Brand new Debian 12 build on a new machine where webmin, usermin & virtualmin where the first and only things installed and was working perfectly fine afterwards. Thought I’d carry on building the mail server side of things today so logged and saw a notification saying there are 67 updates, so install them, now I wish I hadn’t…
It hung on updates for Mariadb server and sat there for ages and ages and ages so obviously something duff in the updates. Rebooted after 10 minutes after being stuck at the same place and now the machine cannot connect to the network, cannot ping anything not even localhost, 127.0.0.1 nor its IP address.
Thank god this wasn’t a production machine. So yeah, basically killed the server and unable to connect to the network now its basically can’t do anything other than rebuild the whole thing from scratch again.
That wouldn’t be virtualmin updates that would be debian updates, why apt failed on mariadb IDK but it’s time to research apt and fix this from the command line or re provision the server and start again
Just built a brand new machine using the same USB stick I used last week, Ran apt update/upgrade and it says there’s nothing new to install. So why did Virtualmin say there were 67 updates when Debian says there aren’t any?
We do not provide the Mariadb package, your OS does.
Virtualmin is showing you updates available from your system package manager. If you run apt update; apt upgrade it’ll do roughly the same thing running updates in Virtualmin does (though Virtualmin gets a list of packages first, using apt-get --just-print upgrade to get a list of available updates, and then allows the user to choose which ones to update).
I’d be surprised if a Mariadb update (or any package update other than kernel) could kill a healthy server, but we haven’t rolled out any new updates in quite some time (two or three weeks back), so if any package killed your server it was not one of ours (I’m even more certain of that than in the usual case, and in the usual case I’m very confident our updates won’t kill your server…we might break our own software from time to time, but we’d literally never break your kernel).
You’ll need to get logged in, maybe in single user mode via a console or remote KVM or whatever, and look at the logs to be sure what happened.
I just got hit by a big upgrade of Debian 12 on my home machine and my Debian 11 server. Both completed OK. Weird that you used the same image and there were no updates. That sounds a little strange.