2 failed disks

[fernandojosepenna]

Hi Guys,

Good Morning. I have 2 disks failed in my storage hp 3par 10400.
But the status os disks is degraded, and don't change to Failed status. How can i change the status to failed, to change this disk.

[fernandojosepenna]

One Doubt, the red edge can i cahnge the or none. i still wait for rebalance stop, to change teh disks?

[fernandojosepenna]

in attach, The status of disks don't change. Can anyone help.
How command can execute to change disks the degraded to failed.

[fernandojosepenna]

please help me i'm really confused cause couldnt find any related doc on the web .
thanks.

[feldsam]

Hello, what following commands show?

showpd -e
showpd -s

I will try to decommission failed disk and try to admit again or replace it.

How to remove PD

[Yavor]

Hello,

there is an interesting question though:
we had similar issues with servicemag
following the procedure above i was able to replace the disks
however after replacement the disks remain with 0 spare chunklets
is there a way to reinstate the spares?
something like a reverse of the removespare PDID:A

[MammaGutt]

admithw will recreate and distribute required spares.

[Yavor]

Thank you MammaGutt for the clarification.

However is there a way to do it, as it should but it does not?

[fernandojosepenna]

Hi Guys,

i resolved this problem:

This one LUN is with one stuck chunklet, and this LUN lost data, and the Chuncklet do not leave this LUN, then i removed this LUN, and chuncklet disappeared. then, i run servicemag again and run with nor problems.

Note: i recover data for this LUN.

is occurred in saturday, and resolved.

being that i have one disk crash, and crash another yesterday. that is, i have 2 disks failed again.
but none with degraded, but is FAILED, and then, i wait for the disks and change disk 1 at a time.

[MammaGutt]

Data loss is never fun but there should always be lessons learned after such incidents.

As a note for the future, you might want to look at RAID6 for your volumes.

* It is not generally recommended to use single-parity on basically anything anymore from any vendor.
* Your system and drives are getting older for each day and are even more prone to errors and failures.
* Most if not all 15k fiber channel drives manufactured after 2010 has a higher failure rate than those manufactured before and higher than the 15k 2.5" SAS drives manufactured now (3PAR has released patches the last years to improve error handling and reduced SMART threshold on some of these drives and trying to gently pre-fail them to reduce the risk of double disk failure. If you're running a very old Inform/3PAR OS version you should consider getting it patched/upgraded.
* The last manufacturer to make (new) FC drives stopped making them in 2013 or so, so any replacement you get is refurbished and probably already have at least 3 to 5 years use on them.

With that in mind and the fact that you've already suffered data loss due to double disk failure it sounds like a no-brainer.