Lost Partition

Lost Partition

[Vincent Sweeney]

I have just upgrade my kernel to 2.4.12 and in the process I've lost a
partition on my secondary IDE drive. Since this is my /usr partition
it's kind of important ;)

Here is the relevant section from my dmesg output and fdisk (note
/dev/hdb5 is the one missing). The output from fdisk in recovery mode
can see the partition and the ID looks correct but the kernel seems to
have an 'issue'.

If you need any more details just ask.

Vince.

---
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
VP_IDE: IDE controller on PCI bus 00 dev 39
VP_IDE: chipset revision 6
VP_IDE: not 100% native mode: will probe irqs later
ide: Assuming 33MHz system bus speed for PIO modes; override with
idebus=xx
VP_IDE: VIA vt82c686b (rev 40) IDE UDMA100 controller on pci00:07.1
    ide0: BM-DMA at 0xd400-0xd407, BIOS settings: hda:DMA, hdb:DMA
    ide1: BM-DMA at 0xd408-0xd40f, BIOS settings: hdc:DMA, hdd:pio
hda: SAMSUNG SV1021H, ATA DISK drive
hdb: SAMSUNG SV1021H, ATA DISK drive
hdc: LG CD-ROM CRD-8521B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 19932192 sectors (10205 MB) w/426KiB Cache, CHS=1240/255/63,
UDMA(33)
hdb: 19932192 sectors (10205 MB) w/426KiB Cache, CHS=1240/255/63,
UDMA(33)
Partition check:
 hda: hda1 hda2 < hda5 hda6 hda7 hda8 >
 hdb: hdb1 hdb2 <  >

---

Disk /dev/hda: 255 heads, 63 sectors, 1240 cylinders
Units = cylinders of 16065 * 512 bytes

   Device Boot    Start       End    Blocks   Id  System
/dev/hda1   *         1         9     72261   83  Linux
/dev/hda2            10      1240   9888007+   5  Extended
/dev/hda5            10       140   1052226   83  Linux
/dev/hda6           141       271   1052226   83  Linux
/dev/hda7           272       337    530113+  83  Linux
/dev/hda8           338      1240   7253316   83  Linux

Disk /dev/hdb: 255 heads, 63 sectors, 1240 cylinders
Units = cylinders of 16065 * 512 bytes

   Device Boot    Start       End    Blocks   Id  System
/dev/hdb1   *         1        66    530113+  82  Linux swap
/dev/hdb2            67      1240   9430155    5  Extended
/dev/hdb5            67      1240   9430123+  83  Linux

Re: Lost Partition

[Alexander Viro]

On Thu, 11 Oct 2001, Vincent Sweeney wrote:

> Disk /dev/hdb: 255 heads, 63 sectors, 1240 cylinders
> Units = cylinders of 16065 * 512 bytes
> 
>    Device Boot    Start       End    Blocks   Id  System
> /dev/hdb1   *         1        66    530113+  82  Linux swap
> /dev/hdb2            67      1240   9430155    5  Extended
> /dev/hdb5            67      1240   9430123+  83  Linux

How about the contents of sectors 0 and 1076355?

[PATCH] Re: Lost Partition

[Alexander Viro]

On Thu, 11 Oct 2001, Vincent Sweeney wrote:

> I have just upgrade my kernel to 2.4.12 and in the process I've lost a
> partition on my secondary IDE drive. Since this is my /usr partition
> it's kind of important ;)

Please, see if the following fixes the problem.

