Did you backup?
calibernode
Provider 
in General
Did you back up your server this week?
And if you don't do it automatically... why not?
How much data have you lost recently if ever?
I want to hear your backup horror stories and don't lie! 🤣
if your data only exists in one place, it doesn't really have a backup. Unless of course your data is stored on this!
CaliberNode.com — 200+ Open-Source Apps, Instantly Deployed on Reliable Hardware
Self-host without the server hassle.
Comments
I lost like a month worth of claude code chatlog (mostly stale conversation). Now, the Jsonl is streamed to 2 boxes on hourly basis. The box was on monthly backup schedule. Oof.
Sadly, I’ve never really had the habit of making backups. I usually only remember how important they are after I’ve already lost something.
Yes !! 👍
Btw Lost three sites hosted on HostDzire vps (Mumbai). Restored two thx to backups. Third… used the opportunity to get out of Wordpress .
P.s : in 2018 one (human) error and I lost all personal data -images, videos… pre-2018. Another blessing in disguise 🥸
blog archives
Sadly my backup horror stories are very very public.
Michael from DragonWebHost & OnePoundEmail
Completely lost data of two hostdzire Indian VPS 😔
Sign up at Dynadot [aff], spend $9.99 within 48 hours, and get $5.00 DynaDollars for domain registrations!
We store our files on a HDD pulled from our Dell Inspiron 600m laptop, purchased in 2005.
It's been a decade since we installed this drive on the USB-PATA adapter, connected to a USB 2.0 port on our home router.
Every night we hear weird noise coming from this drive.
Don't worry, our files are totally safe because it's right next to our pillow.
Daniel fan club 😎 affbrr
Weekly auto backups with email notification with the log as message body, the only way to backup I accept...
20 years ago I was playing with mi PC installing Ubuntu. Somehow I damaged the whole 80GB HDD with all of the photos of my family.
I think I got traumatized and now I always take backups of important stuff.
I've kept my data online since 2002. It's fucking worthless, but everything still works.
"It's a hard life- to be a stick insect." - Karl Pilkington
Don't forget to also test out restoring a backup to make sure it actually works and is backing up what you expect.
As mentioned above, test your backups. I have never lost anything by not backing up, but I have lost things through backups being corrupt.
In 2022, while trying to restore the backup, I found out that my site was backed up perfectly . . . without the database... The latest database backup was 6 months old when I ran it manually. Lucky I still had the main site online when moving to a new server so ran a latest database backup manually and restored it from there.
Ever since then, I no longer trust "complete" backups and will go and manually backup data at least once a month
I have lost things sending a corrupt backup overwrite pristine data because: brain fart.
ok it was not 'lost' thanks to the 3-2-1 rule, more like an awful waste of time and bandwidth
Happy customer at AlexHost, AxusHost, Bakker-IT, Host-C, Ionos, Veesp + NanoKVM ftw.
Yup, done that
No, cause I like the thrill living on the edge.
I used to think that when killing hookers instead of having to pay them, but the thrill is very short-lived.
rclone - rsync - localsend - snapshot - cron
Hetzner 5TB storagebox (to clone OVH-KS2 master and Hostbrr boxes).
Android phone is localsend copied to local pc (nightly)
Total storage costs usually hover around $45 USD a month
It's a bit of maintenance, but I can afford it, and it gives me some peace of mind.
I'm joking, I'm always still stressing about it.
I have been collecting elastic rubber bands since 1992.
Only wimps use tape backup: real men just upload their important stuff on ftp, and let the rest of the world mirror it
— Linus
“Technology is best when it brings people together.” – Matt Mullenweg
@somik that one's the killer. Job reports OK every night, then months later you find it was writing nothing useful.
I look after monitoring for a few small shops, and we got the same shape elsewhere: a customer's suricata crash-looping, 14,177 restarts in 8 days, owner thought he was covered. We only caught it on day 8. Service was "running", so nothing fired.
Now I alert on the output, not the exit code. Dump size vs last week, row count. Exit 0 just means the script finished.
Yes, that reminds me, I already setup a logs monitoring system but have yet to implement it. Just being lazy as I have to update my legacy backup script and since it's been a few years, it's mostly me thinking "wow, I was so stupid when i wrote this script", or "why the f**k would i do it that way" and so on... Well, I should get to it BEFORE the next backup failure!
The rewrite can wait, the check doesn't have to. A cron line that records the dump's row count and mails you only when it drops would have caught the 2022 one in the first week, since there was no dump to count. Across the 25 boxes I watch, output size is the only backup signal I trust.
I've just had to restore a year old backup I had with Aquatis.Host since I lost some crucial data that was pretty much cold storage and it was once stored in the VPS I had with them, and I ultimately needed that info to get some legal paperwork done recently.
If it wasn't for their automated Virtualizor backups, this would've been pretty much lost media, and I would've had to rot in the city hall for like a few months for it to be re-issued, god bless Aquatis (aff)
The data in question is now saved across 4 different servers, 1 NextCloud, 1 seedbox, iCloud, and Google Drive, so I hope I won't need to go through this again. Restoring a live KVM backup, logging in to see a system with dozens of pending updates and "Last logged in Dec 7 2024" was certainly delightful.
datalix's aff #1 fan
Thank you, this can not be emphasized enough!
I've been in IT for 35 years and I can't even count the amount of times I've had customer thinking they have backups because they got the "backup completed" email, just to later find out that the backup job successfully completed copying exactly 0 bytes of the data they wanted backed up.
Verify your backups before you need them.
The 0 byte case is the merciful one. Worse is the dump that is not empty but truncated, mysqldump --force exits 0 after skipping the table it choked on, and an empty dump still gzips to a few dozen bytes so any size greater than zero test passes. Alerting on a percentage of last week's size catches both.
I help my family for their backup. for my own photo and website? hell no, yolo!
Action and Reaction in history