HPE Storage Users Group

Cleanur wrote:However from memory the CPG will attempt to grow into another lower tier to avoid this issue but then you need to manually retune to fix.

That is exactly what I hope the storage would do... uptime > all. Do you recall where you read or learned that? I would definitely like research more into that... as it may help to confirm my theory that in the absence of SSD spare chunklets the system would automatically use FC spares if/when an SSD failed.

I can't remember where I saw this documented but my understanding of this (from a while ago) was that

The underlying LD's within a CPG grow automatically based on the CPG policy, however if the LD's are unable to grow further, because either :-

A. The available space can't satisfy the policies layout e.g cage level availability.
or
B. The tier is genuinely out of space

Then the following SHOULD happen

The LD will attempt to grow on the same tier but using a lower availability level e.g
Cage >>> Mag.

If space is still unavailable then the grow will be attempted in the next lowest tier e.g
SSD >>> FC
FC >>>NL
NL >>||

Since NL can't grow down the only option is to try to lower the availability level from cage to mag.

If this were to occur you would simply re-tune the volumes back once you have new space available in the original CPG.

I believe sparing works in a similar manner where spare space in a given tier has been exhausted, but all of these are very much situations to be avoided.