Re: bug during print

[Jim Meyering]

Grgur Tokic wrote:
> I've encoutered this message when trying to print the information about a disk
> with gpt label, previously formated using the same instance of parted.
>
>
> You found a bug in GNU Parted.  Please email a bug report to address@hidden
> containing the version (1.6.19), and the following message:
>
> Assertion ((PedSector) PED_LE64_TO_CPU (gpt.AlternateLBA) <= 
> disk->dev->length -
> 1) at disk_gpt.c:621 in function gpt_read()
> failed.
> Ignore/Cancel?
>
>
> On a different note, we're having different versions of the similar problem 
> for
> months--we cannot format 3TB disk, not even using parted. On a clean 
> partition,
> we create gpt label in parted and form one primary partition occupying the
> entire disk, also using parted. Then we are trying to create the ext3 file
> system using mkfs.ext3 /dev/sdd1 (where /dev/sdd1 is previously created 
> primary
> partition), but it never seems to work. Are we doing something wrong? Any help
> would be greatly appreciated!

Thanks for the report.

However, your version of parted is very old.
The GPT infrastructure in parted-1.x was not capable
of dealing with a logical sector size other than 512,
so a failed assertion is to be expected, since your disk
probably has 4KiB logical sectors.

To do anything useful with a GPT partition table, you really
should use only the latest version (2.3) of Parted.

I've just tried that, and it seems to work fine.
Here's what I did, using parted-2.3 and Fedora 14:
[this command creates an unaligned "slop" partition from sector 34
 to just before the 1MiB mark, then the rest goes to a "big"
 partition of length almost 3TB. ]

    $ parted -s -- /dev/sdd mkpart slop 34s 255s
    Warning: The resulting partition is not properly aligned for best 
performance.
    $ parted -s -- /dev/sdd mkpart big 256s -34s u MiB p free
    Model: Seagate FA GoFlex Desk (scsi)
    Disk /dev/sdd: 2861588MiB
    Sector size (logical/physical): 4096B/4096B
    Partition Table: gpt

    Number  Start       End         Size        File system  Name  Flags
            0.02MiB     0.13MiB     0.11MiB     Free Space
     1      0.13MiB     1.00MiB     0.87MiB                  slop
     2      1.00MiB     2861588MiB  2861587MiB               big
            2861588MiB  2861588MiB  0.11MiB     Free Space

    $ env time mkfs.xfs /dev/sdd2
    meta-data=/dev/sdd2              isize=256    agcount=32, agsize=22892699 
blks
             =                       sectsz=4096  attr=2
    data     =                       bsize=4096   blocks=732566356, imaxpct=5
             =                       sunit=0      swidth=0 blks
    naming   =version 2              bsize=4096   ascii-ci=0
    log      =internal log           bsize=4096   blocks=357698, version=2
             =                       sectsz=4096  sunit=1 blks, lazy-count=1
    realtime =none                   extsz=4096   blocks=0, rtextents=0
    0.00user 0.21system 0:51.58elapsed 0%CPU (0avgtext+0avgdata 
11776maxresident)k
    280inputs+2863944outputs (0major+914minor)pagefaults 0swaps


Notice that I chose XFS as the file system type, not ext3 (and certainly
not ext2).  If you want to avoid the risk of a very long-running fsck,
you will want to avoid ext2 and ext3.  Ext4 may be better, but mkfs.ext3
on a file system that large would have taken far longer.