DL380 G10 very slow performance on RAID1 with one disk on predictive fail status.

[Psystems]

I have had the same problem with 2 customers the last half year. They have the latest DL380 G10,
and RAID sets with all SSD mixed mode SAS disks. If a disk goes into predictive failure, the RAID set becomes so slow it is unusable, even if there is barely any load on it,  I never experienced this behavior with classic 10 or 15k disks in the past. Is there someting i am missing?

[support_s]

System recommended content:

1. HPE ProLiant Servers - Troubleshooting SATA/SAS Hard Disk Predictive Failures

2. Advisory: HPE Smart Array Controllers: Spare Disk Drives in a Redundant Array Configuration May be Marked as "Predictive Failed"

Please click on "Thumbs Up/Kudo" icon to give a "Kudo".

Thank you for being a HPE valuable community member.

[Erdogan Temur]

Hello,

Have you checked the SSD Wear Status? The slowness is mostly caused by the end of life of the ssd.

[Psystems]

Hi Erdogan,

The disks are brand new, server is installed in december, they are still at 100%. One disk is in predictive fail, the other is fine.
HP is still looking after a replacement disk for a week now. It is a Raid1 set, the other disk is fine, but is unworkably slow.

[BPSingh]

Greetings!

As requested before, please share the following information. 

Could you confirm if the drive with predictive failure is removed, is the server performing okay?

What is the model number of the drive and what firmware is it running?

Storage controller model and its driver/firmware?