If usb-hdd image larger than 2000MB aufs gets trouble (for me)
Um, your problem is here...
# parted mkpartfs primary fat16 0.0 100%
FAT16 is limited to 2 gibibytes ( ;-) )
On Nov 13, 2007 12:01 AM, Peter Skogstr?m <peter.skogstrom at bitrunner.com> wrote:
> Hi
>
> I have done some tests with doing a customized partitiontable in
> lh_binary_usb-hdd, but I got trouble when aufs starts if I have an image
> over 2000MB.
>
> The usb-hdd image boots nice, but when reaching
>
> Begin: Mounting root file system... ...
> aufs 20070924
> loop: module loaded
> squashfs: version 3.2-r2 (2007/01/15) Phillip Lougher
> attempt to access beyond end of device
> hda1: rw=0, want=198357, limit=197265
> attempt to access beyond end of device
> hda1: rw=0, want=198357, limit=197265
> SQUASHFS error: sb_bread failing reading block 0x1545d
> SQUASHFS error: unable to read uid/gid table
>
> This for an 4000MB image.
>
> First I have an usb-hdd image that I know boots all good. That one is
> 101M and and have exactly
> a 196607 sectors FAT16 partition size
> (the working image is built from the lh_binary_usb-hdd parted script
> that says:
> parted mkpartfs primary fat16 0.0 100%
>
> Then with the same config I do a lh_binary
> and build my custom stick/image
>
> But for this custom image I use parted myself and the same size 101M .
> For this its:
> parted mkpartfs primary fat16 0 101
>
> (the rest of the parted stuff is the same)
>
> Except for the size of the image. If the image is up to 2000MB it is
> working but above I get this error. For example with
> an 2000MB image the fat16 partition get the size 197265 sectors and it
> works, but with 3000MB and 4000MB I have the same sector size its still
> 197265 sectors.
>
> But now the above mentioned error starts and the "access beyond end of
> device" offsets is getting larger if the image is larger.
> For 4000MB its 198357 sectors 1092 sectors diff (the case shown above)
> For 3000MB its 198233 sectors. 968 sectors diff
>
> Huh ?
>
> I think that an fat16 partition maximum is 2000MB but in this case its
> just the image that is bigger not the fat16 partition at all. Is there
> some hardcoding with aufs that I have missed ?
>
> Anyone having a clue.
>
> regards
> //Peter
>
>
>
>
> --
>
> bitrunner
>
> 0708-460260
> Andra L?nggatan 28
> 41327 G?teborg
> Sweden
>
> _______________________________________________
> debian-live-devel mailing list
> debian-live-devel at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/debian-live-devel
>
--
Best Regards.
Please keep in touch.
Reply to: