Many of you will remember my articles from 5 years ago where I replaced the Aerohive wireless with Ubiquiti. At the time it was one of the largest rollouts in the Southern Hemisphere and despite a rocky start it was a great success, and the time has come again for me to replace the unifi network.
The existing network was starting to get a little long in the tooth, in part because we had invested in the UniFi EDU AP’s that had worked great but went end of life. Once this happened we had already been moving away from using Bells and we simply turned off the Bell part of the EDU AP and went on with life. Though not having access to upgrade the AP’s mounted in the EDU AP’s was a frustrating thing, especially since we were happy to forgo to the Bell part, the AP was basically an AP Pro. But I understand where Ubiquiti was coming from, while I would be willing to manually flash all of my AP’s and accept it I broke any that was on me. I can imagine the forums filled to the bream with people complaining UI broke their AP.
The idea of it gives me flashbacks to before UI took more control of their cameras and the forums were full of complaining users arguing their camera systems were slow and horrible. Yet when you looked at their spec’s they were on under-spec’d hardware and still complaining about it. It is an area I find funny, you give yours the flexibility to pick and choose their hardware, they don’t do what is recommended and then complain when it fails. You still see it, often people complain they tried to rollout a UniFi network, installed 30 switches and 100 AP’s but then complain the Cloud Keys couldn’t handle it. The product must stuck. But if you read the recommended numbers the Cloud Key’s aren’t designed for that size of rollout.
Pulling myself back into the topic of this post, I’m not writing about users mis-using products and complaining, I’m writing about our rollout that SHOULD be happening here at Christmas, and what I did on Monday night.
In planning for what will be a rather expensive UniFi rollout (yet still half the cost of Aruba). We are looking at over 180 Access Points and only 80 Switches. These will be a mix of U6-Enterprise, U6 Pro’s and USW-48-Enterprise-POE and some USW-Aggregation-Pro. I’ll put more detail in what we rollout at Christmas when that happens.
For now to test that this will be a good solution we have a Phase 1 rollout, this was 12 U6-Enterprise APs and 2 switches. A brand new controller and Radius server was span up. Now last time when I did a rollout it was just CCGS staff installing, and even when we upgraded our new Junior School building we simply had some electricians mount them all for me and I adopted and setup.
This time it is different, and that comes from a desire that I believe that Wireless needs to be considered as Critical Infrastructure in the classroom. So to that end I have partnered with Hubify who is an Australian Company that see’s the benefit of Ubiquiti gear and recommends them. So on Monday afternoon we when swapped the AP’s they were the ones to turned up and installed the new AP’s for me.
And what we thought would take a few hours took about 30 minutes. We discovered that the U6 Lite, SHD and EDU mounting options would work with the U6-Enterprise. Which was great as it meant the rollout didn’t require drilling and patching giant holes (I’m looking at your EDU Mount). And what was even better is that the EDU Speaker Mount while the speaker won’t work (not an issue) it does simple ethernet passthrough and the U6 Enterprise 2.5 Gbps speed is maintained.
All devices up and running and many happy users.
You can see the AP’s connected running with light blue which is 2.5 Gbps speed.
*While I would love to give you some speed tests now with WiFi6E 6GHz speeds… for now I can’t. While 6GHz is approved for release in Australia, I did reach out to a contact at Ubiquiti, though I was only advised to watch this site for information about when it is rolled out. From this site, going purely on guess work here, the fact Australia hasn’t allowed the full 6E GHz range available, and I assume that allowing the smaller bandwidth has larger coding requirements. And since Australia is a small country in terms of total users I can understand why that isn’t give a hit priority.
I wrote the above paragraphs just after the test Phase 1 had been released. Now it is 3 weeks later and I know how it has gone. And it has been a success though with some issues. I’ve spoken to some staff who have given what I consider to be good feedback. Three have said that they can’t actually remember if there were any chances to the wireless, while many would think this is bad. I consider this a success, at this time of year in NSW Schools it is some of the busiest times of the school year, the staff are incredibly busy trying to get everything going. And once I remember them that the wireless had been upgraded you see a click, ‘Hey no wireless issues.’ It’s gone from wireless mostly worked with some staff having issues, to it just works without any complaints.
I have had one issue where users were not able to connect to an AP, though I hadn’t not enabled enough logging to be able to figure out why it was happening, and when Hubify got onto the issue and we turned on the required logging the issue had removed.
For me the trail was a success, and thanks to the partnership with Hubify, even when we run into some issues in the future I still believe it will be a success and work going forward.
I sign off now, though this post was late. So the update will come in a week or two as the rollout of the next stage is almost complete.