, but didnt help. anyone else had this problem with good old AP125 which doesnt want to reset ? 

 

really strange.

Yes, experienced the same problem with 2 AP125's in my lab. They were working fine, converted them to RAP's (make a config mistake :-( ) and now they won't connect. Also tried al the purgeenv, saveenv, printenv even fiddeld around with setenv and saveenv but no luch so far.

 

- Boot process aborted

- AP in same vlan as 3400 controller

- Master is on 10.127.91.100

- DNS record   rap.swdemo.nl -> 10.127.91.100

- dhcp:

Hit <Enter> to stop autoboot:  0
apboot> dhcp
en0:    Up     1Gbs Full duplex, (port  1)
BOOTP broadcast 1
DHCP IP address: 10.127.91.23
DHCP subnet mask: 255.255.255.0
DHCP def gateway: 10.127.91.1
DHCP DNS server: 10.127.3.11
DHCP DNS domain: swdemo.loc
apboot>

 

- normal boot:

ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
10.127.91.23 255.255.255.0 10.127.91.1
Running ADP...Done. Master is rap.swdemo.nl

However, it's not finding the controller.

 

I resolved my RAP config issue and was able to succesfully convert another, still working AP125 from CAP-mode to RAP. Then, out of curiosity, I converted the AP125 in RAP-mode back to CAP-mode and that worked fine as well.

 

Any suggestions would be welcome to get my old LAB equipment to work.

 

Bart

I resolved my problems thanks to suggestions in this post:

 

https://community.arubanetworks.com/t5/Wireless-Access/Tutorial-AP-Boot-Commands/td-p/91734

 

I used the printenv to show current sessions and then used the setenv commands:

- setenv master 10.127.91.100

- setenv name

- setenv group

- setenv remote_ap=0     <old value was 1

- saveenv

- clear os

- saveenv

- reset

 

Conclusion: factory_reset doesn't seem to reset het enviroment settings. Printenv shows the current settings. If you see setting that conflict, use setenv to change or wipe them. Surprising enough in my case the changed environment settings didn't show anymore when executing printenv. Also not after applying saveenv and then retyping printenv. Lastly I used clear os and saveenv. Then the AP125 happily booted up on my controller. In my case it fixed my problem with 2x AP125's.

 

Bart

in old models of AP, like 125, you must use purgeenv not factory_reset, iirc.

By factory_reset I meant to say purgeenv :-)  but thanks for noticing