DL380 G4, unable to boot off USB Floppy

[Orla]

I'm trying to boot a DL380 G4 using a boot disk and USB Floppy drive but so far have been unsuccessful. It gets past the "press ctrl + s to enter setup" screen but then seems to hang. I've tested the drive itself and that is working fine. I've also set the boot sequence to various orders i.e floppy first then usb and usb first than floppy but this doesn't make any difference.

Any responses/ideas would be greatly appreciated.

Thanks,
Orla.

[Robin_37]

Please check if you have the latest system firmware.

http://h18000.www1.hp.com/support/files/server/us/download/21813.html

[Ivajlo Yanakiev]

Are you sure that your USB is bootable ?

[Orla]

Thanks for your replies...I replaced the usb floppy drive and it is since working fine. I now have a new problem though...I'm trying to build the server using RDP. It sees the usb device as a C drive so when the OS loads the usb floppy fails to boot back to RDP as there is already a C drive i.e. the OS?? Haven't been able to get past this yet.

Any ideas?

[Ivajlo Yanakiev]

Do you need floppy for RDP ?
You boot from network (from RDP server ).

[Orla]

I had a look at the network boot, and our setup seems fine but I'm getting a PXE-E53 "No boot filename received" error. I followed some advice I got of this site to ensure all was setup ok but still getting error. We've always used the boot floppy successfully but we got a new batch of servers in without floppies, thus the need for a usb floppy.

[Ivajlo Yanakiev]

If you are getting an error message, it is most likely due to time out issues, or routing issues. You have to remember that PXE is a broadcast protocol and therefore doesn't cross subnets (unless you network gear has helper address' configured to allow PXE to cross subnets). Best to have a PXE server on each subnet though. If you're PXE server and the server doing a PXE boot are on the same subnet, and you are still getting error check the port configuration for the server doing the booting. If you have spanning tree protocol enabled as most gear does by default (Cisco especially) you need to disable it/enable port fast. This tells the port to become active immediately - thus getting rid of the PXE timeout issue. This does have ramification on your network should you cause a network loop on that interface, but that would be uncommon.

Let me know if this helps.

[Ivajlo Yanakiev]

look here :
http://forums1.itrc.hp.com/service/forums/questionanswer.do?threadId=221036

[Orla]

The two servers are on different subnets but there are helper addresses setup to allow them cross. I'll look into it and let you know how I get on. Thanks again for your help.

Orla.