Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Mike Lewis on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

best practices for Clariion?

Status
Not open for further replies.

usao

Technical User
Mar 27, 2010
1
0
0
US
Ive been working with CX for several years, and have always done things as I was shown. Lately, questions have come up which suggest that there are perhaps alternatives which im not aware of. I thouht I would try asking in the forums to see if anyone has pointers for my specific questions below, or can point me to best practices guides which may answer some of the questions.

Q1) It's been suggested that the best way to carve-up luns is to create a set of fixed-size luns on all the raid groups, then create a meta-lun which spans multiple raid groups untill the "desired" allocation size is met. Previously, I have been simply carving a lun of the desired size from a single raid group and presentig that to the host. Any thoughts?
Q2) Does it matter when declaring hot spares which bus they are on? Our existing methods were to spread the HS disks across the bus and on multipe DAEs, but someone else I was talking with suggested putting all the HS luns up-front right next to the vault drives. So, does it matter at all?
Q3) I have typically just declared a hot-spare RaidGroup, but not put any luns on it. Recently, ive noticed a system which said HS was not configured untill I put an actual lun on ths HS disk. I thought since the enttire disk was replacing a matching disk elsewhere that the "lun" on the HS disk was irrelevant, but an config error on a newer system is making me revisit that theory.

Again, if you can offer feedback on these questions, and/or point me to best practice guides for layout out CX arrays, it would be appreciated.
 
1. It depends on the IO that the host needs for that LUN. For a lower IO requirement a LUN on a single RAID group is fine. For higher IO then you'll need a meta LUN. There's no one correct answer here, as it will depend on your environment. Personally I've got both on my system.

2. No it doesn't. I've got my hot spares on both busses simply depending on where the extra disks were in my config.

3. I just checked the hot spares on my array, and each one does have a LUN on it which is the entire size of the physical disk. They are private LUNs so I think that the array created them automatically when I created the hot spare RAID group. Granted it has been a while since I had to create a hot spare raid group. I'll be creating some in a few weeks when my new shelves arrive in a few weeks. I'll try and remember to post info here.

Denny
MVP
MCSA (2003) / MCDBA (SQL 2000)
MCTS (SQL 2005 / SQL 2005 BI / SQL 2008 DBA / SQL 2008 DBD / SQL 2008 BI / MWSS 3.0: Configuration / MOSS 2007: Configuration)
MCITP (SQL 2005 DBA / SQL 2008 DBA / SQL 2005 DBD / SQL 2008 DBD / SQL 2005 BI / SQL 2008 BI)

My Blog
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top