On the server, one of the virtual machines caught the brakes on the virtual disk

[Max]

Hello friends!)

The problem is this.
There are ten virtual machines on the server. All of them are located on WDBlue raid 1 disks. Everything worked stably for more than a year and suddenly one of the virtual machines began to slow down, the disk was loaded at 100% (SQL base is located on the virutal machine). We checked the physical disks by load and by SMART - everything is fine. We urgently solved this issue by transferring this virtual machine to another physical disk (to raid 1 disk WD red) and then everything worked well. But they couldn't understand the reason anyway.

Please help, who has encountered this or knows what the reason may be?

We tried to disassemble:

1. We thought the problem in WD blue was due to the new SMR technology (That they do not withstand the load well and lose performance). But we put together another server and simulated a similar situation, gave the load to the disks
   for 4 days and after the test everything is normal, no brakes and everything is fine according to SMART)

2. Viruses are definitely not the problem. Everything is checked and the system is in order

3. Drivers for better XCP-ng performance are installed. The system has always worked stably

So it remains a mystery what it could be (

Thanks for any help!)

[olivierlambert]

What would be interesting to understand is if the slow down was caused by a VM or by the disk itself.

[Max]

Good afternoon @olivierlambert !)
Most likely caused by the virtual machine itself. Because all the other VMs were fine) and the VMs are all on the same disks. According to the load and smart physical disks, everything was normal.
Also, when restarting the problematic virtual machine, the problem did not disappear. We didn't want to recover from backup, because it would take a lot of time and it was decided to move the virtual machine to another physical disk.(After that, the VM load problem disappeared)