Hardware :: Invalid Partition Table For Swap Partition That Moved With Gparted?
Feb 8, 2010
I was reading another thread about someone with a bad partition table and I decided to join this forum. I'm not going to take any drastic actions with the partition (/dev/sda3) in question. I am going to wait for instructions on what to do first. I am not very good with Linux and need some hand holding. System: DELL 4550 Dual-Booted with XP and Ubuntu. Works OK, just no swap. Well, here's what I did: I deleted a partition for Windows XP Pro because it was a trial, and it ran out. I then decided to slide the swap partition for the Ubuntu Linux that I dual-boot into over. (If this was successful, I was going to try expanding the root partition to take up the unused space.) I used Gparted on a CD to do this, as I figured it was safe to do.
I now cannot mount the swap space at bootup (and have to go into a backup version of the OS), although I can use Gparted in Linux to execute the "swapon" command, and it appears that it worked because I now see "swapoff" as an option on the context menu. (I actually don't even need a swap partition, except to hibernate.) If I highlight the swap partition and click on "Drive" on Gparted's menu bar and select "Create Partition Table", it will erase all data on /dev/sda, so how do I fix the bad partition table non-destructively?
I was installing opensuse 11.2 in parallel with windows xp.but during installation suddenly power has gone and after that opensuse is giving me the error message corrupt partition.i am also not able to login in xp. so I decide to reinstall windows, I got the error saying "invalid partition table" after the first restart of windows xp installation.
I tried to use windows system recovery console and committing fixmbr and fixboot commands, but didn't work. i have 2 window partition(1 for windows and 1 for data).i do,nt want to format 2,nd partition.
How can I installed windows?My plan was first to install windows xp, then opensuse again.
I'm trying to clone a Linux install to a different laptop. It's made a little complicated by two facts:
1) The 'new' laptop I'm trying to copy my Linux installation to is actually older and has a smaller hard drive then the computer I'm copying from
2) The computer I'm copying from has both a windows and Linux installation; I only care about the Linux partition.
I figured I would copy only the Linux partition from my primary computer to the laptop, sense the laptop doesn't have a large enough hard drive to copy everything. So I used the DD commands to copy SDA3 (main Linux partition) from my main computer to SDA2 of my laptop. When I came back a few hours later I was surprise to find my laptop trying to reboot itself (I never turned it off). It would keep starting to reboot, failing, and restarting itself. Not too surprising sense its boot partition wasn't changed so it's trying to boot into centos when I copied a redhat partition to it.
The problem is that when I used a redhat boot disk the rescue mode was unable to find a Linux partition to mount. /dev/sda2 exists, but trying to mount it gets the complaint "No such file or directory". "fdisk -l" lists sda1 (the boot sector) and sda2. Sda2 is the correct size and reports Linux LVM for its system. But "fdisk -l /dev/sda2" gives the error message "Disk /dev/sda2 doesn't contain a valid partition table" Did I not clone the drive correctly, or was an error caused due to the boot sector not being copied yet (the laptops boot sector is smaller then my old computers, so I can't copy from old computer to laptop)? Can I salvage the laptops partition table somehow, or do I have to repeat the cloning process? And if I do have to re-clone my computer can anyone tell me what I did wrong the first time so it works this time? I don't care if I copy just the Linux partition or both windows and Linux. Even though my main computer has a larger hard drive I'm only using about half of its available space so it should be possible to copy both partitions if I could ignore the unused sections of the harddrive.
Edit: I used DD to copy a tiny part of the Linux partition from my laptop so I could look at it. Most of it is illegible binary of course, but I scrolled through till I found some text right near the beginning:
Code:
VolGroup00 { id="F2MWxh-....-BidcLe" seqno = 1
[code]....
So it seems that the DD command did copy everything over to the laptop, which is good to know. I noticed that it says device="/dev/sda3" right in the middle of the code I just posted. The Linux section of my original computer was SDA3 but I copied it to partition SDA2 of my laptop. So is the problem because the boot partition is for the wrong device? I don't suppose if I modified that one line to say SDA2 it would be able to load correctly? (Not that I know how I would modify the line, short of using the DD command again).
