Microsoft wants you to buy 1 PC and license per user. They pretend to support multi-user systems but in reality most of their admin tools suppose a machine is dedicated to a user, or that users are at least generally always using the same machine.
If you can programatically fix the issues causing the calls, put the fix in a login script using traditional mechanisms.
MS support should be able to help with that, or at least point you in the correct direction. we had the same problem and yeah they did struggle a bit but eventually they were able to help.
if you want to roll your solution, your best bet is to is forgo the old mailboxes (export the emails to the new accounts first) and then basically only use the online account synced back to on-prem.
Nextcloud can easily handle 4Tb, I’ve seen guys with 50Tb instances with no issues
The backup problem is eternal. If you can, set up a consumer NAS somewhere else than where your server is and back up daily to it. Even over a slow connection, you should be able to get by after the first full backup.
If you can’t do that, either you go without backups and learn the hard way when your systems fail (I’m currently doing that, still waiting on the find out phase), or you will have to pay someone to store your data somewhere. From what I hear backblaze is affordable, but they do charge you per download. For 4 Tb, they would charge $180/year, with added costs per download which hopefully you can avoid.
A final alternative is to store cold backups somewhere else (at work, for example). Buy 2 external disks, and back up your server to one of them. Store it at work, then do the same with the other disk. Every month, rotate the disks. Your backup solution probably has a function to do exactly that, so you can leave one plugged in, accumulating this months data over time.
Sure. Reddit could, if they wanted, implement activitypub and join Lemmy, and Twitter could too and join Mastodon. Of course, since their server software wasn’t designed to do this it would be a significant investment and they won’t want to do it for a slew of business reasons, but if they really wanted they could.
We did a gitlab upgrade last week to 16.1and it went fine, but I noticed they never put out a vanilla v16.1.0 tag for gitlab-runner. There exist images such as Ubuntu-v16.1.0 but we usually just use the vanilla one. Anyone know what’s up about that?
Did that before. First, ensure you aren’t running on any snapshots. If your VMs have any snapshots, delete them.
If the move function works then use that but I suspect it won’t. You can simply use the “move the vm’s storage” option to ensure the VHD is properly consolidated, then you can put it on the new host and start the VM there (shut it off before moving of course). If your host is running a newer hyper v version you might need to bump the hypervisor version for each VHD in PowerShell, that might take a while but it’s usually fast.
