AIR-AP1141N-A-K9 VID: V04 clients disconnet often

[john.wright]

The environment:

20 X air-ap1141N-A-K9 (these are standalone) in three different buildings with either 1 or 2 units per floor depending on sq. footage. IOS is c1140-k9w7-tar.124-25d.JA1.

Problem

Users running various tpyes of wireless cards on windows XP SP3 get disconnected periodically throughout the day whether they roam or not. This disconnect problem started happening after we replaced older 1200 series units running wpa and tkip and replaced with 1141n running wpa2 with aes.

Users running windows 7 do not get disconnected. However there are very few W7 users on campus as this point in time.

I set up a single test unit shown below and associated two laptops to see if they would disconnect. Not once did I lose connectivity.

NAME: "AP1210", DESCR: "Cisco Aironet 1200 Series Access Point"
PID: AIR-AP1252G-A-K9  , VID: V03, SN: FTX1339903J

NAME: "Dot11Radio0", DESCR: "802.11N 2.4Ghz Radio"
PID: AIR-RM1252G-A-K9  , VID: V01, SN: FOC1337107F

IOS: c1250-k9w7-tar.124-10b.JDA3

Any ideas?

I have attached the config of one of the 1141n in service.

[Leo Laohoo]

Did you try to upgrade your bootstrap to 12.4(23c)JA1?

[john.wright]

Leolaohoo

I saw in another post that you responded to a "how to upgrade bootstrap".

However I am unclear as to where I would secure a bootldr file.

Here is what I currently have on the AP.

BOOTLDR: C1140 Boot Loader (C1140-BOOT-M) Version 12.4(23c)JA, RELEASE SOFTWARE (fc3)

[john.wright]

I decided to try performing this procedure on a unit not in service but the units do not recognize the upgrade cmd.

here is what I get.

kbctest#upgrade ?

% Unrecognized command

kbctest#upgrade rom ?

% Unrecognized command

kbctest#upgrade rom

Even if I type the whole line I get this.

kbctest#upgrade rom tftp//:10.49.2.87/c1140-k9w7-mx.124-21a.JA1

                ^

% Invalid input detected at '^' marker.

[john.wright]

Leolaohoo

I guess I need to use copy cmd.

Howeve I just discovered that I am not authorized to download that boot file that you mentioned in the previous post.

[Leo Laohoo]

To upgrade the boostrap of the 1140 the command is totally different from a router.

The command to upgrade the bootstrap is very simple:  "copy tftp:///c1140-boot-m.124-23c.JA1.bin bs:".  And reboot your WAP.

Talk to the authorized dealer and ask them if they can give you the bootstrap.

[john.wright]

Leolaohoo

Thanks for the command syntax.

I am still waiting on our partner to provide me with the boot loader code.

I was also wondering what the difference is between the 1250 and the 1141N?

I replaced one of my 1141N's with a 1250 that has different code 12.4-10b-JDA3 and so far I have not lost any connections.

[Leo Laohoo]

I was also wondering what the difference is between the 1250 and the 1141N?

I hate sounding like a history teacher (so feel free to tune-out anytime) ...

The 1250 is the first Cisco WAP that offers 802.11n.  It is big, bulky and friggin' heavy as a sack of bricks!  IT's got two detachable radio modules.  The logic behind this is due to it's PoE power requirements.  The 1250 requires 20.0 w of PoE power.  When this model was released, nobody knew about the PoE requirements.  The Data Sheet initially didn't mention this.  To circumvent the issue, Cisco recommended to remove one of the radio modules so you can run this AP on 15.0w of PoE.  But this is not what customers want, so Cisco released an IOS that will allow both radios to operate (and give you 802.11n) on 15.0 w of power.  The "catch" is that the highest two MCS rates were disabled.   Aside from this, the 1250 is designed for "challenging" conditions.  It has an all-metal casing and six external antenna ports.

The 1140, however is suited for office use.  Unlike the 1250, the 1140 has internal antennas and can operate under 15.0w of PoE power.

