On Fri, Mar 8, 2019 at 1:09 AM <no_spam@xxxxxxxxxxxx> wrote: > > BTW: > S_Haran on Reddit suggested that given I have /etc/lvm/backup > I should be able to use and I quote: > " > When a logical volume is created a copy of the config is kept in the > OS volume (possibly /dev/md1) in /etc/lvm under subfolder name backup > or archive. If you can find the config file then you can use it with > command vgcfgrestore to bring the PV /dev/md3 back to life. > " -- > https://www.reddit.com/r/DataHoarder/comments/aws9iv/btrfs_shr_rookie_mistake_dd_bs1m/ehynwr3 > > I've asked him to explain the process. Hopefully he will sometime tomorrow. That was my fallback plan, until just a minute ago... Question: When you backedup the 1MiB of /dev/md3 as a file saved on the main NAS volume, it took you some time to realize the mistake. I'm wondering how long the NAS was running after this wipe, before it got shutdown? Because, it sounds to me like there's some possibility the metadata in the wiped 1MiB section is stale. It might be quite a bit older than the actual state of the file system, because the storage stack isn't going to immediately stop working just because that 1MiB is zero'd - in fact there's a pretty good chance it doesn't reference that section except at mount time. Soooo, yeah we may in fact have to do a reconstruction anyway. But that 1MiB backup might still be useful. -- Chris Murphy
