Hardfile problem - created from amiga HD in PC

[chiark]

Weird one this. I plugged my 2GB drive into my PC and created a 2GB hardfile from it using WinUAE.

Now when I come to use it, the three partitions that are on there all report as uninitialised. If I try to boot from it, I get "not a dos disk in device QDH0" - which is indeed the name of the bootable disk on there.

The system that the disk came from was a 3.9 system.

My 4000 has recently died, and to say I'm a touch nervous now is understating how I feel...

To add the disk, I'm just going to "hard drives", then "add Hardfile", then entering the path, and leaving it at that with the UAE controller.

I've tried this in Amiga Forever 2006 version of WinUAE, and the latest public beta 8 - same result in both.

HDToolBox reports the drives as being there and using standard international ffs as the file system.

Any clues?

[Toni Wilen]

winuaelog.txt (if enabled in misc panel) may have some hints.

Perhaps you have huge buffers but not enough memory. Or perhaps too old Kickstart (try 3.1) or perhaps RDB installed filesystem is 68040+ (is there any?) version and selected CPU is 68020? Or perhaps FS requires fast ram but you don't have any.

Lots of possibilies..

[chiark]

hmm, thanks Toni. I've tried 3.1 Kickstart, and tried setting the filesystem to be the fastfilesystem off the OS3.9 install CD just in case... I shall have another play!

[chiark]

OK, I'm giving it the Cloanto 3.1 ROM, a 68040, 64MB z3 RAM, 2MB chip, 4MB Fast and it can't see the contents of the disk. It's recognising the partitions perfectly, but not reading the contents.

30-375 [8 226x040]: F:\2GAmiDrive.hdf:
30-375 [8 226x040]: RDB: RDSK detected at 0, FSHD=4, C=2500 S=1680 H=1
30-375 [8 226x040]: Mounting uaehf.device 0 (0) (size=2151088128):
30-375 [8 226x040]: RDB: 'QDH0' dostype=444F5303. PartBlock=1
30-375 [8 226x040]: BlockSize: 512, Surfaces: 1, SectorsPerBlock 1
30-375 [8 226x040]: SectorsPerTrack: 1680, Reserved: 2, LowCyl 2, HighCyl 612
30-375 [8 226x040]: Buffers: 30, BufMemType: 00000000, MaxTransfer: 00ffffff, BootPri: 11
30-375 [8 226x040]: RDB: RDB filesystem 444F5303 version 40.1
30-375 [8 226x040]: RDB: 444F5303 in FileSystem.resouce version 40.1
30-375 [8 226x040]: RDB: FS in FileSystem.resource is newer or same, ignoring RDB filesystem
30-375 [8 219x048]: F:\2GAmiDrive.hdf:
30-375 [8 219x048]: Mounting uaehf.device 0 (1) (size=2151088128):
30-375 [8 219x048]: RDB: 'QDH1' dostype=444F5303. PartBlock=2
30-375 [8 219x048]: BlockSize: 512, Surfaces: 1, SectorsPerBlock 1
30-375 [8 219x048]: SectorsPerTrack: 1680, Reserved: 2, LowCyl 613, HighCyl 1861
30-375 [8 219x048]: Buffers: 30, BufMemType: 00000000, MaxTransfer: 00ffffff, BootPri: 0
30-375 [8 219x048]: RDB: RDB filesystem 444F5303 version 40.1
30-375 [8 219x048]: RDB: 444F5303 in FileSystem.resouce version 40.1
30-375 [8 219x048]: RDB: FS in FileSystem.resource is newer or same, ignoring RDB filesystem
30-375 [8 224x056]: F:\2GAmiDrive.hdf:
30-375 [8 224x056]: Mounting uaehf.device 0 (2) (size=2151088128):
30-375 [8 224x056]: RDB: 'qdh2' dostype=444F5303. PartBlock=3
30-375 [8 224x056]: BlockSize: 512, Surfaces: 1, SectorsPerBlock 1
30-375 [8 224x056]: SectorsPerTrack: 1680, Reserved: 2, LowCyl 1862, HighCyl 2499
30-375 [8 224x056]: Buffers: 30, BufMemType: 00000000, MaxTransfer: 00ffffff, BootPri: 0
30-375 [8 224x056]: RDB: RDB filesystem 444F5303 version 40.1
30-375 [8 224x056]: RDB: 444F5303 in FileSystem.resouce version 40.1
30-375 [8 224x056]: RDB: FS in FileSystem.resource is newer or same, ignoring RDB filesystem
30-375 [8 225x064]: F:\2GAmiDrive.hdf:
30-375 [8 223x072]: P96 RESINFO: 0780DB38-0780ED98 (63,4704)
30-437 [11 226x101]: Thread 'hardfile' started (1456)

I'll try with the physical disk - might that make a difference?

[chiark]

And the physical disk works without a hiccough in the same config. That's really rather strange! I'm just making another copy... You have no idea how relieved I am that the disk isn't completely dead too

[chiark]

Well, the original file copied must have been corrupted. I've run an MD5 of both files, and they're different. Seems strange that it managed to identify the RDB though?

The old file was made with WinUAE 1.2, the new one with 1.4.6 beta 8 which might so some way to explaining the issue

[Toni Wilen]

There was compatibility problem with hardfile image creator in 1.2 or so..

[chiark]

Aha, sorry for digging over old ground then!