Wireless network stability

[campbech1]

Running WISMs on 6.0.196 and 4404's on 6.0.196 and seeing many wireless disconnects.

TAC has released to me 6.0.197.10 but still experiencing drops and another unidentified bug that I'm waiting to hear back on.

Anyone running 6.0.196 and seeing the same? Might be downgrading from 6.0.196 to 5.2.193 this weekend.

Kind of at a loss and need to get this resolved.

[Chuck Dillon]

I should be getting a newly generated bug id later today. I'll post

it when I get it.

Basically what we think is happening is that when the badge performs a L2, inter-controller roam, the WLC the client roamed to is not updating the switch CAM tables like it should be.  So far, this does not seem to be affecting laptops. However I heard of this happening with handheld scannersas well. So my earlier statement about vocera was because they were the first to complain this is not a vocera specific bug.

[campbech1]

I just got word from our Vocera analyst that Vocera has released a notice concerning upgrading to 6.0.199. Guess this kills my upgrade planned for this weekend. I was hoping this was an isolated issue.

The Vocera Advisory states:

Vocera is aware of an issue that customers are experiencing after moving to Cisco WLC version 6.0.199 that manifests itself in a substantial increase in difficulty with badge communications to the Vocera Application Server over the network. Badges will display "Searching For Server" or "Searching For AP."

Vocera is working closely with Cisco and its mutual customers on the problem.

Please consult with Vocera Technical Support and Cisco TAC before upgrading to 6.0.199.

[Chuck Dillon]

Yes, stay away from the 6.0.199

Here is the bug that has been filed for this issue: CSCti21621

I'm currently on a special eng release 6.0.196.159 - which fixes both of the following bugs: CSCtf34858 Client can't transmit traffic if it reassociates to an AP within 20 sec & CSCte89891 Radio may stop transmitting beacons periodically among others.
It is prior to the 199 'fix' that introduced the above bug,CSCti21621.

We made this change last night and so far so good.

[campbech1]

Is the bug only present if your default gateway for the wireless user is a HSRP address?

[Darren Ramsey]

So I take it this is fixed in 7.0.98.0?  Not sure I'm ready to take on Clean Air bugs to fix CSCti21621.  Is TAC going to release an engineering 6.0.199.x build?

[Chuck Dillon]

Yes, before the bug was removed they had "Conditions: This is seen in cases where the wireless client's default gateway is an HSRP address." and that is exactly how we are set up.

7.0.98.0 is listed as a work around and not a fix because it was released prior to 6.0.199.0 which included a bug fix that introduced CSCti21621.

I have not heard about a 6.0.199.x eng release yet. We have been on 6.0.196.159 since 8/5 with no related issues that we were plagued with in 6.0.196.0 or 6.0.199.0.

[Darren Ramsey]

Chuck,

Good info, I gave you 4 stars!

We've been waiting for 7.0 almost 2 years, specifically for a GLBP fix (CSCsv21441 and CSCsv21464).  Please keep us posted on CSCti21621.

[campbech1]

Chuck,

How has the testing been going and are you still on verison 6.0.196.159?

I was all geared up and ready to install 6.0.199 until the last bug you posted. With us having some gateways on our wireless network being HSRP addresses, it wasn't worth the risk. I have to do something pretty soon. I can't stay on our 6.0.196 code any longer, I'm still having lots of issues.

[Darren Ramsey]

Yes, we were ready to go with 6.0.199.0 as well, until we found out that 199 would break our Vocera. CSCti21621 is listed in the bug toolkit as catastrophic and the workaround is to disable LAG. Good luck with that on a WISM.

I believe TAC and the WBU are extremely negligent leaving 199 posted on CCO with the AssureWave logo. They should have pulled it down and deferred it. How many unsuspecting engineers have downloaded and upgraded based on the AssureWave testing, only to find issue and have to revert back to a previous version.

Having said that, I'm told 6.0.199.2 should be posted in the next 2-3 weeks.....

[campbech1]

Just got off the phone with TAC and 6.0.199.2 should be released later today, Monday at the latest.

[Chuck Dillon]

That's good to know.

I'm happy to report that the 6.0.196.159 has been stable for us. I'll closely read through the the release notes of 6.0.199.2 but I probably will not upgrade at this time. I need to look forward to 7.x because I have a large deployment coming up at the end of this calendar year that will be using 5508's and 3502's so we will want to take advantage of the new features. I'm hoping by then there will be a solid MD for the 7.x train will be released.

not sure if you looked at the CSCti21621 Bug Details recently but the 'Fixed-In' has been updated since I last looked.

6.0(200.0)
6.0(199.1)
7.0(112.20)

[Darren Ramsey]

6.0.199.4 now on CCO. Anyone running this yet in production?

[andreas.eichhorn]

Yes we did it and we have an improvement with the L2 roaming issue bug (CSCti21621). With 6.0.199.0 we had this "20 seconds problem" and with 6.0.199.4 the problems are solved.

Customer installation:

120 1252 APs

140 7921 Phones

2 WISM

In the last 2 days we could not find any other bugs.

We are very dissappointed about that. We opened a TAC Case last week and the TAC Engineer ist still working on it. We told him yesterdam that we solved the problem with this software and got no more respone. I am just asking me why we have to tell this "wireless expert" that there is a new software available that is the solution for our problem. Without NetPro we would still investigating the problem and sending more and more debugs to TAC. So I would like to thank everyone here for helping because NetPro made our customer satisfied.

[michacollins]

Have other folks that were experiencing wlan drops upgraded to 6.0.199 and noticed an improvement?

Short of end user complaints, is there any other way to detect these drops and capture data when the drops ocurr?

Campbech1,

Is the one building that you upgraded to 6.0.199 still running smoothly with no drops?

Thanks,

Mike.

[campbech1]

I have had three complaints of drops / application freeze ups from them last week. The testing has been positive enough that I am going to upgrade our remaining 7 controllers on Monday morning.

I will continue to post my findings.