--- S11/fs/partitions/msdos.c	Tue Oct  9 21:47:27 2001
+++ /tmp/msdos.c	Thu Oct 11 09:18:22 2001
@@ -103,21 +103,20 @@
  */
 
 static void extended_partition(struct gendisk *hd, struct block_device *bdev,
-				int minor, int *current_minor)
+			int minor, unsigned long first_size, int *current_minor)
 {
 	struct partition *p;
 	Sector sect;
 	unsigned char *data;
-	unsigned long first_sector, first_size, this_sector, this_size;
+	unsigned long first_sector, this_sector, this_size;
 	int mask = (1 << hd->minor_shift) - 1;
 	int sector_size = get_hardsect_size(to_kdev_t(bdev->bd_dev)) / 512;
 	int loopct = 0;		/* number of links followed
 				   without finding a data partition */
 	int i;
 
-	first_sector = hd->part[minor].start_sect;
-	first_size = hd->part[minor].nr_sects;
-	this_sector = first_sector;
+	this_sector = first_sector = hd->part[minor].start_sect;
+	this_size = first_size;
 
 	while (1) {
 		if (++loopct > 100)
@@ -133,8 +132,6 @@
 
 		p = (struct partition *) (data + 0x1be);
 
-		this_size = hd->part[minor].nr_sects;
-
 		/*
 		 * Usually, the first entry is the real data partition,
 		 * the 2nd entry is the next extended partition, or empty,
@@ -196,6 +193,7 @@
 			goto done;	 /* nothing left to do */
 
 		this_sector = first_sector + START_SECT(p) * sector_size;
+		this_size = NR_SECTS(p) * sector_size;
 		minor = *current_minor;
 		put_dev_sector(sect);
 	}
@@ -586,12 +584,13 @@
 		}
 #endif
 		if (is_extended_partition(p)) {
+			unsigned long size = hd->part[minor].nr_sects;
 			printk(" <");
 			/* prevent someone doing mkfs or mkswap on an
 			   extended partition, but leave room for LILO */
-			if (hd->part[minor].nr_sects > 2)
+			if (size > 2)
 				hd->part[minor].nr_sects = 2;
-			extended_partition(hd, bdev, minor, &current_minor);
+			extended_partition(hd, bdev, minor, size, &current_minor);
 			printk(" >");
 		}
 	}

Re: [PATCH] Re: Lost Partition

[Stefan Smietanowski]

Hi. Re partition problem.


> -	unsigned long first_sector, first_size, this_sector, this_size;
> +	unsigned long first_sector, this_sector, this_size;

> +	this_size = first_size;


It seems that's sorta wrong, no?

You just removed "first_size" and then you access it :)

// Stefan

Re: Re: [PATCH] Re: Lost Partition

[Alexander Viro]

On Thu, 11 Oct 2001, [iso-8859-1] Stefan Smietanowski wrote:

> Hi. Re partition problem.
> 
> 
> > -	unsigned long first_sector, first_size, this_sector, this_size;
> > +	unsigned long first_sector, this_sector, this_size;
> 
> > +	this_size = first_size;
> 
> 
> It seems that's sorta wrong, no?
> 
> You just removed "first_size" and then you access it :)

Look carefully at the arguments list.  first_size had just become explicitly
passed to extended_partition().

Re: Re: [PATCH] Re: Lost Partition

[Stefan Smietanowski]

Hi.

>> Hi. Re partition problem.
>>
>>
>> > -	unsigned long first_sector, first_size, this_sector, this_size;
>> > +	unsigned long first_sector, this_sector, this_size;
>>
>> > +	this_size = first_size;
>>
>>
>> It seems that's sorta wrong, no?
>>
>> You just removed "first_size" and then you access it :)
>
> Look carefully at the arguments list.  first_size had just become
> explicitly passed to extended_partition().

Yeah I see that now. Sorry, didn't look closely enough.

// Stefan

Re: [PATCH] Re: Lost Partition

[Vincent Sweeney]

No luck I'm afraid. The patch applied successfully bit I still get
exactly the same problem.

Vince.

