And for some reason my Apple sent an email before I was done… Continuing…
We had issues with rebootstrapping of radios on Aps in ResNet. This is the same problem (I believe) that UW faced. We have turned on CPSec, restored timers to normal, and have seen no issues since doing so. We consider this a big win for the moment. We had issues with failovers not working correctly after a power outage. We were unable to recreate the breakage, so that issue will have to go into the bit bucket. We also have issues of artificial high noise on certain Aps. This was experienced when the team was here, but we have no root cause at this time. The short of it, I feel as though we’ve managed to get to a work around for several of our issues. Aruba has told us they are committed to using whatever resources are necessary to resolve the other issues, but we are going to let them (and us) take a little break, and hopefully enjoy the few wins that we had. We have some plans together to work on the other issues, but we need to push the ball forward a little bit more before asking them to spend more time with us. Keith Miller can reply to this message with all the things I said wrong. For now, I’m about to take a 2 week vacation and forget about all of this 😊. Thank you all for your interest, and we will keep you up to date as we learn more. Thanks, Ryan From: Ryan Turner <rhtur...@email.unc.edu> Date: Thursday, January 16, 2020 at 5:10 PM To: "WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU" <WIRELESS-LAN@LISTSERV.EDUCAUSE.EDU> Subject: Update on our Aruba solution All, Since the thread generated significant interest last week, I wanted to let you know how Aruba responded. After hearing of our issues, Aruba sent a tiger team (5 or 6 folks) that came in to work on the bugs. We had a punch list of things to work on. On the top of the list was the 515 performance issues. This is where people would stay connected, but data wouldn’t flow for a period of time. The symptoms were reproduced many times during the week with everyone present. Aruba found a bug in code that does not handle queuing properly in certain circumstances. They produced code to fix this issue, but we cannot confirm at this time if this will resolve what we are seeing…. We saw a similar symptom immediately after putting the fix on the AP. After seeing the same symptom immediately after putting on the hotfix, they realized that someone on the team has an intel AX adapter which has significant issues with OFDMA. It can essentially wreck the airwaves for other clients. The solution is to TURN OFF OFDMA on AX access points until Aruba releases a build that can selectively ignore Intel OFDMA (while allowing others). I have a release from Broadcom on January 6 speaking to this issue, so they aren’t making that. I confirmed it with a separate wireless vendor that Broadcom has had some issues on the OFDMA front. I plan on keeping it off for likely the next year as we don’t really have a significant quantity of ax clients to make it work the hassle at the moment. ********** 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
