dmj

The best solution would be to pay professionals to do it, schools often do this by hiring 'teachers'

It will automatically create the project id with the project

This might be helpful https://developers.google.com/apps-script/guides/cloud-platform-projects

I do vaguely remember something about this, but don't remember the specifics so forgive me if it's incorrect but it should give you some pointers. IIRC Google sheets scripts create a GCP project associated with them. I think you need to go to the GCP management console https://console.cloud.google.com find the project in there, and have a look around, possible in IAM or perhaps in Authentication and see if you can transfer ownership.

Back when I was working at a school with 1:1 Chromebooks, we saw many, USB-C charging port failures. It's a really common failure mode. It's also really easy to fix. It does require an outlay in equipment: a decent soldering station, microscope and air filtration system. It's not terribly hard to do and we must have saved thousands of pounds replacing these ports when they get damaged. A daughterboard would have been welcomed, but the manufacturer's know it's a common issue and it helps with their planned obsolescence - they are designed to fail.

Debian 8 ( Jessie ) went end of life two years ago. Current release is 11 (bullseye). Was the fix to upgrade the base OS, or did they point apt to a smoothwall hosted mirror of jessie? If the latter, is the smoothwall OS actually getting patched anymore? edit: just read @MrBrow post and it looks like it upgrades to bullseye

I can't even register on Ciscos site to download anything, presumably they want to match my email domain with the contract. Not keen on this sort of thing, knowing the second hand value goes to zero and devices are essentially bricks as soon as you want to transfer.

Purely out of interest, I've been burned in the past with Cisco's support policy, and not been able to get (often critical) security updates for Cisco products without a support agreement for the exact device. Have they 'opened up' at all with regards to this lately? IIRC it was with council or contractors provided devices that they then forgot about and we couldn't patch.

ok. seems a bit like throwing the baby out with the bathwater, but radical changes are sometime necessary. It's quite difficult to do this sort of thing on windows because the application would have to be written from the ground up, whereas *nix systems can rely on a range of subcommands and processes that simply don't exist on windows - packages to handle email, sms are baked in to *nix systems so tying them together is fairly trivial (see Unix philosophy https://en.wikipedia.org/wiki/Unix_philosophy ). You are asking for quite a niche application and I'm not really surprised that there are not many windows only options. The market for this would be tiny, and most people would take the easy (and free) route. As you are asking for a radical change, I would approach the problem form a different perspective: Reduce the risk of internet failure - get a backup line to send sms, email, alerts whatever. It's ultimately a better solution because all you really need is a method of informing you of failure. A backup internet connection with automated failover is a better solution than fabricating a backup solution that can *only* communicate using sms.

Checkout Bitwarden. It should solve this issue as you can share credentials between staff. Pay or self host.

Can I just recap as I'm getting confused: You've got a system that currently works but the operating system it runs on is EOL shortly. You don't want to port the current application to a new OS, but you do want to change OS and change the application. You want to pay for support.

You don't need to change operating system, modern windows can run this in containers, or WSL

Does the windows pc have WSL installed? There's a straightforward blog post explaining how to do this https://blog.danman.eu/e-mail-to-sms I'm sure the author would be delighted to take your money if you needed support.

Is it still an electron JavaScript webapp? Surely that's the most easily deployed app there is - Point a browser at a URL....

I suspect it's from the RIPA legislation, that upon notification from the home office you would be required to hold data on an individual for a maximum period of 12 month. I guess the DfE took that figure and just made sure that systems have the ability to store 12 months data.

AFAIK it's just a judgment call. IIRC we did about a month, and compressed it wasn't huge amount of data so 90 days wouldn't be unreasonable. Personally I wouldn't go beyond that. In the past (at a former employer) we had notices from the Secretary of State under RIPA laws to log internet connectivity for certain individuals. I wasn't given the details but gather it was related to suspected terrorism.

TPLink everytime for warranty/features. Vastly underrated, especially on this forum.

Don't. There are much better ways to solve this as other have pointed out. Putting 50k users as local logins for a web application isn't the answer to this problem, annoying as it might be.

There's some example configurations here: https://developer.mozilla.org/en-US/docs/Web/HTTP/Proxy_servers_and_tunneling/Proxy_Auto-Configuration_PAC_file https://en.wikipedia.org/wiki/Proxy_auto-config

https://schoolsweek.co.uk/revealed-unethical-ofsted-alerts-tip-off-schools-about-inspections/ Revealed: ‘Unethical’ Ofsted ‘alerts’ tip off schools about inspections

Is this the ChromeOS admin course or something new? I've not done ChromeOS admin but probably should do - the vast majority of our servers are ChomeOS based.

Years ago we did this with google sites, which clearly has O365 auth. IIRC we made them login only for our domain, it must have been 10+ yrs ago so the details are a bit hazy.

I'd have a good look at the TPlink Jetstream range as comparison.

504 isn't a browser/cookie issue. It literally means a server couldn't deal with the request in the allotted time. It's either generated by a proxy (on the users network, or cloudflare) or on the webserver itself. edit: I suppose technically it could be DNS related too, but my money is on the proxy or server. I'd correlate the logs of both with the error and see what was happening.

yeah sorry, that wasn't the intention.