mrwoberts

This is more intriguing that an episode of Poirot. So, you established that... Win 8.1 -> 2016 R2 H-V (DELL and HP x 2) works. Win 10 -> 2016 R2 H-V (DELL) works. Win 10 -> 2016 R2 H-V (HP x 2) does not work. Despite the setup of all the Hyper-V hosts being 'identical', and nothing was needed in Windows 10 that you previously had in 8.1. You found that partially disabling some HP software gave different errors, but still no connection. With all that mind, would it be worth stripping back (uninstalling) some of that HP software, or at least posting the errors you received. HP support also sounds like a good option, but I don't know how quick they will be to respond.

Updated my post, added a second link. Agreed, but there most certainly is something different, or not now in place that we haven't thought of yet. Did you test disabling the firewall on the affected servers? purely a process of elimination. Can you identify anything else that has changed during your move from 8.1 to 10. Anything now missing from the new 10 client that the two servers are expecting?

@ITGURU Can you successfully RDP onto the two servers? If you can't then this is not just related to hyper-v management - possibly disable firewall on the servers to test communication. Assuming it is feasible, unplug the lan cable to one of those servers and see if you can ping it - if the IP responds, then there is an IP conflict. Are these Core edition installations? Do the error messages say anything about CredSSP?? Are you ticking the box (for the Hyper-V console) 'connect as a different user' so that you supply the hyper-v admin credentials. I have recently setup a 2016 Hyper-V server (Core) and had to issue these commands to be able to connect to it... On the Server: Enable-WSManCredSSP -Role server On the client: Enable-WSManCredSSP -Role client -DelegateComputer * (if above client cmdlet didn't work, I also issued the -Role server, on the 10 client EDIT: Also found these pages helpful LINK 1 LINK 2

I guess that depends on how important uptime is. RAID isn't a server specific technology, but as you say, it is used less in desktop environments. There are a number of ways to recover, but they all hinge on the weakest link. The server aspect adds a weight of reliability that you'd have to be very resourceful to imitate - sounds like you've got something that suites your needs. Thanks for sharing; I'm always interested in hearing about these kinds of solutions.

If uptime is essential, would you consider RAID ?

I wonder which is likely to fail first, the USB or the internal disk (hdd/ssd). Nice idea, but I'd probably stick the recovery partition on the same disk. I know that defeats the purpose of recovering from a disk failure, but in situations where that is essential, I would install a second ssd inside for backup/restore-only. Incidentally, I've had some USB drives that get really warm when left in for long periods of time; probably just dodgy usb drives though.

What errors/warnings are showing up in the event logs; I'm pretty sure there will be some with this issue. Any events relating to services failing to start? Namely, 'DFSr'. Jet database corruption? LINK EDIT: Event logs on affected server and PDC

Sometimes vendors like Lenovo, Dell and HP have particular utilities, that came pre-installed on the laptop, that essentially provide the logic hook into the OS to know how to treat the battery under certain conditions. Might be worth checking the Lenovo drivers/utilities section for that laptop to see if such an app exists. If it does, install it, along with any related download, possibly even the chipset drivers as there maybe some customisation Lenovo have made. Also worth checking the BIOS/UEFI to see what options you have regarding power/battery.

I'd say no, don't install it. Not that I think it will definitely cause problems, just that it has a higher probability to cause problems than just installing the normal quality rollup. I remember looking at a server once where the preview had been installed and I was a bit nervous for a few days afterwards, partly because I didn't know how it was going to impact things - best not to do it unless you have a particular issue that is addressed in that preview rollup (I'm sure you can check what is included on the MS website).

I'm thinking this will need to be scripted - Powershell, or other... In summary, you'd be resetting the permission of files and folders in a way that maintains who exactly has permission to a specific file/folder.

Thank you for the quick reply @localzuk. You've understood very well. This is an historical thing that I've adopted, and is a fairly common (bad) occurrence in a number of schools I now look after. The more I think about it, the more I just want a single domain, but I just don't know how to go about that. Without asking too much of you, how would I go about merging into a single domain? Ideally, I'd like to get rid of the admin.school domain since all it really does is have the office users (and GPs), and the SIMS database, which doesn't really need to be on a DC. Even more ideally, I'd actually like the domain to be the name of the school, rather than 'admin.school', or 'curriculum.internal' I certainly feel out of my depth with a number of server admin tasks such as this. Can you recommend any good reading? I have a couple of week spare :-)

Forgive the double post, but the original location wasn't that sensible. Folks I've been Googling and I'm struggling to satisfy my mind that the setup is correct at a school I support. Two physical servers. Two domains - curriculum.internal (CS) and admin.school (AS) This school only used to have one server, the AS server. CS was introduced a few years back and has taken responsibility for DHCP, and DNS. However, (AS) has the DNS role enabled and the NIC adapter settings are: Primary 10.1.1.1 (IP of CS) Secondary 127.0.0.1 The AS also has the CS zone in the forward lookup group, and vice-versa The CS NIC adapter settings are Primary 10.1.1.2 (IP of AS) Secondary 127.0.0.1 Every workstation is given the following details for its DHCP request (served by CS) Primary DNS 10.1.1.1 (IP of CS) Secondary DNS 10.1.1.2 (IP of AS) domain suffix curriculum.internal Whilst things seem to be working, I want to make sure I'm following best practise. Question time... Is the DNS role strictly needed on the AS? If so, was is the correct way to configure it? I've wondered whether I should be using conditional forwarders, or just include the others zone in the forward lookup Any good resources, reading I could follow up with to get my head around good DNS and DHCP setup Kind regards, hope you all have a good holiday

