3PAR degraded/failed disk workflow

[Aissa]

Hello,

we have several 8450 and need to know what is the worklow when a disk become degraded and then failed.

What is the array doing when the disk is degraded (relocation ?)

What is the array doing when the disk is failed ?

thank you

[Sheldon Smith · Accepted solution]

Take a look at the HPE 3PAR StoreServ Sparing Technologies technical papers available via the Hewlett Packard Enterprise Information Library. While it was written looking at the 10000, everything still applies.

[AshrafK]

In 3PAR, physical disks are at the lowest level of data structure. The next is Chunklets. In StoreServ, a chunklet is 1GB of contiguous space on a physical disk. RAID is implemented at the chunklet level.

For any RAID implementation, 3PAR takes a specific number of chunklets including parity.

If any PD has an issue, then Servicemag will automatically start moving the chunklets to spare chunklets. During this process, the PD will be reported as degraded.

If the servicemag is in progress, then wait for it to complete and then PD will be reported as failed. Then, it is safe for the replacement.

Few helpful commands:-

cli% showpd -failed – degraded

cli% servicemag status

[Aissa]

Hi, 

thank you for you quick responce, really apreciated. The documentation is very good.

Thanks to the auther response with the command i already know.

i looked once on the event log and cant find the date and time about the end of the relocation after a disk went degraded.

I there a command or a place to find when exactly the relocating end, i faced perfomance issue since the drive went degraded but i cant determine when that issue ended pertubating the array.

thank you

[Torsten.]

The replacement is finished when you see the dismisspd for the old PDID.