I installed Ubuntu as shown in the wiki and when I went to restart it gave me a lovely blinking cursor and nothing else. So I held down option, loaded into osx, reinstalled rEFIt and got my menu on startup. Unfortunately, the partition sync tool doesn't seam to be working, it gives me an error: Status: MBR partition table is invalid, partitions overlap. Error: Not Found returned from gptsync.efi
After I formated my windows partition using GParted it became Unallocated and moved under Extended partitions. I can't create the unallocated partition as primary one, or drag it out of the extended ones. I tried GParted live CD also, but nothing worked.
I upgraded Windows to 7 from Vista. My Acer laptop had a recovery partition with Vista on it. I don't know, what I was thinking, after the update I deleted the recovery partition. Then got in to problem that Partition Table is deleted. Recovered the partition and partition table with LiveUSB and gpart.
So laptop was working again in about 30mins. Now I see the following issue. Laptop boots and works fine both in Ubuntu (default OS) and Win7. In Disk Utility the partitions are shown as in attachment. In Gparted the disk is not recognized as partition table is not recognized (so I guess) Output of fdisk is here for ref;
Code: home@home-laptop:~$ sudo fdisk -l [sudo] password for home: Warning: ignoring extra data in partition table 7 Warning: ignoring extra data in partition table 7 Warning: ignoring extra data in partition table 7
While installing Ubuntu I have created 50MB boot partition and was sure is enogugh, but now with growing initrd and kernel size, an autoupdates which were always killed by no sufficient disk space I decided to increase its size. I had following partitions - in disk order:
Is the problem still in bad begining of the partition saved or data are already broken ? If the first is there any mount option to really not touch filesystem ? Because I've read even if we mount readonly there are some options saved to disk (as last mount time etc) and I'm aware it could destroy entrie filesystem.
Has the ext3 some partition header as NTFS do? Or is there a way to locate real start of partition? because testdisk and gdisk shows only this one starting sector for sda6. When I moved the sda6 partition there was space between begin of extended partition and begin of sda6, now there is no space, so that's the main reason of suspecting partition begining is in fact over 100MB further.Current partition schema in attached PNG. Also the swap partition was deleted, I don't know when, now there is 4GB hole.
I have a 6TB external eSata bay (Lacie BigQuadra). I made a GPT table with only one big ext4 partition. All was ok. I resized the ext4 partition and I created a 1TB NTFS partition. I can use it on Kubuntu but Windows 7 tell me the partition is not formated. When I go back to Kubuntu, parted tell me that the secondary GPT table is not at the end of the disk and tell me it's probably an other OS that thinks the disk is smaller that its real size. It seems Windows 7 thinks the disk size is 2 TB (and modify automaticaly the GPT table and create a secondary GPT table on the middle of the disk).
What can I do to make my NTFS partition visible in Windows 7? What can I do to prevent Windows 7 to move the secondary partition table on the middle of the disk and to modify the primary GPT table ?
This machine has UBUNTU & wINDOWS XP. I'm currently logged into UBUNTU. I was just checking the features of GParted and accidentally clicked Device > Create Partition Table. A default MS-DOS partition table is created. Now if I re-start the Gparted there is nothing. Its showing entire disk as UNALLOCATED space.
Lucky thing is All the drives (C:, D:, E:) are currently mounted and I'm in UBUNTU. I guess its possible to re-create the partition table using current status. how to do this. This is a lab computer. If its not recoverable. I'm completely screwed!
GParted tells me my hard drive is not partitioned and has an unrecognised partition table, but I know it has because i'm using it now to write this on here, and fdisk shows the following:
Quote:
Anyone know of anyreason GParted may not be working or can offer an alternative to create a partition?
I have installed HP g6 notebook from live-cd with 10.04 LTS across multiple partitions, only to find that the partition table is not setup correctly. I place the following mount points on separate partitions:
When I start gparted, to see if I can add extra harddisk where all my data is, I get the next error:
Code: [root# gparted libparted : 1.9.0
The kernel was unable to re-read the partition table on /dev/sda (Device or resource busy). This means Linux won't know anything about the modifications you made until you reboot. You should reboot your computer before doing anything with /dev/sda. I can start fedora 12 x64 amd whitout any problem. My partition layout on a single disk (80 gb sata maxtor) for the system is: Code: / /home swap
When I try to mount one of the extra disks (samsung spinpoint 1 tb, I have 4) I get to next error: Code: root# mount /dev/sdb1 /media/sdb1 mount: you must specify the filesystem type
If I add the option with the filesystem I get: Code: root# mount -t ext3 /dev/sdb1 /media/sdb1 mount: special device /dev/sdb1 does not exist
In gparted, when I select just a flag option, and deselect, press ok, the disk is forced to be reread by the program. After that I can also mount it. When I use any of the programs linked in this post, I am able to mount the disks whitout any problem. Except after a reboot, I get the same problem. SO whit each reboot, or poweroff, I need to do a re-read of the disks. The system is fedora 12 x64 for amd, installed from a live cd, the rest whit yums groupinstall, the latest kernel.
