Serious issue with 27.1, MR55 and Intel Wifi chipset

[cmr]

See below for results on 27.1:

Reply from 192.168.1.254: bytes=32 time=197ms TTL=64
Reply from 192.168.1.254: bytes=32 time=314ms TTL=64
Reply from 192.168.1.254: bytes=32 time=517ms TTL=64
Reply from 192.168.1.254: bytes=32 time=21ms TTL=64
Reply from 192.168.1.254: bytes=32 time=957ms TTL=64
Reply from 192.168.1.254: bytes=32 time=7ms TTL=64
Reply from 192.168.1.254: bytes=32 time=453ms TTL=64
Request timed out.
Reply from 192.168.1.254: bytes=32 time=188ms TTL=64
Reply from 192.168.1.254: bytes=32 time=809ms TTL=64
Reply from 192.168.1.254: bytes=32 time=1320ms TTL=64
Request timed out.
Reply from 192.168.1.254: bytes=32 time=156ms TTL=64
Reply from 192.168.1.254: bytes=32 time=53ms TTL=64
Reply from 192.168.1.254: bytes=32 time=1198ms TTL=64
Reply from 192.168.1.254: bytes=32 time=399ms TTL=64
Reply from 192.168.1.254: bytes=32 time=827ms TTL=64
Reply from 192.168.1.254: bytes=32 time=23ms TTL=64
Reply from 192.168.1.254: bytes=32 time=450ms TTL=64
Request timed out.
Reply from 192.168.1.254: bytes=32 time=51ms TTL=64
Request timed out.
Reply from 192.168.1.254: bytes=32 time=87ms TTL=64
Reply from 192.168.1.254: bytes=32 time=92ms TTL=64
Reply from 192.168.1.254: bytes=32 time=402ms TTL=64

And then 26.8

Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=6ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=62ms TTL=64
Reply from 192.168.1.254: bytes=32 time=74ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=8ms TTL=64
Reply from 192.168.1.254: bytes=32 time=5ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64
Reply from 192.168.1.254: bytes=32 time=5ms TTL=64
Reply from 192.168.1.254: bytes=32 time=4ms TTL=64

Is anyone else seeing this?

Android/iOS devices seem okay

[cmr]

The laptop is an XPS13 with Intel WiFi5 8260 chipset, had it four years and never seen this!

[NolanHerring]

Open a case with support. I have an MR45 I can test with as well. What driver version is your Intel card? Does it only happen with your laptop? What about cell phone, or other devices etc.

[NolanHerring]

@cmr  I think your onto something here, but I suppose its so be expected with a new firmware train.

I have a laptop with an AX200 (21.70.0.6) on an MR45 with 27.1, on 20MHz on channel 100 (no other networks on this frequency), no other clients on this AP, and I'm about 10 feet away from the AP.

I am getting the following on 27.1:

Reply from 172.31.0.1: bytes=32 time=21ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=29ms TTL=64
Reply from 172.31.0.1: bytes=32 time=16ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=14ms TTL=64
Reply from 172.31.0.1: bytes=32 time=15ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=24ms TTL=64
Reply from 172.31.0.1: bytes=32 time=16ms TTL=64
Reply from 172.31.0.1: bytes=32 time=20ms TTL=64
Reply from 172.31.0.1: bytes=32 time=13ms TTL=64
Reply from 172.31.0.1: bytes=32 time=16ms TTL=64
Reply from 172.31.0.1: bytes=32 time=14ms TTL=64
Reply from 172.31.0.1: bytes=32 time=19ms TTL=64
Reply from 172.31.0.1: bytes=32 time=25ms TTL=64
Reply from 172.31.0.1: bytes=32 time=16ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=17ms TTL=64
Reply from 172.31.0.1: bytes=32 time=24ms TTL=64
Reply from 172.31.0.1: bytes=32 time=18ms TTL=64
Reply from 172.31.0.1: bytes=32 time=18ms TTL=64
Reply from 172.31.0.1: bytes=32 time=12ms TTL=64

Downgraded firmware to 26.8, and things look much better:

Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=6ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=6ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=1ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=3ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64
Reply from 172.31.0.1: bytes=32 time=2ms TTL=64

Adding @MerakiDave so he can run this up the flag pole. I was able to reproduce the issue (not as severe as @cmr example, but still easily noticeable) within 10 minutes of testing. 

[cmr]

@NolanHerringit only appears to happen on the laptop, the driver version is 20.70.16.4 which was the latest for the 8260 until this week (the newer package didn't contain a newer driver) but I see 21.90.3 is now available at: https://downloadcenter.intel.com/download/29536/Intel-PROSet-Wireless-Software-and-Drivers-for-IT-Ad... so I am going to see if that makes a difference.

