Problems with Hetzner, need advice?

Question

Hetzner

Greetings. I will say right away - I am not an admin guru, I can install software and somehow set up the server in a basic way, but no more.

Yesterday my VPS at Hetzner suddenly crashed.

The server stopped giving http, ftp \ ssh - does not connect. Ping does not ping. Those. It felt like the server just went down. In the robot, the status is "working".



I poked “restart” in the robot, got an error:

> The server details could not be determined due to an internal error.
> Please contact our support team by using the contact form on the support page.

I wrote to the support, at that time I poked “off” and “on” a few more times, in the robot the above-described error was issued, then it was said that everything was OK, but the server was also down.

I went to the emergency console and saw the following there:

Gave up waiting for root device. Common problems:
- Boot args (cat /proc/cmdline)
- Check rootdelay= (did the system wait long enaught?)
- Check root= (did the system wait for the right device?)
- Missing modules (cat /proc/modules ;ls /dev)

ALERT! /dev/disk/by-uuid/fa3736c9-9aef-45ed-82c5-8f0814a8f399 does not exist. Dropping to shell!

BusyBox… blah blah blah

(initramfs).

While I got to the console and googled what the support says, they said “go to the console, it seems the server cannot boot there.”

Answered they say I already see, but I have no idea what to do with it, help pliz. I sat down to google. Support at that time went home, he works until 18 in Germany.

After googling I did:

blkid

Looked at guides of disks. The guide that was in the alert was listed as sda3.

Then I tried to mount sda3 to root to see what happened to the partition, as a result, the console hung for about 15 minutes and I decided to press “off” in the robot again, then I turned on the server and it booted up as if nothing had happened.

Nothing in the server logs. Those. all recordings end at 16:10, then two and a half hours of downtime and server loading. No suspicious activity from yesterday. Today correspondence with support. Their point of view is that there was nothing and everything works.

I'm trying to figure out what the hell it was. I ask for advice - what could cause the loss of the disk? In which direction can you dig?

Answer 1

Overheat? Broken screw? Where is my coffee grounds? It doesn't really matter.
Make sure you have a backup of everything.

Answer 2

For those who found this topic by searching, I give instructions.
1. Go to the hetznera management system: robot.your-server.de/server
2. Activate the resque system of the version that was installed
3. Launch the web console, vServer tab. If the console has started, send the cntrl+alt+del command,
if not, restart your virtual server with power.
4. After starting the resque system, you can connect to the server, including via ssh. Your client will swear at the substitution of ssh keys, ignore it!
5. In the ssh console of the server, run disk check fsck -C0 /dev/sda3
6. In the web console of the server, run reboot shutdown -r now. You will be able to see what the console of your virtual server writes during the boot process.
More information here:
wiki.hetzner.de/index.php/Hetzner_Rescue-System/en

Answer 3

In the sense of overheating or a defective screw? After all, the virtual disk is vps for me. If the problem was with a piece of iron, then this would affect many users and it would be unlikely that the hetzner would remain silent “we don’t know anything”.
I can’t initiate a check with them in any way - I can only check on my side. Therefore, my question is this - what can theoretically cause disk loss if I have not touched the server for a week or more, and it has the usual lamp set + mail on it.

Answer 4

Chip hosting for those who know how to administer or are willing to pay 21 euros for 15 minutes.
Use chip hosting - solve your problems yourself.

Answer 5

better chip hosting and decide for yourself…