Okayyyy. So. Not to put too fine a point on it, I've officially done my unix initiation. You know, the one where you type 'rm -Rfd stuff' and delete / by mistake?
:D:D:D
There will now be a brief but mandatory pause for 5 million forum users to die laughing
In mitigation, I honestly don't know how it happened, as I was cd'd into a data subdir with a couple of thousand pointless HDD image backups in many dirs and no symlinks, and the command was 'find . -name 'text' -exec rm...{} \; and not mistyped. I'm not completely naïve. Well, maybe today I am..... but I'm also not too upset because
My last backup config is from maybe 3 weeks ago. But I know a few shares changed slightly since then. Not fatally, but still worth looking for the old config. The old boot volume had no snaps newer than my backup when mounted after reinstall. But it did contain a bunch of mountable .system datasets, and in these are one .db file per day, about the same size as a config file, every 24 hours, right up to a few hours ago. The latest is "20180114%s.db" - barely 6 hours before things died.
I don't want to try without checking, in case it kills anything, so pretty please with cherries on top, in return for the mild entertainment and a chance to let the schadenfreude out to play a bit
, could someone confirm if these are what I think they miiiight be, and if so, whether it's safe to rename to "freenas-v1.db" and reinstate the latest of them? 
Thanks!
There will now be a brief but mandatory pause for 5 million forum users to die laughing
#!/bin/sh
pause
lol > /dev/random
other_stuff_here
esuap
In mitigation, I honestly don't know how it happened, as I was cd'd into a data subdir with a couple of thousand pointless HDD image backups in many dirs and no symlinks, and the command was 'find . -name 'text' -exec rm...{} \; and not mistyped. I'm not completely naïve. Well, maybe today I am..... but I'm also not too upset because
- 15 min x 7 day rolling snaps on the entire dataset and before I could hit ctrl-C the output was filled with a ton of "cannot delete from read-only dataset", so thanks to the gods of ZFS and data should be just fine. Just import and rollback to previous snap (Nothing's lost by that, I was just wiping a cluster of dead files)
- I managed to wipe the root directory -R as well. Like, the OS itself and my config file. I honestly don't know how it jumped from a data subdir to the root dir. But.... I have config backups anyway.
My last backup config is from maybe 3 weeks ago. But I know a few shares changed slightly since then. Not fatally, but still worth looking for the old config. The old boot volume had no snaps newer than my backup when mounted after reinstall. But it did contain a bunch of mountable .system datasets, and in these are one .db file per day, about the same size as a config file, every 24 hours, right up to a few hours ago. The latest is "20180114%s.db" - barely 6 hours before things died.
I don't want to try without checking, in case it kills anything, so pretty please with cherries on top, in return for the mild entertainment and a chance to let the schadenfreude out to play a bit
Thanks!