psydii

Do we have thoughts on Ubiquiti kit in this space? https://eu.store.ui.com/collections/unifi-protect

Here, IT "own" CCTV and Access Control, delegated from the Business Manager. Day to day operations handled by general admin staff / SLT (for sensitive matters). 3rd parties do the installation and maintain and support. The 3rd parties say this is one of the normal ways things like this are done, the other most commonly being complete separation from IT and existing infrastructure.

"midspan" I think, is the term you are looking for. Just also to note that the ones in your initial post do not list 802.af .at or .bz standards compliance. These do list (some) of those standards but are more expensive: https://www.networkwebcams.co.uk/index.php?cPath=180 We still have some 14 years old non-managed Powerdsine midspans in service. Never had any problems with them, except power-cycling the attached assets in a physical process.

Most likely, but very much not a given. We’re stocking up on spares just to make sure we can last until its all ten years old and due a refresh under the lifecycle policy (an import ban, or an extension to the ban of new installs at sensitive sites being extending it out to LAs/NHS/Education seems quite possible). That said, we are hoping things don’t deteriorate to the point that it has to be ripped out.

and with a 3-5 year timeframe, I'm not sure that what you buy today will be that relevant when SLT actually cut the cord. In terms of infrastructure and how to frame scoping the servers/storage I completely agree with localzuk above. Who currently has control of your Microsoft 365 tenant? Are the users there provisioned by AzureAD Connect, or 'direct' from your MIS? How are your on prem users provisioned? Who hosts your MIS? Who controls the flow of data from the MIS to other 3rd party services? Working in the dark as a random on the internet, I would say that the most obvious way to move away from a legacy provider is to gradually migrate services to cloud providers where the school is the customer. This will require skill building of the local team to ensure they are able to provide the necessary support and IT leadership services in house (by closely with new providers in the migration and implementation support phases, in addition to technical and soft skills training). In reality it depends on what you discover about who has the ownership and control of the services upon which you currently depend. Bringing the servers in house first may be a good path (if say, outsourced is unreliable or unable to provide the scale and performance you require), but also may be a short trip into a dead end and a world of pain (servers and storage just work, until they don't. I haven't done any server or storage support and development work in over ten years until a few week ago when something went wrong, and oh boy was I glad I had 1st party 24/7/365 6hr break-fix support). While on prem servers are in general a dead end if you are making changes to the structure of your IT Service, I still don't see AD DHCP or DNS or Radius, some file server capability for some app deployments and legacy non-core but critical apps going anywhere soon, so perhaps it is as good a start to your journey that any other. Virtualise, 3 hosts, 1 storage box (or some sort of virtualised storage between the hosts) and a backup system that is able to boot and access the backup storage physically independently of the virtualised system.

I'm told there is something of a ddos happening at the moment? We've had momentary drops of our internet connectivity for the last few days.

This is the third thing I've seen this morning that might be attributable to problems on the wider internet. Who is your ISP?

Our Veeam solution was designed by someone who went off to be a Solution Architect at Veeam. For the huge (too big in her opinion, but customer demanded) VMs there was one job per VM. Then there was one job for the rest of the datacenter. Over time we've broken it down into more jobs, primarily to make it easier to see when a VM has failed its backup rather than it marking the whole job as failed we see individual notifications for each server. Also when moving to backblaze it was easier to lift individual jobs to the cloud rather than the entire datacenter. There is more opportunity here for mistakes though - have we got the right exclusions, have we really backed everything up? With a Datacenter level job you know you've got everything. Generally though I would recommend following the best practice / example guidance from the Veeam site - 3rd party engineers really love it when then encounter systems that were clearly configured following a standard design pattern.

Having seen how two large universities and three large enterprises handle remote access and the updating of their fleets of laptops and LoB applications, we're doing pretty darn well. Also, staff don't seem too upset that three nights a year I tell them they can't work from home. This year we have a timetable structure that every two weeks allows us to push out updates without interrupting work or requiring laptops to be left on site, so it is even less of a problem. I have run an MIS migration in the face of SLT ambivalence. so I'm not going to push for a migration *because* of these minor inconveniences, but I wont object though if there is a push to move to another MIS without this limitations.

