Thanks - I need to get out of having a "representative" disk because I'll be working with many, many applications and disks. I guess I'll have to settle for working out how to extract the information I need from 'sfdisk -lm '
Thanks again, David On Mon, 2009-03-09 at 23:11 -0700, Jordan Share wrote: > David Cottrill wrote: > > See below - at the beginning it gets upset about the formatting and it > > never seems to recover. > > I'm not good enough at partitioning to be able to explain it properly. > > fdisk has no problem and cfdisk has only a little grumble but still > > works ok. > > > > If I follow through with sfdisk it prompts me to zero the first 512b of > > any FAT partition (the boot record). I'm not sure of how to ressurect > > the boot sector afterwards but it certainly doesn't boot whether I zero > > it as instructed or just ignore the warning. > > > > David > > > > > > Debian:/# sfdisk /dev/sda > > Checking that no-one is using this disk right now ... > > OK > > > > Disk /dev/sda: 1009 cylinders, 16 heads, 62 sectors/track > > Old situation: > > Warning: The partition table looks like it was made > > for C/H/S=*/188/22 (instead of 1009/16/62). > > For this listing I'll assume that geometry. > > Units = cylinders of 2117632 bytes, blocks of 1024 bytes, counting from > > 0 > > > > Device Boot Start End #cyls #blocks Id System > > /dev/sda1 * 0+ 146- 147- 303103+ 83 Linux > > end: (c,h,s) expected (146,106,20) found (37,187,22) > > /dev/sda2 0 - 0 0 0 Empty > > /dev/sda3 0 - 0 0 0 Empty > > /dev/sda4 0 - 0 0 0 Empty > > Input in the following format; absent fields get a default value. > > <start> <size> <type [E,S,L,X,hex]> <bootable [-,*]> <c,h,s> <c,h,s> > > Usually you only need to specify <start> and <size> (and perhaps > > <type>). > > > > /dev/sda1 :; > > /dev/sda1 0+ 1008 1009- 500463+ 83 Linux > > /dev/sda2 :,,L > > /dev/sda2 0 - 0 0 83 Linux > > partition ends on cylinder 1023, beyond the end of the disk > > /dev/sda3 :; > > /dev/sda3 0 - 0 0 0 Empty > > /dev/sda4 :; > > /dev/sda4 0 - 0 0 0 Empty > > New situation: > > Units = cylinders of 507904 bytes, blocks of 1024 bytes, counting from 0 > > > > Device Boot Start End #cyls #blocks Id System > > /dev/sda1 0+ 1008 1009- 500463+ 83 Linux > > /dev/sda2 0 - 0 0 83 Linux > > /dev/sda3 0 - 0 0 0 Empty > > /dev/sda4 0 - 0 0 0 Empty > > Warning: partition 2 has size 0 but is not marked Empty > > Warning: no primary partition is marked bootable (active) > > This does not matter for LILO, but the DOS MBR will not boot this disk. > > end of partition 2 has impossible value for cylinders: 1023 (should be > > in 0-1008) > > I don't like this - probably you should answer No > > Do you want to write this to disk? [ynq] > > What are you trying to do, exactly? It looks like you accept the first > partition as is, and then don't specify a start for sda2, although you > do specify the type. I think this is the source of your difficulties. > (Here is my pastie where I do your steps and get similar errors: > http://gist.github.com/76769 ) > > For sfdisk, I usually: > * partition once by hand (on a representative disk) > * sfdisk -d /dev/sda > ~/partitions.txt > * edit ~/partitions.txt, removing the "size = xxxx" section from the > partition I want to expand (assuming it is the last partition on the disk). > * cat ~/partitions.txt | sfdisk /dev/sda > * or whatever, basically dump the edited file back into sfdisk on > the new box. > > A sample file that I use all the time is: > # partition table of /dev/sda > unit: sectors > > /dev/sda1 : start= 63, size= 995967, Id=83 > /dev/sda2 : start= 996030, Id=8e > /dev/sda3 : start= 0, size= 0, Id= 0 > /dev/sda4 : start= 0, size= 0, Id= 0 > > (Note that the "size= xxxxx" bit has been removed from sda2's > definition, but a start is defined.) > > In this case, I am making a /boot partition at the start of the disk > (perhaps overly generously allocated), and then using the rest of the > disk as an LVM volume. You would change the Id=xx to whatever is > appropriate for you. > > Jordan -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
