I have a CX4-120 and we just added another DAE and some additional drives. The disks that we added are bigger than the ones that pre-existed and now I am getting the error alert below:
The storage system has no hot spares that are as large as its largest disk. If a disk drive larger than the hot spare fails, the hot spare cannot substitute for it.
ALERT CODE 0x7206
Resolution
Allocate at leat one hot spare as large as largest drive in the storage system.
The general rule of thumb is to have 1 hot spare for every 30 disks. The Navisphere manager help guide will show you how to configure a hot spare, this should be configured at the end of the enclosure (disk 14).
With newer flare versions you also must bind a LUN to the new hot spare raid group before you actually have a hot spare - the older flare was a one-step porcess.
Thank you everyone. I seem to have corrected it thanks to the great answers provided. I wish I could tag more than one answer as the "correct answer" so I'm sorry if you only got a "helpful answer".
One thing I still am a bit fuzzy on, according to some of the rules of thumb, I should have one hot spare per encosure right? Or I should have one hot spare per bus? I'm wondering why when I created two hot spares for 55 disks ( I already subtracted the 5 for the head) that it sets both hot spares down on Enclosure 2, Bus 0. I have 4 DAE's stacked, 0 thru 3. Shouldnt I have to assign a hot spare to each bus? Why are both hot spares on the same enclosure?
Glen, i think it is the other way around... I like to place my HS in teh 200 RG range and also give it a lun ID of 200's but with the auto creation thing, i first make the RG, then delete the lun and recreate it with a lun# of my chosing.
AgeeTech
You dont need to have a HS for every shelf. THe best practice is to have 1 hotspare for every 30 drives of a certain type. The hotsparea re in that enclosure becuase that is where you selected the disk. It could have been becuase those were the only free disks to select from.
Lets say all your 55 drives are the same (300G FC) my recommendation would be to have 2 300G FC hotspares since you are close to the 60 mark.
I personally try to place the hot spares on different busses while following the 30 drive rule. Other factors that could increase that ratio are drive size and drive type. The way the Hot Spare selection works is that when a drive fails the bus is searched for an available Hot Spare (HS) that meets the HS selection criteria. If there is not a HS, it is already being used or if the criteria is not meant then the search will move on to the next bus and continue until a suitable HS is located (assuming there is one). This all happens very quickly, like in milliseconds. So, if your HSs are all on the same bus it is okay.
The 30 drive rule is a best practice guideline based on research. I have had customers that have environments, like development where they are okay with assuming a higher risk and don't follow that rule. I have had customers that want a HS per tray because it is a HA critical environment. I have also had customers that keep their equipment in an environmental situation that is less than suitable and fosters a higher than usual drive failure rate. One such customer printed newspapers and the data center although in a separate room was located in the printing plant and exposed to chemicals, heat and unclean air.
Christoffer3
27 Posts
1443
0
Posted October 14th, 2010 06:00
Have a look at this; http://www.scribd.com/doc/38738816/LUN-Raid-Group-Storage-Group-Navisphere
Basically you just create a new raidgroup on the new enclosure, set the type to hot spare.
After this you bind it and it should be listed under 'LUN Folders -> Unowned LUN's' as a hot spare.
Usually the last drive in the enclosure is used (ie.14).