ServeRaid BIOS 4.50.05 won't upgrade

[Jman251]

Does anyone know a way to truly force a bios update on a ServeRaid card? I've tried booting from a floppy and hitting Ctrl+F to force the update, but it still fails. The firmware will update to the very latest, but the BIOS will not budge past 4.50.05 - I even tried downgrading to 3.50, but no dice. This is maddening. There is no array set up, no logical drives, I'm totally rebuilding from scratch, so I don't care if I have to use 'destructive' techniques. IBM's site is very little help; all the links besides 3.50 redirect to the latest 7.10. I read something about problems with going up too many versions at once, so I tried 5.10 as well - no luck. I also burned the latest iso of ServeRaid Support CD for Linux, booted to that, and tried updates. As usual, firmware updates to 6.10.24, BIOS update fails.......
Netfinity 5600
86643-3AY (OEM from Chrysler, think it's the same as 3BY, but not dual proc)
ServeRaid 3L card
Server BIOS: 1.16
ServeRaid BIOS: 4.50.05
ServeRaid Firmware: 6.10.24

 

[catorze]

It's not the server, it's the card that determines the version of code that it supports. When you say the BIOS update fails, are you actually getting an indication of a failure, or are you assuming it failed because it wasn't updated? From the flash GUI, you should see the installed version and the "current" version. As you've already seen, the 3L only flashes up to v6.10.24.

 

[Jman251]

I understand it's the card that matters- I gave the server bios for informational purposes. When I go into the ServeRaid update program, there are two components; the firmware updates to 6.10.24 just fine. The program indicates it will update the ServeRaid BIOS, tries, and fails. It actually says Failed. I saved the error file to a disk, and the error code it gave was 0xb00.

 

[Mammoth]

I think I spoke to you on the phone about this. I know this may sound crazy but try the update multiple times back to back. I have certainly seen updates of ServeRAID cards fail multiple times and after multiple retries it takes. I don't have an answer as to why but I can certainly replicate it in a lab.

 

[Jman251]

Tried the multiple updates method - no luck. Does anyone have access to the 4.70 release, or 4.84? I have a feeling that if I step it more gradually, it may work.