View previous topic :: View next topic |
Author |
Message |
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Mon Jul 13, 2015 2:40 am Post subject: [SOLVED] Booting minimal image from USB stick |
|
|
Hi everyone,
I am trying to boot the minimal image in order to install Gentoo Linux on a laptop. I am running into a very strange problem I have never seen before and I really don't know what I can do to fix this problem.
First of all, the laptop is having a Windows 7 image which no longer boot neither. Second, it had a previous Gentoo Linux "version" installed and instead of trying to upgrade it, I decided to reinstall it because it is outdated for a long time.
So, I decided to modify the partitions of my only drive, I then booted with the USB stick and used fdisk to erase the swap partition in order to reduce its size and create a new partition in between.
After I did that, I was no longer able to boot sanely from the USB stick. WTF is going on? The first sign something went wrong is the udev events processing takes almost infinite time. I can interrupt it to proceed further, but when I finally get the prompt after a very long booting process I can use the keyboard only few seconds before it freezes. After waiting for a long time, I can see finally some characters I typed appearing on the display, then it is active for few more seconds, etc. The gpm process is started, but the mouse doesn't seem to work at all even if no error message was issued during the boot process.
I am desesperated to find what is going on.
Any ideas? Hints?
I can trying to boot normally without the usb stick, but the boot process get interrupted (looks like a kernel panic even if I cannot seen the kernel panic message), but this can be somewhat normal since the swap partition number has changed. That's why I need to boot on the usb stick to at least correct my /etc/fstab.
But, why everything seems to freeze?
I managed to issue the fdisk /dev/sda command, with many minutes delay between each small time window I can type something. So, here is what I get just after doing this, the welcome message is displayed followed by a message in red saying:
Code: | Omitting partitions after #60. They will be deleted if you save this partition table. |
What is that?
I requested a partitions table print and waiting for it...
Got the table, although I cannot see the upper part of it. It seems, for an unknown reason the newly created partitions and all subsequent partitions (a total of 5 partitions) are repeated over and over until it reaches the total number of 60 partitions including the partitions before them. There was three primary HPFS/NTFS/exFAT partitions for Windows, then the extended partition and finally in the extended partition I had the boot partition, the newly created partition after deleting the swap space partition, the newly created swap partition (not yet switched to Linux swap type), then my root partition, then a LVM partition and finally another HPFS/NTFS/exFAT partition (newly created as well I forgot to mention it).
So, the section with the following partitions is repeated over and over with the same start, end sectors and size, something like the following is shown:
Code: |
/dev/sda6 152594432 152856575 262144 128M 83 Linux
/dev/sda7 152858624 159150079 6291456 3G 83 Linux
/dev/sda8 168980480 170004479 1024000 500M 83 Linux
/dev/sda9 170006528 374806527 204800000 97.7G 8e Linux LVM
/dev/sda10 374808576 437723135 62914560 30G 7 HPFS/NTFS/exFAT
/dev/sda11 152594432 152856575 262144 128M 83 Linux
/dev/sda12 152858624 159150079 6291456 3G 83 Linux
/dev/sda13 168980480 170004479 1024000 500M 83 Linux
/dev/sda14 170006528 374806527 204800000 97.7G 8e Linux LVM
/dev/sda15 374808576 437723135 62914560 30G 7 HPFS/NTFS/exFAT
/dev/sda16 152594432 152856575 262144 128M 83 Linux
/dev/sda17 152858624 159150079 6291456 3G 83 Linux
/dev/sda18 168980480 170004479 1024000 500M 83 Linux
/dev/sda19 170006528 374806527 204800000 97.7G 8e Linux LVM
/dev/sda20 374808576 437723135 62914560 30G 7 HPFS/NTFS/exFAT
...
/dev/sda60 374808576 437723135 62914560 30G 7 HPFS/NTFS/exFAT
|
Anyone knows how I can fix this? Yes, I used the expert mode of fdisk to reordered the partitions since after deleting the swap partition, the newly added partitions where having numbers higher than the last one which was /dev/sda8 at this time. _________________ Achille Talon Hop!
