The disk on OVH.COM is crumbling. What to do?

Question

Network administration

On ovh.com I rented a kimsufi dadik (Windows 2008).
The server crashed after 3 months. Reinstalling the system helped. Lost a few days (tried to bring up the server).
This time, just in case, I installed CrystalDiskInfo monitoring.
And now, 5 days later, alarming messages began to arrive in email:

TOSHIBA DT01ACA100 / Z29W8L5PS / C: D:
[601] Maintenance: [Good] -> [Alarm!]
TOSHIBA DT01ACA100 / Z29W8L5PS / C: D:
[603] Breakdown: (C4) Remap events [0->56]
TOSHIBA DT01ACA100 / Z29W8L5PS / C: D:
[602] Breakdown: (05) Reassigned sectors [0->46]

What to do? Is it a disk crash? (there is a reyd from 2 disks Toshiba).
If a disk is crumbling and it needs to be replaced urgently, what should I write to support and how to justify it?
Who has experience, please help, otherwise you don’t want to stand idle for a couple of days a second time.

Answer 1

The situation is unpleasant, but tolerable.
Get more information with smartmontools: sourceforge.net/apps/trac/smartmontools/wiki
Pay attention to the parameters: Reallocated_Sector_Ct, Power_On_Hours, Current_Pending_Sector
Reallocated_Sector_Ct - the number of sectors that have been reassigned to the backup area. If the value grows, then the disk has fallen down and needs to be replaced urgently.
Power_On_Hours - and so it is clear. The time that the disk "flew"
Current_Pending_Sector - the number of servers that the disk marked as unreadable. The disk could not read data from a particular sector and marked it as bad. When writing to it, “pending” may go away and the sector will again become readable, or the disk will reassign it to the spare area (Reallocated_Sector_Ct)
It is enough to specify these three parameters as a reason for the support to replace the disk.
To replace a drive, specify the model, serial number, and possible cage number (in which the drive is installed). Get the serial number and model from smartmontools, although you already have them.
An important point
Before unsubscribing a drive for replacement, check the second drive in the mirror. The main condition for a successful build with a new hard drive is the absence of pendings on the "original" disk. If pendings are present, take care of a backup copy.
All in all

Answer 2

Put smart in addition, connect catcti, and see what happens to the disks. And with this data already write to the support

Answer 3

A banal answer, write to the caliper and they will change the disk for you.
The situation is unbearable for any production.