| 1 |
Robert G. Siebeck wrote: |
| 2 |
|
| 3 |
>The only thing which seems strange is an error thrown by fdisk: |
| 4 |
> |
| 5 |
> |
| 6 |
># fdisk /dev/hda |
| 7 |
> |
| 8 |
>The number of cylinders for this disk is set to 116280. |
| 9 |
>There is nothing wrong with that, but this is larger than 1024, |
| 10 |
>and could in certain setups cause problems with: |
| 11 |
>1) software that runs at boot time (e.g., old versions of LILO) |
| 12 |
>2) booting and partitioning software from other OSs |
| 13 |
> (e.g., DOS FDISK, OS/2 FDISK) |
| 14 |
> |
| 15 |
>Command (m for help): p |
| 16 |
> |
| 17 |
>Disk /dev/hda: 60.0 GB, 60011642880 bytes |
| 18 |
>16 heads, 63 sectors/track, 116280 cylinders |
| 19 |
>Units = cylinders of 1008 * 512 = 516096 bytes |
| 20 |
> |
| 21 |
> Device Boot Start End Blocks Id System |
| 22 |
>/dev/hda1 1 130 65488+ 83 Linux |
| 23 |
>Partition 1 does not end on cylinder boundary. |
| 24 |
>/dev/hda2 131 79717 40111848 5 Extended |
| 25 |
>Partition 2 does not end on cylinder boundary. |
| 26 |
>/dev/hda3 * 79718 95960 8186314+ a5 FreeBSD |
| 27 |
>Partition 3 does not end on cylinder boundary. |
| 28 |
>/dev/hda4 95964 116279 10239264 c W95 FAT32 (LBA) |
| 29 |
>Partition 4 does not end on cylinder boundary. |
| 30 |
>/dev/hda5 2212 79717 39063024 83 Linux |
| 31 |
>/dev/hda6 131 2211 1048792+ 82 Linux swap / Solaris |
| 32 |
> |
| 33 |
> |
| 34 |
> |
| 35 |
|
| 36 |
Technically, there is no "error" here. The first message is completely |
| 37 |
normal for large disks. The "does not end on..." messages mean that the |
| 38 |
disk was originally parititoned in "LBA" mode (which represents the disk |
| 39 |
as having 255 heads, and about 1/4 as many cylinders), but for some |
| 40 |
reason, Linux is seeing the disk in non-LBA mode. (side note: I really |
| 41 |
don't understand why BIOSs don't provide an option that says "Force the |
| 42 |
damn disk to LBA!!!") |
| 43 |
|
| 44 |
Again, nothing actually wrong with this, but if you were to delete and |
| 45 |
recreate any partitions with fdisk, it would insist on creatin the |
| 46 |
partitions on a cylinder boundary, meaning they start/end sectors would |
| 47 |
*not* be the same. So, if you see such messages, either be prepared to |
| 48 |
delete the entire table and start over, or don't change anything with fdisk! |
| 49 |
|
| 50 |
>cfdisk doesn't even start but stops with the following message: |
| 51 |
> |
| 52 |
>FATAL ERROR: Bad logical partition 6: enlarged logical partitions overlap |
| 53 |
>Press any key to exit cfdisk |
| 54 |
> |
| 55 |
> |
| 56 |
> |
| 57 |
|
| 58 |
Well, this is most likely because it is extremely rare to have logical |
| 59 |
volumes out of order in the extended volume. So the sanity checks in |
| 60 |
cfdisk might be fairly simple, like making sure the start cylinder of |
| 61 |
hda6 comes after the end cylinder of hda5. Sounds like a bug in cfdisk |
| 62 |
if you ask me.... |
| 63 |
|
| 64 |
I think the only way to really get rid of the error messages would be to |
| 65 |
do a full backup, then blow everything away, and then restore it. |
| 66 |
Probably not worth the effort though... |
| 67 |
|
| 68 |
-Richard |
| 69 |
|
| 70 |
-- |
| 71 |
gentoo-user@g.o mailing list |
Archives