I have a 4402 controller just recently upgraded to code 5.2.178. I have a new 1142 out of the box and I am trying to get it up and running. It finds the controller and starts the image download but then stops with the following error:
capwap_image_proc: problem extracting tar file
It then tries to reconnect to the controller, it finds it and errors out again.
Figured I'd ask for some ideas here before I went to TAC.
Hmm not much to go on here... You know I collect Martin Guitars (to digress here a bit) and at these Bluegrass festivals I like to attend once in a great while they have a bumper sticker that's pretty popular.. it says... "It takes a village to tune a banjo"... at risk of sounding like the village idiot
1. Does another 1140 do the same thing (where's the other banjo)
If no, you have a bad AP. If YES then..
2. Have you tried to download a new version of controller code and flash the controller again? Maybe during the process the code it pushes to the AP got corrupt? If so that's why the AP rejects it remember the controller holds that image for the AP.
If you flash it and it still won't talk to the AP.. then what's your network look like.. maybe the image is getting hosed across the network or through something in the network.. break the network down into something easy see if it still fails.. (if it works that's the answer) if it doesn't... time to find another controller.. From the problem description there just isn't enough stuff here to rule out what is broke...
TAC is a good village to help you tune that Banjo but I don't think you need them for this...
I tried another AP and it does the same thing and give the same error. 1242 and 1252 AP can connect just fine. Being a guitar player myself I can see where you have to take this thing apart. I'm in the process of downgrading to 5.2.157 and see what happens with that.
I can tell you that it SHOULD work. I am using a 4402 with 220.127.116.11. I just took an 1142 out of the box, and it worked fine. This is all in a lab environment, so my network is extremely simple/direct. Power was provided by the wall wart, though, not PoE.
While you have the 1140 still there, can you console into the AP and find out the status of the flash? If you are running the RCV file, can you type the following commands:
clear lwapp private
clear lwapp ap controller ip address
clear lwapp hostname
clear lwapp ip address
Make sure that you followed the upgrade path on the controller properly. Including the ER boot strap upgrades. Also make sure the date and time are correct on the controller or the file extraction will fail due to securtiy certificate issues.
I was originally running 4.2.176, so I could upgrade directly to 5.2.178. I do not have any 1522 APs so I I don't think I need to upgrade the bootstrap image. The controller time is in sync with our time server. I downgraded the controller to 5.2.157 and the AP downloads the image fine. I tried to redownloading 5.2.178 again and am still having the same problem.
As far as our network, I have our controllers connected to a 3750-24T running 12.2(25)SEB4. From there we are allowing only wireless traffic and vlans to be passed to the controllers. The 1142 is connected to a 6509 and that is routed to the 3750. 1242 and 1252 APs connect and download just fine so I don't think it's a routing issue.
When you tried another AP, did you plug it into the same VLAN segment as the WLC?
"I don't think I need to upgrade the bootstrap image."
You're right, to solve this problem, I don't think you need to, however, there are alot of patches the new bootstrap fixes/addresses with the old bootstrap. It's worthwhile upgrading the bootstrap when you are able to. (I know it's a pain and a hassle after doing those bootup.)
After submitting a TAC request and doing debugs on the controller and APs and reviewing the configs on our 3750 stack and 6500's this was filed as a bug.
1142 fails on downloading code from controller with Bad Record MAC alert
LAp 1142 will not downlaod code from controller running 5.2.xxx with error Bad Record MAC alert and Invalid CAPWAP states
WLC running 5.2.xxx connected to stacked 3750 switch. Initially had LAG enabled and controller ports connecting to two different stacks in same switch. Later had no LAG and only 1 port connecting to switch and AP on same switch and subnet as controller nad AP would still not join
Thanks to all who offered suggestions for me to try.
Now I'm trying to figure out what I'm going to do with these 30 1142's I have until they get this issue fixed.