Scaleway have lost my server


#1

Hi there.
29 March 2019 (Friday) and 31 March 2019 (Sunday) I got e-mails from Scaleway:

Subject: “Planned migration impacting your server”
Dear cloud rider,

Due to an issue with the physical server that hosts your instance 81fcb496-2591-40c7-9304-77461ee6919b, we are going to migrate your instance to a new hypervisor.

This maintenance is scheduled the 2019.03.31 14:00:00 UTC, your server will be stopped and then started on another physical server, during this operation you can expect a downtime of a few minutes.

No specific actions are required on your side, if you have any question, do not hesitate to contact our assistance.

We are very sorry for the inconvenience.

The Scaleway Team

I have read it “No specific actions are required on your side” and I have decided that it is a safety procedure and I really shouldn’t worry about it.

01 April 2019 (Monday) I received another message (ticket TEYJ-6033-IAIR):

Subject: Critical issue with your server
Hi rylskov,

Our support team created a new ticket associated to your account.

Dear Cloud Rider,

Your instance ‘amsterdam’ (‘81fcb496-2591-40c7-9304-77461ee6919b’) is running on a hypervisor that underwent a critical failure. Unfortunately, we were not able to reboot the hypervisor. As a result, all files located on your local volume could not be recovered and your node has been stopped.

If you made snapshot of this disk or if you halted your node recently, you will be able to recover your disk to the latest stop state.

We strongly encourage you to snaphot your server regulary to ensure you have a recent version of your files on your server. The Hot Snapshots feature which enables making snapshots while the instance is in production is now available on all our Virtual Cloud Servers: https://www.scaleway.com/docs/backup-your-data-with-snapshots/.

We apologize for the inconvenience and remain available for any question you may have.

Scaleway Team

(The last snapshot of this server is created 2 years ago.)

So. The Scaleway team did some mistake when migrating my server and all server-data was lost.
I tried to ask reasonable questions, why they allow the situation that the customer’s server was lost, why the staff didn’t make backups before dangerous actions, that they couldn’t do successfully? Why they sent messages with no recommendations to make backups? I asked about any compensations too, because this situation is unacceptable for hosting service.

We discussed this issue for 2 days. And they didn’t provide any solution. They just apologize and it all.
Moreover, they tried to shift blame to me!!! They wrote that risk was based on the absence of backup!

  • That the staff didn’t make before migration - WHY?!,
  • I didn’t make a backup too because I got a message from Scaleway where was written that I should do nothing.
    update: I did backups to secure my actions, but these backups were stored on the server.

It looks like they are mocking. It’s very cool to kill a customer server and apologize. Do the Scaleway really think that it is enough?

I post this issue here to warn customers that the Scaleway doesn’t provide reliable service.
Your data could be lost anytime, and it is very easy because the Scaleway doesn’t worry about these. Anyway, they will just apologize for the huge problems that you can get here.


#2

Hello @rylskov,

I am sorry to hear that your data was lost.
After checking your issue with the concerned team, indeed your instance was running on a hypervisor having a failing disk. All data stored on sectors that were readable was backuped automatically, unfortunately your data was stored on sectors that were no longer readable due to the failing state of the disk. We will deploy our new developer and general purpose instances in the Amsterdam region very soon, letting you profit from a RAID-5 on all hypervisors. This new compute platform is already available in the Paris region. Feel free to come back to us by ticket if you have any further question.

Benedikt