Hello, I've had a system crash. I have all the data backed up including snapshots of the db server and the clients folder, the etc folder and the entire ispconfig and vmail folder. Unfortunately the virtual machine that all of this was on became corrupted btrfs file system corruption that is just unrecoverable. Not happy about that. I digress. Is it possible to recover from this by building a new perfect server of the same version and simply laying the configs and database back down on top of this or am I going to have to start from scratch? Thanks, Curtis
OK, good to know. another question. I built the last one on Ubuntu 16.04 LTS. I'd like to do this one as debian stretch. Can I lay all of the configs and data down on that?
Ok, progress ispconfig is up. email is up. Now to restore all of the websites and the fstab. I'm sure I can assume all the uuid's will be changed, correct?
The uuid's in fstab will differ, yes. But you don't have to copy the whole fstab, just copy the bind mount lines of the websites from old to new fstab.
Changed the underlying platform from Ubuntu 16.04 to devuan ascii 2.0. Oh, it runs so much better without systemd. laying the configs down worked pretty well, but not perfectly. turns out I needed to snapshot /var/www, not just /var/www/clients. I also should have backed up /var/log/ispconfig I'm having to recreate all of those by hand, but progress is being made. I also did not merge the passwd, group, and shadow files quite correctly. It's resulted in my having to correct ownership issues. too late to revert to the snapshot and start again. :-( Thank you for your help. It's coming albeit slowly. mail is up. the websites are all broken. I'm noticing that I have to make a change on them all, save it, then change/save it back. that's gotten the php7.0-fpm pools to start. I'm building a new server at a different location. I'm loath to just move the virtual machine and renumber it. I think it will be cleaner.
