Hot answers tagged fdisk
12
Because your old disc was partitioned with a old utility, such as the Linux fdisk, that uselessly implemented track-alignment using the entirely fake disc geometry that you see reported, and your new disc has been or is being partitioned by a newer utility that (by default) aligns to 1MiB boundaries instead.
Further reading
Jonathan de Boyne Pollard ...
5
Setting up partitions on other partitions isn't a supported operation. If you want to subdivide your /dev/sda2 then one option is to use LVM2 (that's Arch Linux documentation but applies everywhere).
For your example of three 512Mb:
pvcreate /dev/sda2
vgcreate TEST /dev/sda2
lvcreate -L 512M -n one TEST
lvcreate -L 512M -n two TEST
lvcreate -L 512M -n ...
4
fdisk reads from stdin so you just need to feed it the appropriate commands. For example, the following clears the partition table, if there is one, and makes a new one that has a single parititon that is the; entire disk:
(echo o; echo n; echo p; echo 1; echo ; echo; echo w) | sudo fdisk
I recommend you do the task you want, recording what you type so ...
4
They tell you exactly what they say. No, you shouldn't care about that. Your cylinders are complete fictions and no actual operating system, not even MS-DOS, has actually itself cared about such alignment. This is a pointless message from fdisk, which is over three years behind other disc partitioning utilities and showing no signs of improving. Get ...
4
If it's a mounted filesystem, you can use df -hT:
$ df -hT
Filesystem Type Size Used Avail Use% Mounted on
/dev/sda1 ext4 5.5G 3.0G 2.3G 58% /
none devtmpfs 229M 252K 229M 1% /dev
none tmpfs 233M 0 233M 0% /dev/shm
none tmpfs 233M 100K 233M 1% /var/run
none tmpfs 233M 0 233M 0% ...
4
sdb is a second disk, given the results I think you have some kind of raid on your computer.
sda2 and sda5 are on the same location because sda2 is an extended partition, it is a special partition which containts the logical partitions of your system. Finally, sda5 is the first and only one logical partition.
More informations about disk partitioning : ...
4
Heh, heh, heh. This brings me back. Back, way back, in the Win3.1 days there was a package out there that was exactly what you're looking for. It was called PCTools and had a hex-editor that could directly edit the partition table (as well as the master boot record). If you had a misaligned partition for some reason, you could fix it in a way that DOS FDISK ...
3
In Linux, hard drives are listed with a three letter description like this:
Interface Type Identifier
With most disks these days, the interface is always recognized as being SCSI so the first letter is S for SCSI. The second letter D means that it's a drive. The final letter is assigned in alphabetical order from A to Z starting from the first ...
3
Clonezilla doesn't support software-based or firmware-based (aka "FakeRAID") arrays.
From what I understand your HP system has an Asus-made MB in it with an Intel ICH9R chipset. If you used that to make your RAID 0 then Clonezilla can't do anything with it.
From their About page (under "Limitations"):
Software RAID/fake RAID is not supported by ...
3
Here is your answer. DOS doesn't allow a partition to start (or end) mid-cylinder. It assumes the partition table is corrupt when it sees this, and not only won't boot off of that partition, it won't boot any partition on the disk.
Basically, unless you're dual-booting, turn DOS compatibility off. Even if you are, unless you're running an old version of ...
3
Since you are Linux savvy, try running both Testdisk and PhotoRec on the drive. Or even better, try to get a dd image of the drive and have them crawl through it. Hard drives that make noise tends to have serious problems and not last long after the noises begin.
3
Finally, I solved this problem! :-P
It turned out to be that I insmod scsi_mod.o and sd_mod.o, but the kernel had support for SCSI already.
sigh~
I also found that scsi_mod should be insmod'ed before sd_mod, and ehci_hcd.o before usb-storage.o. If not, dmesg will say usb.c: USB device not accepting new address=2 (error=-71).
3
If you've already installed linux onto /dev/sda1, you can't put LVM on top of it.
LVM is created first.
The following example creates an LVM layout, which should be created from a livecd, or rescuecd
pvcreate /dev/sda1 # initializes /dev/sda1 for LVM
vgcreate vgname /dev/sda1 # creates Volume Group with one device
lvcreate --name boot --size 1G vgname # ...
2
Maybe it will be useful to add a comment here.
For LUKS partition, it is said to delete and re-create the partition at the same place, but larger before calling cryptsetup resize.
But when you created your partition long time ago, it start at the sector 63. Using fdisk, the partition will be recreated at the wrong offset, resulting in a lost partition.
I've ...
2
fdisk is an interactive partition table editor, so the only argument it needs is the hard disk device you want to work on.
If you resized the partition /dev/hda1, then yes, sudo fdisk /dev/hda is all you need. Or if you resized /dev/sdb2, then you need to run sudo fdisk /dev/sdb.
