So here we are again another one bites the dust and im sitting here at 3am with my cold coffee that I made at 11pm when I got the email and my cat Mr whiskers jumped on the keyboard twice because he knows when im upset and it was raining outside in tucson which never happens in October and I had just eaten a bagel with cream cheese that was probably too old but I wasnt paying attention because I checked my phone and there it was Contabo is shutting down their little ARM boxes that ive been running for four years and I paid for three years upfront in 2023 because they had that deal and now its gone and I had everything on there my plex server my home Assistant my wireguard my mums photo backup that I promised her was safe and the worst part is I only found out because someone posted on reddit I never got the email they sent to my spam folder and now I have seventy two hours to migrate four Terabytes and I cant even be mad at them really because I knew it was too cheap to last but still it hurts and my husband came in and said why are you crying about a server and he doesnt understand nobody understands unless youre here
Another company killed their product and I'm emotionally available
Seventy two hours. Thats your restore window now. Test your backups weekly. I learned this when Hetzner "migrated" my node in 2019 and the snapshot was corrupt. Verify your restores. Paranoid? Maybe. Alive? Yes.
Contabo gave no export tool? Classic. rsync while you sleep. Set two alarms.
Wah lau eh this kind of thing happen before already lor
Last time Vultr die also same pattern, email go spam, everyone kpkb then bo bian migrate
Can one lah, 72 hours enough if you start now leh. Bob correct, rsync loh. My hokkien friend say 慢慢來, steady steady can one
Next time dont pay 3 years upfront, siao ah
Fun little project: I had a similar thing with RackNerd's cheap storage VPS line back in 2024. Ended up building a temporary wireguard tunnel to a OVHcloud box and just let it sync for two days straight.
It actually works. Here's what I ran:
#!/bin/bash
# emergency egress script
rsync -avz --progress /data/ [email protected]:/mnt/rescue/Pro tip: screen or tmux. Your ssh will drop.
Maybe its not really about the server, you know. Its the promise they made, you know. The idea that you could just have this little corner of the internet that was yours, you know.
And then its gone, you know. Maybe we keep doing this to ourselves, you know. Maybe the weekend project is the mourning itself, you know...
Mr whiskers KNOWS tho
Bro this one you ownself check ownself. I also kena before, pay 2 years then company pivot become crypto mining only. Now I pay monthly everywhere, even if more ex.
Your plex library how big? Mine 8TB, took 4 days to rsync to Hetzner storage box. Better start tonight loh.
Mr whiskers has seen things. That cat has a thousand-yard stare and three dead VPS providers in his eyes.
This. Last month I tested my "perfect" Proxmox backup to Hetzner. Turns out the backup was fine but the restore script had a path bug that only showed up when I actually ran it. Four hours of panic at 2am.
Test the restore, not just the backup file. The file can be perfect and the restore still broken.
This is the one that hurts. Your plex you can rebuild. Home Assistant you can rebuild. Your mum's photos?
I keep a copy in Auckland with my brother now. Not because I don't trust the cloud. Because I don't trust myself to not pick the wrong provider twice.