Reliably reconizing netraid drive failures

[bill johnson_4]

I just recovered from a 2-drive raid5 failure and need to prevent such a situation from recurring. I'm using a netserver LH3R with integrated netraid and 12 36GB scsi drives in two cages under NT Server 4.0 SP5. I would appreciate guidance in two areas.

First, is anyone familiar with a level of firmware or driver that will solve this problem: the drive status LEDs are always green, even after the controller has failed a drive. During POST, the LEDs properly flash red/amber/green. But from then on amber/red never show. I'm asking, rather than just updating to the latest of everything because I fear losing the configuration again - management is suddenly very sensitive regarding this server.

Second, I need to be notified of future failed drives. We're a small business and would prefer not to develop a complete TopTools infrastructure. I have the megaserv NT service running, but it just reports to the event log. Is there a simple tool that can alert one's workstation when a server drive fails?

[Theo Hill_1]

I don't know of an issue where an older version of firmware would cause the LED's to appear correctly, but updating everything is the way to go. This may solve the issue, and it will prevent some issues from occuring in the future. All f/w, BIOS and driver updates are made to better the functionality of the server. Most admins don't realize the importance of updating as they become available, but they know how to dial support when something isn't right. Most issues could have been prevented if only the admin would take the time to update. Make sure to update the HDD firmware as well. In most cases HDD do not fail completely, they just get knocked off line. Most times this is due to firmware. As far as being notified of HDD failure, HP does not offer anything other than using TopTools.

[Marino Meloni_1]

hi
if you doesn't see a led alert on the failed drive, probably you haven't installed a specific agent, it is the agent which usualy control the leds.
to be notified about a disk allert, you can use the "Netraid monitoring Utility"
to avoid a 2drive failure, you can use a hot-spare disk, when the first hdd fail, the raid autorebuild on the spare disk.
if two hdd fail in the same time, you can force on-line one of the two hdd, because the data in this case are usualy consistant.
bye
marino

[Alanado]

Hi Bill,

I think maybe you can check the firmware of the HDD. If you are using HDD with firmware started with D84N or P84N, it should be better if you upgrade it to D94N or P94N. I faced similar issue before, and found that is the HDD firmware causing the problem. Now after I have change the HDD with firmware H008 and F612, I never encounter the same thing again.

Maybe you can try to schedule a weekly consistency check on the RAID 5 in case both HDD fail together.

Normally when 2 disk in RAID 5 fail, it should be actually due to 1 of them. while another should be a false failure.

Regards.