cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
10373
Views
10
Helpful
15
Replies
henrikgeorgson
Beginner

Packet loss and high pings

Hi,

In my office we have a WLC 2504 and three 2702i. They works pretty well but not perfect. The normal user thinks that it works fine but our sys admins are complaining over SSH sessions that drops, Skype is dropping calls and Voip isn't working very well. When I try to ping our firewall over the wireless network I have quite rough response time and about 1% packet loss. 

We have the latest fw.
We have about 100 clients connected to three different wlans.
I've been tweaking data rates after reading other posts here.
I've disabled session timeout.
We are only using WPA2/AES.
I've disabled Client Load Balancing
I've disabled Client Exclusion
We are seeing the same behavior on both Macs and Chromebooks, even when the computers are like one meter from the AP.

We had two 2602i before and had the same issues there.

 

Any ideas? 

15 REPLIES 15
Erik Witkop
Cisco Employee

The first thing to do is to isolate the issue. Let's make sure that the problem is definitely related to the wireless network. Take a PC and connect to the same switch that the WLC is connected to. Run the same SSH, ping, skype tests... If that is working fine, then lets proceed.

Check the statistics on ethernet port that connects to the WLC. On the switch check the ethernet port for packets drops etc...

If that all checks out, I would run the WLCCA found here:

https://supportforums.cisco.com/document/7711/wlc-config-analyzer

 

 

Forgot to mention that I've 0% packet drops over our wired network and all is connected in same switch.

 

I will run WLCCA and come back with the result.

 

Thanks!

Have you tried updating the drivers on the device?  What code are you running on the WLC?  Another test you can do is to create a open SSID and test from that and see if you still are having issues. That can eliminate an issue in the controller, and network side. 

-Scott

-Scott
*** Please rate helpful posts ***

To add to Scott's post .. What wireless nic ? 

 

What model aps 

"Satisfaction does not come from knowing the solution, it comes from knowing why." - Rosalind Franklin
___________________________________________________________

Hi,

I got the same results on many different kind of devices. So I really don't think that this is a driver issue. I've seen it on Mac OS X, Chrome OS, Linux and Windows.

The controller is running 8.0.110.0.

I just tried to create an open SSID and I got the same result.

Anything else that I can try? Any input on the wlc config analyze report?

Thanks

Have you tried your SSID on 2.4 GHz and 5 GHz only . Not mix, but one or the other and compared results?

"Satisfaction does not come from knowing the solution, it comes from knowing why." - Rosalind Franklin
___________________________________________________________

No LAG activated anywhere in the network.

I did try to set up a test ssid on 5GHz with same result as usual.

Anything else that I can try?

Okay so LAG is disabled on the WLC and you only have one port off of the WLC connected to the switch.  Can you provide the output of the show run-config in a text file for us to review?

-Scott

-Scott
*** Please rate helpful posts ***

I have attached the output in a text file.

 

Thanks alot!

/Henrik

Henrik,

I will look at it more later, by what I see is your wlan config looks good, your data rates maybe can be tweaked a bit. If you need 11mbps because of old 802.11b clients then you would need to keep that mandatory. If you just want 11mbps, then you could set that as supported and maybe have 24 and 54 as mandatory.  You can follow that on the 5ghz also or try 18 and 54.  

The one thing I noticed is the client count. You have a lot of clients on a given AP and if that is where your testing, then that can be the issue. I have schools I have done work for see around 30-35 clients before users start complaining. Take some readings when there are less clients and see how that goes. What your trying to determine is if it's client counts that is causing the slowness or not.

-Scott

-Scott
*** Please rate helpful posts ***

Great.

 

Well, the problem is that we used 2602i before and there was around 50 clients per each AP. So I upgraded to three 2702i with the exact same result.

I've also been trying after office hours with less connected devices, still with the same result.

Some of my latest tweaking have done it impossible to connect with Android devices while iPhones are working fine. Any idea there? :)

 

Scott,

Have you had the time to look into this?

I solved the Android problem.

I'm really stuck!

 

 

/Henrik

You have LAG enabled?  if so, you should disable LAG and only connect one interface to the network and see.  I run a 2504 in my home lab and had issues with LAG in the past, but not on 8.0.110.0.  This is just to test things.  If you do have LAG enabled and have the WLC interfaces connected to the same switch, make sure that the etherchannel is set to channel-group mode on and your ehterchannel load balancing is set to src-dst-ip.

-Scott

-Scott
*** Please rate helpful posts ***

Hi again,

Here's the output from the analyzer. There's a lot of small things that it suggest. Is there anything particular that I should dig deeper into?

