psydii

It might pass the initial test but fail as other nearby links come up and the noise gets too much. Then the link goes down and auto-negotiation starts again. I can imagine this getting not pretty quite quickly for some sites. See NBase-T downshift.

I do wonder if the increased prevalence of 2.5gb switches and devices might be a bit of a sleeper problem for sites with longer cat5e runs. Particularly if they weren’t installed and tested to spec.

I concur, there is nothing customer-side connected to that infrastructure. You can turn it off.

A fun exercise is getting the finance office to give you all of the bills for all of the telecom/alarm/service lines and cross checking those against what you can find around the building. Is anything actually plugged in/punched down to this BT termination point? If so, what is it?

I had an almost identical mistake in a firewall rule. It was there for ten years before we had a device with an IP address the wrong side of the bit mask. If I hadn't been looking at the firewall config for another ticket at the time the new device failed to work properly, I don't know whether I'd have ever got to the bottom of it.

I'm confused. Why in a hub and spoke arrangement would you not have spanning tree on the core? The middle of the network should be the arbiter of the topology, otherwise you are going to have a 'random' edge device being the root, and any connectivity failure there and you might end up with a full spanning tree recalculation happening or things going weirdly split-brain?

Maybe I missed the memo, but this caught me by surprise... https://find-and-update.company-information.service.gov.uk/company/03416996/persons-with-significant-control https://arbor-education.com/blog-welcoming-timetabler-to-the-key-group/

I've always read it as the salary will be pitched to be competitive for where they perceive you perceive your value/worth to be. I would expect that the collective wisdom is to not advertise salary in order to make it possible to snaffle up previously underpaid talent without paying true/commercial market rates, while still also attracting talent from the commercial sector who know their value.

If that doesn’t yield results it might be worth asking noctua?

Internal charging for full cost recovery: assets (assuming a 5 year life), maintenance, toner, paper, staples etc. We assume there are some "cost centres" that are not actually funded (student printing being one) we set the charges below so those cost centers do not leave us in the red. A4 BW 0.02 A3 BW 0.04 A4 Colour 0.04 A3 Colour 0.08 These are our MFD print charges with 15 devices across a staff of 200. We do not provide direct financial incentive to use the devices in the central reprographics facility, however if they submit jobs to the repro tech, they can (depending on the job) run them on the Riso's which give colour at significant discount. All our machines in the central facility are run at close to capacity. Most of the other machines do not do volumes that make economic sense, however they provide instant access and a backup is a machine or two fails in the central facility. Printers are charged at roughly twice/three times the MFDs (we use the same calculations across the printer to set charges as we do for the MFD except we fudge the volumes in the calculations because they actually do so little compared to the MFD we'd end up charged £1 per page of BW if we put the real values in!). Following a rejig of leases these prices actually generate a significant surplus, and the finance office apply discounting to departments. The idea currently is this discount will be reduced year on year until they are paying full price, at which point devices will start to age out, and our lease cost will need to increase again by perhaps a third. Hopefully without causing pain to the departments.

Lenovo ThinkPad is not the same as "Lenovo", except when it is. The V range is low-end value for small business who don't yet need/realize the benefits of the 'enterprise' firmware/driver support. A few years ago they were physical reasonable, just below the lower end HP ProBook, in my opinion. The ThinkPad range is where quality has definitely been a bit variable recently. Their value-enterprise range (L Series) is definitely at the lower-edge of reliability/robustness in the hands of teachers, and a fleet wouldn't last 18 months with students. Interestingly the E series (which is below enterprise-grade in terms of software/driver support) has proved somewhat more physically robust that the L Series. We had a few T series, all of which developed faults due to inadequate chassis strength. The non-Thinkpad 11e/11w N100 / 13 series devices - aimed at K12 only are the sweet spot for us. Incredibly reliable and quite robust - but we don't have kids taking them home in school bags.. so my opinion might be different if we did. These units also seem to have all the benefits that are supposed to come with the ThinkPad brand, but without the price. Dell dropped me in a hole in 2009 with a complete lemon of a product, which consumed 80% of the budget over 5 years. They would literally have to give me a fleet before I'd consider them again. HP, fine no issues, but no recent experience. Looked promising with their unified new EUFI firmware back in 2016 or so.

Microsoft are make big changes ahead of Server 2025, I wonder if some of this new code has found its way into the client-side Monthly updates. Last time they made such big updates (12 years ago) the problems caused/were caused by bugs at almost every level the systems, client tcp/ip, filesystem redirectors, client-side-cache, memory management, the smb server components, and then finally Office apps and the way they handled files too.

Here's a quiet little announcement by the Powershell Graph SDK PM , it suggests Microsoft are (finally) building powershell cmdlets useful getting actual things done in Entra/(and 365?) again: We should see something in the next couple of months.

