Help save me...

[BBB]

I won't go into details on a public forum (I'll leave it to the reader to guess why I'm asking such questions) but I wanted to know if anybody has ever experienced any of these:

a) a RAID5 array suffer a double disk failure and lose data

b) a RAID6 array suffer a triple disk failure and lose data

c) a modern array with fully redundant dual controllers fail and lose data (in particular interested in EMC Clariion - but note I've NOT experienced this with an EMC (or any vendor actually), to avoid starting any unfair rumours!!!).

Thanks!!!!!!

 

[gimlet]

I ran into situation 'C' a few months ago where several LUNs on multiple hosts 'disappeared.' Only FC-connected LUNs were affected, and the SAN controllers had nothing in their syslogs, so I presume a fibre switch freaked out momentarily. Our OS, HBA, and SAN combination did not support multipathing.

Luckily, the only data loss was the temporary tablespace for Oracle, which was easily recreated by our DBA.

 

[sandeepkontham]

I never encountered this but one of my frnd experienced data loss due to option (a)

 

[matt.kaleta]

I've ran one time into "C", but in fact after failed firmware upgrade. Strange was, that the same upgrade in EXACTLY same Storage system went OK. manufacturer's support had no clue what happened, so I restored data using AIX LVM Mirroring