Folks I've been Googling and I'm struggling to satisfy my mind that the setup is correct at a school I support. Two physical servers. Two domains - curriculum.internal (CS) and admin.school (AS) This school only used to have one server, the AS server. CS was introduced a few years back and has taken responsibility for DHCP, and DNS. However, (AS) has the DNS role enabled and the NIC adapter settings are: Primary 10.1.1.1 (IP of CS) Secondary 127.0.0.1 The AS also has the CS zone in the forward lookup group, and vice-versa The CS NIC adapter settings are Primary 10.1.1.2 (IP of AS) Secondary 127.0.0.1 Every workstation is given the following details for its DHCP request (served by CS) Primary DNS 10.1.1.1 (IP of CS) Secondary DNS 10.1.1.2 (IP of AS) domain suffix curriculum.internal Whilst things seem to be working, I want to make sure I'm following best practise. Question time... Is the DNS role strictly needed on the AS? If so, was is the correct way to configure it? I've wondered whether I should be using conditional forwarders, or just include the others zone in the forward lookup Any good resources, reading I could follow up with to get my head around good DNS and DHCP setup Kind regards, hope you all have a good holiday

Possibly, but for general maintenance such as this, it can be done during an idle moment - keep those workstations busy and they'll look after themselves Before shutdown is when you may want to manage other tasks, such reboot & update...

@ScottD536, I would avoid adding it as part of your boot/login process, for the reason Oaktech states, it can slow down things slightly - depends on your environment. I'd create a scheduled task to run your script once per day, when the PC has been idle for 5 minutes; or some other similar criteria. You can do that with Group Policy. In fact, there was an article not long ago (can't remember the link), where I read that a number of maintenance jobs are now better served through a scheduled task - things which I used to do via GPO. A definite improvement.

No problem. I'll see if I can write a few instructions later tonight/tomorrow. Am I right in assuming you want to create a Windows 10 install ??

Yes! You need to read the syntax and example instructions... I think this will achieve what you want (NB: it excludes the local administrator account. Add any other users you don't want deleted /ed:) This assumes you have placed the DelProf2.exe in the windows folder. ::Run on a local machine %windir%\delprof2.exe /ed:administrator /q /d:5 /i You can also run the app from the domain server and point it at a domain workstation. ::Run on a Domain Server %windir%\delprof2.exe /c:\\> /ed:administrator /q /d:5 /i

I presume it will, since this is a one-off and workstation based, rather than user based. In which case, the workstation would need a reboot before it fully takes effect.

One to keep in mind, but as others have indicated, MDT, or a standard install USB (with a custom install.wim) is your best bet. If you're needing help with any suggestions here, just say and we'll break things down - not physically ;-)

Here's what I would do. Jot down all the settings that you see - on paper, or in notepad. Ping each server, using the IP address, hostname and FQDN - you are more interested in the IP address replies, but the others are for good measure, but you may get a response back from an old DNS entry. This is to make sure there isn't a real server tucked away somewhere that you don't know about! Remove the entries that are known to be old and redundant Run the Best Practise analyser on the DHCP and DNS roles - it will report if things aren't set to their best settings (some can be ignored) Scavenge entries from you DNS, and DHCP - you may have to do so manually, but you'll need to reconcile forward and reverse entries. IPCONFIG /FLUSHDNS - this perhaps isn't so necessary, and will lead to a repopulating of the DNS cache of the server Be good if others can chip in and see if there is anything wrong with these instructions before you fire them at your server

You could go down a DIY scripting solution - a bit of a hack, but I love that stuff. --Simple overview-- 1. So, depending on the size of the internal SSD, you could create a partition 'deployment' 2. Push/Copy the wim and bootable PE files to the deployment partition 3. Make deployment partition bootable at the next reboot - perhaps more tricky than I'm making it sound... 4. Once booted into your deployment environment, it will automatically apply the .WIM image to the C drive and restore the boot priority back to the C drive 5. Make yourself a cuppa and sit back to admire your resourcefulness - just before it all fails and you're back to doing it manually. The above would work, in theory, and may be a great project, but if you want something quick, the MDT and WIM image on usb would probably be your best bet.

Old servers should definitely be removed - assuming you're 100% sure they are not needed for compatibility or as a workaround for something... For the multiple IP addresses, do you have a VPN setup, or perhaps the Remote Access role (not doing much). If so, take a look at IP settings, particular the DHCP settings in the sub section of RA.

Use a scheduled task with something like DelProf2 LINK

Hi Will I've recently embarked on a little rebuild project myself (thread). One thing I kept reading in threads regarding HP equipment was that they were 'fussy' about what drives were connected - obviously preferring HP branded disks. This may be linked to particular controller cards, or particular systems. Worth double checking though.

Similar thoughts of setting up a pFSense box. Does this provide PPPoE so I can remove my Netgear modem and connect straight to my BT Openreach VDSL modem? [Wall socket] --- [bT VDSL Modem] --- [EdgeRouter] --- [switch] ??? Current setup... [Wall socket] --- [bT VDSL Modem] --- [Netgear WNR2200] --- [switch]