Last edited by AchilleTalon on Sun Jul 26, 2015 4:01 am; edited 1 time in total |
|
Back to top |
|
|
Keruskerfuerst Advocate
Joined: 01 Feb 2006 Posts: 2289 Location: near Augsburg, Germany
|
Posted: Mon Jul 13, 2015 6:21 am Post subject: |
|
|
Do you have a second computer?
Try to update the Bios of the not working computer.
Maybe the hardware of the not working computer is defective.
HDDfirmwarevirus? |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Mon Jul 13, 2015 9:05 am Post subject: |
|
|
AchilleTalon,
Something is broken. I suspect its fdisk.
Try a clean install of System Rescue CD on your USB stick. Its Gentoo based, so you will feel quite at home.
Its unlikely to be failed hardware - there are too many checks along the way before the hardware boots.
The logical partitions in an MSDOS partition table are stored in a linked list. If you were to get loops in the list,
the list output would never end.
Do you need to salvage any data from /dev/sda5 and later ?
If so, post the first two loops ot the partition table starting at /dev/sda5. We can mount the partitons without using the partition table. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Tue Jul 14, 2015 3:09 am Post subject: |
|
|
@NeddySeagoon,
I finally managed to get the partition table as shown by the fdisk command and yes, I need to salvage some data from partitions in the extended partition. I agree with you it is unlikely to be a hardware problem. I believe the problem is actually I am getting a loop in the linked list, the command stop at partition 60. When booting with the USB stick with the minimal distro it scans /dev/sda0 to /dev/sda255 before switching to /dev/sdb1 (the USB stick drive). This number of 256 is looking like a limit once reached to give up indicating in fact the loop is infinite.
Here is the output of the fdisk -l /dev/sda command:
Code: | Warning: deleting partitions after 60
Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 1828 14680064 27 Unknown
/dev/sda2 * 1828 1841 102400 7 HPFS/NTFS
/dev/sda3 1841 9490 61440000 7 HPFS/NTFS
/dev/sda4 9490 60802 412163072 5 Extended
/dev/sda5 9490 9499 71680 83 Linux
/dev/sda6 9499 9515 131072 83 Linux
/dev/sda7 9516 9907 3145728 83 Linux
/dev/sda8 10519 10583 512000 83 Linux
/dev/sda9 10583 23331 102400000 8e Linux LVM
/dev/sda10 23331 27248 31457280 7 HPFS/NTFS
/dev/sda11 9499 9515 131072 83 Linux
/dev/sda12 9516 9907 3145728 83 Linux
/dev/sda13 10519 10583 512000 83 Linux
/dev/sda14 10583 23331 102400000 8e Linux LVM
/dev/sda15 23331 27248 31457280 7 HPFS/NTFS
/dev/sda16 9499 9515 131072 83 Linux
/dev/sda17 9516 9907 3145728 83 Linux
/dev/sda18 10519 10583 512000 83 Linux
/dev/sda19 10583 23331 102400000 8e Linux LVM
/dev/sda20 23331 27248 31457280 7 HPFS/NTFS
/dev/sda21 9499 9515 131072 83 Linux
/dev/sda22 9516 9907 3145728 83 Linux
/dev/sda23 10519 10583 512000 83 Linux
/dev/sda24 10583 23331 102400000 8e Linux LVM
/dev/sda25 23331 27248 31457280 7 HPFS/NTFS
/dev/sda26 9499 9515 131072 83 Linux
/dev/sda27 9516 9907 3145728 83 Linux
/dev/sda28 10519 10583 512000 83 Linux
/dev/sda29 10583 23331 102400000 8e Linux LVM
/dev/sda30 23331 27248 31457280 7 HPFS/NTFS
/dev/sda31 9499 9515 131072 83 Linux
/dev/sda32 9516 9907 3145728 83 Linux
/dev/sda33 10519 10583 512000 83 Linux
/dev/sda34 10583 23331 102400000 8e Linux LVM
/dev/sda35 23331 27248 31457280 7 HPFS/NTFS
/dev/sda36 9499 9515 131072 83 Linux
/dev/sda37 9516 9907 3145728 83 Linux
/dev/sda38 10519 10583 512000 83 Linux
/dev/sda39 10583 23331 102400000 8e Linux LVM
/dev/sda40 23331 27248 31457280 7 HPFS/NTFS
