Garacesh

Medical data is one of those things where we respect people's privacy. In theory, the person with 'rona informs who they've been in close contact with. Management here have made a huge fuss about seating plans being accurate because of contact tracing, can't fault it really, not the worst idea.

You might be able to repair your own device, but you're an ICT professional with transferrable skills. For the rest of the populace, there's third-party repair stores, except apple go after those because they want you to go to the Genius Bar[tm] and get ripped off to high-heaven told it's not repairable/has water damage/etc. And they actively try to quash any kind of repair information about the hardware. Apple are so incredibly anti-right-to-repair it's nuts. (If I'm being reasonable, most consumer electronics organisations are now, but most of them just say "Nah, warranty's void now, mate. Soz." whereas apple take it to a whole new level) Louis Rossmann has done a lot of videos about this, and about their refusal to fix even the simplest of problems (I remember one where it was just a pin on a ribbon cable being bent, that was a very common issue on that model and Apple said was "unfixable", 30 seconds later, repaired) As for the stifling innovation, I do admit that I can see that argument. Except the latest iPads use USB-C over Lightning. So clearly it's not that terrible, eh? The solution is obvious: If they want to innovate, they can do that, but they have to submit their innovation to the USB consortium for the greater good of all consumers (and device manufacturers). That being said, whoever is in charge of USB needs to pull their finger out and get a better grip over their standardisation. The bajillion different implementations need to go. Something like yearly major revision versions that are backwards compatible and not ten million different tweaks of it being realised whenever the hell a manufacturer feels like it. It needs to be properly enforced. On the best end, the current setup just really muddies the water and allows for sub-par gear to be sold via lying by omission. At the worst-case end, there's stuff like the Nintendo Switch charger being USB-C-except-it-isn't and frying phones because it's pumping them with too much juice.

We have some information to send off to the NHS about certain pupils and obviously this should be sent encrypted. However, when in GMail (browser version), all encrypted attachments (password protected .zip file, created with 7Zip, if it matters) fail to upload. Tested in both the latest Chrome and Edge. I've been digging around and I've narrowed this down to the problem is definitely caused by the web filter. If I check the log viewer I can see that Sophos is tagging it as Malware because it can't scan it (Protocol: HTTPS, Detection: Unscannable, Log subtype: Virus, Message: Malware 'Unscannable' was detected and blocked in a download from mail.google.com, Message ID: 08001) About the only Malware settings I can find in Sophos are under Configure / System services / Malware protection, and the only option there is to set what my 'Primary antivirus engine' is. I don't really want to start blanket not-inspecting traffic to GMail for certain members of staff if I can help it, but I'm not sure what other options I have at the moment. I guess it's not the worst thing in the world, we still have local AV, AppLocker, and FSRM blocking certain filetypes being saved to user-writeable areas, but still.. Anybody dealt with this before?

Yeah but isn't that like £4 a user? I remember them showing it off at, I think it was EDIT Salisbury? And it did seem incredibly useful and impressive in scope, but the cost would be too great here.

Usually I would use GAM, and have done in the past, but it just wasn't working for me this morning. Kept failing. No idea why. I finally managed to make it play ball just by keep repeating the command over-and-over until it succeeded, but that's hardly the best method. (Error was EOF occurred in violation of protocol (_ssl.c:1123), if anybody can make sense of that) Only results I found was to make a nocache.txt in the same folder as the GAM exe but that didn't work, and to lower the maximum concurrent threads, which it would still fail even on 1.

After an instance of something being accidentally sent to a pupil mailing list this morning (and the subsequent faff that followed purging the emails) I've spent a bit of time whipping up a script that should do all this for us. This script does require PSGSuite. tl;dr it hunts through all active users' inboxes for emails matching the specified RFC822 Message ID, if it finds one it performs the lookup again, this time checking if the email has been opened. If it has, it prints out a warning. Then (regardless of opened or not) it deletes the email. Simply find the sent email in Vault, click the blue 'Original' link, and find the Message-ID (it's usually just below From: and Date:, and should be a full email address, e.g. [email protected]), and feed that value including the domain.gtld into the $RFC822 variable, then go! Import-Module PSGSuite $RFC822 = "[b]Your Message ID goes here[/b]" ForEach ($User in (Get-GSUser -Filter "isSuspended -eq '$false'")) { ForEach ($Email in (Get-GSGmailMessageList -User $User.User -rfc822msgid $RFC822)) { If (((Get-GSGmailMessageList -User $User.User -Filter "is:read","rfc822msgid:$RFC822").count) -gt 0) { Write-Warning ("Email marked as Read for user " + $User.User + ".") } Remove-GSGmailMessage -user $User.User -id $Email.Id -Method Delete -Confirm:$false Write-Host ("Removed email ID " + $Email.Id + " from user " + $User.User + ".") } } One slight caveat: It also deletes the email from the sentbox of the person, but I can live with that.

Almost the same as everyone else's, only we put the number at the end, so Surname, First initial, Year of entry (assuming that they started in year 7 if they join us late) e.g. BloggsJ21 I just dump a report from SIMS and feed that CSV to a script that does it all for me, flags up any username collisions, and I just use the next character of their name (so let's say we had Joe and Jon Bloggs, whoever was first in the list would get BloggsJ21, and whoever was second would get BloggsJo21).. I'd prefer to use middle names, but they're basically never recorded in our SIMS.

The plot thickens, much to my chagrin So I've gone into the BIOS, disabled Secure Boot, even wiped the Secure Boot keys, logged in as an admit, executed bcdedit.exe /set nointegritychecks on from an administrative powershell prompt, rebooted, installed ActivDriver 5.18.16 x64 and... ActivManager doesn't see a board at all.

Alrighty. Thanks very much for your time You've been a great help.

Okay, I'm aware that I might be coming across as a little argumentative, and if I am I sincerely apologise, it's not my intention, it's just that I don't feel that my question has actually answered. Also I respect that you're acting as an individual and not a member of promethean support (as per your signature earlier in the thread), so I appreciate if the only answer you can give is "Get in touch with support". I don't want to be seen as hostile or pushy or anything. So do I need to leave it off? I've been trying to find actual information about signature enforcement in between tickets coming in, but all I'm finding is loads of articles about how to turn it off, none of them are actually mentioning how it works. Is signature enforcement an "enforce signature checks when a driver is installed" or an "enforce signature checks when the driver is run" type of deal? If it's the former, then I can turn it back on, if it's the latter then I can't. Also we do use Secure Boot here, so will I need to reimage the machine with secure boot disabled, or does the previously posted commands override that? I guess lessons will be over soon, so I can test turning it back on myself, I was just hoping it was an easy yes-or-no answer. Yeah but that firmware doesn't seem to be available anywhere. The newest I can find on Promethean's website is the 3.75 firmware previously mentioned, which I'm currently using. But the instance of ActivManager installed on that device was on (and even still states) a higher version, a version which apparently doesn't exist if I use the 'Internet' option - it says there are no updates available.

....... $Diety damnit. How did I miss that? And I mean, I'm not even gonna try and save face. I have been to that article several times (although admittedly most of the revisits were just *click*, "No, I've seen this page already", *close tab*, but still!) I guess this is what happens when you have several folk breathing down your neck to fix it right this instant huh? The guide you have linked permanently disables signature enforcement.. Does that have to be the case? Or is signature enforcement a check at the point of install? Do I absolutely have to leave it turned off, or can I turn it off, install the driver, then turn it back on again? I know technically it's not much of a risk, as a user would still need administrative rights to install a driver, signed or unsigned, but the amount of privesc vulns we see floating around the wild just makes me a bit uneasy about the idea. Edit: also is there a way I can get the V3.97 Build 13 or later firmware back?

We don't have any spares, so I'd have to lop one off the wall and ruin someone else's day for that (though, given how many of these IWBs we have, and how few people have reported crashes, I'm not convinced that many are actually using the interactive bit of their board, so I could probably get away with it...) I'm doing some digging into how exactly they're cabled. The main room I'm doing my tests in (because it's pretty much always got the same 2 members of staff in it, so I get proper feedback about any changes) has both a USB connection and a serial to... I have no idea, it looks like a DC12V pin but I don't know if that's it's actual purpose. Consequently, that computer has COM1 and COM3 available in ActivManager. I'm unsure if this is related, and I'm waiting for other classrooms to be free to investigate further. Oh, and to add insult to injury, I know that V3.97 Build 13 exists, as ActivManager clearly displays the firmware downgrade, but if you try to do an internet upgrade, it says there isn't a newer version. WAT. It's worth a try, I shall drop you a line Didn't know if you'd get shouted at for providing historical versions we all know some company or other that are like NO YOU HAVE TO GET IT FROM US YOU CAN'T GET IT ANYWHERE ELSE DON'T EVEN THINK ABOUT IT RAAAHHH so didn't want to risk it lol. I have actually replaced the workstation, as initially I was using a different BSOD analyser that didn't give me as much info, so I had no idea what the stop code or erroring process was.

Title, really. I don't expect there's any help to be had, since the hardware is so old, and Promethean's official line (at least as per this thread) is "It's not an us problem it's a Windows problem", but it's worth an ask. So we still have a lot of 2nd generation Promethean IWBs that are intermittently crashing with IRQL_NOT_LESS_OR_EQUAL blue-screen-of-deaths. Going off reports, they seem to work just fine for a bit, and then kablooey. I have tried using the latest ActivDriver (5.18.19, from this page). This is the initial driver we installed. I have tried the 'unsupported' ActivDriver listed on this page. Using that, there's no crashes. There's also no 'interactive' functionality, either. The pens stop working when this driver is used. I have obtained a much earlier version of the driver (5.15.10) from another school who shall remain nameless. This version still produces BSODs. I have confirmed we are using the latest board firmware, according to ActivManager (Update/Internet option said no new version) I have rolled back to older firmware (3.75, from this page) to no avail BSOD data obtained from WinDbg: SYMBOL_NAME: ACTIVhidmini+c0f MODULE_NAME: ACTIVhidmini IMAGE_NAME: ACTIVhidmini.sys STACK_COMMAND: .thread ; .cxr ; kb BUCKET_ID_FUNC_OFFSET: c0f FAILURE_BUCKET_ID: AV_ACTIVhidmini!unknown_function OS_VERSION: 10.0.19041.1 BUILDLAB_STR: vb_release OSPLATFORM_TYPE: x64 OSNAME: Windows 10 FAILURE_ID_HASH: {6af976ac-df4f-6003-027e-f55fb1e09168} I'm really hoping there's something I've missed, or some kind of fix someone can throw my way. Rolling back to older versions of Windows isn't really an option at this rate, considering crashes have been reported as far back as 1903. Thanks for reading.

Ahh I misunderstood. My bad! I have no idea if you can double-book slots, let alone deca-book them.

Office 365 natively has Powershell support, although you might need the Azure module. Combine that with PSGSuite, a third-party Powershell integration for G Suite, and you could probably do a ForEach loop to download each user's OneDrive contents to local storage, upload it to their Google Drive, delete from local storage, then move on to the next user. I used it to do similar (well, the second half, at least) at the start of Lockdown to dump everyone's on-site user area into their G Drive. I'd imagine you'd be able to do it for SharePoint/Shared Drives too, though that's just a guess. Your gripe about converting inboxes/folders to labels is probably fair. Effectively they work exactly the same way for 99% of users once set up - you drag a folder from your inbox to the label, it disappears from your inbox, and appears under that label - but it's still a lot of faff to set up for people that utilise it heavily (safeguarding/student support making a folder/label for each kid, for example).. The upside is for the people that do make heavy use of it, giving an email multiple labels is actually super useful. But users that would actually benefit from that are few and far between, I suspect you'll find most of your issues are 'Why do I have to...' or 'I don't get it It's different and that frightens me'

We use https://www.schoolcloud.co.uk/ It seems decent enough, even when parents evenings are physical, I'm pretty sure parents still book their slot on there. Only connection issues we've had has folks at home with terrible WiFi or dying laptops. Only gripe from a technical standpoint I can say is it seems to consistently forget your password. And then you say OK, I'll reset it. And then you reset it to exactly the same thing, and login. Then next time you log in it says lolno that's wrong, and you do the same song and dance. Couldn't tell you about the pricing, though. Nowt to do with me.

