EWC AP upgrade failed - stuck in boot loop

[wififofum]

Hi,

I upgraded an EWC AP to 17.6.1 and since then the AP has been continuously rebooting without fully initializing. (see attached the 9120-bootlog.txt).

I interrupted the boot process at the uboot> prompt and based on community feedback changed the boot variable to partition 2 (part2), upon which the AP got to beaconing but quickly shutdown and started the boot cycle again. Changing the setenv BOOT part1 repeated the reboot cycle.

I also performed a TFTP boot of an AP image to no avail. I've since performed a factory reset on the unit.  After this, another part 2 boot got me to the EWC controller config dialog (showed as 17.6.1) before it eventually interrupted the setup and resumed the reboot cycle.

It certainly appears the upgrade failed. Everything was predownloaded via the GUI/Activate. Can the EWC/AP code be restored from the uboot prompt?

Thanks all!

[Leo Laohoo]

---

@wififofum wrote:

Hit ESC key to stop autoboot:  8 MG link 1G

---

Pay close attention to the AP when it boots up.  When it hits this stage, hit the ESC key and the curser should be "u-boot>>".  

Then do the following: 

setenv BOOT part2
saveenv
boot

The AP should boot into 17.5.1.  When this happens, load a different version of eWLC like, say, 17.3.4.  Once this one works, upgrade to the desired version.

[wififofum]

Thank you Leo!

I did set the AP to boot from part2.  I also have a 9130 EWC on 17.6.1 code which it appeared to discover and join and as a result select its matching (backup) code.   After rebooting though it does not seem to boot from part2 and recycles (same as previous log).

[*10/04/2021 14:27:30.0090] CAPWAP State: Discovery
[*10/04/2021 14:27:30.0100] Discovery Response from 192.168.0.62
[*10/04/2021 14:27:34.5720] Start: RPC thread 2978673536 created.
[*10/04/2021 14:32:02.0000]
[*10/04/2021 14:32:02.0000] CAPWAP State: DTLS Setup
[*10/04/2021 14:32:02.7580]
[*10/04/2021 14:32:02.7580] CAPWAP State: Join
[*10/04/2021 14:32:02.7600] Sending Join request to 192.168.0.62 through port 5254
[*10/04/2021 14:32:02.7680] Join Response from 192.168.0.62
[*10/04/2021 14:32:02.7680] AC accepted join request with result code: 0
[*10/04/2021 14:32:03.2960] Received wlcType 2, timer 30
[*10/04/2021 14:32:03.3810]
[*10/04/2021 14:32:03.3810] CAPWAP State: Image Data
[*10/04/2021 14:32:03.3810] AP image version 17.5.1.12 backup 17.6.1.13, Controller 17.6.1.13
[*10/04/2021 14:32:03.3810] Version does not match.
[*10/04/2021 14:32:03.3810] AP backup image version matches controller. No download, activate backup.
[*10/04/2021 14:32:03.4300] status 'upgrade.sh: Script called with args:[ACTIVATE]'
[*10/04/2021 14:32:03.4700] do ACTIVATE, part2 is active part
[*10/04/2021 14:32:03.5130] status 'upgrade.sh: activate part1, set BOOT to part1'
[*10/04/2021 14:32:04.0720] status 'upgrade.sh: AP primary version after reload: 17.6.1.13'
[*10/04/2021 14:32:04.0960] status 'upgrade.sh: AP backup version after reload: 17.5.1.12'
[*10/04/2021 14:32:04.0980] Backup image activated. Rebooting AP...

I have two logs attached - one with the 9130 online and one without.  Is there a relevant debug I can issue on the 9130 to see what the 9120 is doing as far as downloading code, etc.?

Thanks a lot for your support!

--Bruce Johnson

[Jitendra Kumar]

Possible for a Bug..

my suggestion is open Tac ticket for that.

https://bst.cisco.com/bugsearch/bug/CSCvr28291 

[JPavonM]

As side note to this chat, tehre is a U-Boot recovery procedure for all Catalyst APs here.

HTH