/dev/sda41 9499 9515 131072 83 Linux
/dev/sda42 9516 9907 3145728 83 Linux
/dev/sda43 10519 10583 512000 83 Linux
/dev/sda44 10583 23331 102400000 8e Linux LVM
/dev/sda45 23331 27248 31457280 7 HPFS/NTFS
/dev/sda46 9499 9515 131072 83 Linux
/dev/sda47 9516 9907 3145728 83 Linux
/dev/sda48 10519 10583 512000 83 Linux
/dev/sda49 10583 23331 102400000 8e Linux LVM
/dev/sda50 23331 27248 31457280 7 HPFS/NTFS
/dev/sda51 9499 9515 131072 83 Linux
/dev/sda52 9516 9907 3145728 83 Linux
/dev/sda53 10519 10583 512000 83 Linux
/dev/sda54 10583 23331 102400000 8e Linux LVM
/dev/sda55 23331 27248 31457280 7 HPFS/NTFS
/dev/sda56 9499 9515 131072 83 Linux
/dev/sda57 9516 9907 3145728 83 Linux
/dev/sda58 10519 10583 512000 83 Linux
/dev/sda59 10583 23331 102400000 8e Linux LVM
/dev/sda60 23331 27248 31457280 7 HPFS/NTFS
Partition table entries are not in disk order
|
_________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Tue Jul 14, 2015 8:51 am Post subject: |
|
|
AchilleTalon,
Heres the analysis ... Code: | Disk /dev/sda: 500.1 GB, 500107862016 bytes
255 heads, 63 sectors/track, 60801 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 1828 14680064 27 Unknown
/dev/sda2 * 1828 1841 102400 7 HPFS/NTFS
/dev/sda3 1841 9490 61440000 7 HPFS/NTFS
/dev/sda4 9490 60802 412163072 5 Extended |
Thats the easy bit ... the four partitions in the MBR. Nothing overlaps and all of the space on the drive is described there, so its probably OK. Now it gets interesting ...
Code: | Device Boot Start End Blocks Id System
/dev/sda5 9490 9499 71680 83 Linux
/dev/sda6 9499 9515 131072 83 Linux
/dev/sda7 9516 9907 3145728 83 Linux
/dev/sda8 10519 10583 512000 83 Linux
/dev/sda9 10583 23331 102400000 8e Linux LVM
/dev/sda10 23331 27248 31457280 7 HPFS/NTFS
/dev/sda11 9499 9515 131072 83 Linux |
It looks like the last block of sda5 is also the first block sda6. It can't be both, at least, not without some data loss.
sda6 and sda11 describe the same space on the drive, so that will be the start of the repeat.
If you really had a sda11 (and later), its likely the data is still there but it can no longer be accessed using the partition table.
There is a gap between sda7 and sda8, that can be correct as deleting a partition leaves the space unallocated.
sda8 and sda9 share a block.
The highest ending number is 27248 but the extended partition ends at 60802, so there were either other partitons or lots of empty space or the partition table has more damage than a circular linked list.
Run testdisk on the drive. It will try to spot partitions. Do not allow it to write the partiton table. Do not use the drive read write.
Post the testdisk results. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Tue Jul 14, 2015 1:17 pm Post subject: |
|
|
NeddySeagoon,
there is not partition after sda10. From sda11 it is a repeating pattern of partitions sda6 to sda10. Yes, the gap of unallocated blocks is expected since I didn't reallocated all the blocks after deleting the partition. The only partition I really care about is the Linux LVM partition where I have data I would like to salvage if possible. sda10 is a new empty unformatted partition.
As soon as I have extra time I will run the required test. _________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Tue Jul 14, 2015 2:07 pm Post subject: |
|
|
AchilleTalon,
At face value, sda8 and sda9 overlap by one block.
Code: | Device Boot Start End Blocks Id System
/dev/sda8 10519 10583 512000 83 Linux
/dev/sda9 10583 23331 102400000 8e Linux LVM |
That may or may not be true. The 10583 could be incorrect in both places.
