10-08-2010 06:53 AM - edited 07-03-2021 07:16 PM
As we transition buildings to our lightweight infrastructure we're installing new autonomous 1142 APs to fill in gaps and swapping out older 1242s with autonomous 1142s. What we're seeing is that some of these APs, running the latest IOS code available, are shutting down their 2.4 GHz radios. The only thing in the logs is:
%LINK-5-CHANGED: Interface Dot11Radio0, changed state to reset
%LINEPROTO-5-UPDOWN: Line protocol on Interface Dot11Radio0, changed state to down
The only residual elements that are found (most times) is on the client association page where we'll see what amounts to stale client sessions. There are two flavors of such stale sessions. Associations to open networks show clients with IP addresses of 0.0.0.0, the MAC addresses are still listed, and the state is "Associated". For 802.1x based associations clients show their assigned IP addresses, MAC addresses, and the state appears as "EAP-Associated".
I was looking at some of the open caveats with IOS release 12.4(21a)JY to see if anything matched up with the problem I’m seeing. In the bug toolkit, bug CSCtg09159 (might be related to our problem?) is stated as being “Fixed-in” release 12.4(21a)JHB.
Anyone seen the problem we're seeing? Anyone know when 12.4(21a)JHB is scheduled for release?
10-12-2010 05:44 AM
Hi,
Are these APs joined to the WLC?
If yes, have you checked the trap logs for that timestamp?
Can you share it with us so we take a look?
Thanks,
Tiago
10-12-2010 06:59 AM
These are autonomous 1142s not managed by a WLC.
10-12-2010 05:48 PM
Plug a console into the AP and reboot the AP. Post the bootup logs.
10-12-2010 07:50 PM
What color is the LED on the AP? What are you using to provide PoE power to the AP?
10-14-2010 08:21 AM
All of our APs are powered by POE switches of different flavors. The AP's light appears to always be blue.
Here's an example of one of the problem 1142s.
System Software Filename: c1140-k9w7-tar.124-21a.JY
System Software Version: 12.4(21a)JY
Bootloader Version: 12.4(18a)JA3
The 802.11g radio shows as “Dot11Radio0 is reset, line protocol is down” while the 802.11a radio appears to be up and running.
You’ll note in the documents attached the stale client connections also manifest themselves off of the 11a radio, but we’re yet to see that interface reset and shutdown. Clients listed with valid IP addresses are not actually associated to the network.
The attached files include:
Image of the Association GUI display (176 associated clients).
Output from the AP’s “show tech” command
10-14-2010 03:06 PM
That's wierd ... In a sense that your PDF files and your issue don't "gell". You say your "b" or 2.4Ghz radio shuts down, right? Your association attachment shows a number of clients successfully associated.
And when I mean by "successful" associated these are clients with a valid IP address. Clients with an IP address of 0.0.0.0 means they are associated but NOT authenticated.
10-15-2010 12:33 AM
Well, the show tech shows indeed that the interface is "Dot11Radio0 is reset, line protocol is down".
And the configuration looks ok...
I would do a debug on the radio itself.
Can you pelase connect via console and enable the following debugs:
debug dot11 dot11Radio 0 carrier busy
debug dot11 dot11Radio 0 tprint
debug dot11 dot11Radio 0 trace print
Then do shut/no shut and send us the output.
Also, one thing i would try is to set the channel dynamic so that the AP chooses the best channel to transmit:
On the radio remove "channel 2437" and add "channel least-congested".
HTH,
Tiago
--
If this answers your question please mark the question as "answered" and rate it, so other users can easily find it.
10-21-2010 07:20 PM
I'm glad to see that I'm not going crazy.
%LINEPROTO-5-UPDOWN: Line protocol on Interface Dot11Radio0, changed state to down |
I have had what seems to be the same issue with my 1142N (autonomous; no controller). It will work for a day or so and then all of a sudden the radio(s) will go down (e.g., be disabled). For example, when I came back to the site Radio0-802.11N2.4GHz was disabled while Radio1-802.11N5GHz was enabled. Devices were able to connect to the AP an hour earlier can now no longer find the AP. When this happens the LED on the AP will usually remain light blue which implies, I believe, a connection which is odd considering there are none when this occurs.
I won't be able to reconnect to the 1142N until I reset the AP. Usually I have to pull the power (POE through Cisco injector) to achieve this as issuing a reset from the web management page does not always work.
Product/Model Number: | AIR-LAP1142N-A-K9 |
System Software Filename: | c1140-k9w7-tar.124-21a.JY |
System Software Version: | 12.4(21a)JY |
Bootloader Version: | 12.4(18a)JA3 |
Note: "channel least-congested" was enabled by default and is currently enabled.
I've read that the AP can shut down radio(s) based on power but I don't see any power messages in the log. Under System Power Settings of the web management page it reports NON_CISCO-NO_CDP_RECEIVED but the radio(s) work fine for at least a day when the AP is reset. I'm currently using a Cisco AIR-PWRINJ4 injector and I've also tried using the POE port on a Cisco ASA5505. Same issue.
I've had a pending SR with Cisco TAC for over a month and have made it no where fast. I've sent them logs with show controllers d0, show controllers d1, show log and show tech per their request. Having to manually reset my AP every other day is ridiculous!
10-22-2010 06:12 AM
One question What is the "power inline negotiation" configuration of your AP?
If CDP is failing maybe the AP gets into a "low power" profile and shutdow the radio. I suppose you should see a message log indicating this behaviour but is possible.
Best Regards.
10-22-2010 07:18 AM
r-manzano wrote:
One question What is the "power inline negotiation" configuration of your AP?
If CDP is failing maybe the AP gets into a "low power" profile and shutdow the radio. I suppose you should see a message log indicating this behaviour but is possible.
Best Regards.
These are my settings. I'm currently using a Cisco AIR-PWRINJ4.
Power State: | FULL POWER |
Power Source: | NON_CISCO-NO_CDP_RECEIVED |
Power Settings: | Power Negotiation Pre-standard Compatibility |
Power Injector: | Installed on Port with MAC Address: (HHHH.HHHH.HHHH) |
10-22-2010 08:07 AM
Is connected to a Cisco Switch?
Can you set the "power inline negotiation" to pre-standard source??? It can fail if the Switch does not use CDP (or it´s disabled) and the switch does not support "Intelligent Power Management" when using the actual power inline negotiation.
Best Regards.
10-22-2010 08:11 AM
r-manzano wrote:
Is connected to a Cisco Switch?
Can you set the "power inline negotiation" to pre-standard source??? It can fail if the Switch does not use CDP (or it´s disabled) and the switch does not support "Intelligent Power Management" when using the actual power inline negotiation.
Best Regards.
I tried setting "Pre-standard Compatibility" via the web managemnt page but the setting does not seem to take. The AP defaults back to "Power Negotiation."
The 1142N is currently powered by a Cisco AIR-PWRINJ4 power injector which is connected to a NetGear switch.
Thanks.
10-25-2010 05:00 AM
Can you do through CLI??
Casually today I had two Autonomous 1142 Ap´s reporting the same message error and setting this command line by CLI I did not see it anymore.
Best regards.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide