1, Download Scratch

2, Unzip the .exe file

3, From the $PLUGINSDIR, unzip the app-32.7z file

4, Copy the contents of the app-32 file into a shared network folder

5, Create a script

Robocopy /copyall /s /e "\\SERVER\NETWORKSHARE\Scratch3.6" "C:\Program Files\Scratch3.6"

6, Create a GPO to create the folder Scratch3.6 in C:\Program Files\

7, Create a GPO and assign the start up script created in step 5

That should do the trick!

So, it's been a little while since I rolled out Windows 10 to staff, and needed a new image, for both staff and students. I'll use this blog to note my successes and inevitable failures on the journey to getting a working, student safe Windows 10 image.

Getting Windows 10 without inBuilt Apps

First up, we need a Windows 10 image, and since the latest one seems to not delete My Documents on a whim, I plumped for Windows 1903, and duly grabbed the .ISO from VLSC and imported it into MDT. Previously I had mounted the .wim and stripped out the inbuilt apps that way, however I wanted to try something new, and so decided on a method that removed the apps as part of the MDT process. There are also scripts for earlier versions of Windows 10 as well. Following the guide was pretty straightforward, and on a test build the apps were removed. Success!

Getting Office 2019

Next up, I wanted to get Office 2019 installed. At some point I should move to Office 365, but installing that for students PCs seems a bit more complicated, so for the time being I'll stick with Office 2019. That said, installing Office 2019 is no longer a simple process.

First up, you'll need the Office Deployment Tool from Microsoft

When prompted for an installation folder, make a new folder on your machine, say Office2019-Install

You'll then need a config file that can be created using the Office Customization Tool Export and save into the folder created in the previous step. I've called mine StudentConfig2019.xml

Then run a powershell session, and browse to the Office2019-Install folder and run:

.\setup.exe /download .\StudentConfig2019.xml

This will download Office 2019 using the parameters setup in the xml file created using the Office Customization Tool.

Copy this folder to your MDT server.

Installing Office 2019 as a MDT task - Adapted/Taken from this guide

In your MDT Workbench, open your build deployment share and expand the Applications Folder. Create a new folder called Microsoft, then right click the Microsoft folder and choose New Application

Choose Application with source files. click Next

In the Application Name field enter Install - Microsoft Office 2019 Pro Plus click Next.

Click the Browse button to navigate to the Install - Office2019-Install folder you copied over. Tick the box to Move the files to the deployment share. Click Next.

Name the folder Install - Microsoft Office 2019 Pro Plus. Click Next.

The Command Line to enter in the field is:

setup.exe /configure configuration-Office2019Enterprise.xml - I should say that it didn't work using this, I had to put in the full path, including server name, for this to work. So I put in this, but YMMV:

setup.exe /configure "\\<WDSSERVER>\deploymentshare$\applications\microsoft office 2019 Pro Plus\May2019.xml"

Click Next. Click Next.

Open the task sequence and add the application. With your build task sequence open, click Add -> General -> Install Application

Choose Install a single application. Click Browse and choose your Install - Microsoft Office 2019 Pro Plus application. I like to move the install application tasks into a separate folder for Custom Tasks, before a Windows Update task to ensure the build has the latest office updates.

Hopefully this helps and it's helped me realise I need a different Office 2019 config file for staff and students!

In a nutshell, tried to rebuild some laptops using MDT/WDS. Kept failing to load the boot image, kicking to Boot manager with the error message: A required device isn't connected or can't be accessed.

Tried rebuilding the deployment share - No dice.

Deleted and readded the boot wims into WDS - No dice.

Went through various drivers (we have FAR too many) and removed a great swathe - no dice.

I was contemplating recreating the entire WDS server, when I came across this thread:

https://social.technet.microsoft.com/Forums/en-US/cb267f4e-a704-44ad-a444-a8160c30e775/pxe-fails-each-time-with-0xc0000001-quota-required-device-isnt-connected-or-cant-be?forum=configmanagergeneral&prof=required

The actual answer ticked mentioned boundaries, which I think is a SCCM issue. The actual answer for me came a few posts afterwards:

I happened upon this error as well. The solution that fixed it for me was the "Variable Window Extension" Checkbox un TFTP on the Server properties in WDS. I would post a picture, but my account is too new.

This box was CHECKED for us. Not sure how as I do not believe this was checked by default, but UNCHECKING this box solved the issue for us. Hope it's this simple for anyone else who stumbles across this thread and doesn't lead to needing to rebuild WDS.

SUCCESS!

Apparently it's a known Microsoft thing:

"After installing this update, there may be issues using the Preboot Execution Environment (PXE) to start a device from a Windows Deployment Services (WDS) server configured to use Variable Window Extension. This may cause the connection to the WDS server to terminate prematurely while downloading the image. This issue does not affect clients or devices that are not using Variable Window Extension."

I just wanted to spread the word in case this causes as much confusion/aggravation to anyone else.

However, it's a perfect candidate for Chromebooks, so a quick install of Neverware and its good to go. That is, until the damned thing won't boot... Powers on, screen flashes, reboots.

Since I had a spare HP Stream, my first thought was to swap the hard drives. So opened up the laptop, and, erm, no hard drive. Or rather, the hdd is embedded into the motherboard, so swapping them over wasn't really an option.

Tried a HP hardware reset (unhook battery, hold power button for 30 seconds, turn back on). No dice. Ran the start up check on the RAM, no issues. Couldn't run the HDD check as it needed HP UEFI Support Environment installed. Trying to boot from the Hard Drive from the Boot Options menu gave the error message "Selected boot image did not authenticate", which pointed to Secure Boot. At this point, I had forgotten this laptop was no longer running windows, but figured that turning off Secure Boot was probably a good idea. Found a website with a handy guide, but following Method 1, Secure boot would not stay turned off. I would save the setting in BIOS, it would reboot, and Secure Boot would turn itself back on.

Out of frustration, I turned off Secure Boot, turned on Legacy Support, and also turned on POST Hotkey Delay to 5 seconds. Then on a reboot, a message appears asking for a code to be entered to turn off Secure Boot. Entered the code, and ChromeOS is back!

Long story short, if your HP Stream 11 isn't booting, it's worth:

1, Pressing esc on boot to bring up a start menu

2, Enter BIOS - System Configuration - Boot Options

3, Turn off Secure Boot

4, Turn on Legacy Support

5, Change the POST Hotkey Delay (sec) to 5 (or higher)

6, F10 Save and Exit

7, Enter the code to disable Secure Boot.

So, the initial issue was that I was getting the error PXE E23 when trying to PXE boot a Gen 2 VHD with Server 2012r2 and MDT 2013. Helpfully, @l15ard had pointed me in the right direction with his thread here:

Pxe-e23-error-gen2-hyper-v-fix

However, I had no idea how to 'Force x64 UEFI architecture'. Indeed, the file the VM was trying to pull down had 0 bytes. Which I thought was unusual, so I dived onto the MDT Server to see if I could find the file. For those interested, the file it was trying to pull down was boot\x64\wdsmgfw.efi and this is stored in the RemoteInstall share. Unsurprisingly, the file wasn't there.

A bit of Google-Fu and I wasn't completely the wiser as to why the file wasn't where it was supposed to be. There was a lot of talk about DHCP options but this didn't seem relevant to me (but might be to someone else with this issue) as I didn't actually have the files it was looking for.

Now some of you are probably way ahead of me, but eventually inspiration from somewhere hit. I hadn't installed a x64 Boot.wim into WDS.

I haven't needed to install a boot.wim for quite some time, but the process is pretty simple:

1, Mount a Windows disk that has a x64 Boot.wim in the sources folder (Server 2012r2, Windows 10 etc)

2, Load up the WDS Console

3, Right click Boot Image and Add Boot image, browsing to the mounted disk in stage 1

Then bingo, the required file is there!

But, I've been using MDT and Lite Touch as my deployment method of choice. So I needed to create a x64 version of my Lite Touch deployment:

1, Open up MDT Deployment Workbench

2, Right Click MDT Deployment Share

3, General tab, tick x64 as supported

4, Windows PE tab, platform x64, tick Generate a Lite Touch bootable ISO image.

5, Click OK and then right click the MDT Deployment Share again and 'Update Deployment Share'. This should create you a Lite Touch Bootable ISO you can then import into WDS.

But this tale doesn't end here. PXE booting was working, but by heavens it was slow. @mdrabble had experienced the same issue and it tickled something in the back of my mind too. The fault is with the Broadcom NetXtreme NICs that came with my lovely HP servers. HP have released drivers that fixed the problem, or you can manually go into the properties of the NICs and turn off VMQ (Virtual Machine Queuing). Since my NICS were only connected at 1gb, VMQs were not going to improve the performance at all, so I turned them off on the NICs. Problem solved, PXE booting on Gen 2 Hyper-V VHDs at rapid speed now attained.

In something of a rarity for voice commands, the Echo, or Alexa as I call her, is actually pretty decent at picking up and understanding the human voice. But I didn't want to turn this into advertisement for the Echo, but rather ask the question of whether a voice controlled device like the Echo has a place in the classroom?

Students use devices in classrooms to access the internet to answer fairly simple questions. What if you could ask a device that sits in the classroom that question? What year was the Battle of Hastings? How long do you bake a Mushroom for? How much wood does a woodchuck chuck if a woodchuck could chuck wood?

I like the idea that facts are just a question away, and could see it being useful in lessons where skills are more important to learn than just straight facts. If we acknowledge that learners are always going to have Google to find facts, an Echo device would make it near instantaneous to find out information, rather than waiting for laptops to login, browsers to load or relying on batteries to be charged.

So will I be introducing an Echo to a classroom any time soon? No, I don't think the current Echo is quite prepared for the classroom just yet. As it has access to radio stations and music that you wouldn't necessarily want, and no ability to restrict what the Echo can do. Also it's voice recognition, whilst very good, isn't perfect, and it could lead to some frustration for students and staff alike. You have to quite specific to get the answer your looking for. For example, asking 'How to cook a Mushroom?' doesn't get an answer. But ask 'How long should I cook a Mushroom for?' Does get the desired answer.

Then there's the questionable issue of having a microphone listening to everything that happens in a classroom...

But niggles aside, I think there might be a market for a Education Echo. As Google, Apple and Microsoft start to move in this direction, a voice enabled electronic assistant in the classroom might not be too far away.

Helpdesk

It's not an original sentiment. There's a lot of forum posts on Edugeek that will hammer home the need for a helpdesk, so it's perhaps not a big surprise that it's the first segment in the TNTG. At it's most basic, a helpdesk is a way of recording all the work that the IT department is conducting. Even for a single person department, a helpdesk is invaluable for:

• Tracking your workload
  
• Keeping tabs on longer projects
  
• Spotting recurring patterns
  
• Organising your time
  
• Setting individual/departmental priorities
  

Some IT Departments like to only work to the helpdesk (no ticket, no help), which is a valid methodology, but not one I've ever stuck to. I've always preferred to be use the Helpdesk as a guide to work that needs to be completed, and as a way for me to ensure myself and my technician are on the same page.

Regardless of how it's used, my advice to any new technician is to make sure you know how your school's helpdesk works, and to keep it updated with everything you do. If there isn't a system in place, I would recommend looking at Spiceworks or Freshdesk, or one of the numerous threads on the subject on Edugeek. Or, at the very least, some kind of spreadsheet noting what you've done, and when, just something digital to refer to for use at appraisals, job reviews or just for your own good practice.

I'm perhaps not the best person to write about Helpdesks, as I'm not a massive devotee. I find it incredible useful for setting tasks for my techie, but I'm not great at adding my own jobs to the list as I tend to use an old fashioned diary unless it's a job that affects the both of us. Staff are encouraged to use the helpdesk, but there's certainly no punishment for them not using it (more on 'Managing teaching staff' in a future segment I think). My appraisal tends to me far more strategic in view, rather than a count of how many tickets I've completed (for which I am very grateful, as I find target setting of that kind not suited to the job). But that said, it can be useful to show if an IT department is overworked, and it can be used to highlight the need for an extra technician, extra hours or overtime.

As usual, any comments or constructive criticism gratefully received.

Just as a warning this blog post is not going to particularly well structured. It's more of a writing down of some ideas that have been floating in my head for awhile, and I would certainly appreciate some of the ideas being discussed with other technicians and Network Managers. Disclaimer out the way...

It's always struck me as odd that there's no 'official' guide or training offered to Technicians. Yes, there's the FITS Foundation but it costs money, doesn't run all that often and I'm not sure it covers everything an IT technician working in education needs. But even if we took FITS as the required training, the only place I see FITS mentioned is on EduGeek. I've not seen it as a requirement on any job posting, nor did it come up at any County level technician conferences (back when Oxfordshire could afford to host such events). It leads me to a key question:

Where, as a new or inexperienced technician, are you meant to go to find out how to do your job?

Also, Where do you find best practice to ensure that what you are doing is actually what is required for your school?

I guess this all stems from a realization I had a few days ago, if I didn't read the edugeek forums, there would be great swathes of my job that I wouldn't even know were part of my responsibility. I took this job as a junior technician, under a brilliant ICT-Coordinator, but I was the only member of the tech team. For months I didn't even realise there was another server dealing with the admin side of the network, hidden away in another office! As I continued, I slowly realized that there were things that I should probably be taking care of, but never explicitly told about (back ups, web filtering, software updates etc etc). Now some of these were largely dealt with at County level, but now with the advent of acadmies and MATs and free schools, these areas are now down to the school (and their IT support) to figure out, maintain and control. Thinking back, this was around 10 years ago, and there was a lot less pressure on the computer system back then. Only around 4 members of staff used email on a daily basis(!) and the reliability of computers was such that, if they didn't work on that day, it didn't matter, teachers would just do something else that lesson. Of course, this has changed dramatically now, and IT is heavily relied upon by all aspects of the school.

I digress, my point is that there is no central body, website or guide to being a technician, no one stop shop that says what you should be doing, and offer solutions on how to achieve it. Don't get me wrong, edugeek is a fantastic resource, but the answers it provides are scattered over hundreds of different threads and blog posts. What I would propose is the creation of the "Technicians non-technical guide to being a technician". I don't know what form this would take, possible blog, wiki or old fashioned book, but more importantly I do know what I would want the content to be: useful guides on how to setup, maintain and troubleshoot an education based network. What I wouldn't want is a bunch of technical how-tos, I would want it to be as device/OS agnostic as possible, but to go into as much detail as possible where it suits.

Of course, the trouble with such a guide is who would write it. Whilst I consider one of my skills is the ability to translate the technical into the non-technical, I've only ever worked in one school, and whilst I have 10 years here, I have no real authority on what is best practice. I'd like to think I'm doing a good job, but then I get the worm of self doubt after I read another edugeeker has created a some amazing server infrastructure that I completely fail to understand, or the fact it took me well over two weeks to get my head around vLanning...

But perhaps I should aim smaller, and just write about my experiences, and not claim to be anything other than a Network Manager who's managed to get a network into a reasonable state, rather some authoritive body on the best practice of a education network.

Ultimately, I feel that if I had been given some kind of guide when I started this job, it would have made my life a lot easier. Not a specifically technical guide, like a Mastering Server 2003 book (which I did have, and was pretty useful!), but something that explains:

• What is expected of you as a technican/network manager in a school
  
• What your responsibilities are
  
• What isn't your responsibility, but what you should probably keep an eye on and report to your line manager
  
• What software could help in your job (Helpdesk, Server elements such as WSUS etc)
  
• The importance of documentation
  
• The importance of budgeting
  
• Any other bits and bobs that would be useful to know
  

As I said at the start, any feedback on this would be appreciated, would you find this kind of thing useful? Are there better sources of information out there that I don't know about? Please comment below. I also realise that this could very well just be me, and I'm unique in the way I found out all the ins and outs of my job, however I suspect that's probably not the case... In the mean time, I think I'm going to start sketching out how to write this. My inital idea is to do a series of blog posts, and then maybe try to tie them together, but we'll see how it goes...

I spent some time reading around http://www.edugeek.net/forums/internet-related-filtering-firewall/145867-google-moving-ssl-searches.html and Google's enforced moving all searches to SSL and the issues this was causing for Web Filtering. Fearing that our web setup would again force me to block Google searches, I read through and found, to my relief, that all would be well.

Or so I thought.

I have redirected our Google searches to enforce Google safe search, using the Google VIP

However, two pieces of information came together and made me realise that I needed to make a change to the network or face the possibility of uncomfortable questioning. The first was that Ofsted are asking questions about web filtering, with this specific issue in mind (google-moving-ssl-searches Page 5l

And secondly a post from @Irritable Tech Google-safe-search which reminded me that I had, in fact, only set the redirection to Google Safe Search for google.com and .co.uk as these were the only two domains that the students ever used. Of course, with BYOD, even with SECURUS onsite, I couldn't know for sure that students weren't using other domains for their searches on their phones or other devices around the school without Securus (iPads, tablets etc).

Then came the mammoth task of adding all 190 or so countries to my own DNS. I don't have a proxy filter, instead relying on our Palo Alto provided by Virgin, and the Web Filtering inbuilt therein. For the most part it does a good job, and with SECURUS backing it up, I'm sure it will pass any Ofsted/Child safety inspection. However, for all it's good things, it can't force Google safe search without the DNS 'hack', which I have to do on a local server.

I started entering them one by one, and figured it would take me the best part of a day, and thought there must be a better way. Command prompt could do it, but I figured that Poweshell would be the best bet.

Technical Bit - How to add lots of Forward Zones and Host A Records using Powershell

First up, I needed to know what commands would allow me to create Forward Look Up Zones - Technet - Add Zones to DNS via Powershell

In this case, I needed

Add-DnsServerPrimaryZone -Name "www.google.com" -ReplicationScope "Domain" -PassThru

Next, the code to create an A Host record Add-dns-a-records-via-powershell/

Again, the code required is

Add-DnsServerResourceRecordA -Name "." -ZoneName "www.google.com" -AllowUpdateAny -IPv4Address "216.239.38.120"

So, two pieces of script, now I needed a list of google domains - Google supported_domains

And to combine them all, I needed Excel. Paste the list of domain names into Excel, then copy them again, and this time paste - special - tick transpose to get a nice long list of domains. Place into Column B.

In Column A type "Add-DnsServerPrimaryZone -Name "www" and then in column C type "" -ReplicationScope "Domain" -PassThru" (without outside quote marks) Then in column E type =A1&B1&C1 This should mean, in column E, you'll get the text "Add-DnsServerPrimaryZone -Name "www.google.com" -ReplicationScope "Domain" -PassThru". Extend the series' in Columns A, B and E all the way to the bottom of the domain list. Then copy colum E, paste into Notepad and save it as 'DNSZones.ps1'. That's part 1 sorted.

Part 2 is similar, just in Column A make it "Add-DnsServerResourceRecordA -Name "." -ZoneName "www" and Column C "" -AllowUpdateAny -IPv4Address "216.239.38.120"" (again, without the outside quote marks). Extend the series and copy Column E into notepad and this time call it 'DNSAHost.ps1'.

If you're feeling brave, you can run the scripts on your DNS server, ensuring to run DNSZones first, then DNSAHost, but I would copy one line from each file first and run them as a test. All going well, you should have every Google domain covered and redirected to safe search.

Wireless has, for the most part, been the bane of my existence as a Network Manager. It's flaky, it's unreliable and it can ruin your otherwise lovely day. Combined with some odd behaviour with Windows 7 and domains, student logins tend to be a bit of a lottery, with around a 1% chance of getting an error "there are no logon servers available..." and another 1% chance of logging in fine, but having no My Documents or Start Menu as they haven't mapped properly on login.

This 2% error rate has fluxuated depending on model of WAP, it was a lot higher with the Netgear and stand alone Cisco WAPS, but since the introduction of Aerohive, I can be relatively assured that the issue is going to be with the laptop/Windows, rather than the wireless itself. Not that Aerohive is 100% faultless, it has a tendency to put APs that are next to each other on the same channel, and the odd WAP occasionally needs to be prodded with a paperclip, but it's generally a good, reliable system.

The interesting bit

But to the more helpful aspect of this blog. After receiving a complaint from a teacher that everytime she closed the lid of her laptop to move lesson, she would lose network connectivity and thus any files she had open would be made read only, causing her some grief. It wasn't insurmountable, but if we could look into it, that would be appreciated. My first thought is power settings, change it so that when the Lid is closed, nothing happens and problem is solved. Whilst digging around in the power settings, I can across a setting I hadn't seen before, and is definitely not in the GPP for power settings: Wireless Adapter settings. On the default power settings for Windows 7, on battery, Windows will reduce the performance of your wireless adapter.

After a bit of research to find out if I can change this setting via Group Policy, I found out that this has caused other people grief with poor wireless performance, and that is isn't possible to change via GPO or GPP. Indeed, our very own EduGeek has a thread on the subject:

http://www.edugeek.net/forums/windows-7/83351-wireless-power-gpo-settings.html

I wanted a simpler solution to the one offered by @VeryPC_Tom_M and wondered whether the 'high performance' plan built into Windows 7 had enable 'Maximum performance' on Wireless adapters, which it has. This being the case, I've set a script to run on start up for laptops for the following:

Powercfg -setactive 8c5e7fda-e8bf-4a96-9a85-a6e23a8c635c

This should ensure that all laptops are set to High Perfomance, and hopefully reduce that 2% down to 0.2% or lower.

Now, if I could only get laptops to receive Group Policy updates over wireless I'd be onto a real winner...

User Profile Service failed the sign-in - Fix in Windows 8

However this one laptop was not playing ball. I'd ran through all of the above steps, and nothing, no new accounts could be created, the guest account wasn't working and I was in a state of confusion. After reaching the third page of Google, I found an unlikely solution here:

Windows 7: "The User Profile Service failed the logon. User profile cannot be loaded

THe problem was, Windows Live wasn't installed on this laptop, but I thought the idea was sound, so I tried this:

Browse to C:\Users\Default (ensure you can see hidden files and folders)

Right click AppData

Choose Security

Advanced

Change permissions

Edit Everyone and give them Full Control

Click Okay

Then click on the tick box next to Replace all child permissions with inheritable permissions from this object

Click Apply then Yes to the Security warning

Amazingly, this solved the issue, so my assumption is that if a file in the Default User profile has it's permissions screwed up, new accounts can't be created. Since every other solution was either to go back to a system restore point or rebuild the machine, I thought it worth blogging on here as it might save someone a lot of time and effort!

I had thought I had prevented disaster using this guide:

http://www.redmondpie.com/how-to-stop-windows-8-updates-from-force-restarting-your-pc/

and then running gpupdate /force

However, this turned out to be a false hope, as a phone call from reception to tell me that SIMS wasn't accessible lead to a cold sweat and realisation that, yep, the server had decided to restart itself. After many promises of violence against the person at Microsoft who thought it was a really fantastic idea to take control of a server's restart time away from the administrator of the system, 20 minutes later all is back up and running.

For future reference, this would appear to be a better bet to prevent Microsoft from ruining your day:

http://blog.powerbiz.net.au/fixes/stop-windows-server-2012-and-windows-8-from-automatically-rebooting-the-server-after-logging-in/

My first piece of advice if considering this laptop is either to buy an mSATA SSD drive and fit it yourself and clone the C: drive onto it, or buy the next model up with a SSD drive. This alone will save you a lot of grief.

However, if, like me, you've bought the £750 model with standard drive, and have found a fair few issues with the GE60, then let me explain what the problems are, and how to fix them.

The first problem is Windows 8, indeed most problems with this laptop stem from having Windows 8 as an OS. This isn't to bemoan Windows 8 as a whole, but on this laptop, and the MSI configuration of Windows 8 on this laptop, has made the performance woeful. So my second piece of advice is to either reinstall Windows 8 from scratch, or downgrade to Windows 7 if possible. Again, this will save you a lot of time, sweat and tears. All drivers for Windows 7 can be found on the MSI website (apparently, I didn't go down this route).

Again, if the above isn't an option, then read on.

Windows 8 likes to do something called 'Write-caching'. In theory, this speeds up performance of your machine, and prevent data loss in the event of a power outage. All very sensible. The problem is, the data loss part of the equation absolutely kills the performance as the hard drive is in near 100% usage all the time. From the moment I turned the laptop on, disk usage was at 100%. After around 5-10 minutes, it would dip to 90% or so, and the laptop became usable.I had come to accept this as part of the laptop usual routine, blaming the slow speed of the hard drive. However, this does not need to be the case!

After my laptop became slow and unresponsive loading up the basic Mail client, I started to look around and noticed something in the system tray called 'Intel Rapid Storage Technology'. Thinking that my storage wasn't being particularly rapid, I loaded up the application. Nothing seemed out of the ordinary, however there was a message talking about the Windows write-cache buffer flushing policy, and that disabling it can improve performance. Thinking anything is worth a try, I followed the instructions, written below for your convenience:

1. Right click the bottom left hand corner of your screen, this should open a drop down menu, choose Device Manager.
   
2. From Device manager, expand Disk drives.
   
3. Right click your disk drive, and open properties.
   
4. Choose the Policies Tab.
   
5. Tick the tick box next to "Turn off Windows write cache buffer flushing on this device."
   
6. Click OK.
   

All of a sudden, my disk usage dropped to 5% and my laptop has never felt so fast and responsive. Seriously, it's finally performing like it should have from the start. A word of warning, should your laptop suddenly lose power, then you are far more likely to lose anything you were working on. But, I figured this is a gaming laptop with a battery life of two hours tops. Most of the time, it's going to be plugged in anyway, and the performance gain is well worth the price.

There are other issues with this laptop, primarily the fact that MSI in their wisdom decided to install every language display language as standard. This means that every update you download from windows update has to be configured for each language, meaning every update takes an incredible long time to install and configure. In addition, Windows 8 is clever enough to realise you're not likely to use these extra language packs, so tries to uninstall them every once and awhile after a reboot. Sadly, this process can also take hours, leaving your machine with the "configuring windows features" with either 15% or 100%. I have successfully removed all language packs, and my laptop was still chugging (primarily due to the above write-caching issue), though windows updates weren't quite so painful.

So to sum up, if you've bought this laptop and are just a little bit disappointed with it's speed, then do the following:

1. Turn off Windows write-cache buffer flushing on your hard drive
   
2. Remove all the extra display languages from the laptop (http://techmell.net/how-to/uninstall-language-packs-windows-8/)
   

I hope this helps someone, somewhere, as the laptop itself is a fine bit of kit, and with the above tweaking, can be the laptop it should be.

Anyway, having updated to Office 2010 over the summer, my previous group policy objects aren't going to work with Office 2010, so time for a new way of doing things.

First of all, I'll credit the website where I got the info, and feel free to follow them rather than my ramblings! This site will pretty much guide you through the entire process:

http://wgsnetman.blogspot.co.uk/2011/04/disable-ms-word-2007-proofing-tools-via.html

It states its for Office 2007, but it works just fine for Office 2010 as well.

You'll also need to download the admin templates from here:

http://www.microsoft.com/en-us/download/details.aspx?displaylang=en&id=18968

And there is another method you might want to try here:

http://www.migee.com/2011/12/06/disable-grammar-spell-check-thesaurus-proofing-in-office-2010/

Having read through the above links, I downloaded and installed the Admin templates into my Sysvol group policy directory, and then carried out the instructions from the first link, including the extra instruction in the comments section you might miss:

"Office 2010 no longer has the Group Policy option to turn off the Spell Checker. However it this can still be achieved by writing a DWORD value to the registry: HKEY_CURRENT_USER\Software\Microsoft\Shared Tools\Proofing Tools\1.0\Office

Value name "WordSpellingOptions" Value=0 No Spell Checking. Value=5 Spell Checking Enabled."

With the group policy and registry setting in place, I tested it. Spelling had definitely been disabled, and I couldn't access any of the options to turn it back on. Also the shortcut to run spellcheck had been disabled. Sadly, Grammar tests were still being carried out, and my Word document filled with little green squiggly lines.

I edited the GPO, and went to User Config - Polices - Admin Templates - Microsoft Word 2010 - Word options - Proofing and disable anything and everything to do with spell and grammar checking, including everything under AutoCorrect and AutoFormat as you type.

This completed, I tested again, and success, the squiggly green worms had disappeared.

However, there is one last remaining issue I can't work out, which is that the first letter of each paragraph is being automatically Capitalised. I should be able to solve this by manually logging onto each Exam account, and unticking the 'capitalize beginning of table or cell’ option. I would far prefer a network soloution to this, if anyone finds one, please let me know!

Please note this was tested on a Windows 7 PC, with Office 2010 in a Server 2008R2 environment.

My preferred method of shutting down is though Group Policy Preferences - Scheduled tasks. I set a scheduled task to run the inbuilt shutdown command within Windows 7 to run a 15.15, with various settings to wait for idle for 15 minutes, and only shut down if the PC had been idle for 10 minutes. It was this second part that didn't seem to kick into gear, no matter how much I tinkered with the settings. It appears, from my experience, that if the inital command can't run, the scheduled task is then set as 'Queued' and then refuse to run, even after the set idle time had elapsed. I had set this task to run again at 16.15 and again at 20.00. However, by this point the scheduled task had become a touch unwieldly in size, and was having a negative impact on boot up times. Also, if PCs missed the 16.15 shutdown time, PCs were left on for a long time before finally shutting themselves off. A better way was needed.

My settings are now as follows:

Boot up Group Policy Management

Create new GPO

Computer Configuration - Preferences - Control Panel Settings - Scheduled tasks - Scheduled task (Windows Vista)

Name the task

I would strongly recommend using a domain admin as the user to run the task as, but I leave it up to your own security procedures as to whether you do this or not.

Triggers - New... Set for every weekday at 15:15 then Repeat task every 1 hour for the duration of 12 hours.

Actions - New - Start a program "C:\Windows\System32\shutdown.exe" Add arguments /s /t 90 /d p:04:01 (Arguments are shutdown, after 90 seconds, with a reason - I find the later prevents 'This computer shutdown unexpectedly' errors for students first thing in the morning').

Conditions - Start the task only if idle for 10 minutes - wait for idle 15 minutes - stop if computer ceases to be idle and restart if the idle state resumes (I keep the wait for idle time and restart if idle state resumes more out of hope than any knowledge that they will actually work) - Wake the computer to run this task (not really useful as I turned off sleep mode as I found Windows 7 wasn't always 100% reliable in waking up again).

Settings - Can leave blank, I've ticked if the running task does not end when requested, force it to stop as it seemed like a good idea.

Common - Remove this item when it is no longer applied - I find this very helpful should I want to change the settings later, but again YMMV.

Once all this is set up, click on OK and apply to relevent OUs.

What this will do is run the shutdown script at 15:15 and then quarter past the hour every hour until 04:15. If the machine is in use, then it will wait until the next time to try and shutdown again. I believe it will also pick up on machines that have been turned back on after being shutdown, but I haven't 100% tested this.

So there we have it, I hope someone finds this useful!

First issue with codeacademy is that it doesn't like IE and flatly refuses to play ball with it. One option is to install Chrome across the site, but I'd had issues getting Chrome to play ball with my network before, and wasn't too keen on trying again. So I choose option 2, install Google Frame. Installing it locally was easy, but a network install could prove a bit trickier. This is when I found an old gaming friend's website, with a handy guide to network installing Google Chrome Frame on a school network.

http://lofi-gaming.org.uk/blog/2012/01/25/managed-network-install-of-google-chrome-frame/

Very useful, and I'm about to test to see if it works!

First up, I should point out that much of this blog can be found at the following link, and it also adds a cool way of adding printeres to specific Active Directory components, such as Groups, OUs or even specific users...

http://www.msserveradmin.com/the-one-reason-you-should-use-group-policy-preferences/

In any case, my goal was to fire out our network printers, and to be able to set up a default printer. The Printer Management function in Server 2008 could do the first, but not the latter. With a bit of searching I found the above link, and all was well. Until it came across my Canon printer in one of our ICT suites, which refused to play ball. Checking the event viewer, it turned out the Canon really wanted to install its drivers before printing, and even when logging in as an Admin, wouldn't install via GPO.

Another quick google search found the answer, disable Point and Print Restrictions via GPO. This, I think, can only be done by a Server 2008 R2 machine, although I'm sure there is a GPO template around somewhere.

In any event, that Printer now works with a student account, the PCs are named something obvious for all the students (as my printer naming convention got a little convuluted) and all is a success (so far). So a quick step by step on how to achieve this, bearing in mind I'm untrained and all this is from google, so attempt at your own risk! Also, all GPOs were created with Server 2008 R2.

First of all, we need to enable loopback processing, as we will be using User preferences to set up the printers. The reason for this is because only by setting printers by users can we set a printer as default. A pain, but there you go. To do this, create a GPO and go to Computer-> Policy -> Admin Templates -> System -> Group Policy -> User Group Policy loopback processing mode = enabled mode = merge

Then, in the same or different GPO, we need to make sure that the workstations use the server drivers, rather than try and pull down the printer driver to each machine. This step may not be necessary for all printers, but I don't think it does any harm either. (Please correct me if I'm wrong!). Go to Computer -> Admin Templates -> Printers -> Point and Print Restrictions = Disabled.

Finally, we go to User -> Preferences -> Control Panel -> Printers -> Add TCP/IP Printer

Enter the Printers IP Address

Give it a local name, I prefer BW Printer or Colour Printer, but whatever you want to students to see the printer as

You can then set it as a default printer if you desire

In Common setting, tick "Remove this item when it is no longer applied" This is optional, but will mean the printer is removed should the PC move OU.

If you want to target the printer to specific AD objects, use the targetting tick box and options. Fairly self explanatory, but I would read the link above just in case.

Once completed, assign the GPO to the OU of choice.

----------------------

Update

----------------------

As a general rule, I now assign printers to machines rather than students, as the above method means that printers are installed each time a student logs on, adding arounf 10 seconds to each logon. If I assign it to the PC, it installs on boot, so will only increase login time once, or not at all if the computers are set to wake up automatically before the school day. The downside to this method is not being able to set a default printer, but there's an easy script that will allow for that:

wmic printer where name='PRINTERNAME' call setdefaultprinter

All you need to know is the printer name (the friendly printer name the PC knows it as, so if you installed it via GPO calling it Colour Laser, put 'Colour Laser' in the script), and assign the script to apply to all students logging onto computers in that OU.