I have just installed the newest Debian Stable 7.8 release on my new notebook. Before installation I had to free some disk space from the preinstalled Windows7 with ntfsresize and fdisk. In addition to the existing three primary partitions I created an extended one with three logical partitions for /(root) /home and swap, see the output of 'fdisk -lu'
For some reason I put a bootable flag on sda7, and the only small concern during installation was that some BIOS systems might not work with boot-flag no logical drives. Now, every time I boot I get this "Invalid partition table!' message which I must 'enter" away before I get to the GRUB menu.
I installed openSUSE on a XP runnng computer and every thing is OK.When I tried to Install a new XP, I got this error:Code:Invalid Partition Table I fixed it by reinstalling GRUB using Rescue System.I tried fix mbr on XP Repair System and I got that error again.Now I want to keep XP and remove openSUSE. But computer only boots by GRUB and XP's boot loader is unable to boot and shows that error
Is there a difference between using GPT partition table when formating hard drives and MS-DOS partition table? What are the advantages/disadvantages of using either?
I usually repartition a disk by backing up, deleting the partitions, formatting them and repartition. I just did a 200 gig backup (so i am safe) and i want to join 2 (ext3) partition together, sdb1 (data4) and sdb5 (data5) into one big partition. Is there a way to do it without scraping the data in sdb5 (data5). It would save me from rewriting the data back to that new partition (200 gig is time consuming).
I have tried to automate the configuration of a usb drive with not much success.
The problem that I have is that I have a large amount of usb drives that have a partition table of type "loop" and I need to change them to "msdos". The size of the drives vary and I need to use FAT32 or FAT16 file system.
I've tried various partitioning commands and gui applications but cant find one that I can give a one line command to to set the partition table, maximum partition size and file system.
Everything is installed and setup on my system, but when I setup my partitions I chose my Windows partition to be bootable. Can I just use cfdisk to toggle the bootable flag so my linux partition is bootable and rewrite the partition table?
I want to change my sda2 partition to ntfs type. i have installed GParted but it is returning a strange type of error. Here is the error dump file...
[Code]...
WARNING: the kernel failed to re-read the partition table on /dev/sda (Device or resource busy). As a result, it may not reflect all of your changes until after reboot. WARNING: the kernel failed to re-read the partition table on /dev/sda (Device or resource busy). As a result, it may not reflect all of your changes until after reboot.
Is there a program that will reread the partition table and update the kernel even if one of the unmodified partitions is mounted? I installed my system on one partition, then I added another with free space. Now I want to format the second partition, but the kernel doesn't know about it yet. I tried sfdisk -R /dev/sda, but it refuses while the root partition is mounted. Is there anyway I can avoid rebooting?
USB flash disk partition disappeared as well as partition table I'm not sure about the cause
Code:
root@u# less /var/log/syslog usb 5-1: New USB device found, idVendor=058f, idProduct=1234 usb 5-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[code]....
Where did the partition table go? The device had one ext3 partition something around 4GB(size of USB storage device). I need to restore few files from this device.
What I believe has happened is that I've corrupted the partition table. Essentially one of my partitions' ending point exceeds the maximum number of cylinders/sectors on my drive.
Essentially I have the same problem as on the thread @ [url] but do not know how to fix this and am afraid to reformat/partition based on sectors without really knowing what I'm doing here.
[code]...
When I try to look at SDA in GParted everything shows up as unallocated (though it's obviously not) and it says
In my efforts to resize my BTRFS Partition, I accidentally unmarked my BTRFS partition as being BTRFS, and can't mark it back as I can't find the numeric ID for BTRFS and how to apply.