If you meant me, I've just been looking but I can't find them, it was six years ago. Also the chipsets and firmware and standards have all changed since then. So vendor/configuration performance on current standards cannot be extrapolated. I will (and probably should have) acknowledge that there are significant changed in ax that may make 2x2 as or more performant than 4x4 in previous generations, I would love to test (or see test done) today.

One thing I will note (which is more closely aligned with the OP's question) is that 4x4 AP's in our last bake-off were noticeably better at handling large numbers of clients that had a variety of radios. Those vendors who showed up with APs with only 3x3 or 2x2 saw performance crumple much more quickly in our real-world tests (120 clients over six rooms with 3APs, client mix of netbooks(1x1), ipads (various), cheap laptops(1x1), good laptops(2x2) and a couple of mac books(4x4). Also not all vendors implemented some of the channel management tricks of the standard (e.g. temporarily rolling back 80 to 40 or 20 in the presence of overlapping broadcasters) and instead went with very conservative set of capabilities that had lower cpu requirements (which manifested in lower cost APs), this again we saw a performance difference between hardware that used these capabilities and hardware that did not.

Back in a/b/g times Aruba, Chantry, Cisco (with their early acquisition of Aironet) were the best, Trapeze and Aerohive were competitive in some markets and were among the best in the 'b' days. Meru also had an approach that solved many things that the MCA of these three could not. Once Autocell was either licenced or "freely available" all the big players got really good (and chantry disappeared into Extreme via a very round about route). Over time as the newer standard rolled out the smaller players, despite having some of the talent that defined these standard, simply did not have the deep pockets to compete. The value proposition of each of the remaining vendors (and newer entries like Unifi) was always built around their management tools and support. Over time their secret firmware sauce has been replaced by Atheros / Broadcom silicon, and now it is all about the management tools and support. That said, the next time I do a procurement I am going to do a head to head performance check (if I have the facilities), since I've not seen any real-world demonstration that Wifi 6/6e (and extrapolating the trend, Wifi 7) will actually improve empirically the performance the user experiences. As time goes on Wifi (like Layer 2/3 ethernet before it) is becoming commoditized. They are increasingly all using the same silicon and, while the silicon is crucial, the overall product is what matters. Different customers have different priorities and those priorities drive their decisions. [edit: rewrote last sentence to better express my point.]

Today our NM had to retrospectively do a DPIA because "teachers will be teachers", and I had to seek permission to purchase some A4 folders on an order for which I already have an authorised PO from the finance system. So make of that what you will.

I can understand that a specialist networking company is somewhat reluctant to recommend/sell kit where support is not good (in reality it may be fine, it may be enough, but on paper it might as well say "yeah good luck with that"). It is their reputation on the line if there is a problem with the installation. We once went with Procurve for our switched network when the vendor's standard design was cisco (for this national scale project). Cisco was their preferred solution because it wasn't just reputation but financial penalties could be incurred, linked to underperformance. We went procurve because we got more bang for our buck (and fwiw most of that kit is still in use today). However there was a problem (bugs in switch firmware) and when it was first installed the network would occasionally become unusably unstable. It took months before a firmware patch was available. A stressed Solution Architect on the project observed with Cisco they'd have had actual named engineers on the phone working through the issue with private patches shipped within days, but with Procurve all they could do was send an email and wait... It put the school back 10 years in terms of leadership's trust of IT, it was only after Aruba took over the Procurve division that this vendor moved back to promoting non-cisco switches again. 15 years later I've *needed* to update switch firmware three times across four vendors. However with wireless (again across four different vendors) firmware has needed updating almost bi-annually to maintain (or restore) stability. The only time I have had to raise a P1 ticket Aruba already had released the fix. From time to time I wonder if we could get away with Unifi, but then I remember how much damage an unstable network can do to trust.