The difference between the two, other than external/internal antenna, physical appearance and weight, is the PoE. 

[john.wright]

Leolaohoo

I really apreciate the history lesson.

I did receive the boot loader file from our partner and I am currently running three laptops to the 1141n with the 12.4(25d)JA1 ios with the new boot loader 12.4(23c)JA1.

Hope this works!

[john.wright]

Leolaohoo

I was able to apply the boot file to our APs this morning but we still have the drops.

The drops are selective. One user reported that he lost connection to internet browsing and live meeting but he stayed connected to outlook and microsoft communicator.

Sometimes he loses connection to communicator but other connections stay alive.

When he reported the dropped connections to me today I looked to see if he lost association to the AP but he did not.

The only thing he did was walk from his desk to a conference room on the same floor and I saw that he roamed to another AP. He said that he actually did two live meetings and the drop occured after he tried to move to the second live meeting that he had scheduled. While waiting to see of he could make the live meeting conenction he tried to connect to outside web site and he was unable to that.

Any ideas?

[Leo Laohoo]

Hmmmm ... If there's one thing I hate about wireless it is "wireless".  The most common complaint I get from clients are "wireless ain't working".  Almost exact description to what your have now.  They sit down, loose connectivity to applications and internet and immediately MS tells them it's a network issue.  The problem to this is they don't look at the signal strength indicator which is showing them 3/5 to 5/5.

Back to your issue, if there was an issue with wireless signal is it evident?  Do you see logs like "%DOT11-6-DISASSOC: Interface Dot11Radio1, Deauthenticating Station Reason: Sending station has left the BSS"?  For autonomous, this message indicates that the client has lost wireless connectivity to the specific WAP.

Have you looked at the Systems Logs of the client?  I mean wireless is pretty consistent.  If you loose wireless connectivity, you loose the lot.  You don't loose Outlook but internet works (or something like that). 

[john.wright]

Leolaohoo

Thanks for taking the time to help.

I do see a lot of the messages that you posted. Here is the log from the AP from which I tracked my own laptop which I left running on my desk when I left yesterday. When I arrived this morning this laptop was connected to the AP but these messages were in the log.

Dec 15 15:21:24: %DOT11-6-ASSOC: Interface Dot11Radio0, Station

KBCOH1WAPAD3S 5894.6bdf.6ab0 Associated KEY_MGMT[WPAv2-CP]

Dec 15 22:55:41: %DOT11-6-DISASSOC: Interface Dot11Radio0,

Deauthenticating Station 5894.6bdf.6ab0 Reason: Sending station

has left the BSS

Dec 15 22:55:42: %DOT11-7-AUTH_FAILED: Station 5894.6bdf.6ab0

Authentication failed

I am not sure what those signal strength number are that you posted but people tell me that when they look at their signal strength it is almost always excellent.

Here is what I see for this laptop that I mention here in the post.

Signal Strength   : -47  dBm           Connected for    : 2621 seconds

Signal to Noise   : 45  dB            Activity Timeout : 9 seconds

Power-save        : Off                Last Activity    : 11 seconds ago

I did not look at the signal for user mentioned in this post but he and others have said that it is excellent. The unit close to me is about 25 feet from me. The user I mention in the post sits right under the AP and when he goes to the conference room on the same floor the unit is no more than 25 feet from everyone in the room. I have three on this floor using chan1,6 and 11. On the floors above and below I have two each and I have them using either 1,6 or 11 depending on where they are in relationship to the one below them and above them. The unit on the floor above me which happens to be almost directly above me is using chan 6 and the unit right over me desk is chan 1.

Thanks again for your efforts.

[john.wright]

Leolaohoo

I want to correct something from my last post.

The unit that I mounted is the AIR-AP1252G-A-K9.

[john.wright]

This issue was resolved when we moved the whole wireless network to its own subnet. it was determined that there were too many broadcasts and the subnet on which it was running.