Once you get fdisk running, you'll have to delete the partition you ...
2
I do all of my partition resizing with a Parted Magic Live CD or USB stick and it's not let me down yet - it's a single-step operation in a GUI environment and there's no need to shrink manually and then use fdisk.
Might be worth reinstalling your image and giving it a try.
PS: Kudos for actually keeping a backup before trying the operation - many don't!
2
This is true, you can not resize a mounted partition.
Download the gparted LiveCD, put it on a USB stick using unetbootin and boot from it. That way, the filesystem is not mounted and you can resize it.
Edit: with /tmp, you may also get away with rebooting the machine in single user mode or just booting to a shell. For example, if you pass init=/bin/bash ...
2
A quick format just lays out the partitioning structure for the disk, while a format will typically overwrite the whole disk.Thus a quick format is not as secure as a full format. That said, a full format is not a secure as doing a proper full disk wipe. (For a start, a full format may not always be programmed to overwrite all the bits on the partition, ...
1
If this is actually a Micro/Mini SD card in an SD card adapter, try to get a different adapter. I had 3 Kingston adapters that failed exactly like that - they became read-only.
Also try to do these operations in another computer/card reader. It is possible you have issues with your reader.
If these things don't help, I think this card is somehow defective ...
1
The Intel ICH7R and ICH10R onboard SATA controllers are not true hardware RAID controllers; they are actually software controllers. The problem you are experiencing is that the RAID functionality is only supported under Windows, not Linux.
Instead of letting the ICH7R manage the RAID volume, switch the controller to AHCI (instead of RAID) in the BIOS and ...
1
there is no need to resort to an liveCD. I mostly follow those steps:
resize the partition as needed
if the resized partition can't be reread because it's active umount it
if it can't be umounted because it's vital to the running system we must reboot
now the resized partition can be online-expanded by 'resize2fs -p /dev/xxx'
1
While I realize that the actual question was answered, a quick fix for the 63 vs 2048 first cylinder issue is something like:
sfdisk -d /dev/sdb | sfdisk --force /dev/sda
(if you're sure you've got drives that are the same size)
You can then proceed with adding the various partitions back into the RAIDs where you got complaints about the partitions not ...
1
The Gparted Live CD is a great free solution for formatting drives. Also, the Ubuntu Linux live cd boots into a desktop environment that has partitioning software installed.
However, if your disk setup isn't particularly special/complicated, and you're installing Windows, the Windows install process includes a partitioning step that allows you to create ...
1
For details, see: http://en.wikipedia.org/wiki/AT_Attachment
Linux treats even parallel ATA drives as SCSI devices, they're close enough, and the SCSI code was far cleaner than the IDE code, so the change was made from /dev/hd* to /dev/sd* a while back.
ATA only allows two drives per channel, so there's typically four two channels, each with a master and a ...
1
The default alignment in order to fully ensure compatibility with both 4K sectors and SSD cell boundaries (typically 128 or 256 or 512K) would be 1MByte into the disk.
2048 (sectors) * 512 = 1048576 (or exactly 1MByte)
So to me your disk looks to be properly aligned, both with your flash-erase cell size and the page write size.
Remember that sector ...
1
The current anecdotes suggest to have 512KB alignment (or greater integer value of log2n), therefore you want to make sure all your start sectors are evenly divisible by 1024 sectors (i.e 512Kbytes).
/dev/sda1 * 2048
2048 / 1024 = 2 OK
/dev/sda2 73244670
73244670 / 1024 = 71527.99
So this if off by 2 sectors. However since it is an ...
1
If you've never used fdisk at all, this might be a little scary. It's good to practice for a while without (w)riting. After you get into fdisk ("sudo fdisk /dev/hda" or "sudo fdisk /dev/sda" or "sudo fdisk /dev/hdb" or something similar depending on what disk you want to partition), use 'p' to print the partitions.
Look at them carefully. Maybe write ...
1
Assuming this is a straight disk image (e.g., your virtualization solution is not adding some extra metadata to things), this document (also here) offers some suggestions for accessing your partitions on /dev/sda2 (summary: use the -o offset option to losetup to create a new block device referencing the appropriate offset into sda2).
You can also use the ...
1
http://www.grc.com/spinrite.htm
I've heard so many good things about this program.
It costs a pretty penny ($ 89.00 for a lifetime, i believe, licence), and i cannot talk about it from personal experience.
However i hope the community will downvote me if it isn't a good solution to your problem - wait for other responses (and feedback) before spending ...
1
I don't think you have a problem.
If your partitioning tool is seeing the RAID array properly, the RAIDness of the disk being partitioned doesn't matter.
I have heard that the actual layout of a modern disk is nothing like what is reported in the partitioning tool - so what is seen as a cylinder boundary in the partitioning tool probably isn't one anyway. ...
Only top voted, non community-wiki answers of a minimum length are eligible