Alexander Viro wrote:
> 
> On Thu, 11 Oct 2001, Vincent Sweeney wrote:
> 
> > I have just upgrade my kernel to 2.4.12 and in the process I've lost a
> > partition on my secondary IDE drive. Since this is my /usr partition
> > it's kind of important ;)
> 
> Please, see if the following fixes the problem.
> 
> --- S11/fs/partitions/msdos.c   Tue Oct  9 21:47:27 2001
> +++ /tmp/msdos.c        Thu Oct 11 09:18:22 2001
> @@ -103,21 +103,20 @@
>   */
> 
>  static void extended_partition(struct gendisk *hd, struct block_device *bdev,
> -                               int minor, int *current_minor)
> +                       int minor, unsigned long first_size, int *current_minor)
>  {
>         struct partition *p;
>         Sector sect;
>         unsigned char *data;
> -       unsigned long first_sector, first_size, this_sector, this_size;
> +       unsigned long first_sector, this_sector, this_size;
>         int mask = (1 << hd->minor_shift) - 1;
>         int sector_size = get_hardsect_size(to_kdev_t(bdev->bd_dev)) / 512;
>         int loopct = 0;         /* number of links followed
>                                    without finding a data partition */
>         int i;
> 
> -       first_sector = hd->part[minor].start_sect;
> -       first_size = hd->part[minor].nr_sects;
> -       this_sector = first_sector;
> +       this_sector = first_sector = hd->part[minor].start_sect;
> +       this_size = first_size;
> 
>         while (1) {
>                 if (++loopct > 100)
> @@ -133,8 +132,6 @@
> 
>                 p = (struct partition *) (data + 0x1be);
> 
> -               this_size = hd->part[minor].nr_sects;
> -
>                 /*
>                  * Usually, the first entry is the real data partition,
>                  * the 2nd entry is the next extended partition, or empty,
> @@ -196,6 +193,7 @@
>                         goto done;       /* nothing left to do */
> 
>                 this_sector = first_sector + START_SECT(p) * sector_size;
> +               this_size = NR_SECTS(p) * sector_size;
>                 minor = *current_minor;
>                 put_dev_sector(sect);
>         }
> @@ -586,12 +584,13 @@
>                 }
>  #endif
>                 if (is_extended_partition(p)) {
> +                       unsigned long size = hd->part[minor].nr_sects;
>                         printk(" <");
>                         /* prevent someone doing mkfs or mkswap on an
>                            extended partition, but leave room for LILO */
> -                       if (hd->part[minor].nr_sects > 2)
> +                       if (size > 2)
>                                 hd->part[minor].nr_sects = 2;
> -                       extended_partition(hd, bdev, minor, &current_minor);
> +                       extended_partition(hd, bdev, minor, size, &current_minor);
>                         printk(" >");
>                 }
>         }

Re: [PATCH] Re: Lost Partition

[Alexander Viro]

On Thu, 11 Oct 2001, Vincent Sweeney wrote:

> No luck I'm afraid. The patch applied successfully bit I still get
> exactly the same problem.

Please, do the following:

dd if=/dev/hdb of=p0 bs=512 count=1
dd if=/dev/hdb of=p1 bs=512 count=1 skip=1076355

and mail the contents of p0 and p1.

Re: [PATCH] Re: Lost Partition

[Vincent Sweeney]

[-- Attachment #1: Type: text/plain, Size: 365 bytes --]

Alexander Viro wrote:
> 
> On Thu, 11 Oct 2001, Vincent Sweeney wrote:
> 
> > No luck I'm afraid. The patch applied successfully bit I still get
> > exactly the same problem.
> 
> Please, do the following:
> 
> dd if=/dev/hdb of=p0 bs=512 count=1
> dd if=/dev/hdb of=p1 bs=512 count=1 skip=1076355
> 
> and mail the contents of p0 and p1.

Files p0 and p1 attached.

[-- Attachment #2: p0 --]
[-- Type: application/octet-stream, Size: 512 bytes --]

[-- Attachment #3: p1 --]
[-- Type: application/octet-stream, Size: 512 bytes --]

nú Edición), y las imágenes son reemplazadas por iconos. Sirve para quitar los iconos de sustitución y ver las imágenes reales de la página.

<LI><B>Imprimir</B>
<BR>Sirve para imprimir el contenido de la p&aacute;gina que est&aacute; en pantalla. Se abre un cuadro de di&aacute;logo en el que pueden elegirse las caracter&iacute;sticas de la impresi&oacute;n.

