We have seen several Mavericks now that constantly go on and offline. This is not the 30 second EAP roaming issue, I am on top of that one. The symptom on this one is constantly going on and offline in our 802.1x wireless network. Packet capture does not show anything strange, just the Macbook re-requesting an IP every time it comes back online. The VERY strange thing is that is also happening on the wired interface, almost like a bad jack or cable. No auth or encryption of any kind in the test wired network I'm using. The only stable connection these machines have is with our open wireless network. I think this is probably particular to our network somehow, but thought I throw it out there. We have seen about 4 (and one this morning running 10.8.3) so far showing this behavior.
We did experienced really weird issues (non-roaming ) with 10.8 but haven't seen many issues with 10.9 yet .
- Device will have session going and out of nowhere it drops and it takes a while to reconnect , and according to the packet capture the device starts the EAPOL converstation and it never responds to the identity request send from the controller
firstname.lastname@example.org - Brandeis University Mail_2013-11-06_13-38-24.png" alt="Mac OSX WiFi Issues - email@example.com - Brandeis University Mail_2013-11-06_13-38-24.png" src="https://higherlogicdownload.s3.amazonaws.com/HPE/MigratedInlineFiles/56c4ce14dd014b93b712e22d5d3c40f7_d9481f9b5a194b1fa6e1e0acbd15d8b4" align="center" border="0">
- The other issue we saw was the device is able to associate / authenticate gets a valid IP address but then is unable to reach anything on the network, but in the packet capture you can see that the device is sending the traffic but not replying
I saw this issues on 220.127.116.11 and on 6.3.1.x after users upgraded to 10.9 we haven't these issues again
From our release notes...
Symptom: 802.11ac MacOS clients are unable to pass traffic to APs in tunnel forwarding mode. Scenario: This issue may be triggered by issues in the client Broadcom drivers, when there are three MPDUs in an AMSDU packet.
Workaround: Change the Maximum number of MSDUs in an A-MSDU parameters in the high-throughput SSID profile to a value of 2.
wlan ht-ssid-profile <profile>
Sorry for going slightly off topic, but regarding the "known" 30 second EAP roaming issue, could someone tell me if this behavior happens with EAP-TTLS or is only EAP-PEAP?
We are using 2048 bit InCommon certs with AddTrust (intermediate) and have many reports of Macs going "off the air" repeatedly.
Any info is appreciated.
I saw it on both, PEAP and TTLS.
I was able to fix the roaming issue with this knob under ssid profile: "Rate Optimization for delivering EAPOL frames"
@mlew2433 wrote:I saw it on both, PEAP and TTLS.I was able to fix the roaming issue with this knob under ssid profile: "Rate Optimization for delivering EAPOL frames"
From another thread I saw you're on 6.2, right? This knob didn't seem to help me with 18.104.22.168 or 22.214.171.124.
In my testing - also with a 2048 inCommon cert - the ~ 30 second roaming issue happens for both EAP-TTLS and EAP-PEAP
In both cases having the Macbook trust the cert for SSL - removes the cert validation delay.
@mldickson wrote:Sorry for going slightly off topic, but regarding the "known" 30 second EAP roaming issue, could someone tell me if this behavior happens with EAP-TTLS or is only EAP-PEAP? We are using 2048 bit InCommon certs with AddTrust (intermediate) and have many reports of Macs going "off the air" repeatedly. Any info is appreciated. MIke
At Aruba, we believe that the most dynamic customer experiences happen at the Edge. Our mission is to deliver innovative solutions that harness data at the Edge to drive powerful business outcomes.
© Copyright 2020 Hewlett Packard Enterprise Development LPAll Rights Reserved.