There are couple of things missing like you haven't mentioned total how many drives present in the system and how many vdisk present in the system.
I see you have mentioned " 1 disk failed. I replaced the disk (Disk 5 in RAID5) with a new disk. I cleared the META DATA and Added the disk as spare." -> If the Disk 5 failed and you have replaced it then the status of the new drive should be in AVAIL state. I am not able to understand how you have cleared metadata of the new drive? are you sure you have cleared metadata of the new drive or some other drive in the same RAID5 vdisk?
Coming to next portion of the issue statement "
The rebuild started and after a while 2 disk become on status LEFTOVR. I thought the new disk is a defect one and I buyed a second new disk. This give also this message. In the log I see :
Vdisk reconstruct failed. Command failed. (error code: 1). (vdisk: VDISK_RAID5, SN: )" -----> If the rebuild started then possible you have done everything properly but Vdisk reconstruction failed due to other drives having issue. So eventhough you replace failed drive again and again vdisk reconstruction will never complete until we figure out what is the other drive part of the vdisk having issue which is not allowing vdisk reconstruction to get complete
It's difficult to say anything without checking complete log file. So I would suggest you to log a support case with HPE support or else provide me the MSA log in private chat
Note: Never clear metadata when you find your vdisk in QTOF or Offline state. Also if there are situations where multiple drive having hardware errors then also you shouldn't clear metadata instead involve expert to check further.
Hope this helps!
Regards
Subhajit
If you feel this was helpful please click the KUDOS! thumb below!
***********************************************************************************