Wow I did briefly once wonder what happened to xirrus. I mean, once the Propogate/AutoCell stuff was out in the wild (initially through licencing deals, but later with Cisco getting the patents invalidated) products like xirrus arrays and Meru SCA were dead-ends.

We should open a ticket with them every time a user hits this problem. Tie up their service desks, hurt their KPI’s. And tweet them calling it out. Every single time.

some things that spring to mind: Clubs, lunchtime duties, being back up for the duty member of staff, working parties, chairing working parties, outreach to other schools, training staff on behalf of other teams (e.g. finance systems, hr processes etc), running procurements, first aid, fire marshalls, and whisper it quietly *teaching*.

Don't give departments enough budget to spend on big-ticket IT items? Also, did they complete a DPIA? No? well they need to... this gives then pause for thought. (our NM does the DPIAs* so they are forced into the loop whether the purchaser wants them there or not) *supervised by the DPO.

I have a few "special little flowers" of devices on our network. I just tidy them up by hand. The rest if they don't come into compliance automatically or with a quick and dirty script, we wipe and reload. Registry clearing tools are generally not a good idea and may often break a machine permanently in subtle ways that cannot be recovered.

Wasn't even true in 2014. Back then it would have been a brave decision to move from SIMS to Bromcom as a secondary school, but Serco/Progresso was imploding and many of the others were still too new or too stagnant, it made a suitable non-SIMS replacement to any of the competition. These days I can't speak from personal experience, but bringmoredata's analysis shows continuous growth and low churn, which implies it's definitively good enough for many. (Arbor, IMHO was "too new" back in 2014, also shows these properties, now with a comparable customer base to Bromcom, and its growth rate this year well outstripping Bromcom. (caution: variation in LA/MAT wins can dramatically skew this value year on year, and of course the SIMS 3 year lock in will have a chilling effect on growth rates for all its competition)

This is something I have been wrestling with for a while. We are good at SIMS. Updates aren't really a problem (a bit of work for the helpdesk for a few days every four months, but mostly they just work - - and our RDS server is there for anyone whose local sims install isn't working during the first hour of the day after an update because they took their laptop home last night, against advice). It is relatively snappy, it collects the data and produces the reports we need, we have three third-party 'best of category' products that fill in gaps, but they write back to SIMS. None of the other MIS's are as good at these things as these third party products. It all might be costing more, but it is better. (well so we think) So I suppose what I'd be interested in hearing is, if you are a school that is good at SIMS, and already has good/great 3rd party products improving on areas you feel SIMS is lacking, and you have decided to (or have completed) a move away to an alternative MIS solution, leaving SIMS and 3rd party's behind - what was your motivation.... what drove the move... how did you bring those users who were happy with the status quo along, particularly where the new was not as good for them as the SIMS/3rd Party? I'm thinking the data team, the timetabler, the behaviour "team" (if you have a good third party product for this already), did you find you had to make changes to staffing (in admin roles) in order to respond to how the MIS made some tasks easier / more difficult? Related: You have to run a procurement because the contract is up for renewal... how do you capture and evaluate your requirements when you are potentially comparing a collection of products to a single party offering of the alternative MIS's?

Montegu Private equity want their dividends (or perhaps, as is common in these sorts of things, their loan interest repayments)

I did this once "for fun" about 17 years ago. It stopped being fun quite quickly and I rolled it back. They aren't wrong... that would improve the security of your network... and good for you if that is the most effective way for you to improve your security! I do suspect though that almost every school has lower hanging fruit in that regard.

I'm wondering if Bromcom et al. can make can help their customers to make use of this: https://www.notifications.service.gov.uk/

Wifi6e does additional funky things that make this even harder to generalize accurately... https://www.numerousnetworks.co.uk/noversight/optimising-wi-fi-6e-networks-for-apple-devices/