Oaktech Posted May 25, 2016 Posted May 25, 2016 I've got a 2012r2 server core install VM as a DC. It shares duties with a physical DC running 2012r2 gui. Everything is fine between the 2, AD, sysvol, DNS and DHCP failover all replicate/work. This may be a rhetorical question: Am I brave enough to take down the core VM, move it to a new VM host and fire it up again in the middle of the day? Am I an idiot for even suggesting it? I can't see a technical reason why I couldn't, but the old school sysadmin in me says this is an out of hours job.
3s-gtech Posted May 25, 2016 Posted May 25, 2016 Only thing is, if it goes down while people are logged onto it you may get some brief interruption where users are logged onto the sysvol from that machine, until DFS-R takes over and shoves them across. That may mean some policy oddness if someone is logging on at the time - perhaps Start menus disappearing if you redirect them to the sysvol for instance. Still may be worth doing even as a test - it shouldn't actually cause any harm beyond a couple of Windows error messages for a few users. 1
Oaktech Posted May 25, 2016 Author Posted May 25, 2016 Only thing is, if it goes down while people are logged onto it you may get some brief interruption where users are logged onto the sysvol from that machine, until DFS-R takes over and shoves them across. That may mean some policy oddness if someone is logging on at the time - perhaps Start menus disappearing if you redirect them to the sysvol for instance. Still may be worth doing even as a test - it shouldn't actually cause any harm beyond a couple of Windows error messages for a few users. We don't redirect to sysvol, so that's not an issue... But, as @Patrick suggests, murphy's law!
Chuckster Posted May 25, 2016 Posted May 25, 2016 (edited) I don't envisage it being a problem because I install monthly updates to both DCs during work hours and restart them as and when required - again, during work hours but not both at the same time. Also took one one of the DC's offline for an hour while it was transferred over to a new storage device. Edited May 25, 2016 by Chuckster Forgot to add "don't 1
3s-gtech Posted May 25, 2016 Posted May 25, 2016 I don't envisage it being a problem Edited for what you meant 2
dapaulio Posted May 25, 2016 Posted May 25, 2016 (edited) I have always tried to leave all server maintenance until the end of the school day with prior notice to the users. i wouldnt like to suggest doing it any other time to anyone. if i have had to reboot a server during the day i would have to have a bloody good reason if disruption if any was noticed. I run that risk myself but i would always suggest server maintenance to be carried out at the end of the day with prior notice to your users. then you are covered. This scenario does not sound like an emergency and therefore could wait Edited May 25, 2016 by dapaulio 1
Chuckster Posted May 25, 2016 Posted May 25, 2016 If your DCs are simply DCs and nothing much more than that, i.e. file server et al, then you can reboot a DC one at a time. No noticeable affects are seen. 1
pete Posted May 25, 2016 Posted May 25, 2016 Live migrate (DC stays running with a very brief interruption for clients) or actually shutdown and move? I've done the former lots of times without issue, I tend to avoid the latter during the working day unless I have no other option. 1
Oaktech Posted May 25, 2016 Author Posted May 25, 2016 Live migrate (DC stays running with a very brief interruption for clients) or actually shutdown and move? I've done the former lots of times without issue, I tend to avoid the latter during the working day unless I have no other option. Live migrate is not an option... Current host is 2008r2, it's moving to 2012r2. I've wimped out and scheduled some time on Tuesday during half term. I've got a 3TB physical box to p2v as well (gulp).
dapaulio Posted May 25, 2016 Posted May 25, 2016 Live migrate is not an option... Current host is 2008r2, it's moving to 2012r2. I've wimped out and scheduled some time on Tuesday during half term. I've got a 3TB physical box to p2v as well (gulp). Good option. No point adding more stress to your working day then you necessarily have too. I assume your 3tb p2v is your file server. Piece of p!ss. You'll be fine.
Oaktech Posted June 1, 2016 Author Posted June 1, 2016 Good option. No point adding more stress to your working day then you necessarily have too. I assume your 3tb p2v is your file server. Piece of p!ss. You'll be fine. The DC came over no problem... I shut it down, copied the VHDx over, preserved the MAC address and it came up and is replicating fine. The 3tb box - not so much. The OS Disk-2-VHD worked a treat, but the data drive VHDx mounts as RAW and is unreadable. It's possible I captured it in the wrong format (didn't select VHDX from the type drop down), and it's possibe I connected it to an IDE before I read about >127gb VHDx needing scsi, so I'm capturing it again to have a play. The go live will have to wait until either a weekend or the summer now. Which is annoying. I never had this grief with VMWare, but I can't afford to licence all the features I get for free with HyperV
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now