Lets assume its correct for now, in which case, your LVMs may already appear in /dev/mapper/...
If so, try and mount them one at a time, read only and copy the data somewhere else. Recovery in place is not possible.
If you don't have any Logical Volumes, try and look again.
If there is stil nothing, the sda9 start of 10583 is probably incorrect. Do a testdisk scan but do not allow testdisk to write to the disk.
Have you ever extended any of your logical volumes or are they all single fragments? _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Tue Jul 21, 2015 11:55 pm Post subject: |
|
|
Just an update. I cannot boot with a USB stick or even a USB drive without being trapped into a never ending process. I guess it has something to do with the linked-list loop for the disk partitions.
What kind of boot media and what stuff on it will make my laptop to boot? Do I have to disable something about the drive to avoid the boot process trying to figure out everything on that disk? _________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Wed Jul 22, 2015 7:42 pm Post subject: |
|
|
AchilleTalon,
How did you boot to get to fdisk?
If that works so should other forms of booting.
As long as that was Linux, it will do to attempt to recover your data. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Thu Jul 23, 2015 12:03 am Post subject: |
|
|
I booted with my USB external hard drive. However, it was a very only kernel version. So, I decided to update it to get the latest version of the tools and since that time I run into this problem, almost the same with the USB stick which is having as well a recent image. _________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Thu Jul 23, 2015 4:32 pm Post subject: |
|
|
AchilleTalon,
That you were able to boot at all shows its not a problem with your broken partition table.
Try with System Rescue CD on a usb stick. Follow the instructions to install it onto the stick. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Sat Jul 25, 2015 6:06 pm Post subject: |
|
|
I finally got it and here is the output from the analysis from testdisk:
Code: | TestDisk 7.0, Data Recovery Utility, April 2015
Christophe GRENIER <grenier@cgsecurity.org>
http://www.cgsecurity.org
Disk /dev/sda - 500 GB / 465 GiB - CHS 60801 255 63
Current partition structure:
Partition Start End Size in sectors
1 P Windows RE(store) 0 32 33 1827 181 18 29360128
2 * HPFS - NTFS 1827 181 19 1840 117 5 204800
3 P HPFS - NTFS 1840 117 6 9489 98 17 122880000
4 E extended 9489 98 18 60801 80 15 824326144
5 L Linux 9489 130 50 9498 111 21 143360
X extended 9498 111 22 9514 224 54 264192
No ext2, JFS, Reiser, cramfs or XFS marker
6 L Linux 9498 143 54 9514 224 54 262144
6 L Linux 9498 143 54 9514 224 54 262144
X extended 9514 224 55 9906 161 47 6293504
No ext2, JFS, Reiser, cramfs or XFS marker
7 L Linux 9515 2 24 9906 161 47 6291456
7 L Linux 9515 2 24 9906 161 47 6291456
X extended 10518 107 22 10582 73 51 1026048
8 L Linux 10518 139 54 10582 73 51 1024000
X extended 10582 73 52 23330 159 61 204802048
9 L Linux LVM 10582 106 21 23330 159 61 204800000
X extended 23330 159 62 27247 1 18 62916608
Invalid NTFS or EXFAT boot
10 L HPFS - NTFS 23330 192 31 27247 1 18 62914560
10 L HPFS - NTFS 23330 192 31 27247 1 18 62914560
X extended 9498 111 22 9514 224 54 264192
Logical partition must be in its own extended partition
X extended 9498 111 22 9514 224 54 264192
X extended 9498 111 22 9514 224 54 264192
|
Code: | TestDisk 7.0, Data Recovery Utility, April 2015
Christophe GRENIER <grenier@cgsecurity.org>
http://www.cgsecurity.org
Disk /dev/sda - 500 GB / 465 GiB - CHS 60801 255 63
Analyse cylinder 60800/60800: 100%
HPFS - NTFS 0 32 33 1827 181 18 29360128
HPFS - NTFS 1827 181 19 1840 117 5 204800
HPFS - NTFS 1840 117 6 9489 98 2 122879985
Linux 9489 130 50 9498 111 21 143360
Linux Swap 9498 143 54 10518 107 5 16383984
Linux 10518 139 54 10582 73 51 1024000
Linux LVM 10582 106 21 23330 159 61 204800000
Invalid NTFS or EXFAT boot
0 D HPFS - NTFS 23330 192 31 27247 1 18 62914560
HPFS - NTFS 23330 192 31 27247 1 18 62914560