<LI><B>Seguridad</B>
<BR>Sirve para ver la p&aacute;gina <B>Informaci&oacute;n sobre seguridad</B>, en la qu

Re: [PATCH] Re: Lost Partition

[Jeff V. Merkey]

Al,

Please check the NWFS code posted at vger.timpanogas.org, specifically,
nwconfig.c (top area of file).  One thing NetWare did that I thought
was rather clever was to stamp a copy of the boot sector into the 
second sector of every new partition in the event someone ever 
wrote to the boot sector and trashed someone's partition table.  I
do not know if Linux has something like this, but it was sure a lifesaver
when errant NLMs stepped all over the hard drive and trashed the 
boot sector, causing all the partition info to get wiped.  Each 
partition header was time stamped so someone could just scan the 
first two sectors of every partition (after locating the fist one
via brute force) and determine which copy was most recent.  

I realize this may not have been the problem here, but it would sure
make it easier for EXT3, REISER, etc. to recover trashed partition
tables a lot easier than custom patches if Linux could pick a sector
on one of these volumes, and make it a bogus file or something by
default and stamp a copy of the boot sector into every partition
that gets created along with a time stamp of some sort so you know 
which one is most recent.  Then fsck would be able to recover hosed
partition tables very quickly.

:-)

Jeff

On Thu, Oct 11, 2001 at 09:19:58AM -0400, Alexander Viro wrote:
> 
> 
> On Thu, 11 Oct 2001, Vincent Sweeney wrote:
> 
> > I have just upgrade my kernel to 2.4.12 and in the process I've lost a
> > partition on my secondary IDE drive. Since this is my /usr partition
> > it's kind of important ;)
> 
> Please, see if the following fixes the problem.
> 
> --- S11/fs/partitions/msdos.c	Tue Oct  9 21:47:27 2001
> +++ /tmp/msdos.c	Thu Oct 11 09:18:22 2001
> @@ -103,21 +103,20 @@
>   */
>  
>  static void extended_partition(struct gendisk *hd, struct block_device *bdev,
> -				int minor, int *current_minor)
> +			int minor, unsigned long first_size, int *current_minor)
>  {
>  	struct partition *p;
>  	Sector sect;
>  	unsigned char *data;
> -	unsigned long first_sector, first_size, this_sector, this_size;
> +	unsigned long first_sector, this_sector, this_size;
>  	int mask = (1 << hd->minor_shift) - 1;
>  	int sector_size = get_hardsect_size(to_kdev_t(bdev->bd_dev)) / 512;
>  	int loopct = 0;		/* number of links followed
>  				   without finding a data partition */
>  	int i;
>  
> -	first_sector = hd->part[minor].start_sect;
> -	first_size = hd->part[minor].nr_sects;
> -	this_sector = first_sector;
> +	this_sector = first_sector = hd->part[minor].start_sect;
> +	this_size = first_size;
>  
>  	while (1) {
>  		if (++loopct > 100)
> @@ -133,8 +132,6 @@
>  
>  		p = (struct partition *) (data + 0x1be);
>  
> -		this_size = hd->part[minor].nr_sects;
> -
>  		/*
>  		 * Usually, the first entry is the real data partition,
>  		 * the 2nd entry is the next extended partition, or empty,
> @@ -196,6 +193,7 @@
>  			goto done;	 /* nothing left to do */
>  
>  		this_sector = first_sector + START_SECT(p) * sector_size;
> +		this_size = NR_SECTS(p) * sector_size;
>  		minor = *current_minor;
>  		put_dev_sector(sect);
>  	}
> @@ -586,12 +584,13 @@
>  		}
>  #endif
>  		if (is_extended_partition(p)) {
> +			unsigned long size = hd->part[minor].nr_sects;
>  			printk(" <");
>  			/* prevent someone doing mkfs or mkswap on an
>  			   extended partition, but leave room for LILO */
> -			if (hd->part[minor].nr_sects > 2)
> +			if (size > 2)
>  				hd->part[minor].nr_sects = 2;
> -			extended_partition(hd, bdev, minor, &current_minor);
> +			extended_partition(hd, bdev, minor, size, &current_minor);
>  			printk(" >");
>  		}
>  	}
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/