Blue_Cookeh

That's exactly it, I can't pinpoint it either. I've not done anything fancy, it's just a (relatively clean) standard mandatory profile. It's starting to become a real problem for us now so I've started building a Win8.1 image to roll out it's happening for us on the standard Windows task bar and (now) with classic shell also.

The profile is a roaming profile for a generic account a younger class use, so I'd have thought if it was a profile issue I'd have seen it across all devices. Also what I mean is, the user cannot actually click anything along the task bar (including start button/cortana). Either way I'll give Classic Shell a go, was just hoping someone might have a more definitive answer before I waste time pursuing it Might end up going back down to W8.1

Does anyone know if this fixes an issue where users cannot click on anything along the task bar? I'm getting reports that occasionally our kids can't click on the task bar/search bar/start button.

You should use the WinPE boot disk instructions posted a few posts up and mount the BitLocker-suspended drive. You can then use the trick to rename cmd.exe to the sticky keys executable, reboot into Windows, and then hit shift five times to bring up a SYSTEM command prompt and reset the Admin password. Make sure you put the sticky keys and cmd executables back once finished.

Look at Amazon Glacier. Very cheap to transfer up to, a bit more expensive when you need to pull it down (hopefully never).

Apple stuff is usually available in the UK around the same time as the US. It doesn't look like they'll limit to to >16GB devices, but I can understand why they would recommend you don't use them, particularly if you have a lot of students storing a lot of work.

Hopefully Google will come up with something similar to compete so we have a bit of choice Excited to have a bit of a play around with this on some of our iPads when it comes out!

We use DirectAccess and an IPSec/L2TP VPN as a fallback.

Blocked for student accounts, unblocked for iPads and staff accounts since we run school accounts the kids can post to.

I'd still vote for three virtual hosts with local storage. Forget "Enterprise" SSDs and get say 6 large "standard" SSDs RAIDed in each system, create a Hyper-V cluster, setup Veeam to back the VMs up to a NAS (I like Synology), and then ship a copy off to Amazon Glacier if your connection is fast enough.

I see our IT as more like a typical business' than a school. In your typical business environment a laptop is a key tool for the job nowadays, so we do the same. We don't have whiteboard desktops in our classrooms, so teachers dock their laptops wherever they are instead. We also have some staff that travel around to other schools, so we've deployed DirectAccess and given them laptops with 4G modems built in. It means we have full control over what data leaves our network, and staff can still work in a secure environment at home. I wouldn't expect to do my job using my own laptop, so why should I expect our staff to? Saying this the HT and I did decide iPads would be cheaper/easier for some staff, people like site agents that don't use much apart from e-mail for example.

FSMO roles can be offline for quite a while, if you suspect it's going to be more permanent then seize the roles onto the online DC whenever you have time. Also if those DCs are DNS servers your clients may be really slow resolving DNS queries. Another thing to be wary of is your tombstone lifetime for DCs that are uncontactable. I would also advise bringing up 1 DC at a time when they do come back online.

That changelog doesn't really seem worth the pain of upgrading again.... yet again

I could be completely off track here but those look like standard DB files to me. What if you copy the library.bak to a new file called library.mdb? EDIT: Saw the date but my point still sta

If you do 'nslookup email.domain.co.uk' in Command Prompt on an affected device what result do you get?

What OS are you deploying? I had this problem whilst deploying Windows 10 a while back and never found a solution, despite pouring over countless CCM logs I think actually it ended up being something to do with my product key, either way after recreating the TS about twenty times it eventually worked for long enough to deploy the OS to my netbooks. http://www.edugeek.net/forums/enterprise-software/158020-sccm-not-completing-task-sequence-all-sudden.html

There's no reason not to use UEFI nowadays.

I really ought to become a Sophos salesperson... Loving our Sophos UTM here. Very competitive on pricing. Does everything in one box. Top notch support. Just not keen on the way their new XG firewall is going, but I believe they're slowly fixing that one You can download the Home version which will run everything as if it were licensed to play with, just limited to 50 IPs. You can run it on Sophos hardware, your own server, or within a VM.

I wouldn't think like that . RAID is amazing when you come to have a failed disk, but you ran into the exact trap I did when I didn't really know much about it. If you're not monitoring disk health/failures then it just breaks things further, if you are, it's a god send. There should be some management agents you can install like Intel RAID Utility, or Dell OpenManage that will alert you via e-mail when the health of a drive degrades. This is the time to get a RAIDed drive replaced, not when you come to find out both have failed Funnily enough it was this scenario that led me to implementing DFS-R, failover DHCP etc. Bit stupid I didn't in hindsight really since we had the hardware ready for it. Now we can lose one of our servers (with RAID, and backups, and replication) and I don't even need to restore it since everything will automatically fail over to our secondary DC.

What did you use to take the backup? Despite it being "corrupt" you may well be able to salvage your Active Directory database and avoid having to build an entirely new network. It's possible, but a mega pain in the ass. Funnily enough it's what got my budget for a Hyper-V cluster approved after a know-it-all governor said no to me getting a secondary server

Where are you seeing this? Your AD DNS should forward on to a public DNS service such as Google (8.8.8.8, 8.8.4.4) or your county's DNS. DCs should be pointed to other DCs first, and then it self second. Workstations should be pointed at your DCs.

As a client? Install SNMP via the Add Roles/Features Wizard and then configure it (the community name/allowed hosts) through Services.msc > SNMP Client > Properties. Make sure it's set to Automatically start.

Oh how true. You can request features/report bugs/request device additions via their GitHub at https://github.com/librenms/librenms. I ended up liking it so much I started to contribute to it.

Yes! It's a free fork of Observium.

I know you've probably spent a lot of time on Nagios but can I suggest something a) easier and b) more up to date? LibreNMS I ditched Nagios for it about 6 months ago and haven't looked back.