|
After scan:
Code: | TestDisk 7.0, Data Recovery Utility, April 2015
Christophe GRENIER <grenier@cgsecurity.org>
http://www.cgsecurity.org
Disk /dev/sda - 500 GB / 465 GiB - CHS 60801 255 63
>* HPFS - NTFS 0 32 33 1827 181 18 29360128
P HPFS - NTFS 1827 181 19 1840 117 5 204800
P HPFS - NTFS 1840 117 6 9489 98 17 122880000
L Linux 9489 130 50 9498 111 21 143360
L Linux Swap 9498 143 54 10518 107 21 16384000
L Linux 10518 139 54 10582 73 51 1024000
L Linux LVM 10582 106 21 23330 159 61 204800000
L HPFS - NTFS 23330 192 31 27247 1 18 62914560
L HPFS - NTFS 56976 152 9 60801 47 46 61442048
Structure: Ok. Use Up/Down Arrow keys to select partition.
...
NTFS, blocksize=4096, 15GB / 14GiB
|
BTW, when I am on the command line, I can rarely type more than a dozen of characters at once and I have to wait few minutes before I can continue. So, that is a bit tedious. I also checked and I have a backup for my filesystems on the LVM, not current, but not too old. So, I can in fact delete all Linux partitions if needed, including the last NTFS partition which has not been formatted at this point anyway.
Anyway, I can bring on-line the VG and access all the LV within it now. So, no problem here to recover latest data.
The system on the rescueCD is now responsive. I don't know why. But it is definitely usable. _________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Sat Jul 25, 2015 7:47 pm Post subject: |
|
|
AchilleTalon,
The simple way forward is to wipe everything, make a new partition table and reinstall from backups.
If you feel lucky you can delete all of your logical partitions (5 and on) then recreate them with the exact same starting and ending values as they have now.
That way, your data will be preserved in place. Because of the on disk data layout, after the first logical partition, (sda5) you have to get this right first time.
If you feel really adventureous, you can fix the loop in the linked list with a hex editor. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Sat Jul 25, 2015 8:39 pm Post subject: |
|
|
I feel a bit adventurous. I mean, I am curious how I could fix the linked-list. Do you have more information, a URL to a wiki, website or something to help me understand and get started? I may not walk down this road, I will see. _________________ Achille Talon Hop! |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Sun Jul 26, 2015 4:00 am Post subject: |
|
|
BTW, after doing the testdisk scan, suffice to write the partition table to restore everything in a working state. I still have to recreate my second Linux /boot and shrink my swap space to do so, however, the partition table is now in a sane state. I will avoid using fdisk next time and rely on parted instead. _________________ Achille Talon Hop! |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54300 Location: 56N 3W
|
Posted: Sun Jul 26, 2015 9:19 am Post subject: |
|
|
AchilleTalon,
If you are happy there are no false positives, you can let testdisk rewrite the partiton table for you.
If it gets partitions 1..4 wrong, there is no harm done.
If it gets partition 5 wrong there will be no damage to the data in partition 5 until it writes the entry for partition 6, whick will be written at the end of the incorrect partition 5 entry.
This may be anywhere on disk.
Be sure that the partition table testdisk wants to write is a truncated version of your posts here. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
AchilleTalon Guru
Joined: 11 Apr 2004 Posts: 368 Location: Montreal, Quebec, Canada
|
Posted: Sun Jul 26, 2015 1:54 pm Post subject: |
|
|
NeddySeagoon,
since nothing has been written to the disk after the partitions were created and testdisk discovered the partitions before the change, writing this partition table fixed the problem.
I am still curious about using an hex editor to edit the partition table. Maybe another time I will explore this with a disk I have no fear to lost anything on it.
Thanks for you help, very much appreciated. _________________ Achille Talon Hop! |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|