Yes, two weeks ago. We've got enough of SIMS out on cloud platforms that there is no reason to maintain a publicly accessible RD solution. Thing is, bad things still sneak in via email and social engineering.

Thanks James for those numbers. My thoughts were more around how the ratio of primary:secondary as a function of MAT vs Non Mat effect the product development demands. If Arbor's 30% increase in Secondary schools is significantly out pacing increase in primary end users, your account and product managers might just be about to start to experiencing the stakeholder management pressures to which I was referring. I must say though, it is noticeable that Arbor has managed to outpace Bromcom's growth while avoiding the perception of the performance and feature delays that seem to have plagued them for a decade. Long may that continue - it will certainly make any decision i need to be involved in easier. About a decade ago I had a great chat with one of your founders. They had a vision and gave the impression of competence that to me (sitting on the sidelines of the MIS wars for now) seems to be part of the core values of the company 10 years later. It will be interesting to watch how things grow over the next few years as sims likely continues to shed its customers.

Guys and St Thomas', Royal Brompton and the Evelina Children's hospitals have declared a major incident https://www.bbc.co.uk/news/articles/c288n8rkpvno

We sometimes see similar in 365. For us the trigger is the email from a parent (or colleague who is also a parent acting in their parental role), that because it references internal data (student names, job titles, other colleagues etc) and comes from an email address that is similar to internal addresses (Jo Smith the Head Teacher's private email is "Jo Smith" which is somewhat Close to "Jo Smith" their 365 email) it gets flagged either as potential unsolicited commercial email, and other times as a phishing/first-contact event. I just have to trawl quarantine twice a day and release incorrectly caught mail. 365 does appear to be quite good at adapting, and we've had fewer emails getting blocked recently.

This is a dotnet clickonce app. I am really not sure why this sort of problem isn't pre-solved by Visual Studio's default build process. Users not being admin is hardly uncommon within the set of users who might need to run one of these things. The other gotcha (this year) is they (AQA/DRS) changed their signing certs. So if you have a publisher rule in Applocker, you'll need to update it by digging out the actual exe post install (buried in the appdata\local\apps\2.0\..... folder. Check your applocker event logs for the correct path).

Microsoft *signed* exe files. Because sometimes we take shortcuts because it's better than not doing anything. Slapping an applocker deny rule using the rule suggested by @GeorgeScott seems like the quickest win to get this absolutely daft problem closed off. I really can't understand why Microsoft is working against the administrators who are the last bastion of their dwindling advocates. Though just writing that out does lead one to a potential conclusion, and if that's true I'm hoping that in 15-20 years time knowing how to admin a Windows 2000 era AD and GPOs will become like having COBOL skills is today.

I'd be interested how Arbor seems to have navigated this (feature request/ bug prioritisation) problem. It seems to me inevitable for any product of this class. I wonder if perhaps its the nature of the customer base? Arbor is bigger with Primary heavy MATs (where the customer is the MAT and the leaders of the MATs can ensure they don't over promise to their users), where as Bromcom has seen more success in Secondary schools and these are might be more independently minded and less prepared to accept roadmaps and prioritisations strongly influenced by the needs of other organisations?

Bromcom aren't alone in taking a while to address some bugs/requests. This Minecraft bug was opened in 2012 and was finally addressed yesterday (12 years later): https://bugs.mojang.com/browse/MC-67

Restore the page from a previous version?

All in on A5. The integration of telemetry between clients and infrastructure gives a really comprehensive view of the estate. We've not had an mail-based compromise since moving to A5 in 2019. For email its not got many (any) knobs and dials really. It just does its job and (almost entirely) gets out of your way. I spend 5 minutes in the morning and 5 minutes in the afternoon sweeping Quarantine to rescue email from parents whose vanity domains don't have dmarc etc configured, or emails from staff personal accounts to work accounts that get caught as "impersonation attempts". But most of those ten minutes are spent sipping my coffee tbh.

A *very* long time ago, I found the support/warranty on the x450 to be disappointing. The hardware was also a little prone to failure. It was replaced by some procurve kit which is still going strong 15 years later. On the networking side though, the extreme kit (and wireless) was way way better. With two EAPS rings we had load balanced traffic and extremely fast convergence on topology changes (no interruption to voice calls etc etc). The wireless tech used by extreme at the time was Chantry and AutoCell, and that IP has been round the houses a bit since then, but ended up purchased by Extreme a few years back. We have Aruba 315 APs right now, and these are as good as the Extreme wifi we built a school around in 2006, nothing else had come close in terms of reliability, resiliency and performance.

I've suggested we go back to paper and chalk boards a few times.