This week, our #SysAdmin syllabus covers backups and restores, including use of dump(8), #rsync, and flux-capacitors (e.g., ZFS snapshots, Apple TimeMachine, NetApp's WAFL). We also were supposed to talk about #syslog and monitoring, but honestly, chances are we'll spend most of our time on the #xz#backdoor.
So very sad to see this happen to three instances that seem to be doing pretty well. And now to read the news that all three of them are gone, and two of them won't be coming back at all.
Instances affected are: musician.social outdoors.lgbt and firefish.lgbt
Looking for a backup solution for both Windows and Linux targets (about 20/80% respectively). We're building a new backup server (Currently using BackupPC), and figured this was a good time to start researching other options.
I'd prefer a solution that's on-prem (e.g.no cloud solutions).
Anyone else use #rclone with #mega? Are your #backups working? Seems mine suddenly stopped on ~28th Feb. It is now failing to log in and hanging up my #RaspberryPi.
I've uninstalled Debian's stable version (V1.60.1+dfsg-2+b5) of rclone and installed from git (V1.66.0). Still having the issue. Tried removing all configs, tried with --debug and --verbose flags both show nothing other than trouble logging in.
I don't have 2FA enabled as it's unsupported on rclone.
Since all the search engines have gone to shit, I'll resort to the old-fashioned way of finding information: Asking people.
So, what's the easiest, most foolproof way to do encrypted offsite #backups on #Linux?
I have a few TB at Hetzner, and a few personal laptops that need automated backups.
The whole thing needs to be fairly foolproof to run (because at least one kid is going abroad for a few months with their laptop), and foolproof to set up up (because it's me setting it up)
An interesting read, and I have to say I do lean more towards building your own. I did buy a bare-bones NAS by D-Link many years ago because you could add your own hard drives, but what I discovered after a few years was there were no more software ...continues
Exploring restic for backing up my workstation (to local external volumes and probably "the cloud" too) at the suggestion of a friend. https://restic.readthedocs.io/en/latest/
Maybe just in time, because...
[605358.398403] nvme0n1: I/O Cmd(0x2) @ LBA 131640760, 1024 blocks, I/O Error (sct 0x2 / sc 0x81) MORE
[605358.398428] critical medium error, dev nvme0n1, sector 131640760 op 0x0:(READ) flags 0x80700 phys_seg 88 prio class 2
Now I'm replacing the primary NVMe SSD tomorrow too...
On the positive side, the files reported unreadable were all various things under /work/xilinx/Vivado/2019.2/...
So, if I'm lucky I haven't lost anything important and Vivado has finally proved itself useful... as a 25GB ablative shield for the rest of my project data on that volume...
Well, that was fun - thought I'd do a little filesystem cleansing and ended up accidentally (I did not check my work, tut tut tut) removed fedi VM ( I thought it was on a different disk, whoops!) - joy!
#Backups are available of course and restoring now. #Snac2 being the easiest to restore so that's back online first! I've lost today, but that's OK I've been pretty quiet!
I have long used backups that are monthly full backups, weekly full, and daily incremental (or sometimes also full if smallish). Yesterday I learned this is traditionally called the “Grandfather-Father-Son” scheme. This was a surprise, at least to me.
Seems a bit patriarchal and has nothing to do with backups. Is there a better name currently in use?
I briefly looked at AWS S3 Glacier storage, thinking maybe having a second cloud host for my backups would be good.
I can't actually figure out how much this will cost me because they charge per operation (you know, like PUT, GET, etc.) in addition to the storage costs (which I easily figured out).