WLC Config Analyzer - Report

 


 

Controller Messages

FYNDIQWLC

30084,General: Virtual Gateway IP is not on 192.0.2.0/24 , 198.51.100.0/24 , 203.0.113.0/24 networks, change to recommended to avoid overlapping with Internet Allocated addresses. RFC5737

30057,General: Disabling low data rates/11b can help to optimize the channel utilization on the 2.4 band. Depending on RF coverage, or if using legacy clients, this may cause problems. Please validate before enforcing the changes, as this may have important RF dependencies. Global Configuration

30058,General: Multicast unicast mode is suboptimal transport for networks with IPv6, mDNS, etc. Multicast mode is recommended. To use it, you also need multicast routing between WLC and Aps

30065,General: EAP request retries lower than 3. EAP requests may benefit for faster recovery, and better behavior on bad RF, by using higher counts, lower retry timeout. Please validate on your specific client types before enforcing the changes

30067,General: Minimum Rogue RSSI detection threshold should be set to -80 or higher, unless mandated by your security policies. Current Value: -128

30076,General: Controller without time source, please configure a valid NTP server

30077,General: Controller with telnet enabled, this is not advisable for security issues

30092,General: For enterprise environments, it is recommended to use DCA with 40 MHz channel width, except for High Density deployment scenarios

30098,General: ED-RRM is not in use. It is recommended to enable for enterprise environments. Band: 2.4 GHz

30098,General: ED-RRM is not in use. It is recommended to enable for enterprise environments. Band: 5 GHz

30101,General: Detected channels on band 100-140 as not in use for DCA. If country regulations allows it, it is advisable to enable to improve channel distribution on 802.11a band

120003,Security: It is recommended to monitor all channels for rogue detection. Band:5 GHz

120003,Security: It is recommended to monitor all channels for rogue detection. Band:2.4 GHz

30081,Enterprise: Load Balancing is a recommended best practice for high density environments

30091,General: Band Select is not in use on any WLAN. it is a recommended feature when there is a good AP density in Enterprise deployments, Avoid on voice WLANs

30082,General: Local Profiling is a recommended best practice for better client visibility

30037,General: Non default RRM timer in use. This is not recommended unless directed by Cisco support. Coverage Hole Interval 11a:90

30037,General: Non default RRM timer in use. This is not recommended unless directed by Cisco support. Coverage Hole Interval 11b:90

 


 

AP Messages

AP74a0.2fff.eb40

20012,AP: Empty primary controller. It is recommended, to have a primary controller name configured, for better/more predictive AP join process. This is not mandatory

20017,AP: Syslog messages are sent to broadcast address, if there are errors reported by many APs, and there are too many APs per vlan, this can cause broadcast storms. For best practices, it is better to configure to individual server

120008,Security: AP Local credentials to access point CLI are not configured. For best security practices, it is desirable to configure to Username/passwords to all APs

60014,RF: AP has channel utilization for 2.4 GHz radio higher than a threshold of 29%. Effect depends on RF conditions

60013,RF: AP on channel 56 has a neighbor on a side channel for 5 Ghz radio, channel: 60 with power -58, Mac: f4:cf:e2:b2:61:10. Effect depends on RF conditions

60008,RF: Load Profile Failed in radio 5GHz, per controller profile settings

60008,RF: Load Profile Failed in radio 5GHz, per controller profile settings

AP74a0.2fba.ff5c

20012,AP: Empty primary controller. It is recommended, to have a primary controller name configured, for better/more predictive AP join process. This is not mandatory

20017,AP: Syslog messages are sent to broadcast address, if there are errors reported by many APs, and there are too many APs per vlan, this can cause broadcast storms. For best practices, it is better to configure to individual server

120008,Security: AP Local credentials to access point CLI are not configured. For best security practices, it is desirable to configure to Username/passwords to all APs

60014,RF: AP has channel utilization for 2.4 GHz radio higher than a threshold of 29%. Effect depends on RF conditions

60013,RF: AP on channel 60 has a neighbor on a side channel for 5 Ghz radio, channel: 56 with power -58, Mac: f4:cf:e2:c3:62:80. Effect depends on RF conditions

60008,RF: Load Profile Failed in radio 5GHz, per controller profile settings

AP74a0.2fb1.c760

20012,AP: Empty primary controller. It is recommended, to have a primary controller name configured, for better/more predictive AP join process. This is not mandatory

20017,AP: Syslog messages are sent to broadcast address, if there are errors reported by many APs, and there are too many APs per vlan, this can cause broadcast storms. For best practices, it is better to configure to individual server

120008,Security: AP Local credentials to access point CLI are not configured. For best security practices, it is desirable to configure to Username/passwords to all APs

60014,RF: AP has channel utilization for 2.4 GHz radio higher than a threshold of 29%. Effect depends on RF conditions

 

Thanks!

Content for Community-Ad