HPE Alletra AF6xxx series - High rate of SSD failures

[yesort]

Hi, I was wondering if anyone experienced a high rate of SSD failures inside their AF6xxx arrays. Over a year ago we bought our first Alletra and everyhting went fine (AOS 6.0.0.400). When we bought our second array, it came with AOS 6.1.1.100. We then updated both arrays to 6.1.1.200 to group them together and from that point on, things got crazy. The older array had 3 complete freezes caused by faulty SSDs. Updating them to 6.1.1.300 solved the complete freeze but the SSDs keep failing. Support got some of them running again by resetting them. None of the already changed SSDs failed a second time so clearly it has to do some with the workload they have seen. We got some insider information from support which I cannot share, but I was wondering if anyone else experienced similar issues.

[Satish04]

Hi yesort,

I have not been able to face these issues, but I would like to suggest the workarounds below.

Identify any patterns or common factors in SSD failures by keeping a detailed log, including serial numbers, failure modes, and error messages.
It is important to keep all SSD firmware up-to-date. Sometimes, firmware updates are necessary that are not included in general AOS updates.
It may be worthwhile to analyze the specific I/O patterns and data operations that are causing the problem since you said the issue appears related to the workload.
SSDs can be stressed differently by high write or read amplification, excessive wear leveling, or other workload characteristics.

If the issue persists, I recommend you contact HPE Tech support for better assistance.

Hope this helps.!

Regards,
Satish