I wanted to delete the Snow Leopard partition and format the Swap Disk partition to something else. exFat was causing major file size bloat on small files. QT sdk bloated to like 11 gigs or something ridiculous like that. Anyways, I loaded up an Ubuntu 10.04 LTS live cd and gparted then deleted the Snow Leopard partition. Gparted said "Mission Accomplished" and tried to rescan the drive, but never found it. At this point I restarted the computer, a dell laptop, which didn't boot with an unable to find a bootable device error. The ubuntu live cd doesn't see the drive anymore. gparted scans for drives indefinitely and fdisk -l has no output.
I am having issues with Grub 2 after installing Debian 7.8.0.The computer is a HP Pavilion 500-307nb. I made the original harddrive /dev/sdb and inserted a Samsung Evo 840 as /dev/sda. From the original hard drive (/dev/sdb), I wiped the windows partition, but left all other partitions unchanged (in case I would ever want to recover the desktop to its original state). I replaced the wiped windows partition with a swap partition and an LVM partition.These are my hard drive partitions:
/dev/sda (Samsung Evo 840)
Number Start End Size File system Name Flags 1 1049kB 3146kB 2097kB primary bios_grub 2 3146kB 944MB 941MB ext4 boot 3 944MB 94.4GB 93.4GB host lvm 4 94.4GB 1000GB 906GB guests lvm
[code]....
The partition /dev/sda3 has 2 logical volumes with filesystem ext4 that I mount to / and /home.The partition /dev/sda2 is mounted to /boot..When I install like this, Debian installs fine, however Grub2 is not installed correctly.Debian installs grub-pc which seems not able to boot the gpt partition. So I boot the Debian CD in rescue mode and execute:
mount /dev/sda2 /boot aptitude purge grub-pc aptitude -y install grub-efi
After rebooting, I come in the grub rescue shell, which says: error: no such device: 986f2176--4a4b-4222-83b9-8636a034b3c7.
When I then enter in the grub rescue shell: set boot=(hd0,gpt2) set prefix=(hd0,gpt2)/grub insmod normal normal
Grub and Debian start up correctly.why can Grub not start up automatically correctly? Where does the UUID 986f2176--4a4b-4222-83b9-8636a034b3c7 come from? I have reinstalled Grub several times, I have reinstall Debian several times, I have even wiped all partitions from /dev/sda and recreated a new gpt table with parted and manually set the partitions in parted. Still on each reinstallation, Grub fails because it cannot find exactly the same UUID. Since this UUID is always the same, it must be stored somewhere, but it cannot be the partitions, I have wiped them and the partition table several times.
I did though a firmware update of the Samsung Evo 840 before reinstallation, could this be a cause?Also the problem is not in grub.cfg. Grub starts correctly if I enter the commands above in the grub rescue screen and the UUID value does not appear there.
I wonder whether to place swap partition on LVM or on standard fdisk partition which will not be in LVM.What is better and more often used on production ?
I have 3 Ubuntu installations & a PCLINUXOS, plus Windows XP installed on one hard disk. I still can boot to each one of them and can mount each one using Ubuntu.
The problem "may" have occurred when I reduced the size of some linux partitions using gparted. I still have plenty of space in each of those partitions.
When I started gparted all of the HD was unallocated. I did that from each ubuntu installation and the PCLINUX installation, plus LIVECDs. All indicated the space was unallocated.
When I did an fdisk -l from a Puppy Linux LiveCD I got a normal start and ends of each partition.
When I tried it from Ubuntu installation or live cd, I received the following types of responses:
Code: ubuntu@ubuntu:~$ sudo fdisk -l /dev/sda5
Disk /dev/sda5: 28.5 GB, 28566397440 bytes 255 heads, 63 sectors/track, 3473 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Disk identifier: 0x00000000 Device Boot Start End Blocks Id System ubuntu@ubuntu:~$ sudo fdisk -u /dev/sda5
The number of cylinders for this disk is set to 3473.There is nothing wrong with that, but this is larger than 1024,and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO) 2) booting and partitioning software from other OSs (e.g., DOS FDISK, OS/2 FDISK) Plus the Windows partition seems to go over its limits.
Since all of my OS installations are still working, I don't know how critical this is. From reading another post, I understand this might be able to be fixed by making some changes in fstab.