A couple of our machines are throwing this error reasonably frequently since updating to 20H2 over the summer, and I'm struggling to track down why. I've used NirSoft's BlueScreenView to analyse the memory dump but the 'Caused By Driver', 'Caused By Address', and 'Crash Address' fields are all blank. Is there any reliable way to chase down what's causing the problem? Edit: @Katy and @tgpips have pointed me towards WinDBG which has satisfied my need. Leaving thread/link here for future searchers!

In a small home lab, I'd probably end up giving my servers names from The Matrix, it seems rather fitting. Neo would be the DC, as he's the most important, The filer server would probably be called Tank, Dozer or Link as they're the ones uploading data.. In a big domain though? Functional names. DC01, File-Server, HyperV-01, Print-Server, MECM, etc

I sussed it out, with a little help from @Boredguy and @localzuk, admittedly. Can't take all the credit! Essentially, I was confused about tagged vs untagged, I was thinking you tag the port as VLAN X to say "Okay, all traffic on this port for VLAN X", which made sense in my head. When in fact, its more like "Packets arriving at this switch will be tagged already, so don't worry about it.".. So in fact, tagged and untagged are the exact opposite of what I thought they were.. That, and I was completely overlooking LAG settings on the core switch. So hurrah, mission accomplished, albeit with a significant detour.. Chalk that one up to experience!

Okay, so we've recently had two permanent 'temporary' portacabins installed. I've had fibre run from the core switch, got our SFP's, and connected them up, but try as I might I just can't seem to get them to play ball. The switches can see the SFP's and detect them as 10GBASE-SR SFP+ modules, they see the port as active, there's even lights, but.. that's it. Our core switch is a pair of Dell N4032F's The new switch is a Dell N322P-ON Comparing the new switch to other (non-core) switches, I have: Set up time synchronisation with SNTP (ntp.org) (System/Time Synchronization) Given the server an OOB address that can be connected to if that's wired into the rest of the network (System/IP Addressing/Out of Band Interface) Set up the DNS and domain name (System/IP Addressing/Domain Name Server and System/IP Addressing/Default Domain Name) Enabled the Captive Portal (System/Captive Portal/Global Configuration) Ensured the Gi1/0/xx ports are enabled (Switching/Port Configuration) Created the same VLAN identifiers, using the same numbers, as the other switches in the network (Switching/VLAN/VLAN Membership) Given the fibre link (Te1/0/1) interface the correct PVID of the management VLAN (Switching/VLAN/Port Settings) Observed that the Global Default Gateway has correctly automatically set itself to the IP address of the core (Routing/IP/Configuration) Given that VLAN a static IP address and subnet mask (Routing/IP/IP Interface Configuration) Things I'm confused about: LAG Settings (Switching/VLAN) - There seems to be no rhyme or reason that I can discern as to why these are set and what they are set to across the other switches in the network, but some of them have been set as they're not all on the default Access mode with PVID 1 What I can only assume is VLAN tagging at the bottom of the VLAN Membership (Switching/VLAN/VLAN Membership) settings. A page full of U's and F's and T's that's just going way over my head, I'll admit it. So, I can now get onto the new switch via its IP on the management VLAN instead of the OOB IP, so it looks like traffic can flow between the switch and the rest of the network, but anything you plug into the switch can't identify the network (Windows gives ye olde Unidentified Network) and there's no connection, e.g. you can't log in because there's no DC. Banging my head against the wall here now, doesn't help that the last time I programmed a switch was in Packet Tracer, about 5 years ago...

Awake on Netflix. Called the plot/conclusion about 20 minutes in, but it was still an interesting movie. Would recommend if you've got 90 minutes to kill. Wait they made a sequel?! I need to get in on this.

The consistency of McD's milkshakes? I think the term is 'furiously shlurp'

Scam calls are out in force today. Still number spoofing.

I dunno, I'm not convinced. Think about how IT literate the average person is. How many of our users panic just because a normal run-of-the-mill error message has a big scary word in it? It would be pretty easy to tell a not-very-IT-literate person that stopped services are bad and they should all be running. The same with netstat, tell someone there should only be one established connection (I dunno, say, between the pc and the router?) and it would be pretty easy to believe. The whole running-tree-and-typing lol virus found giv munnee plox is a dead giveaway though. And I do reckon most of them know what's going on. But still, not so cut-and-dry.

Oh yes, very much the same here. I'm more referring to the issue now where the filtering is being shut down and we're looking at reprovisioning them.