[cmr]

FYI the new package takes the 8260 to 20.70.17.1, so it does update, it looks from the .inf that the AX200 will be upgraded to 21.90.3.2

The fix list is below and whilst not directly mentioning Meraki, it does list some performance and disconnect issues with other vendor's APs.  Also I thought my laptop was overheating as I have a few BSODs and, oh look, looks like Intel was the culprit!

Key Issues Fixed and Changes: Package Version 21.90.3

• When system connected to Wi-Fi 6 networks, Wi-Fi 6 enabled cellphones fail to connect to
  Mobile hotspot at 5 GHz.
• A disconnection might happen under specific scenarios, when connected to specific access
  points Apple* Airport or Asus* AX.
• Intel® Wireless-AC 9560 connection p to NETGEAR* Access Point Nighthawk RAX80
• After resuming from Sleep mode on systems with Modern Standby, the Wi-Fi network scan
  list is empty.
• When streaming online videos, audio gets interrupted.
• Windows* Stop Errors (BSOD) caused by the Intel Wireless Driver.
• When hibernate or Shutdown followed by resume performed and Airplane mode is enabled,
  Bluetooth does not turn on upon exit Airplane mode.
• Reduced Wi-Fi throughput on specific channel.
• Intel® PROSet/Wireless Software 21.90.3 has been updated to include functional and security
  updates. Users should update to the latest version.

[NolanHerring]

Well my concern is that on 26.8 its fine, and on 27.1 its got some latency issues. The only thing that changed is the firmware on the AP, so by default I'm leaning towards it being an issue on the AP side

[cmr]

@NolanHerring I think for me, on the WiFi5 chipset, there is a bit of both.  The BSOD is clearly a driver issue but I'm pretty certain some of the performance is an AP firmware issue, as on 27.1 I even see radio dissociation messages during the timeouts on the AP and that doesn't happen at all with 26.8... 🤔

[cmr]

Interestingly on an MR36 it performs just fine...:

Sam laptop with same drivers, hmm

[NolanHerring]

Not surprised, the MR45/55 should be discontinued IMO. I don't think I've seen any successful deployments on that hardware, and there is no reason to ever buy it now that 2nd gen is out.

[cmr]

I'm inclined to agree, I guess they are still on sale as they have unsold inventory.  Oddly 26.8 stopped working with Android phones and going back to 27.1 seems to have fixed it... 🤔

[cmr]

It looks like the laptop keeps getting booted off and reconnecting, the client history tab looks like this:

I tried 20MHz and 80MHz channels, no band steering, 2.4GHz only, 5GHz only, 802.11ax disabled, all behave the same, except 2.4GHz only where the AP no longer broadcasts SSIDs at all, so I've logged a support ticket...

[cmr]

I bought an AX200 M2 card for $17 including postage and installed it, now I get this:

Pinging 192.168.1.254 with 32 bytes of data:
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=2ms TTL=64
Reply from 192.168.1.254: bytes=32 time=2ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=2ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64
Reply from 192.168.1.254: bytes=32 time=2ms TTL=64
Reply from 192.168.1.254: bytes=32 time=3ms TTL=64

Ping statistics for 192.168.1.254:
Packets: Sent = 11, Received = 11, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
Minimum = 2ms, Maximum = 3ms, Average = 2ms

So it is only a major problem with 802.11ac Intel cards and MR55s with 27.1, personally I think that is money well spent rather than more hours debugging Meraki code...😫

As a bonus it connects at 1.2Gbps rather than the ~800Mbps of the 820 chipset 😎

[cmr]

More information for anyone else seeing the issue with MR55s and 27.x

27.2 doesn't fix it, but plugging the power into the laptop does!

Unfortunately the only remaining laptop I have on-site with an Intel Wi-Fi5 chipset, I have forgotten the admin password for, so cannot see the power settings 🙀 but I have two more arriving later this week so will see if there is a fix there.

Good ol' Intel!

[cmr · Accepted solution]

Found another laptop with an Intel 802.11ac chipset, this time 8265, it had the exact same issue and indeed if you change the power settings for the wireless adapter to remain on maximum performance when on battery, it works fine.

I have informed Meraki support and given them the instructions to reproduce, so hopefully it will be fixed in 27.3!

[Ammar-Tanveer01]

Hi All,

We are facing very serious issue with Intel Wireless - AC 9560 Adaptor in lenovo X series of laptops. The users are dropped suddenly and sometimes disconnected from the network. but we have other series of laptop which is T4xxx from lenovo thats working perfectly.

Can anyone help and share his experience. We just need to understand that the issue is with clients or its with our wireless setup.

Also, we are getting the below logs on the wireless health portal:

Wireless Health logs

[cmr]

@Ammar-Tanveer01 what adapter is in the other Lenovo laptops?

If this is happening when on battery power try disabling the wireless adapter power management, set it to "Maximum performance"

If this is happening when connected to the power supply then it is the authentication issue you are suffering from, see those threads for suggestions.