You can seamlessly update from RAID-4 to RAID-DP, but not the other way around. While the reprotection time will be relatively minimal (it changes depending on the size and speed of the disk as well as how busy the controller is) with a spare disk being available, you are also increasing your risk of "silent bit rot" (exhaustive details) with only a single parity check being available on the aggregate(s).
It is possible to have the system operate without spare disks, however you will receive alarms about being in that state. Have you reached out to your NetApp account team to discuss your capacity concerns?
If you're not already, make sure you are using Advanced Drive Partitioning (ADP), which uses shared root and data disks increasing system efficiency.
Hope that helps,
If this post resolved your issue, please help others by selecting ACCEPT AS SOLUTION or adding a KUDO.
Andrew, many thanks for this quick and detailed reply.
In fact i should have enought space but it makes me crazy to loose 6 disks, this is really too much security for SMB enterprise. I prefer having more space.
You 're totally right about silent bit rot, i have totally forgot such concern, then raid dp without spare seems a better choice.
On dot 7-mode there were a "raid timeout option" that shutdowns the controller after an amount of time (24 hours by default)... i had it at a customer 🙂
I do not want such a thing on the system, for sure.
We are using ADP to avoid "wasting" system disks.
edit : i just read an excelent paper on the new ADP that cdot 9 uses : root-data-data (instead of 'old' ADP root-data). It seems good as i can use active/active controler AND have 2 parity and one spare for all the 20 disks !
I do not know if this can be used on a fas 2552 (it is writen 'not on entry fas') and if mixed hdd/ssd could be used.