I just inherited a dell 2900 & want to add a more HD space by adding another RAID array, leaving the existing array & Windows 2003 r2 OS in place (it was an Exchange Server).
My current, primary, insurance agency management system (AMS) production server has 265 GB Appplication Data on the "D drive" with the data consisting of a Foxpro database & many .msg, xls, ppt, pdf. doc files (Total "D" Partition on 1 array = 618GB HD w/280GB used).
My newly inherited 2900 box will become a high-avaialbility replica "secondary" server running both the AMS & CA Arcserve High Avialability Server & need the same or more HD space as the primary server so I've already picked up 2 - 1 TB HDs. I figure the simplest way for me to add space is the add a second array.
So HERE are the QUESTION(S):
do I set up a RAID 0/1 or a 0/1/5 or something else? I am interested in CONTINUITY (Arcserve HA delivers that) however arcserv will be my backup solution too (dedup during the week, tape weekends).
Since this will be my backup system I'm interested in data preservation at ALL Cost - as opposed to the AMS performance on the secondary server with will only come into play if the primary server fails. So the RAID setup should favor DATA PRESERVATION.
& with the final answer, how many total HDs do I need for each RAID solution?
And I could really use just some very basic, general overview of the RAID setup procedure such as: power off, add drives, get into PERC Bios & setup 2nd array, etc... I"ve done Adaptec SCSI RAIDS but it's been a while.
Dell's no help, their docs suck (although I did see one video w/ BIOS Screen shots) so I'll also ask you for some real life, best practice information so that I'll know what I'm doing. Thanks!
A) if on production serevr Raid 5 is configured, try configring Raid 5 on H/A (secondry)
why------
1) since H/A node will write same data as per primary better you match block-sector evevry thing same.
2) you may be using Replication software or you may require in future.
3) once you configure same raid leavl teh cache size will be same
B) i would also suggest you that select same RPM HDD for H/A node
why-------
1) will help application to replicate smoothtly
2) timeout will not occur from H/A node
C) Use One HDD as Global Spare
Why-------
1) if One HDD fails the Global Spare will take whole operation without any manual intrevention.
i am not Dell user, what i suggested based on my Personal experience, i am having 3 Sites and i did same for Primary & H/A site.