My RAID (dis)array

[yert33]

My client has an intermittent "network resiliency" problem with a network share. A stable connection with this share cannot be guaranteed from any workstation on the LAN.

For example: Open an Excel worksheet from the share. Let it sit idle on the workstation for a couple of minutes, then try to save it. Error: Cannot save.

There are many other examples I can give including the nightly backup not "seeing" the drive during the backup routine (backup runs on another server), etc.

We have replaced NICs, switches, even swapped out everything but the 2 P3 processors on the server in question and still no good.

The share is on an IBM 4500R, with a ServeRAID-4M card controlling 3 8GB internal drives and 10 36GB drives in an EXP200 external expansion chassis. Sorry to say this box is still NT 4.0. I have the latest RAID BIOS, firmware and device drivers (all 7.00.14) installed.

Before I ask them to buy another RAID system what can I check?

Should I be running both internal and external arrays off the same ServeRAID card?

Where can I check to see if these components are all compatible with each other under NT?

What replacement RAID would you suggest?

I appreciate you taking time to read this.

Thanks.
Trey

 

[catorze]

A1: Tell them to buy a current Operating System Seriously, when the share is unavailable, what does the disk configuration look like from the host server? Anything unusual in ServeRAID Manager?

A2: As long as you are not running internal and external off the same channel, you should be fine.

A3: ServerProven Compatibility should be your guide. You can get there from here:

http://www.pc.ibm.com/us/compat/

A4: Your RAID solution is not the limiting factor (most likely your OS is). What is your current RAID configuration. I read above 3x9.1GB HDDs and 10x36.4GB HDDs. Is the hot-swap backplane (internal) connected to a different channel of the 4M than the enclosure (external)? Are the HDDs in two separate Arrays?