Subj : Backup Nightmares @ 2oFB..
To   : All
From : paulie420
Date : Thu Jun 29 2023 07:49 pm

I'm bad at running at the mouth in posts, so I'll try to stay on track:

I have a BBS VM. I mount some NFS share drives on the BBS VM. I also have a PBS backup server.

Last week some users mentioned they were having DL issues; I thought it was a separate issue with filenames - but I've been upgrading many systems to Debian 12 this past week and finally got around to my PBS machine. I noticed in the PBS dashboard that only 2TB of a 16TB drive (thats usually close to full) was being used. Wait, whaaa??? So I checked its NFS backup's DIRs and they were all empty!! HUH?

The rsync command I was using to backup had the --delete tag. I thought that if I modified my BBS VM and removed some /mystic files that I'd want that to be replicated on my backups....

Well; come to find out, my NAS went down sometime last week... and, my BBS VM and PBS servers NFS mount points went down. Historically, I thought this fixed itself - I'm almost sure of it, but...

The weekly rsync backup ran - and... --delete made sure ALL the backup data was - deleted. And, users weren't able to download this past week and the bug reports started flowing in.

When I found it, a simple reboot and remount fixed things... I'm currently backing up some 32TB of data; great for the drive, right? Delete all the data and then re-write it. Awesome. 

I've removed the --delete tag, and am rethinking how I'm doing things... guess I need to put in REDUNDANT checks on the mountpoints, and maybe check in my backup script that if /WHATEVER/MOUNTPOINT is empty, don't backup.

Ugh; stupid, Paulie...



|07p|15AULIE|1142|07o
|08.........

--- Mystic BBS v1.12 A48 (Linux/64)
 * Origin: 2o fOr beeRS bbs>>>20ForBeers.com:1337 (21:2/150)

.