All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Hetzner HDD/NVMe SATA Disconnecting Issues
Hello,
I have a couple of AX101 and SX63 dedicated servers with Hetzner. (No, I'm not farming/plotting Chia, I rent them out to different clients. They're targeted for game servers and high-capacity slow storage).
My servers have had their SATA cables or NVMe connectors disconnected around 9 times.
- 3 of them happened on one of my SX63 servers (main drive).
- 2-3 of them happened on one of my other SX63 servers (main drive).
- 1 happened on an SX63 (main drive).
- 1 happened on an SX63 (main drive).
- 1 happened on the secondary NVMe drive of an AX101
On the ones with repeated disconnects, I've asked them to completely replace the SATA cables and they're still experiencing the same issues. The rest haven't had an issue since. I doubt they're cloning my drives or anything.
I have noticed that on the ones with repeated disconnects, once they boot back up, Proxmox reports a higher error count.
Device: /dev/sdc [SAT], ATA error count increased from 4 to 5
Does anyone know why this is happening on the ones with repeated disconnects? Why is this so common among my servers?
Thank you!
Comments
You get what you pay for bruh
Ticket them, ask for cable replacement and port switch.
If it consists, replace drive or motherboard, simple.
Hetzner does that all for free anyway.
Considering the widespread issues and it occurring on NVMe, it sounds like a software issue or you are misinterpreting whatever it is makes you think they are disconnected
Sounds like the early stages of drive involucration..
Asked them to replace the entire server other than the drives. Perhaps it's an HDD failure.
Running Proxmox VE. When the drives disconnect, I typically check the BIOS and Rescue. Both don't show the drive showing up.
So the disconnections survive a reboot? What about a power cycle?
Hello
You need contact Provider for that issue I am sure hetzner will sorted out asap
Update on this:
The frequent disconnects stopped happening on one of my servers, and I ended up replacing the main drive on one of my other ones. Seemed to stop.
Voidlinux infra guys faced some hardware issues this week as well - https://github.com/void-linux/void-infrastructure/commit/35ee4ec9306607e364be7af8162268985e0f649a