Few months back we came across an issue, not sure how relevant it is. My iPhone (and colleague's MacBook) would be fine for 5 minutes and then it would lose connectivity to internet. The workaround would be to reconnect the device or renew the IP address and then it would work for 5 minutes and then stop working again and the cycle would repeat.
To troubleshoot it, I started continuous pings from the phone to 8.8.8.8, AP saw the traffic, WLC had ARP entry for my phone, the gateway (Juniper switch) had Arp entry for the first 5 minutes of the fresh connection and then all of sudden it removed the entry although my phone was constantly sending traffic and was not idle. Since the issue started after the reboot of one Juniper device in the pair (the traffic going through another device was fine, the traffic is load balanced across the pair), we had to restart that device and that fixed the issue. Jordan: you could reproduce this issue on both 8.5 codes? Thanks, - Cheers, Tariq From: The EDUCAUSE Wireless Issues Community Group Listserv <[email protected]> On Behalf Of Cox, Jordan D Sent: Tuesday, 27 October 2020 2:38 AM To: [email protected] Subject: [WIRELESS-LAN] MacOS Disconnections on Cisco Controllers Good morning, We have been working with Cisco TAC to troubleshoot an issue where our MacOS clients will randomly lose connectivity to the default gateway (and thus internet etc.). The wireless will stay connected in the run state, but the Mac will send out repeated ARP requests for the default gateway during the outages. The outages last between 20 seconds to 5 minutes and is resolved once the client gets an ARP response from the gateway. We have packet captures showing ARP requests going through the CAPWAP tunnel to the controller but NOT leaving the controller to the gateway during the outages. TAC has acknowledged the problem is on the controller, and I'm waiting to hear back from them. I'm wondering if anyone else has seen similar issues? More details: * WLC is two 5508 in HA configuration * WLC was running 8.5.161.0 and we upgraded to 8.5.161.7 to troubleshoot * 250 APs are running in local mode (the issue does not happen when testing in Flexconnect mode with local switching) * Default gateway is a Palo Alto firewall * The MacOS client sends an ARP broadcast to find the gateway every 20 minutes but the outage doesn't happen every 20 minutes * It seems like the issue appears during high utilization on the controller since I didn't see any issues when testing over a campus break when many students were gone * I've seen the issue on multiple SSID's including a test SSID which only had my clients on it * Client debug on the controller shows no issues * This doesn't seem to affect Windows machines Thank you! [cid:[email protected]] Jordan Cox Network Admin II, Information Technology P: 651-882-3995 [email protected]<mailto:[email protected]> | www.unwsp.edu<https://protect-au.mimecast.com/s/ca3nCANpgjC4oPKyuGUwPj?domain=unwsp.edu/> Equipping Christ-centered learners and leaders to invest in others and impact the world. ********** Replies to EDUCAUSE Community Group emails are sent to the entire community list. If you want to reply only to the person who sent the message, copy and paste their email address and forward the email reply. Additional participation and subscription information can be found at https://www.educause.edu/community<https://protect-au.mimecast.com/s/XXgsCBNqjlCjmkXQt67G4M?domain=educause.edu> ********** Replies to EDUCAUSE Community Group emails are sent to the entire community list. If you want to reply only to the person who sent the message, copy and paste their email address and forward the email reply. Additional participation and subscription information can be found at https://www.educause.edu/community
