I have 2 poweredge Servers on an SQL Cluster setup.  The 2 servers are a PE6400 (Tag# HF1K401) and a PE2400 (Tag# 2J1K401) and were purchased at the same time for this project.  The 2 servers are attached to a PV220S that has 3 3drive arrays.  Each array is 3 18gig HDs in a RAID5 configuration.

For some reason that has not been adequately explained to me, when we loose power or the servers have to be restarted, the first array A0 looses its setup and the PERC2/DC on the PE6400 does not recognize that container as valid, this also applies to the PE2400 PERC2/SI.  To resolve this problem we have to boot up the PE6400, go into the PERC2/DC's bios and force it rebuild the array A0.  Sometimes we have to rebuild A0, A1, and A2 as well.  Once the array(s) are rebuilt, then the PE6400 & PE2400 will recognize the first array.

We now have this process down so that when we loose power (and that happens more often then I care to admit due to an old building and overloaded UPS batteries) we can restore the cluster in under 30 minutes.  However, I do not understand why the PV220S is losing this configuraiton or if the PERC2/DC-SI are losing this information.  All controllers and Servers have their original BIOS levels, whatever those are.  We can not take these down for maintenance like I would hope to since they contain our primary database that our subscribers use (Live data feeds.) 

Thank you for any information you may be able to provide.