cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
3175
Views
20
Helpful
13
Replies

Cisco Wireless and HP Laptops dropping DHCP address

brian.bushmaker
Level 1
Level 1

A couple weeks ago, it was been reported that some HP Laptops have been getting disconnected from wireless. The common complaint was that they needed to "disconnect" from WiFi and then "connect" to wifi again on the business SSID. I was able to obtain one of the laptops with the issue and could duplicate the issue. After running wireshark, what is really occurring is the laptop is staying connected to wireless, but dropping it's DHCP address and is unable to obtain a DHCP address until the WiFi is reset. The error is "no access to the internet". I've been troubleshooting the last day, but unable to find anything substantial that can explain this issue. The problem occurs with multiple different models of laptops, drivers, and wireless chipsets (all Intel though). Also occurring in different areas of our facility so it is not limited to one AP. Looking for some guidance to troubleshoot this. Our Wireless controller is cisco wlc 5550, firmware version 8.5; HP Laptops; Windows DHCP. Any assistance or direction is greatly appreciated.

1 Accepted Solution

Accepted Solutions

JPavonM
VIP
VIP

If your HP laptops are using Intel AX adapters, then there are some news about a defect that has been fixed in the latest Intel drivers 22.110.

As you can see on Intel's community, there are some threads about L3 connectivity issues when roaming. Most of times the end-user perceive a drop in the signal followed by a disconnection, but this is only happening in L3 and it is solved upon bouncing the Wi-Fi interface or waiting for the next re-auth. Intel moderators informed in those threads about a defect seen on Intel's drivers and they have finally fixed it on this release.

I've been testing all kind of codes in C9800 since October trying to find a way to fix this behaviour, but at the end it seems that is related to Intel (again).

HTH
-Jesus
*** Please rate helpful responses ***

View solution in original post

13 Replies 13

Scott Fella
Hall of Fame
Hall of Fame

"No access to the internet" is typically when DNS is not working.  What setup do you have, flexconnect?  If so, check your trunk ports on all the ap's and make sure the vlans are defined.  Also check the vlan mappings if using flexconnect.  We have seen those mappings disappear at times causing issues.

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

Thanks for your reply. We are not using flexconnect. However, I wasn't perceiving this as a DNS issue, but will review those logs for any errors.

 

Thanks,

Brian

If a device can't get an ip address, usually it will show that or just disconnect, depending on the device.  If "no access to internet" or another error depending again on device, that is for sure DNS.  On a device that gets that info, if you can ping the gateway, then you have a valid ip address and access to the LAN.  That is what I would first test to make sure that device can access the LAN.

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

It's connected to business SSID with the error, "connected, no access to the internet". If I look at at the wireless adapter, an autoconfigured private IP is assigned (169.x.x.x) so it's unable to communicate with any other devices on the network at that point.

So the device is connected to the SSID, but DHCP is failing.  Check the DHCP server and make sure its functioning for that dhcp scope.  Don't know if its a setting on the controller without really seeing what you have configured.

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

I did review DHCP and didn't notice anything that would be a red flag. Particularly since we have other wireless devices on the same subnet(s) that don't have this issue. Our controller is using an interface group consisting of 3 24 bit subnets. It's my understanding based on reports that it is happening on select laptops on any of those subnets. That's why this is leaving me perplexed and leaning toward a device issue. Even if it is device related, we still need to determine why this is happening.

Well interface groups can put device on specific subnets, which, if the device believes it has the correct address will not request dhcp.  So at the end, the controller places the device on a different subnet, which can cause issues, but in reality, its hashed so typically the device should get the same interface unless your subnets are getting full.  Maybe your lease times are too low or too high.  

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

marce1000
VIP
VIP

 

           - Have a sanity check of the controller configuration with : https://cway.cisco.com/tools/WirelessAnalyzer/

 M.



-- Each morning when I wake up and look into the mirror I always say ' Why am I so brilliant ? '
    When the mirror will then always repond to me with ' The only thing that exceeds your brilliance is your beauty! '

I wasn't aware that this existed. I was working from home and didn't have the setup I needed to do this and will perform it today. I am now leaning toward a driver issue. We were able to duplicate the problem on two "test" laptops. We used an old driver for one and a new driver for the other. The old drive did not disconnect overnight, while the other did. It's too early to determine if that is the official fix, but is promising.

 

Thanks for the tip.

Take a look at the following command: netsh wlan show wlanreport

How to Generate a Wi-Fi History or WLAN Report in Windows 10 (howtogeek.com)

This can show you a lot of details even driver issues.  You also can customize how long to keep track of the reports.  It seems like you are heading in the right direction.

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

JPavonM
VIP
VIP

If your HP laptops are using Intel AX adapters, then there are some news about a defect that has been fixed in the latest Intel drivers 22.110.

As you can see on Intel's community, there are some threads about L3 connectivity issues when roaming. Most of times the end-user perceive a drop in the signal followed by a disconnection, but this is only happening in L3 and it is solved upon bouncing the Wi-Fi interface or waiting for the next re-auth. Intel moderators informed in those threads about a defect seen on Intel's drivers and they have finally fixed it on this release.

I've been testing all kind of codes in C9800 since October trying to find a way to fix this behaviour, but at the end it seems that is related to Intel (again).

HTH
-Jesus
*** Please rate helpful responses ***

Good to know. Of course, I administer the network, so had been looking at it from that point of view.  Now that I've seen some posts I rolled back one of the drivers to 22.40.0.7 which for the moment seems to have corrected the problem. Definitely appears to be a bad driver (22.80.x.x). I'll give 22.110 a trial run on that same device to see how it behaves.

 

Thanks for your reply.

Rich R
VIP
VIP

So first check the Intel driver issues mentioned above.

 

So these were all working fine then all started having problems at a certain date?

When they lose IP does that correspond with DHCP lease expiry time?

If the answers are yes and yes then might be similar to a problem I had a few years back previously mentioned on these forums.

IT team pushed a new version of Windows firewall config via group policy which prevented DHCP renew so when lease expired then customer lost connectivity until disconnect/reconnect -> new lease.

Test with Windows firewall disabled and if that solves the problem then talk to IT team about Windows firewall policy.

If it is this problem then you can use Wireshark and firewall logging to see which packets are getting dropped.

Review Cisco Networking for a $25 gift card