cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2197
Views
8
Helpful
9
Replies

Problem within Autonomous AP Migration process

IIS Network
Level 1
Level 1

Dear all,

hopefully someone can help me. We have a problem within the process to migrate autonomous access points to a LWAPP on the WCS.

We have several branch offices with Cisco 5500 Series Wireless LAN Controller (AIR-CT5508-K9). We are using,

among others new Cisco Access Points, also "old" Cisco AIR-AP1242AG-E-K9. These access points have to be

migrate from autonomous to LWAPP before we can use them with the WLC.

The WCS system is located in our headquarter and the branch offices are connected via different WAN technics

to the headquarter.

We successfully migrated 1242 Access Points (same config and IOS) before, also over WAN lines. The Recovery

image is normally located on a TFTP server at the branch office so that there a not any delays to prevent the migration.

During the last migration we encountered a strange problem. We successfully added the autonomouse access points

to the WCS (Configure -> Access Points -> Add Autonomouse APs). Then we created a "Autonomous AP Migration Template"

as usual, select the correct Access Points and started the convert process.

Everything is running fine until the message "Updating Environment variables." appears. This message will stay for

about a minute in the display and nothing is happen. Then it looks like the migration process is canceled because

the status message "Not Initiated." appears in the display. If we leave the migration dialog and browse back to

the list of  "Autonomous AP Migration Templates" the used template shows the status "Failure".

Because the migration process failed we tried to access the Access Point via ssh and it is working as before the migration.

We checked the Access Point and the recovery image was successfully uploaded and configured on the Access Point.

Also the boot variables were setup correctly. We then tried a restart. On our VLAN 1 a DHCP scope with the WLC option

is configured. It took very long (about 5 to 10 minutes) but after that time frame the Access Points started up again and

downloaded the Controller firmware and we can configure the Access Points as usual via WLC or WCS.

If the Access Points is "really" migrated after this it might be o.k. But still it is a bigger and longer effort as the

normal WCS migration procedure. We have to login to every Access Point and reload it. Because the reload command needs

a confirmation it is also not possible to do this via script or similiar. Also we are not sure if the Access Point is

now "really" migrated or maybe some old config is still on the Access Point and will maybe produce errors in the future.

The 1242 Access Points have an IOS version (see below) installed that should be o.k. for the migration. To be sure we updated

an Access Point before the migration with a newer IOS (c1240-k9w7-tar.124-25d.JA1.tar) but it didn't help.

The migration process of an 1142 and 1252 Access Point in the same branch office was done without any problem!

- Has someone experienced the same behaviour and has maybe a solution to fix the problem?

- Is there another methode than WCS or Update Tool to migrate autonomous Access Points to LWAPP ?

- Is there maybe somewhere a newer version of the Update Tool (current is 3.4)?

Our WCS Version is 7.0.230.0.

Our WLC Version is 7.0.116.0.

The IOS of the autonomous 1242 Access Point is "c1240-k9w7-mx.123-8.JEA2".

The recovery image for the migration process is "c1240-rcvk9w8-tar.124-21a.JA2.tar".

We are aware that there a newer version of WCS (->NCS) and WLC out. Because we cannot upgrade in the moment and

the process was running before it would be great if someone can help.

Thanks a lot in advance.

If there further information needed please don't hesitate to ask.

Regards

9 Replies 9

Scott Fella
Hall of Fame
Hall of Fame

I have always used the achive command with no problems at all.

archive download-sw /overwrite /reload tftp: //location/image-name

Thanks,

Scott

Help out other by using the rating system and marking answered questions as "Answered"

-Scott
*** Please rate helpful posts ***

Hi Scott,

thanks a lot for your quick answer.

I will test this but what image exactly should we use for the download, the recovery image "c1240-rcvk9w8-tar.124-21a.JA2.tar"?

I am asking because under the download for the 1242 I see also a "WIRELESS LAN LWAPP" image and the mentioned

"WIRELESS LAN LWAPP RECOVERY" image.

Thanks a lot

Regards

Peter

Hi Scott,

you can ignore my last post because I got the change to test it directly. I have done it with the

"WIRELESS LAN LWAPP RECOVERY" image and it was working without a problem.

I rated your answer but I will leave this discussion to answered because I still don't know why

the normal migratin process is not working.

Thanks again for you quick response.

Regards

Peter

Scott Fella
Hall of Fame
Hall of Fame

To be honest, their was an upgrade tool in the past but was limited to certain model APs and I believe it stopped being supported on the 1252's and newer models. That didn't work well. The WCS migration didn't work well for me either in the beginning. The archive command is what I use no matter what. You can script the CLI commands to convert APs which also makes it easy. I always shake my head when I see post regarding using the upgrade tool or WCS:)

Sent from Cisco Technical Support iPhone App

-Scott
*** Please rate helpful posts ***

rupert.wever
Level 1
Level 1

The process fails because it cannot write the ENV variables to your tftp server. Are you using the flash in your branch router with tftp-server command? Haven't found a way to make iOS tftp server read/write so if you can use a different server with read/write access, process should work fine.

Sorry to disagree about simply using the archive command on iOS ap: the conversion tool (laptop or WCS) does more than just that. If you have very legacy AP's that don't have their own certificates, the tool generates ssc's and imports them to the wlc as well as WCS.

Please rate this to let us know if this answers your question.

Sent from Cisco Technical Support iPhone App

I understand and its preference. Since the tool never worked well for me, the archive was the easiest and grabbing the hash from the debug was pretty easy. What I have had issue with with the tools is sometime it doesn't properly grab the hash and you would have to run a debug anyways. I still have clients who want to use the tool and they do... I just tell them to run a debug in case any AP's fail to join. Plus if the configuration is huge on the AP, the tool will fail due to memory, in which you have to basically wipe the configuration.

Sent from Cisco Technical Support iPhone App

-Scott
*** Please rate helpful posts ***

We are using the "tftpd32" tftp server and not a tftp server on a cisco device. The connected client

are allowed to write on the tftp server. It is working all the time when we download config files from

e.g. a switch to the tftp server. So it is strange why the access point cannot write anything. As I

said the migration was working before without any problems.

I also want to say that we have migrate about 20 access points with the archive command and

it is working without a problem till now. If I compare "WCS migrated" access points and

"archive command migrated" access points on a wireless lan controller I cannot see a difference.

George Stefanick
VIP Alumni
VIP Alumni

Not to mention the software tool isnt supported on win 7. Archive command is the way to go ..

Sent from Cisco Technical Support iPhone App

"Satisfaction does not come from knowing the solution, it comes from knowing why." - Rosalind Franklin
___________________________________________________________

Good point. My current client is using the tool and he actually had to find an XP machine:)

Sent from Cisco Technical Support iPhone App

-Scott
*** Please rate helpful posts ***
Review Cisco Networking for a $25 gift card